Evaluation of HEVC decoders (SW, Hybrid and HW)

[NikosD]

Extremely low CPU usage.

Did you see the GPU usage using GPU-Z ?

Looks like a native, pure HW acceleration of VP9 which is impressive, if it's true.

[nevcairiel]

Neat they fixed it, and my implementation apparently works out of the box, always nice to know :) time to benchmark myself later

[wanezhiling]

NVIDIA is always the best.

[NikosD]

Nvidia is always second the last few years.

Second to Intel.

Second in speed, quality (transcoding) and adoption of new codecs.

[NikosD]

You ignored two basic categories of facts (speed, transcoding quality) which Intel is the undisputed king since 2011 till now and you made a post about this year's and future adoption by Nvidia of new codecs, followed by Intel after a few months.

Facts mean history and sustainability, things that have already been done and last for good, not future assumptions that last for a few months.

[huhn]

are the fan boys done?

Neat they fixed it, and my implementation apparently works out of the box, always nice to know :) time to benchmark myself later


DXVA checker isn't seeing VP9 from lavfilter. i would have made a small test if this would be the case.

but your implementation looks flawless so far.

[nevcairiel]

I have no idea how DXVA Checker collects this information of which formats are supported by a given codec, the author often had to fix that part in the past.
Speaking of DXVA Checker, I wish he would make it possible to test secondary GPUs. Right now its always tied to the primary GPU, which is my Titan X, not my 960. Unless I'm blind and never found a GPU selector. :)

My desktop icons always get jumbled up when I switch the primary display for benchmarking the 960. :D

To augment some earlier numbers, GPU-Z actually shows Video Engine Load, and barely any GPU load (and very little CPU load as well), so it appears to be a proper HW decoder indeed!

[NikosD]

Not a bad idea.

I asked the developer if it's feasible.

[huhn]

I have no idea how DXVA Checker collects this information of which formats are supported by a given codec, the author often had to fix that part in the past.
Speaking of DXVA Checker, I wish he would make it possible to test secondary GPUs. Right now its always tied to the primary GPU, which is my Titan X, not my 960. Unless I'm blind and never found a GPU selector. :)

My desktop icons always get jumbled up when I switch the primary display for benchmarking the 960. :D

would only work for decoding not playback benchmark.

but yeah would be nice.

[nevcairiel]

would only work for decoding not playback benchmark.


Why not, it could just open the window on the second screen instead of the primary one.

[huhn]

yes if it is connected.

even through UHD at native resolution could still make problems on a 1080p screen.

so generally not working is wrong but iGPU are offend not connected.

[nevcairiel]

so generally not working is wrong but iGPU are offend not connected.

DXVA2 requires a connected display, since its based on D3D9, which doesn't work on fully headless adapters.

[NikosD]

But D3D11 supported on Win 8 and above doesn't have this restriction.

I don't know how difficult could be for LAV Video to support both D3D9/D3D11

[nevcairiel]

Its something I might do at some point, but no priority right now. Its certainly a lot of effort. Not decided yet if worth it.

[P.J]

would only work for decoding not playback benchmark.

but yeah would be nice.

How? =T

[huhn]

the microsoft decoder should support it. you can read the rest.

[NikosD]

I' ve just tried the new DXVA Checker version - not publicly available yet - with a GPU selector.

It works OK on my SandyBridge system with Radeon 6750 as primary and iGPU as secondary under Win 10 Nov update.

Although I have both GPUs connected on my monitor at the same time, playback mode doesn't work for secondary iGPU with any decoder.

Using LAV Video and MS DS H.264 decoder, it falls back to software in DXVA native decode mode.
But DXVA copy-back works OK for LAV.

Unfortunately, QuickSync decoder is not available for SandyBridge on Win 10.

MS MFT H.264 decoder works fine in DXVA native decode mode, probably because it's a D3D9/D3D11 decoder.

[XMonarchY]

So Skylake is better at decoding HEVC than my GTX 980?

[NikosD]

It seems that I managed to got some more interesting results.

First of all, with my two cards both connected to the display, if I select "Extend these displays" I can select any GPU and work it like primary, main GPU.

Everything works fine, including connection to EVR and Playback mode.
The results in benchmarking are the same like only one display is connected.

Now, if I choose as primary the Radeon with "Show Desktop only to 1" and select the iGPU from DXVA Checker, then the driver of SandyBridge doesn't provide D3D11 components in Processor Device List and the Decoder Device list is empty.

Also, DXVA copy-back of LAV is not actually using the iGPU but Radeon even with the iGPU selected.

The DXVA native performance of MS MFT H.264 is about 50% in D3D11 mode (as secondary iGPU) compared to directly connected iGPU card as primary.

Using the opposite order, iGPU as primary and Radeon 6750 as secondary, the LAV DXVA copy-back still uses primary GPU even if the secondary is selected by DXVA Checker.

This time the Device Decoder list and the Processor Device List are not empty for secondary Radeon, because the driver provides D3D11 components.

The DXVA native performance of MS MFT H.264 is the same in D3D11 mode (as secondary Radeon) compared to directly connected Radeon card as primary, but the CPU usage went from 1% to 25%.

[nevcairiel]

Also, DXVA copy-back of LAV is not actually using the iGPU but Radeon even with the iGPU selected.

Copy-Back always uses your primary GPU, only Native switches based on which screen is active.
Thats because Copy-Back is internal to LAV Video, any external component doesn't actually know that its using hwaccel (unless it specifically asks LAV what its doing).

[huhn]

So Skylake is better at decoding HEVC than my GTX 980?

for high quality HEVC you kind of need an fixed function decoder.

all future video cards like pascal, Arctic Islands and intel cannonlake should have a more than powerful enough fixed function decoder for HEVC and VP9.

current hybrid decoder are just nice to have so yeah the 980 is pretty bad in HEVC decoding.

[edigee]

Sorry for a little offtopic!
A friend of mine asked me to built for him a HTPC. He's interested in HEVC decoding ,he has a lot of stuff from internet. So ,here's the options: a Skylake i5 platform or a GTX 950/960 plus older harware(CPU ,mobo). He's not interested in gaming at all. Don't know if Skylake has HDMI 2.0 support ,though. Skylake plus discrete video card is above the budget.

[Yups]

The new LAV nightly build 0.67.23 from here http://files.1f0.de/lavf/nightly/ has added VP9 DXVA2 acceleration.

It would be interesting to test it on your Skylake HD 530 and make a comparison with VP9 CPU Skylake decoding using this clip:
https://www.sendspace.com/file/pr4m1c

LAV 0.67.0.31 DXVA2
i7-6700k, HD 530 20.19.15.4352
Windows 10

Decode
94-112-183 35% CPU


Playback
71-79-108 21% CPU


LAV hardware decoder disabled:

Decode
121-183-330 62% CPU

Playback
75-85-90 27% CPU


For some reason the playback benchmark without LAV hardware decoder is using my HD 530 by about 80% and it runs a bit faster.


edit: Same for HEVC 10 bit videos. Playback runs quite a bit faster. Any idea why this is?

[NikosD]

Can you disable NV12 from LAV properties and try again ?

[Yups]

LAV hardware decoder disabled:

Decode
121-183-330 62% CPU

Playback
75-85-90 27% CPU



Playback
88-102-107 (skin tone improvement disabled, it was enabled in the driver)

It says "unsupported" when I disable hardware decoder in LAV by the way.

http://fs5.directupload.net/images/151228/evyi5mhh.png

[Yups]

Can you disable NV12 from LAV properties and try again ?

Playback
47-62-79 62%

Even though HD530 is used according to GPUz, it runs much worse.

[NikosD]

Doesn't it always say "unsupported" when CPU Decoding is used, if I remember correctly ?

What are the figures with DXVA VP9 enabled, now that you disabled the VPP setting ?

Normally Benchmark Decode mode should have the same performance, but I would test that too using DXVA and CPU decode mode.

[Yups]

What are the figures with DXVA VP9 enabled, now that you disabled the VPP setting ?



Actually there were more video improvements enabled under advanced. I did a Windows 10 reinstall recently, I forgot all this.


VP9 hardware decoding Playback
70-79-108 21% CPU


No Hardware decoding playback
116-158-219 58% CPU

GPU load decreased from ~80% to 10-15% without hardware decode.

[NikosD]

Sorry, but I'm little confused by your figures.

Can you repeat the numbers like your first post ?

With all VPP off

DXVA VP9
Decode mode (CPU/GPU usage)
Playback mode 720p (CPU/GPU usage)

Software Decoding
Decode mode (CPU/GPU usage)
Playback mode 720p (CPU/GPU usage)

[Yups]

VPP all off with Big Buck Bunny-70s


LAV 0.67.0.31 x64 DXVA2
i7-6700k @4,0 Ghz 1.15V
HD 530 1.15Ghz 20.19.15.4352
Windows 10 x64

DXVA2
Decode
95-113-180 /37% CPU/100% GPU

Playback
71-79-109 /22% CPU/100% GPU


LAV hardware decoder disabled:
Decode
130-184-332 /62% CPU/0% GPU

Playback
118-159-214 /58% CPU/10% GPU


Obviously CPU only is much faster in my case with the fastest desktop Skylake. On a mobile ULV with half the cores and clock frequency it should be a turnaround. I did the same test with 2 cores+HT, 2.0 Ghz and HD530 lowered to 1.0 Ghz which is closer to a 15W mobile SKU. Results in my next post.

[Yups]

LAV 0.67.0.31 x64 DXVA2
i7-6700k @2.0 Ghz 2 cores+HT 1.15V
HD 530 1.0 Ghz 20.19.15.4352
Windows 10 x64

DXVA2
Decode
79-97-146 /74% CPU/85% GPU

Playback
61-71-100 /51% CPU/100% GPU


LAV hardware decoder disabled:
Decode
37-65-235 /77% CPU/0% GPU

Playback
37-62-151 /75% CPU/5% GPU


Hardware decoding is quite a bit faster expecially in the most complex scenes. CPU playback falls below 60 fps for several times even if it is for a second only.

[NikosD]

Perfect posts, thank you!

Now, I'm waiting some benches by Nvidias like 950/960 or other cards with DXVA VP9 support on the same clip.

No further CPU tests are needed when Skylake's results have already posted.

[Aleksoid1978]

Nvidia GTX 960.
decode:

Decoder: LAV Video Decoder
[1]
Decoder Device: VP9_VLD_Profile0
Frames: 4256
Time: 26,727 sec
FPS: 159,242 [124-179] fps
CPU Usage: 3 [1-5] %
GPU Usage: -
[2]
Decoder Device: VP9_VLD_Profile0
Frames: 4256
Time: 25,787 sec
FPS: 165,041 [150-188] fps
CPU Usage: 3 [1-6] %
GPU Usage: -
[3]
Decoder Device: VP9_VLD_Profile0
Frames: 4256
Time: 25,718 sec
FPS: 165,485 [149-187] fps
CPU Usage: 3 [0-5] %
GPU Usage: -


playback 720x:

Decoder: LAV Video Decoder
[1]
Decoder Device: VP9_VLD_Profile0
Processor Device: 6CB69578-7617-4637-91E5-1C02DB810285
Frames: 4256
Time: 26,715 sec
FPS: 159,308 [125-180] fps
CPU Usage: 2 [0-5] %
GPU Usage: -
[2]
Decoder Device: VP9_VLD_Profile0
Processor Device: 6CB69578-7617-4637-91E5-1C02DB810285
Frames: 4256
Time: 25,851 sec
FPS: 164,635 [147-189] fps
CPU Usage: 4 [2-7] %
GPU Usage: -
[3]
Decoder Device: VP9_VLD_Profile0
Processor Device: 6CB69578-7617-4637-91E5-1C02DB810285
Frames: 4256
Time: 25,833 sec
FPS: 164,749 [149-187] fps
CPU Usage: 3 [0-6] %
GPU Usage: -

[NikosD]

Thanks!

A GPU-Z report would help for GPU load in both Decoding and Playback mode.

Also, an easy readable form that we use here is the average of about 3 runs as (removing the results of the first run):

min/average/max fps

without all that extra info and specific results of each run.

[nevcairiel]

It's a full hardware decoder, barely any GPU usage, VPU usage instead.

I've been told the GPU uses the same hardware decoder as the Tegra devices, hence the VP9 support, it just was not exposed on PCs until now.

[NikosD]

OK, I assumed that in a previous post due to the extremely low CPU usage, but then it doesn't look as a fast fixed-function decoder.

I mean it should be a lot faster, but that mentioning to Tegra and the fact that probably we need more bitrate to "push" the HW decoder, maybe could explain the results.

I have to check VP9 encoders and come back with some real challenges for this HW decoder then.

[nevcairiel]

An average of 160 FPS 4K decoding isn't exactly slow, though. HEVC is in a similar performance category.

[NikosD]

We have to check it out with 100Mbps and more 4K VP9 to be sure that those two HW decoders are comparable (HEVC & VP9)

In real world situations, besides Chrome browsing YouTube, I don't know if there is any use of VP9 and especially with requirements like BluRay UHD for HEVC.

Perhaps VP9 decoder doesn't have to be that fast anyway, especially in high bitrate clips, because nobody will ever use those in real life.

[huhn]

and VP9 profile 0 is not really competitive with HEVC main10.

and profile1 can't be decoded with this card.

[NikosD]

OK these are my three VP9 4K clips for testing:

50Mbps
https://www.sendspace.com/file/2w3cib

100Mbps
https://www.sendspace.com/file/4l1w5l

250Mbps
https://www.sendspace.com/file/fvrvp7

[huhn]

with 250 mbit it is still running with up to 80 % video engine usage.

I have no proper test tool installed sorry.

[NikosD]

Thanks!

But DXVA Checker doesn't need installation.
I always download the without installer version (portable)

You just unzip it and run it.

[huhn]

i didn't got a notification for a new version. so they fixed the issue or was it limited to me that VP9 wasn't seeing any decoders?

i try it soon.

[NikosD]

I haven't checked that, but judging from the posts above, if you follow the "normal" way I think you will manage to benchmark it because three different systems (Braswell, Skylake and Nvidia 960) have already done it.

Note, that it's a different procedure to see what decoders are available for the VP9 format, than actually go to test a VP9 clip via the the well known method.

[huhn]

using a file to check doesn't work for me.

still the same issue for me. i don't have the will to dig deeper than reinstalling lavfilter and DXVA checker.

maybe an issue with the new insider build maybe not.

[Aleksoid1978]

Test GTX 960 on crowd_run_2160p50_VP9_250Mbps.mkv

Decode:
68-77-80 /3% CPU/94% VPU

[NikosD]

In Decode mode GPU should be 0% for fixed-function HW decoder.

Are you sure it's GPU and not VPU ?

[Aleksoid1978]

Yes - VPU.

[NikosD]

Ok...Then it should be 100% because you are testing it in benchmark mode :)

The clip is short and probably the monitoring tool doesn't have the time to reach the 100% value, but VPU is 100% used.

[huhn]

100% VPU is very unusually even for benchmarks. and there could be a lot of reasons for this.

[NikosD]

With Intel HW I always remember having 100% use of QuickSync MFX Decode engine, using GPA monitoring tool.

Also, my Nvidia monitoring tools always show me a slow progress in reaching 100% or 99% use of VPU.

Hybrid decoding on Intel HW, like HEVC decoding using Haswell, also uses GPU 100% when decoding is done by PowerDVD 15 decoder or MS DS/MFT H.265.

LAV can't keep up to 100%, as I have already posted in LAV filters thread.

[edigee]

I'm getting only black screen on those VP9 videos. Latest LAV ,GTX 960 ,DXVA (native or copyback), MPC-HC ,madVR or EVR CP ,W 7 64bit.
On software mode they work but with high CPU load.

[huhn]

you need the nightly lavfilter version.

edit: you wouldn't get an blackscreen with the older ones so you should have it...

[edigee]

you need the nightly lavfilter version.

edit: you wouldn't get an blackscreen with the older ones so you should have it...

I have 0.67.0.30
VP9 box is checked
Someone said here earlier it might be a driver issue as long as GTX 960 is credited with VP9 decoding capabilities.
I.m using 359.00 .The latest one gives me higher temps on video playback so I rolled back.

[nevcairiel]

VP9 only works in 361.43

[edigee]

VP9 only works in 361.43

Thanks ,good to know!
I'll wait for the next one though, with 361.43 my card runs at 60 degrees in some videos where "the old" 359.00 gives much lower temps. Only 40-45 degrees. Same settings ,same version of LAV ,MPC-HC ,madVR. I found on the specific driver thread on guru 3d forum that I'm not the only one reporting that behaviour.

[wanezhiling]

http://www.fudzilla.com/images/stories/2015/November/rtg-polarisgpu-2.jpg

Polaris’ display block will support features like HDMI 2.0a and DisplayPort 1.3, and its multimedia features will include H.265 Main 10 decode at resolutions up to 4K, and 4K H.265 encoding at up to 60 FPS.

[jones1913]

Polaris’ display block will support features like HDMI 2.0a and DisplayPort 1.3, and its multimedia features will include H.265 Main 10 decode at resolutions up to 4K, and 4K H.265 encoding at up to 60 FPS.
Finally a big step forward after years of stagnation.

Here are some more slides including the "H.265 Main 10" statement:
AMD announces Polaris architecture (http://videocardz.com/58031/amd-announces-polaris-architecture-gcn-4-0)

OT:
Some cool things may come at AMD side the next quarters:
Zen based APU with HBM to be AMD Carrizo successor (http://hexus.net/tech/news/cpu/89246-zen-based-apu-hbm-amd-carrizo-successor)
AMD Further Embraces Open Source with GPUOpen (http://wccftech.com/amd-further-embraces-open-source-with-gpuopen-for-games-and-compute)
AMD/Radeon Has Continued Making Much Linux Graphics Progress (http://www.phoronix.com/scan.php?page=news_item&px=Top-AMD-RTG-2015-News)
AMD @ SC15: Boltzmann Initiative Announced - C++ and CUDA Compilers for AMD GPUs (http://www.anandtech.com/show/9792/amd-sc15-boltzmann-initiative-announced-c-and-cuda-compilers-for-amd-gpus)

[aufkrawall]

Well, this was to be expected. It'd be more interesting if there is native VP9 support or any at all. I can't watch 4k 60fps VP9 in Chrome or Firefox since they are too slow. LAV works fine with my 2500k@4.8Ghz.

[sexus]

sup , anyone able to help me out with this issue , ive got a question about HEVC cuvid support

so ive tried to play a HEVC movie yesterday and my videoplayer application crashed , once ive unchecked HEVC support in cuvid it worked , im using 350.12 nvidia drivers and lav filters 0.65, please help

[edigee]

If your card can decode HEVC (GTX950/960, Intel Skylake) you only have to check the HEVC box on the LAV Video settings and you can use DXVA (copyback or native) for playing the videos. AFAIK Cuvid is not suitable for that ,didn't use it for ages anyway.
If you're on other platform without or with limited support for HEVC decoding check the HEVC box and "none" under "Hardware Decoder To Use". The video will play in software mode with higher CPU load ,of course.

[aufkrawall]

Test GTX 960 on crowd_run_2160p50_VP9_250Mbps.mkv

Decode:
68-77-80 /3% CPU/94% VPU
Are the first values GPU load? If yes, it splits load between GPU & VPU part instead of GPU & CPU like with older Maxwell HEVC hybrid decoder?

[huhn]

MIN, AVG and MAX FPS. there is no GPU load it is a real hardware decoder.

[aufkrawall]

Ok. I could swear that at first it has been said it would be hybrid.

[nevcairiel]

We assumed so, but it turns out the GTX960 has the full decoder chip from the Tegra line, which has had VP9 support for a while - it was just never exposed on PCs yet. So yes, its full hardware acceleration.

[aufkrawall]

Thanks for clarifying!

[CruNcher]

If your card can decode HEVC (GTX950/960, Intel Skylake) you only have to check the HEVC box on the LAV Video settings and you can use DXVA (copyback or native) for playing the videos. AFAIK Cuvid is not suitable for that ,didn't use it for ages anyway.
If you're on other platform without or with limited support for HEVC decoding check the HEVC box and "none" under "Hardware Decoder To Use". The video will play in software mode with higher CPU load ,of course.

In certain Edge Cases cases CUVID can actually be usefull depending on your hardware config but most probably if that ist he case the CPU will be even more efficient and overall most probably then even consume less power alone ;)

Because a huge Difference in CUVID decoding mode is that the Frequency isn't going to change dynamically like it does in DXVA this can mean depending on the Bitstream complexity on a Hybrid Decoder that it might be more efficient if you use CUVID with a higher Frequency then DXVA with a changing Frequency, because for the Bitstream complexity the DXVA Frequency might be to low to maintain a Stable Playback result without dropouts ;)

It's a pretty funny thing on the other side the Frequency for a lower Complexity Bitstream (or if that Bitstream is not Hybrid Decoded) might be way overshoot and you get very inefficient compared to DXVA fot that same Bitstream ;)

How this looks in Practice :)


DXVA Native Changing Frequency (Hybrid Decoder)
http://i2.sendpic.org/t/wU/wUfliF0cQPUXzKg8yVjEjFgURhj.jpg (http://sendpic.org/view/2/i/zfWRonzJtl5Sz0yD5Io5BE9PaEg.png)


CUVID Fixed Frequency (Hybrid Decoder)
http://i2.sendpic.org/t/q8/q8udcrT0AIOfMLdf3rPgBbqwb8z.jpg (http://sendpic.org/view/2/i/ffz97YluX0ijnPqsCHEkzPKcukQ.png)

For DXVA the Frequency of the GPU it can be 200 mhz constant wouldn't matter then the Frequency of the DSP but for a Hybrid Decoder that uses the Shader Units that obviously doesn't applies anymore and when you then come with DXVA ideas for such a Bitstream complexity (heavily fluctuating frequency or to low frequency) it goes heavily wrong ;)

So for the Hybrid Decoder it would need very very good Dynamic Frequency Tuneing for the most efficient results, the same Strategy as with DXVA can't be used here ;)

[NikosD]

It would be the first time ever that they extended a codec beyond its "first" profile, and after releasing the spec at that. Which is why I think its unlikely - but like I said, maybe they'll surprise us!
It would probably require someone other than YouTube starting to pick up VP9 first, so it gets more wide-spread usage.

The new DXVA Checker added that change:

Version 3.12.0 (2016/01/20)
Added "VP9_VLD_10bit_Profile2" as an alternative name of Decoder Device
Minor fixes

Is it Microsoft's or Vendor oriented ?

[nevcairiel]

Microsoft doesn't define such a device, no.

[huhn]

7. Restricted-Mode Profiles
The following restricted-mode profiles for DXVA operations for VP8 and VP9 video decoding are
defined. The GUIDs that identify these profiles will be defined in the header file dxva.h. Additional
restricted-mode profiles for DXVA operations may be defined in the future.
7.2 DXVA_ModeVP9_VLD_10bit_Profile2 Profile
This profile supports the features necessary for a decoder that conforms to VP9 profile 2 for 10 bit video.
In this profile, the accelerator performs bitstream parsing, inverse quantization scaling, inverse transform
processing, motion compensation, and deblocking with the support of 4:2:0 chroma subsampling.
All data buffers shall contain only data that is consistent with the constraints specified for VP9 profile 2
for 10 bit video (including the constraint that BitDepthMinus8Luma and BitDepthMinus8Chroma shall
both be equal to 2).
The associated GUID definition for the corresponding entry in the dxva.h header file is as follows:
// {A4C749EF-6ECF-48AA-8448-50A7A1165FF7}
DEFINE_GUID(DXVA_ModeVP9_VLD_10bit_Profile2,
0xa4c749ef, 0x6ecf, 0x48aa, 0x84, 0x48, 0x50, 0xa7, 0xa1, 0x16, 0x5f,
0xf7);


source: https://www.microsoft.com/en-us/download/details.aspx?id=49188

the pdf it self has a date from 2015 12 14
the site said 2015 01 13

[nevcairiel]

Evil silent updates!
I didn't bother to re-download it because the page still had the same date.

Wonder which hardware will support it however.

[huhn]

unprofessional. the page still shows it is version 1.0

[NikosD]

I was right then ;)

Microsoft surprised us and for the first time made a second profile for VP9 (10bit) after the support of the first one.

All of this year's new architecture GPUs will support it, i think, and Intel and Nvidia will support it in hybrid mode even in the today's architecture, probably.

Wait for the new drivers for Skylake and Nvidia 950/960 GPUs!

[Nullack]

Anyone got a VP9 10 bit sample to trial for this stuff? Thanks

[P.J]

This player is back again (H.265 decoder coming soon)
https://mirillis.com/en/products/splash-free-hd-video-player-all-features.html

:rolleyes:

[wanezhiling]

http://i.imgur.com/Nus4VGT.png

Off topic, is the 'Video Acceleration Settings' option valid on ATI only?

[NikosD]

Probably yes.

I'm not in front of my PC right now, but ATI has some other exclusive features in DXVA checker, like GPU load.

[ionutm80]

Hi to all,

Just a dumb question: is lavfilters fully optimised for Haswell hevc 8bit hybrid decoding?
I'm asking this because I have tried playing a hevc file: http://demo-uhd3d.com/fiche.php?cat=uhd&id=126 , 23.8 Mb/s, 30 fps on my Intel NUC Core I3 4010U, HD 4400 and I have noticed something interesting: GPU was hoovering around 60% with random spikes to 99% and CPU usage varied between 40-60%. Indeed the video was choppy but my impression was that with a little more usage of both CPU and GPU it would have been playable. Tried the same with PowerDVD 15 but the spikes to 99% in GPU usage were more often than using MPC-HC + LAV DXVA native.

Any thoughts on this?

[NikosD]

I have myself reported a few times that LAV seems more optimised for Nvidia's hybrid HEVC decoding than Intel's

The behavior you describe is common to all Intel iGPUs.

Using PowerDVD 15, the opposite is true.
It seems that it is more optimised for Intel than Nvidia.

Generally speaking, those 4K HEVC 8 bit files are a little bit out of hybrid's decoding capabilities using common iGPUs from Intel.

Maybe those rare Iris iGPUs of Haswell and Broadwell could handle them better.

Of course for pure HW decoders from all brands (AMD, Intel, Nvidia) those 8 bit HEVC clips are piece of cake.

[ionutm80]

Of course for pure HW decoders from all brands (AMD, Intel, Nvidia) those 8 bit HEVC clips are piece of cake.

I know :). I have another computer with GTX 960 and I was able to play smoothly all the samples from http://demo-uhd3d.com/.

What I still hope is for an add-on mini pci-e card specifically dedicated for HEVC decoding to emerge (as was the Chrystal Broadcom HD for H.264). A lot of people with Haswell and Broadwell laptops, NUCs will probably not want to upgrade their PCs just to be able to play H.265 files.

[NikosD]

Very interesting new version of DXVA Checker v3.13.0, including a few things requested by many users:

Version 3.13.0 (2016/02/11)
Added support to select GPU
Added support to confirm GPU Usage and Video Engine Usage for NVIDIA GeForce series
Removed some D3D11 TraceLog events
Updated DSF/MFT Viewer
Minor changes

[P.J]

Very interesting new version of DXVA Checker v3.13.0, including a few things requested by many users:

Version 3.13.0 (2016/02/11)
Added support to select GPU
Added support to confirm GPU Usage and Video Engine Usage for NVIDIA GeForce series
Removed some D3D11 TraceLog events
Updated DSF/MFT Viewer
Minor changes

It doesn't work, can't use iGPU :rolleyes:

[nevcairiel]

Switching seems to work fine between my two NVIDIA cards, neat.

[P.J]

For iGPU HD4600, Playback doesn't work at all and Decode works only in CPU mode.
GTX960 (connected to monitor) + HD4600

[NikosD]

As of new Nvidia drivers family 361 (December 2015), using a Windows 10 November Update OS you can get Fermi DX12 (WDDM 2.0) support.

Nvidia, silently, updated their drivers to support Fermi cards in single GPU and non-SLI modes, but not SLI-mode yet.
Maybe that is the reason they didn't advertise their achievement so much, they wrote it in release notes though.

I tried latest 361.91 released yesterday on my GT610 and it works!

I got a WDDM 2.0 driver model with Feature Level support 11.0 and D3D API 11.2/11.3

[nevcairiel]

Note that those 361 drivers only provide WDDM 2.0 support, not DX12 support yet. Its an important step on the way, but its not the final goal yet.
Full DX12 support would show the appropriate D3D API in dxdiag, but right now its still 11.2/3

[NikosD]

Are you sure that Fermi will get full DX12 support ?

[nevcairiel]

Are you sure that Fermi will get full DX12 support ?

That depends what you call "full", but NVIDIA has said it will get DX12 support, that means at least API support to benefit from the lower CPU overhead.

[NikosD]

I thought that WDDM 2.0 was enough for low CPU overhead.

[NikosD]

Speaking of DX12, the other low level API based on AMD's Mantle API - Vulkan - released yesterday by Khronos, as the Next Generation OpenGL successor.

In this case the table of GPUs supporting Vulkan 1.0 is this:

AMD GCN 1.0 & Later

Intel Broadwell (Gen 8.5) & Later

NVIDIA Kepler & Later

[chros]

As of new Nvidia drivers family 361 (December 2015), using a Windows 10 November Update OS you can get Fermi DX12 (WDDM 2.0) support.
What does it mean exactly? only 8bit @ fullHD resolution?
Is it only on Win10?
(I'm asking because of my GT650m on Win8.1)

Speaking of DX12, the other low level API based on AMD's Mantle API - Vulkan - released yesterday by Khronos, as the Next Generation OpenGL successor.
Sorry, what's this for?
Thanks

[NikosD]

Never mind.
It's off - topic and I wouldn't like to be discussed any more.

[Yups]

Speaking of DX12, the other low level API based on AMD's Mantle API - Vulkan - released yesterday by Khronos, as the Next Generation OpenGL successor.

In this case the table of GPUs supporting Vulkan 1.0 is this:

AMD GCN 1.0 & Later

Intel Broadwell (Gen 8.5) & Later

NVIDIA Kepler & Later


Broadwell is Gen8

[vinnytx]

Please, P.J or someone else can tell me how to show this control panel?

CLCvd.ax? No luck with 750Ti too :(

http://i.imgur.com/h9FyKY0.png





I have only this

http://i.imgur.com/Uu9HyYO.jpg

[P.J]

Press Ctrl+Shift before opening it.

[NikosD]

Latest DXVA Checker v3.13 doesn't look compatible with Nvidia Optimus technology.

To be more exact, my Nvidia 740M card in a laptop using Optimus technology, falls back to SW decoding using both LAV and MS DS decoders in DXVA2 native mode.

I even forced DXVA Checker to run on Nvidia card through Nvidia's Control Panel, but still with no success.

It falls back to SW mode.

Could anyone else test if GPU switching technology from Nvidia or AMD is compatible with latest DXVA Checker v3.13 ?

[wanezhiling]

http://bluesky23.yukishigure.com/en/DXVAChecker.html
v3.14 should have fixed that.

[NikosD]

I've been talking with DXVA Checker's developer about the problem and when I saw the new version, I thought the same thing.

Now the problem is that I don't have the laptop available to check it out :)

[P.J]

https://mirillis.com/en/products/splash-free-hd-video-player-history.html

25/03/2016 Splash 2.0.2

NEW FEATURE: Hardware accelerated HEVC/H.265 video decoding support added for NVIDIA and Intel graphics cards

Old right now :rolleyes:

[P.J]

Can someone play a 4K 10bit HEVC 59.940fps video like Samsung_UHD_Ride_on_Board.ts in EVR/EVR-CP mode with GTX 950/960 2GB on a native 4K monitor/TV?
Someone told me that it requires at least GTX 970 or maybe GTX 960 4GB :rolleyes:

[clsid]

2GB is more than enough. A decoded 4K frame is ~32MB. The person that told you it isn't enough probably uses a too high GPU buffer in madVR.

[huhn]

even with an GPU queue of 4 you get somethining like 1700+ mb Vram usages.

for save deinterlacing a GPU queue of 8 is kind of needed which is not possible with 2 GB Vram.

i'm pretty sure i can play the Samsung_UHD_Ride_on_Board.ts with my 960 on my UHD screen i will try it later.

[huhn]

playback with EVR (not EVR-CP) is no issue at all.

madVR playback is doable with bilinear chroma scaling.

Vram usages
EVR ~1100
madVR ~1700 with a GPU queue of 4 and present frames in advance at 3.

[clsid]

crowd_run_2160p50_Main10-127Mbps.mkv

Played on a 1920x1200 screen. But should not make huge difference since most buffered data will be in the full 4k size.

EVR-CP with standard buffer size of 5:
OS (no playback): 180 MB
No HWA: 438 MB
DXVA2 native: 927 MB
DXVA2 cb: 1296 MB

MadVR (GPU queue 4, present 4):
DXVA2 cb: 1072 MB

MadVR (GPU queue 8, present 4):
DXVA2 cb: 1270 MB

MadVR (GPU queue 8, present 8):
DXVA2 cb: 1299 MB

[huhn]

native: http://abload.de/img/vramusagef2bqr.png

copyback: http://abload.de/img/vramusage2kqjp6.png


the OS eats ~650 MB Vram with an open browser and i don't have unusual programs running in the back ground.

you get the Vram usage without browser later.

most likely the 200% scaling is eating a lot.

~350 mb after reboot with and without 200% scaling.

after closing madVR Vram usage was normal.
than i was running madVR with an 3D LUT and after closing i got 521 mb Vram usages.

[Rza11]

I am getting micro stuttering while playing x265 @ 24p.

PC: I7 920 @ 4ghz with GTX 670
TV: 1080p @ 24p
Source file: 1080p

Shouldn't my PC be enough to play these files correctly? As you guys can see i'm using latest lav with madvr and the cpu doesn't seem to be taxed much.

Does Strongene have the same quality as lav for 10 bit ?

http://i.imgur.com/1VDyXmC.jpg

http://i.imgur.com/gTMaN25.jpg

[nevcairiel]

From the madVR screenshot, its not a decoding problem, madVR receives enough frames, but the GPU seems to not handle them fast enough.

[Rza11]

From the madVR screenshot, its not a decoding problem, madVR receives enough frames, but the GPU seems to not handle them fast enough.

Really ok, I will try update drivers (currently 361.91) and all software to absolute latest and report back.

[Rza11]

I installed latest nvidia drivers and checked all codecs and kodi are latest.

I'm not getting any presentation errors but i still get micro stutters.

Anyone know what else should i look into?

Updated Stats:

http://i.imgur.com/UjbHyho.jpg

http://i.imgur.com/7mRdqrc.jpgc

[chros]

I installed latest nvidia drivers and checked all codecs and kodi are latest.

I'm not getting any presentation errors but i still get micro stutters.
I think the problem is that the refresh rate of your screen (23.971) is lower than the content's (23.976).
Solution: use 24p in madvr config or in OS not 23p, OR try to tweak the frequency in 23p to get as close to 23.976 from above.

[nevcairiel]

Your GPU just can't keep the GPU queues filled, all GPU queues are nearly empty and you have tons of dropped frames. There are no obvious answers. Reset madVR settings to defaults, try again. See if some particular setting causes it to break again, etc.
D3D11 10-bit output is problematic on some systems, so try to avoid that. In general, greatly simplifiy the setup until it works, then try adding more options.

[Rza11]

I have disabled exclusive mode which disabled 10 bit and everything is fine now, presentation is filling and no dropped frames. I think this will have to do until i find out the real problem.

in general is a 670 gtx enough for 10 bit or is it borderline? maybe i should upgrade gpus

[krille]

I installed latest nvidia drivers and checked all codecs and kodi are latest.

I'm not getting any presentation errors but i still get micro stutters.

Anyone know what else should i look into?

Updated Stats:

http://i.imgur.com/UjbHyho.jpg

http://i.imgur.com/7mRdqrc.jpgc

This happens to me too in exclusive mode with my Nvidia 960, the queues doens't fill up. Disabling exclusive mode solves the problem for me.

[chros]

Your GPU just can't keep the GPU queues filled, all GPU queues are nearly empty and you have tons of dropped frames. There are no obvious answers. Reset madVR settings to defaults, try again. See if some particular setting causes it to break again, etc.
D3D11 10-bit output is problematic on some systems, so try to avoid that. In general, greatly simplifiy the setup until it works, then try adding more options.
Oh, I missed that part completely :)
I have disabled exclusive mode which disabled 10 bit and everything is fine now, presentation is filling and no dropped frames. I think this will have to do until i find out the real problem.
in general is a 670 gtx enough for 10 bit or is it borderline? maybe i should upgrade gpus

This happens to me too in exclusive mode with my Nvidia 960, the queues doens't fill up. Disabling exclusive mode solves the problem for me.
That's interesting. With my GT650m (optimus system! , using DXVA-copyback in LAVvideo) I CAN use D3D10 10bit FSE output up to 1080p@30p material (4K is a different beast, and 50p/60p content (like sports caps) are more demanding). So that means GPU isn't the problem, it has 2GB GDDR5 VRAM.
If you use dxva-copyback hardware acceleration, then try out software decoding, maybe it helps.

[Rza11]

Interesting,

With my 670 gtx most stuff >90% plays correctly @ 24p 10 bit but i have a few action vids that stutter.

Maybe it has something to do with the way there encoded.

I tested all the lav accelerator settings but there all the same.

I may try overclock the evga 670 and see if it makes a difference.

very interesting it does that too with the 960 i thought the hardware decoding would of taken care of all the x265 issues

[chros]

very interesting it does that too with the 960 i thought the hardware decoding would of taken care of all the x265 issues
You can upload a sample file of the problematic part somewhere so we can check it.

[Rza11]

I might do, I'll look into making a sample

[DragonQ]

Which cards support full hardware HEVC decoding nowadays? Is it just the GTX 750 Ti, 950, and 960? Thinking of upgrading my GT 430 since it's useless for anything in MadVR beyond Bicubic and Lanczos, and I may as well ensure I can play UHD video encoded as HEVC whilst I'm at it.

Also, I notice some people in this thread using Kodi DSPlayer with MadVR. I am testing that on my setup but have an issue with TV - when switching from an SD to an HD channel, the screen goes entirely black. I can still Alt-F4 or even change channel using keyboard shortcuts and if I change back to an SD channel, the video and UI reappear. Anyone else see this issue?

[NikosD]

10bit HEVC decoding in fixed function hardware (full) is supported only by Nvidia cards - 950 and 960 (+2nd edition of 750 ti).

But 8 bit HEVC full hardware decoding is supported by Skylake integrated iGPU and latest iGPUs and discrete cards from AMD.

Personally, I'm waiting for the new cards of Nvidia and AMD which are just around the corner.

They will support 8&10 bit HEVC and probably VP9 full fixed function HW decoding.

[clsid]

Yes, those are the only ones with 10bit. For 8-bit there is also AMD Fury/Nano and Intel Skylake.

[P.J]

... and another one collegue with Braswell NUC (Pentium N3700 - HD Graphics). Seems to be better, like GTX960.

http://i60.tinypic.com/nej7yq.jpg

No benchmark for Braswell/Cherry Trail? :rolleyes:

[wanezhiling]

http://www.anandtech.com/show/10256/intel-unveils-apollo-lake-14nm-goldmont
The new graphics core is listed as being more powerful (most likely due to both better architecture and a higher count of execution units), but will also integrate more codecs, enabling hardware-accelerated playback of 4K video from hardware decoding of HEVC and VP9 codecs.

[wanezhiling]

It's Gen9 GPU though, same as Skylake so no HEVC Main10 hardware decoding, only hybrid/partial VP9 hardware decoding, no HDMI 2.0/2.0a support, no HDCP 2.2 support.
This Gen9 is probably a little different from skylake's.
Because braswell/Cherry Trail's Gen8 has totally same decoding ability with skylake's Gen9 (8-bit HEVC).;)

[P.J]

Skylake's iGPU can decode 10-bit HEVC in hybrid mode but Braswell/Cherry Trail =/

[nevcairiel]

It's Gen9 GPU though, same as Skylake so no HEVC Main10 hardware decoding, only hybrid/partial VP9 hardware decoding, no HDMI 2.0/2.0a support, no HDCP 2.2 support.

The media support of the Atom platform typically is a generation ahead of the desktop platform. Braswell has a Gen8 GPU but still VP9-Hybrid support, which on the desktop is a Skylake feature.
Its important not to confuse the actual GPU with the media capabilities, which are not directly tied together.

[wanezhiling]

Yes, as nevcairiel said, I'm sure that apollo lake already supports hevc 10-bit & vp9 8-bit(maybe also 10-bit?) hardware decoding, not hybrid any more.
Of course the next desktop gen10 gpu(Kabylake) will has same supports like apollo lake.

[P.J]

http://www.tweaktown.com/news/51746/amds-official-2016-2018-gpu-roadmap-released-vega-hbm2-2017/index.html

http://imagescdn.tweaktown.com/news/5/1/51746_01_amds-official-2016-2018-gpu-roadmap-released-vega-hbm2-2017.png

[Yups]

https://lists.freedesktop.org/archives/intel-gfx/2015-October/077255.html



Kaby Lake is Gen9.5. Cannonlake should be Gen10.


Kabylake is also often referred to Gen9 from Intel. There are some hints that Broxton supports HEVC 10 decoding in hardware.

https://lists.freedesktop.org/archives/libva/2015-December/003756.html

Broxton being Gen9 doesn't mean it can't have HEVC 10 bit decoding in hardware. Intel wrote that Broxton Gen9 differs slightly to Gen9 from Skylake. I can't find the link at the moment but found a similar one.

http://permalink.gmane.org/gmane.comp.freedesktop.xorg.drivers.intel/55947/gmane.comp.freedesktop.xorg.drivers.intel/55947

Very similar means it's not identical.


Here is a good overview

https://build.opensuse.org/package/view_file/X11:XOrg/vaapi-intel-driver/vaapi-intel-driver.changes?expand=1

+ Add support for Skylake.
- Decoding: H.264/MPEG-2/VC-1/JPEG/VP8/HEVC

+ Add support for Broxton:
- Decoding: H.264/MPEG-2/VC-1/JPEG/VP8/HEVC/HEVC 10-bit/VP9.

* Add support for Kabylake
- Decoding: H.264/MPEG-2/VC-1/JPEG/VP8/HEVC/HEVC 10-bit/VP9/VP9 10-bit

[leonccyiu]

I can't wait to find out what the hardware decode capabilities of the new Nvidia Pascal cards are. Since they have displayport 1.4 I wonder if they'll do Hevc 10bit at 8k. I assume that they'll have VP9, in 10bit as well, although I doubt google chrome will able to take advantage of that.

I believe the reviews are coming on May 17th.

Hopefully we'll hear something about Kabylake at computex.

[Yups]

http://fs5.directupload.net/images/160515/o9qx4kyv.png
http://videocardz.com/59962/nvidia-geforce-gtx-1080-final-specifications-and-launch-presentation


Even 12 bit decoding.

[nevcairiel]

Even 12 bit decoding.

I doubt that until I see it. The DXVA2 HEVC spec includes nothing for it. Leaks will be leaks.

[aufkrawall]

Maybe Microsoft will again silently edit some knowledge articles? :D

[nevcairiel]

Maybe Microsoft will again silently edit some knowledge articles? :D

Perhaps, but 12-bit is not part of the Main/Main10 profile, so adding that would involve quite some additional changes, and not just add a new profile specific for this mode.
So far they did not, anyway.

But I'll be getting a 1080 myself, so I can see whats what.

[Yups]

I doubt that until I see it. The DXVA2 HEVC spec includes nothing for it. Leaks will be leaks.


Who is talking about DXVA2 HEVC specs? It doesn't say anything about the hardware capability of Pascal. If DXVA2 doesn't support HEVC 2.0 Main 12 I'm sure it will be supported sooner or later.

[nevcairiel]

Who is talking about DXVA2 HEVC specs? It doesn't say anything about the hardware capability of Pascal. If DXVA2 doesn't support HEVC 2.0 Main 12 I'm sure it will be supported sooner or later.

It can have whatever hardware capabilities it wants, but if its not exposed through documented interfaces, we can neither verify it actually has them, nor use them.
But we'll see.

[nevcairiel]

Date Published:
2/17/2016

Here we go again with the silent updates? Or nothing actually changed?

Thats 3 month old and only contained some extra clarifications, no new things.

[JohnLai]

http://fs5.directupload.net/images/160515/o9qx4kyv.png
http://videocardz.com/59962/nvidia-geforce-gtx-1080-final-specifications-and-launch-presentation


Even 12 bit decoding.

Hmmm? 10bit HEVC encode? I wonder how many 'specification' being skipped for the encode again.......

Last time maxwell v2 hevc is missing SAO, 64 CTU, B-frame, what else?

[leonccyiu]

http://fs5.directupload.net/images/160515/o9qx4kyv.png
http://videocardz.com/59962/nvidia-geforce-gtx-1080-final-specifications-and-launch-presentation


Even 12 bit decoding.

Wow I didn't expect the specifications to leak tonight
I can't remember from several posts back, someone testing Intel's Skylake decoder for 8k HEVC benchmarks, did it manage 60?

A concern of mine is regarding VP9 decode, it lists 4k and 2x @120hz, the same as HEVC 8bit, but no mention of 8k. Youtube is going to be our prime source of 8k content for the foreseeable future, and with google's lackluster libvpx codec, at least a 5820k will be required for playback.

Youtube's 8k VP9 is monstrous, the bit-rates range from 60-120, and potplayer uses >2gb ram during playback.

This is going off topic but here are some videos that are encoded in 8kvp9 and not the low bit-rate h264 they originally used.

https://www.youtube.com/watch?v=QPdWJeybMo8
https://www.youtube.com/watch?v=VXUdY5P0VJA&authuser=0
https://www.youtube.com/watch?v=_KoGjIDAFk8&authuser=0

if anyone wants to try playing them back by CPU or downloading them for MPC and ffvp9.

[nevcairiel]

Hmmm? 10bit HEVC encode? I wonder how many 'specification' being skipped for the encode again.......

Last time maxwell v2 hevc is missing SAO, 64 CTU, B-frame, what else?

All these features are optional for an encoder, there is nothing saying it has to use them. Their primary purpose for these encoders is high-bitrate realtime encoding of you gaming, it doesn't need all sorts of fancy coding tools to do that.
In fact, for the purpose of streaming to a shield tablet or a TV, you wouldn't want some of those either way to reduce latency.

Such hardware encoders will never really be able to produce high quality encodes on low to medium bitrates. x264 and x265 will always be the best choices there.
The only thing hardware has going for it is that it is fast. To be so fast, it doesn't use all fancy tools.

[Yups]

http://www.computerbase.de/2016-05/geforce-gtx-1080-test/8/#diagramm-videos-abspielen-h264-codec-lav-filters


Much faster decoding.

[NikosD]

Looks very fast, but from the comparison tables Skylake is missing.

Of course, Polaris (10 & 11) and Kabylake are the real rivals for Pascal, regarding video decoding/ encoding.

[huhn]

even if polaris is only half as fast as this new pascal decoder it doesn't matter. it is so fast you are not run in real work issues with it.

the comparison is pretty pointless anyway there are clearly using an 8 bit encode.

[NikosD]

Maybe DXVA checker version of the screen shot is old and represents 8K as 4KUHD.

You need a newer version or even better the latest.

I'm not in front of my PC to check it out.

Also, DXVA checker displays Microsoft's or sometimes OEM's (rarely) device decoders which obviously are not available for HEVC Main12 by both of them (I mean Microsoft and Nvidia)

[nevcairiel]

Like I said on the previous page, MS needs to release an updated specification to allow Main12 decoding through "standard" means, which would mean it gets new GUID, and would not show up as a "named" device in DXVA Checker unless this GUID is known and DXVA Checker updated to support it (or NVIDIA cheats and allows 12-bit decoding through the Main10 device, but that would be evil and hard to detect in software).

We'll have to wait for MS, or NVIDIA, to make a move there, and more information to surface.

For that matter, their DXVAChecker does seem really old. 33FCFE41 is MJPEG_VLD_PureVideo and 463707F8 is VP9_VLD_Profile0.
The two entries below HEVC_VLD_Main10 are new however, so maybe one of those is supposed to be HEVC Main12.

[v0lt]

http://www.computerbase.de/2016-05/geforce-gtx-1080-test/8/#diagramm-videos-abspielen-h265-codec-lav-filters

[baii]

http://www.computerbase.de/2016-05/geforce-gtx-1080-test/8/#diagramm-videos-abspielen-h265-codec-lav-filters
Hmm, all the amd and old Maxwell, can I assume they are h.265 8 bit?
It is good that the performance is better but decoding is not exactly a how fast question but more a can it do it type of thing, which a 960 is enough for now.

Sent from my 306SH using Tapatalk

[ashlar42]

Such hardware encoders will never really be able to produce high quality encodes on low to medium bitrates. x264 and x265 will always be the best choices there.
The only thing hardware has going for it is that it is fast. To be so fast, it doesn't use all fancy tools.
Maybe sooner or later we will be able to have different "profiles" for the hardware encoders. Or are you saying there would not be spped benefits if encoding with the same quality through GPU?

I would say there would but I am not sure. There seems to be a lot of headroom for raising quality while mantaining a speed advantage.

[huhn]

they have different profiles already.

these encoder are missing way to many well tweak feature that x265 or x264 have.

[NikosD]

For those interested, I uploaded a 8K 19s clip from Youtube in two versions.

Original 8K VP9 and converted 8K HEVC (8bit)

Skylake owners and maybe Pascal or probably even future Polaris owners are welcomed to post their pure decode benchmark (or even playback) results, using LAV x64 0.68.1 and their HW decoders in DXVA native mode for both HEVC and VP9 formats.


8K HEVC - 24fps@38Mbps
https://www.sendspace.com/file/80sclj

8K VP9 - 24fps@66Mbps
https://www.sendspace.com/file/qlujz9

[Yups]

Only 8 bit support in hardware for Bristol Bridge?

[v0lt]

@GTPVHD
8k nobody needs.
This is just a measure for the GPU manufacturers. It is not practical value for consumers.

[Stereodude]

Has anyone seen what hardware decoding capabilities are in Polaris?

[leonccyiu]

@GTPVHD
8k nobody needs.
This is just a measure for the GPU manufacturers. It is not practical value for consumers.

I'd disagree, watching youtube's 4k vp9 on a 55 inch 4k tv, it's clearly a step back from the high bit-rate demo clips. I watch 4k on youtube downscaled to 1080p on my monitor and it looks like perfect 1080p, any combing/aliasing artifacts are likely present in the source. Since the video is 4:2:0, when downscaled it'd be 4:4:4 as well.

Since youtube already has 8k vp9, and that'll likely remain the only source of 8k for a while, 8k hevc and vp9 would provide a sharper 4k image.

[leonccyiu]

Only 8 bit support in hardware for Bristol Bridge?

Has anyone seen what hardware decoding capabilities are in Polaris?

I certainly hope Polaris is capable of 10 bit HEVC decoding at >60fps.

[vivan]

I'd disagree, watching youtube's 4k vp9 on a 55 inch 4k tv, it's clearly a step back from the high bit-rate demo clips. I watch 4k on youtube downscaled to 1080p on my monitor and it looks like perfect 1080p, any combing/aliasing artifacts are likely present in the source. Since the video is 4:2:0, when downscaled it'd be 4:4:4 as well.The only reason it looks better is because it has higher bitrate. Still it's bitstarved to death and any challenging content turns it into shit regardless of the resolution. If that 1080p video had that bitrate it would look much better.

So no, stressing hardware and wasting bandwitch just because one service cares only about marketing is a terrible reason. And it's really sad how upscaling and watching downscaled upscaled content becomes a norm because of this.

[Yups]

A confirmation from the updated Media SDK release notes that Kabylake supports 10 Bit HEVC and VP9:

HW HEVC Encode: Full HW HEVC encode MAIN10 profile supported on 7th Generation
Intel® Core™ and Core M platforms

HW HEVC Decode: Full HW HEVC decode MAIN10 supported on 7th Generation Intel® Core™ and Core-M platforms.

HW VP9 Decode: Full HW VP9 8 bit and 10 bit decode supported on 7th Generation
Intel® Core™ and Core-M platforms

[NikosD]

A more interesting detail in release notes is this:

MFX_CODEC_VP9 and MFX_PROFILE_VP9_0, MFX_PROFILE_VP9_1, MFX_PROFILE_VP9_2, MFX_PROFILE_VP9_3 added for VP9 HW decode support.

If Kabylake supports all VP9 profiles, then according to wikipedia:

Profile 0 allows for a bit depth of 8-bits per sample and supports 4:2:0 chroma subsampling.

Profile 1, which is optional for hardware, adds support for 4:2:2 chroma subsampling, 4:4:4 chroma subsampling, alpha channel support, and depth channel support.

Profile 2 allows for a bit depth of 10-bits to 12-bits per sample and supports 4:2:0 chroma subsampling.

Profile 3 adds support for 4:2:2 chroma subsampling, 4:4:4 chroma subsampling, and alpha channel support.

[leonccyiu]

— Next-generation display technologies – Radeon™ RX Series includes support for next-generation HDR gaming and video on new HDR monitors and TVs. The Radeon™ RX Series also supports HDMI™ 2.0b and DisplayPort 1.3/1.4 supporting the new generation of high-resolution HDR and high-refresh displays. The Radeon™ RX Series features exceptional accelerated H.265 encoding and decoding, enabling effortless streaming or recording of 10-bit 4K video at 60 FPS4.

http://videocardz.com/61028/amd-announces-radeon-rx-470-and-rx-460

Confirmation that the next generation AMD cards do 10 bit H.265 decoding. Will have to wait for more details, and about VP9

[NikosD]

http://videocardz.com/61028/amd-announces-radeon-rx-470-and-rx-460

Confirmation that the next generation AMD cards do 10 bit H.265 decoding. Will have to wait for more details, and about VP9

And 10bit HEVC encoding too

[P.J]

Too late :rolleyes:

Waiting for 8K...

[Stereodude]

http://videocardz.com/61028/amd-announces-radeon-rx-470-and-rx-460

Confirmation that the next generation AMD cards do 10 bit H.265 decoding. Will have to wait for more details, and about VP9
There is no clear statement about the decode capabilities. It says, "The Radeon™ RX Series features exceptional accelerated H.265 encoding and decoding, enabling effortless streaming or recording of 10-bit 4K video at 60 FPS." I guess "streaming" means playback... :confused:

Is 12-bit HEVC decode necessary for any existing content / format yet?

[NikosD]

More and more HW manufacturers are moving to VP9, favoring their components with full decode/encode support of all or main (8bit/10 bit) VP9 profiles.

So, after Intel which seems to support all VP9 profiles (for decoding) with Kabylake, now ARM releases Egil Video processor with full support of 4K decode/encode VP9 (both 8bit/ 10bit)

According to anandtech and a lot of other people, the patent-free nature of VP9 vs the strict and complex license scheme of HEVC could be a reason for that.

http://www.anandtech.com/print/10428/arm-announces-mali-egil-video-processor

[Yups]

Looks very fast, but from the comparison tables Skylake is missing.

Of course, Polaris (10 & 11) and Kabylake are the real rivals for Pascal, regarding video decoding/ encoding.


https://www.computerbase.de/2016-06/radeon-rx-480-test/7/#diagramm-videos-abspielen-h265-codec-lav-filters


There are some decoding benchmarks of Polaris. AMD can't compete with Pascal.

[NikosD]

We need DXVA screenshots of both Pascal and Polaris in order to check video HW acceleration capabilities.

Like HEVC 8bit/10bit & VP9 8bit/10bit and resolution too, like 8K.

And then we can talk about our favorite subject, benchmarks of course.

[Yups]

http://scr3.golem.de/screenshots/1606/Radeon-RX-480-Test/thumb620/Test-Radeon-RX-480-17.png


It's pretty clear that HEVC 8/10 bit decoding will be supported and VP9 only 8 bit, both up to 4k.


Encoding is unclear: http://scr3.golem.de/screenshots/1606/Radeon-RX-480-Test/thumb620/Test-Radeon-RX-480-18.png

There is no mention of 10 bit HEVC.


DXVA screenshot is here: http://www.notebookcheck.net/AMD-Radeon-RX-480-Review-The-fastest-Polaris-desktop-card-at-launch.168250.0.html


The new H.265 decoder handles a 4K 10-bit video at 72 fps and 2 % CPU load.

[NikosD]

Someone has to tell them to update the DXVA Checker version.

They are using an ancient one where 8K is not represented separately, but as 4KUHD.

Not that I think that Polaris supports 8K though.

[nevcairiel]

DXVA screenshot is here: http://www.notebookcheck.net/AMD-Radeon-RX-480-Review-The-fastest-Polaris-desktop-card-at-launch.168250.0.html

There is no VP9 in that list yet, at least not under the official Microsoft DXVA VP9 specification.
Guess future drivers will tell.

[huhn]

this german page claims it supports VP9 and HEVC 10 bit.

but this page is talking about 2 pass encoding for live streaming and that is an "interesting" claim...

http://www.pcgameshardware.de/AMD-Radeon-Grafikkarte-255597/Specials/RX-480-Test-1199839/

performance and power consumption are underwhelming. i had so much hope for this card...

[huhn]

and the decoder is slower than maxwell(950, 960) but still so fast that the speed doesn't really matter.

[nevcairiel]

6GB is plenty on that level of performance, honestly. The real world performance and price in the end is what matters then.

[mandarinka]

this german page claims it supports VP9 and HEVC 10 bit.

but this page is talking about 2 pass encoding for live streaming and that is an "interesting" claim...

http://www.pcgameshardware.de/AMD-Radeon-Grafikkarte-255597/Specials/RX-480-Test-1199839/

performance and power consumption are underwhelming. i had so much hope for this card...

In the architecture press materials, it is stated that VP9 is "to be enabled in a later driver release". It is not working yet (and is probably shader/GPGPU-based).

[huhn]

In the architecture press materials, it is stated that VP9 is "to be enabled in a later driver release". It is not working yet (and is probably shader/GPGPU-based).

that would be a little bit sad.

most CPU can handle VP9 easily because it is mostly used on youtube with very low bitrate. not a deal breaker in my eyes.

[P.J]

Waiting for 1050/1060 :rolleyes:

[sneaker_ger]

If my main use is video, I'd choose the 1060 for VP9 10 bit and better efficiency (imagine Youtube would offer 10 bit for all new VP9 streams...).
Who said it will support 10 bit VP9? :confused:

[aufkrawall]

I confused it with 12 bit HEVC, nevermind.

[NikosD]

From Tom's Hardware it seems that Polaris could support VP9 Profile 2 (10bit)


AMD lists HEVC decode at up to 4K60 using the Main 10 profile, which allows for a 10-bit 4:2:0 format (it all goes back to making HDR possible).

VP9 decode is supported by the hardware, though AMD’s drivers don’t expose it yet—we only know the feature is planned for a future update.

Assume at least profile 2 compatibility if AMD wants to match HEVC’s 10-bit/4:2:0 chroma subsampling for HDR.

Less consequential, perhaps, is hardware-based acceleration of the M-JPEG format at up to 4K30.

[Yups]

From Tom's Hardware it seems that Polaris could support VP9 Profile 2 (10bit)



I don't think so, otherwise AMD would have told so.

[huhn]

only 6 GB every should throw there flag ship fury nano fury X out of the window they only have 4 GB...

[P.J]

GTX950/960 are fine for people who don't game :)

[P.J]

GTX950/960 are fine for people who don't game :)

I meant for 4K HEVC 10-bit playback since i5/i7 can't handle it :)

[Plutotype]

So what is the status of DXVA on RX480 for 10bit/12bit HEVC?
Thanks

[Trevonn]

10-bit videos are decoded in hardware just fine. I don't have any 12-bit videos to test sorry.

[Plutotype]

10-bit videos are decoded in hardware just fine. I don't have any 12-bit videos to test sorry.
HEVC 2160 10-bit?

[Trevonn]

HEVC 2160 10-bit?

Yes that works too

http://i.imgur.com/xqQ7WSt.png

[NikosD]

Can you give us some benchmark figures in order to show us the performance ?

Do you know how to do that ?

[Trevonn]

Can you give us some benchmark figures in order to show us the performance ?

Do you know how to do that ?

I have DXVA Checker etc but I don't know of the top of my head. If you could point me to a tutorial I'd be more than capable of following that.

p.s I'm a long time lurker

[NikosD]

http://uploads.tapatalk-cdn.com/20160717/8bc6b1c07063febe3f1ab30879c3ade8.jpg

It seems from the official presentation of Polaris cards that they support 4K VP9 acceleration.

In the picture the 9 note of the slide says:

"Hardware-accelerated VP9 decode support is planned for enhancement in a future Radeon Software driver release"

I hope that hardware-accelerated VP9 decode support to be fixed function and not hybrid though.

[Trevonn]

Can you give us some benchmark figures in order to show us the performance ?

Do you know how to do that ?

I have figured out how to benchmark. Are there any clips you would like to me test?

[NikosD]

I have figured out how to benchmark. Are there any clips you would like to me test?

OK, that's good.

We haven't seen any fast H.264 decoder since R9 285 (Tonga), the first 4K120fps H.264 decoder from AMD, so let's start with H.264 files first.

1) H.264 -1080p-108Mbps
ftp://helpedia.com/pub/multimedia/x264/testvideos/2012%20-%2001%20-%20QuickSync%20vs%20UVD%202.2%20vs%20VP4/9.Ducks.Take.Off.1080p30fpsRef5-108Mbps.mkv

2) H.264 - 2160p-370Mbps
ftp://helpedia.com/pub/multimedia/x264/testvideos/2160p%20samples/DucksTakeOff_2160p50.x264.CRF24.mkv


3) H.265 8bit -1080p-80Mbps
http://www.imaging-resource.com/PRODS/samsung-nx1/FULLRES/yvid_SAM_0230-ProQuality.MP4

4) H.265 8bit -1080p-80Mbps
http://imaging-resource.com/PRODS/samsung-nx1/FULLRES/yvid_SAM_0235-UHD_Sample1.MP4

5) H.265 10bit - 2160p-140Mbps
http://jell.yfish.us/media/jellyfish-140-mbps-4k-uhd-hevc-10bit.mkv

6) H.265 10bit -2160p-400Mbps
http://jell.yfish.us/media/jellyfish-400-mbps-4k-uhd-hevc-10bit.mkv


Please use latest DXVA Checker v3.14.0 - LAV x64 video 0.68.1.x in DXVA native mode and in decode mode.

If you have time, include also playback mode results using the 1280 x 720 resolution

Run each benchmark 3 to 5 times and always exclude the first run, because it reads the file from the disk.

For each file, post the average of your runs in this form min (fps) /avg(fps) /max (fps)

For example:

1) H.264 -1080p-108Mbps

Decode:
RX 480 160/164/168

Playback:
RX 480 147/155/160

Have fun!

[NikosD]

According to the latest official docs from Nvidia, Pascal supports on selected models 8K x 8K HEVC and VP9 HW decoding.

Also Pascal has a lot faster H.264 HW decoder than the fastest Maxwell 960 GTX and even faster HEVC HW decoder.

Maybe for the first time since Sandybridge, Nvidia managed to outperform Intel in HW decoding speed on all codecs.

[P.J]

http://www.guru3d.com/articles_pages/geforce_gtx_1060_review,1.html
https://www.techpowerup.com/reviews/NVIDIA/GeForce_GTX_1060/

[Trevonn]

I have sold my Reference RX 480. So I won't be able to provide benchmarks for it sorry. In the near future I will purchase a GTX 1060 as I feel it will better suit my encoding, decoding needs.

[nsnhd]

Hi All, my first post here.
After updating the latest Intel driver, I did benchmark this file on my htpc with Win 10 x64 (10586.494)-I5-4440-iGPU HD 4600 (v.4463)

H.265 10bit - 2160p-140Mbps
http://jell.yfish.us/media/jellyfish...hevc-10bit.mkv

Results with DXVA Checker v3.14.0 LAV x64 video 0.68.1 as follows:
Decode:
CPU 12/21/27
Playback:
CPU 10/18/22

However, when this file was played with MPC-HC 1.7.10, LAV x64 video 0.68.1 using Enhance Video Renderer output, it managed to reach
29.97 fps steadily in CPU mode.

So, what may cause this difference of result between two apps ?

__________________
Win 10 x64 (10586.494) - Core i5-4440- iGPU HD 4600 (v.4463)

[kral2008]

I have download a small cut of "Men In Black 1997 WEB DL" 2160Pp, 90Mbps, HEVC 10 bit, 23.976FPS
Here is some result:

CPU Core i7 740QM
VGA: 425M Nvidia
Native DXVA2 Enabled:

Decoder: LAV Video Decoder
[1]
Decoder Device: -
Processor Device: 6CB69578-7617-4637-91E5-1C02DB810285
Frames: 746
Time: 43.808 sec
FPS: 17.029 [0-25] fps
CPU Usage: 92 [26-100] %
GPU Usage: 1 [0-7] %
Video Engine Usage: 0 [0-0] %
[2]
Decoder Device: -
Processor Device: 6CB69578-7617-4637-91E5-1C02DB810285
Frames: 746
Time: 42.552 sec
FPS: 17.532 [0-24] fps
CPU Usage: 95 [32-100] %
GPU Usage: 1 [0-7] %
Video Engine Usage: 0 [0-0] %
[3]
Decoder Device: -
Processor Device: 6CB69578-7617-4637-91E5-1C02DB810285
Frames: 746
Time: 42.673 sec
FPS: 17.482 [0-26] fps
CPU Usage: 95 [28-100] %
GPU Usage: 1 [0-3] %
Video Engine Usage: 0 [0-0] %

As you see I can't reach to 23.9 FPS and that mean I can't play smoothly.

One of my friend bought a laptop with Core i5 6200U processor equipped by the Intel Graphics 520 [which can decode HEVC and HEVC Main10] and 920M Geforce VGA. I decided to test my file on his device. But when I checked it with DXVA checker and Native DXVA2 option enabled it gives me only 15.5 FPS.

What happened there? Why using H/W accelerating of HD 520 graphics gives me less FPS than my old core i7 740QM [which has not any integrated GPU].
Does the power of the CPU effect the decoding performance?

I have old Core i7 920 in my desktop PC, I planned to sell it and buy a Xeon X5690 processor. What is your opinion about this processor for decoding 4K HEVC? Is it strong enough to decode high bitrates of HEVC encoded videos with acceptable FPS?

PS: I'll provide you some information about decoding performance using intel xeon 2600 V3 , 2600 V4 and 1200 V5 series if I can install a Geforce 210 GTX vga card in HP Proliant Dl380 G9 server. I have installed windows 10 in DL380 and I couldn't get any good info because its Matrox VGA can not give me proper output.

[sneaker_ger]

At least desktop Skylake only does CPU/GPU hybrid 10 bit HEVC decoding. Only 8 bit is using powerful fixed-function hardware. Kaby Lake will improve on that.
(I'm not 100% about the mobile chips. They can sometimes be ahead of the desktop ones in such things.)

I have old Core i7 920 in my desktop PC, I planned to sell it and buy a Xeon X5690 processor. What is your opinion about this processor for decoding 4K HEVC? Is it strong enough to decode high bitrates of HEVC encoded videos with acceptable FPS?
No, you can run into troubles with high bitrate/60fps/50fps material. Don't buy such outdated hardware.

[P.J]

Don't waste energy to decode via CPU, get a GTX 950 instead ;)
Intel Graphics 520 can't decode 10bit, only hybrid.

[kral2008]

At least desktop Skylake only does CPU/GPU hybrid 10 bit HEVC decoding. Only 8 bit is using powerful fixed-function hardware. Kaby Lake will improve on that.
(I'm not 100% about the mobile chips. They can sometimes be ahead of the desktop ones in such things.)


No, you can run into troubles with high bitrate/60fps/50fps material. Don't buy such outdated hardware.

@ P.J

But in the first post of this topic it was mentioned that Skylake 530 gives more FPS than 960.
Is there any discrepancy between 520 and 530?

[sneaker_ger]

The first post makes it abundantly clear it's talking about 8 bit decoding. 10 bit is a different matter. Skylake does not have the hardware for it, GTX 950 does. (New Pascal cards even have 12 bit support)

[Trevonn]

1) H.264 -1080p-108Mbps (ftp://helpedia.com/pub/multimedia/x264/testvideos/2012%20-%2001%20-%20QuickSync%20vs%20UVD%202.2%20vs%20VP4/9.Ducks.Take.Off.1080p30fpsRef5-108Mbps.mkv) - This video is decoded in two seconds that is why the results are a bit weird

Decode:
GTX 1060 - 249/257/249

Playback:
GTX 1060 - 249/259/249

2) H.264 - 2160p-370Mbps (ftp://helpedia.com/pub/multimedia/x264/testvideos/2160p%20samples/DucksTakeOff_2160p50.x264.CRF24.mkv)

Decode:
GTX 1060 - 93/99/104

Playback:
GTX 1060 - 93/99/104

3) H.265 8bit -1080p-80Mbps (http://www.imaging-resource.com/PRODS/samsung-nx1/FULLRES/yvid_SAM_0230-ProQuality.MP4)

Decode:
GTX 1060 - 171/172/176

Playback:
GTX 1060 - 172/174/177

4) H.265 8bit -1080p-80Mbps (http://imaging-resource.com/PRODS/samsung-nx1/FULLRES/yvid_SAM_0235-UHD_Sample1.MP4)

Decode:
GTX 1060 - 183/185/188

Playback:
GTX 1060 - 185/187/190

5) H.265 10bit - 2160p-140Mbps (http://jell.yfish.us/media/jellyfish-140-mbps-4k-uhd-hevc-10bit.mkv)

Decode:
GTX 1060 - 149/168/190

Playback:
GTX 1060 - 148/168/189

6) H.265 10bit -2160p-400Mbps (http://jell.yfish.us/media/jellyfish-400-mbps-4k-uhd-hevc-10bit.mkv)

Decode:
GTX 1060 - 60/76/99

Playback:
GTX 1060 - 61/77/100

7) LG_4K_View-the-Feeling.mp4 (http://demo-uhd3d.com/fiche.php?cat=uhd&id=101)

Decode:
GTX 1060 - 211/218/234

Playback:
GTX 1060 - 215/221/224

[P.J]

@ P.J

But in the first post of this topic it was mentioned that Skylake 530 gives more FPS than 960.
Is there any discrepancy between 520 and 530?

I think it was i7-6700K

[CruNcher]

I have sold my Reference RX 480. So I won't be able to provide benchmarks for it sorry. In the near future I will purchase a GTX 1060 as I feel it will better suit my encoding, decoding needs.

Makes no sense and is not logic @ all without proper testing to many are brainwashed currently by the Overall Power Output results im afraid.

Can't believe you had the chance and missed it, anyways thx for the GTX 1060 VPX results :D

Which card was it exactly the Reference ?

[Trevonn]

Makes no sense and is not logic @ all without proper testing to many are brainwashed currently by the Overall Power Output results im afraid.

Can't believe you had the chance and missed it :D

I didn't care about the power output issue it had no factor in my decision at all. AMD sucks at supporting their cards media capabilities it's that simple. Their media SDK was last updated when? 01/29/2015. NVIDIA? 7/18/2016.

Already now I can encode 10-bit 4:4:4 HEVC, decode VP9 up to 8K. The applications I use like staxrip, hybrid, OBS all have first class support for NVENC whereas with VCE, support is less than satisfactory or non-existent.

Finally based on the benchmarks I've seen the 1060 is way ahead of the 480 in terms of decode performance.

[huhn]

if you reached a decoding speed of way more than real time than it doesn't matter if a card can decode faster or not.

and because a video card has an encoder that doesn't mean it is any good...

[CruNcher]

jep especially if the encoder/decoder can operate more efficiently from the heart of the system instead of shuffling data inefficiently around the pci-e bus ;)

Though some would very well argue here being the VRAM closer to the GPU for Recording and keeping the overall latency low.

Thus also why the new AMD SSG concept is interesting, moving the whole data as close as possible to the GPU for decoding/transcoding avoiding most of the latency :)

@huhn
though it can very well be a indication of better overall efficiency doesn't have to but can be which overall can result in lower power requirements at realtime playback.

And Nvidias whole architecture since the inclusion of their Tegra R&D and major introduction with Maxwell is based on Power Efficiency and High Frequencies Serial Execution at it's very core.

[Trevonn]

if you reached a decoding speed of way more than real time than it doesn't matter if a card can decode faster or not.

and because a video card has an encoder that doesn't mean it is any good...

Fair points but for what I do it is simply the better card.

[CruNcher]

true to be told for Video Playback i would prefer a Nvidia Shield and not use my PC @ all ;)

Same most probably for game recording i would build myself a Tegra Based Recording Platform and not use the PC, directly ;)

Only when we come to stuff that can't be really executed from the outside and that would be as main applications for example video editing result acceleration from within a PC Workflow as with the SSG Model for example :D

Hard todo a 4k playback platform as efficient as the Nvdia Shield in 11W, without windows bloat ;)

Which basically reflects the idle requirement of the P8 State on the GTX 1060 very well ;)

[Trevonn]

true to be told for Video Playback i would prefer a Nvidia Shield and not use my PC @ all ;)

Same most probably for game recording i would build myself a Tegra Based Recording Platform and not use the PC ;)

Only when we come to stuff that can't be really executed from the outside and that would be as main applications video editing result acceleration from within a PC Worflow as with the SSG Model for example :D

Hard todo a playback platform as efficient as the Nvdia Shield in 11W ;)

Which basicly reflects the idle requirement of the P8 State on the GTX 1060 very well ;)

I Just did a quick run through of some gameplay clips I've recorded and looked at the power usage in the application HWINFO. The whole time it was in the P8 State consuming 9-11 Watts.

Extra: With the 4K 10-bit 400 Mb/s jellyfish clip it was around 25 watts.

[CruNcher]

Gameplay clips aren't really complex at all they only carry enough complexity @ constant 60 FPS

4K 60 FPS

H.264
H.265

are far more interesting ;)

Though i would be very well interested in a 4K 60 FPS Encoding result of Star Wars Battlefront Endors Level (Speeder Bike Chase) with the improved Encoder ;)

Also be advised the drivers you have in your signature have some severe Nvdiia acknowledged DPC issues with Pascal and its PWM which can cause problems for Encoding/Decoding related tasks.

https://forums.geforce.com/default/topic/941579/geforce-1000-series/gtx-1080-high-dpc-latency-and-stuttering/post/4944053/#4944053

there is a hotfix driver for this and a new driver coming though the results look rather odd

calling this good seems like blasphemy from that Manuel ;)

WDDM 2.0 gives them big headaches currently ;)

https://a70ad2d16996820e6285-3c315462976343d903d5b3a03b69072d.ssl.cf2.rackcdn.com/18f8f7b290aafc463e12e29f5c6688d5


PS: You didn't answered which card yet if you are a "Gamer" it's sure no Reference and it sounds like you are mainly using it for that Purpose ;)

[huhn]

Fair points but for what I do it is simply the better card.

i'm most likely going for the 1060 myself.

but decoding speed is not a point for the nvidia card in my book.

[Trevonn]

I am using the Zotac GTX 1060 Mini. I also have the high DPC latency but it does not seem to affect anything from what I can see audio/visually

[Yups]

if you reached a decoding speed of way more than real time than it doesn't matter if a card can decode faster or not.


Polaris might struggle to hold 60+ fps in 4k with very high bitrate content. Based on this test: https://www.computerbase.de/2016-06/radeon-rx-480-test/7/#diagramm-videos-abspielen-h265-codec-lav-filters

Pascal is factor 2.7 faster than the RX 480 there and also slower than the shader based solutions from AMDs prior generation. And they obviously don't use a demanding sample. That would mean the H.265 10bit -2160p-400Mbps sample is too demanding when even Pascal can barely hold 60+ fps over the entire video (#1226). RX 480 might struggle with the H.265 10bit - 2160p-140Mbps sample assuming there is a factor 2.7 gap in this sample as well. Although we need more high bitrate 4k decoding tests to see how it copes with demanding 4k videos.

[JohnLai]

I didn't care about the power output issue it had no factor in my decision at all. AMD sucks at supporting their cards media capabilities it's that simple. Their media SDK was last updated when? 01/29/2015. NVIDIA? 7/18/2016.

Already now I can encode 10-bit 4:4:4 HEVC, decode VP9 up to 8K. The applications I use like staxrip, hybrid, OBS all have first class support for NVENC whereas with VCE, support is less than satisfactory or non-existent.

Finally based on the benchmarks I've seen the 1060 is way ahead of the 480 in terms of decode performance.

Nice, so, I have question on NVENC for you.
Using staxrip default CQP and lookahead 32, what is the percentage difference between HEVC 8bit and 10bit file size? (Make sure ignore the audio size) If possible, average FPS for 8bit and 10bit encode result too (and resolution of encoded video)?

[easyfab]

Nice, so, I have question on NVENC for you.
Using staxrip default CQP and lookahead 32, what is the percentage difference between HEVC 8bit and 10bit file size? (Make sure ignore the audio size) If possible, average FPS for 8bit and 10bit encode result too (and resolution of encoded video)?

+1 I'm also interested in encoding performance.
If possible some samples and/or PNSR/SSIM.
You have some samples here : https://media.xiph.org/video/derf/
park_joy would be a good choice

[huhn]

Polaris might struggle to hold 60+ fps in 4k with very high bitrate content. Based on this test: https://www.computerbase.de/2016-06/radeon-rx-480-test/7/#diagramm-videos-abspielen-h265-codec-lav-filters

Pascal is factor 2.7 faster than the RX 480 there and also slower than the shader based solutions from AMDs prior generation. And they obviously don't use a demanding sample. That would mean the H.265 10bit -2160p-400Mbps sample is too demanding when even Pascal can barely hold 60+ fps over the entire video (#1226). RX 480 might struggle with the H.265 10bit - 2160p-140Mbps sample assuming there is a factor 2.7 gap in this sample as well. Although we need more high bitrate 4k decoding tests to see how it copes with demanding 4k videos.

400 mbit has nothing to do with real world performance.

edit: my 960 plays it fine too...
hardware decoder have a "max" speed and they only loose speed with really high mbit. you can assume bit rate about the 100 mbit are no problem for 2016 GPU of cause a test would be better.

BTW. the quality of that 400 mbit file is just terrible. it has a lot of banding it is super unsharp and i can see something like macro blocks...
just a bad source pumped with way to many mbit. even a hardware encoder would do a way better job at 1/4 the mbit if the source was fine.

[CruNcher]

Polaris might struggle to hold 60+ fps in 4k with very high bitrate content. Based on this test: https://www.computerbase.de/2016-06/radeon-rx-480-test/7/#diagramm-videos-abspielen-h265-codec-lav-filters

Pascal is factor 2.7 faster than the RX 480 there and also slower than the shader based solutions from AMDs prior generation. And they obviously don't use a demanding sample. That would mean the H.265 10bit -2160p-400Mbps sample is too demanding when even Pascal can barely hold 60+ fps over the entire video (#1226). RX 480 might struggle with the H.265 10bit - 2160p-140Mbps sample assuming there is a factor 2.7 gap in this sample as well. Although we need more high bitrate 4k decoding tests to see how it copes with demanding 4k videos.

Assumptions don't bring us far we need Real results ;)

though i would agree 400 mbps surely not but that is also a extreme target no avg user needs to decode @ all ;)

based on the Computerbase test i would say most current consumer content will be fine as well.

Though i wouldn't expect todo multistream decoding on UVD as high as possible as with Nvidia VPX ;)

Normal PIP usage but most probably that's it resources left on VPX are massiv no doubt :)

http://i1.sendpic.org/i/5V/5Va4r9EajGDMMOYvX8z7IcczjoU.jpg


Now the question is how the improved 2 pass encoder does :D

And who knows maybe AMD could improve it's decoding Performance in an entire DX 12 DXVA2 Render scenario over time.

We need raw frame decoding results to no render surface @ all.

[Yups]

400 mbit has nothing to do with real world performance.



Nobody denied this.


Assumptions don't bring us far we need Real results ;)



For both sides. Huhn assumed that Polaris is more than fast enough which has not been proven. We only have a Computerbase and Notebookcheck decoding test where Polaris does much much worse for H265. And VP9 decoding is still not supported by the driver.

GTX 1080:

The new video decoder was also convincing in our review. Both 2K H.264 as well as 4K H.264 playback was smooth at high frame rates and the CPU load was low (4K 100 Mbps 138 fps, 4% CPU). HEVC was no problem, either (1080p at 565 fps, 4% CPU and 4K 267 fps at 1% CPU).

RX 480:

The new video engine can actually convince in the test. A 4K video (H.264, 100 Mbps) was played at 84 fps and with 3 % CPU load in the DVXAChecker Playback benchmark. In comparison: The GTX 1080 managed 138 fps and 4 % CPU load.

The new H.265 decoder handles a 4K 10-bit video at 72 fps and 2 % CPU load.

http://www.notebookcheck.net/Nvidia-GeForce-GTX-1080-Desktop-Review-Pascal-has-arrived.165500.0.html
http://www.notebookcheck.net/AMD-Radeon-RX-480-Review-The-fastest-Polaris-desktop-card-at-launch.168250.0.html


Factor 3.7 in the Notebookcheck test, most likely a higher bitrate video, maybe 100 Mbit like H264? The gap seems to grow with higher bitrate content.

[CruNcher]

So all in all it just doesn't certifies the DRM Requirements Nvidia can better handle this with their Falcon Core :)

So we don't know yet how much additional overhead the encrypted transmission will cost AMD though the Blu-Ray test seems to give some insight into this.

Also the AMD DXVA Checker list of uknown GUIDS seems pretty damn long compared to Nvidia :D

Polaris 10

http://www.notebookcheck.net/fileadmin/Notebooks/Sonstiges/AMD/RX_400/dxvac.png

15 uknowns

GP104

http://www.notebookcheck.net/fileadmin/Notebooks/Sonstiges/Grafikkarten/NVidia/GTX1000/dxvachecker.png

10 uknowns


@Trevonn

Could you post a updated shoot from your GTX 1060 GP106 and current driver release :)

[Trevonn]

http://i.imgur.com/vsV8VZv.jpg

That GP104 shot looks to be from an older DXVA Checker

[P.J]

400mbps is useless for H.265 but Prores

[CruNcher]

yes but in Trevonns updated shoot you can pretty well see that the Performance overhead is most likely for Simple 8K Decoding goals and Multi Decoding 2 Simple 4K Streams.

Nothing of this has @ all relevance for the target audience not even in the distant future ;)

Though we most probably not only see Nvidias VPX Core advancements but also their Internal Framebuffer Compression Advancements that allow to reach this Goal, which is pretty interesting the topic of their DCE R&D and how much they are away from AMD here saving bandwith cutting costs and still be more efficient or at least as efficient with a cut down Hardware Design on the Memory side of things i find that a very impressive achievement of Nvidias R&D :D

Its like they have a ultra efficient CABAC running internally over everything and in Pascal they started to improve on that further with 3D Data :D

Now imagine Nvidia is a AOM Member, so if they might have a lucky day sometime and think "hey why not contributing our research for the greater good" that would be quiet cool ;)

Most Probably Nvidia started so many Neural Network Algorithm improvement stuff by now that their own GPU is improving it's Design Partly already so to speak itself ;)

[Trevonn]

Now I just need to win the lottery so I can buy an 8K display

[Trevonn]

Nice, so, I have question on NVENC for you.
Using staxrip default CQP and lookahead 32, what is the percentage difference between HEVC 8bit and 10bit file size? (Make sure ignore the audio size) If possible, average FPS for 8bit and 10bit encode result too (and resolution of encoded video)?

Percentage difference: 5.07%

I hope this has all the information you need

http://i.imgur.com/odoq2IL.jpg

[huhn]

no b frames AGAIN...

[JohnLai]

Percentage difference: 5.07%

I hope this has all the information you need



:thanks:
6% decrease in encoding performance and 5.07% file size saving. Reasonable tradeoff.

[kral2008]

Tested by i5 6500 Desktop cpu (intel HD 530 graphics)

Astra.SES.Demo.HEVC.ts
Size: ~ 350 MB
Resolution : 3840x2160
Overall bit rate : ~ 24 Mb/s
Bit depth : 10 bits
Frame rate : 50 fps

Codec LAV filter : DXVA2 Native
Test intervals: 3 times
FPS: 28 - 39
CPU usage: 39 - 47 %

I wonder why still CPU usage is under 60% and FPS is lower than 50.
In my opinion it should load more CPU to give higher FPS than 39 frames per second.

All other acceleration technologies over loads the CPU and delivers unacceptable results.

Edit:
I couldn't install any VGA to HP ProLiant DL380 G9 or others. Only specific cards can be compatible with these servers. This means I can test xeon v4 series. Maybe I can benchmark only the 1225 V5 processor in near future.

[nevcairiel]

I wonder why still CPU usage is under 60% and FPS is lower than 50.
In my opinion it should load more CPU to give higher FPS than 39 frames per second

Not everything multi-threads fully to use all of your cores - hybrid hardware decoders are one of those things.

[CruNcher]

Yeah and it takes time ;)

See for example CoreAVC vs FFMPeg it took some time before they reached the Assembly Optimization state of Picards CPU Mastery ;)

i wonder if Mainconcept is still running behind the deblocking performance today :D

ANd still Intels result is nice compare it to the Nvidia Hybdrid H.265 8 bit Decoder and you would be shocked how bad that is and especial how good ffmpeg is on 4 CPU cores vs it by now :D
y
Though to be fair Nvidia doesn't really invest to much time into the hybdrid decoder Performance just always low complexity goals and moves on fast then to the DSP Core which we saw first in Tegra X1 and later the GTX 950 and now every Pascal could theoreticaly do 8K low complexity and also Intel always has the advantage of faster data transfer with their hybrid ;)

You can guess which is which no words or even numbers really needed and yeah you don't want to see that spikes end (even can see it at all) @ 60 fps Broadcast Target ;)

http://i1.sendpic.org/t/2y/2yvEcUL2Vt5Eb7rGfivinfdNX7x.jpg (http://sendpic.org/view/1/i/55UW30l8Oq3DEbmBUu7m3LKXAIl.png)

http://i1.sendpic.org/t/uM/uMFXoYafHHD0qz9ef2sHYqpzwIm.jpg (http://sendpic.org/view/1/i/Agsiq4mgBUQeEgEyF5ds6bH0lED.png)

CUVID looks a little better but overall even the shaders don't really seem to be utilized that much and the higher clock alone doesn't do that much more magic @ all so totaly starving :(

http://i1.sendpic.org/t/8D/8DZ8fYtygrAGCM4NFZcjm3MWtek.jpg (http://sendpic.org/view/1/i/dsRVO7fZEMzfYW8cSEK0Ld9s9k2.png)

[alfianadjeh]

anyone have been tested RX460 for hardware decoding HEVC 4k 60fps ?
I really confusing, buying GTX950 or Rx460 2gb.

[huhn]

if you want to display at UHD you should use a card with 4 GB Vram.

2GB is at it's limited when used with UHD. i'm just telling you this because i made the huge mistake and got a 2GB 960 and not the 4GB version and i got a lot of problem thanks to the 2GB Vram.

[nsnhd]

if you want to display at UHD you should use a card with 4 GB Vram.

2GB is at it's limited when used with UHD. i'm just telling you this because i made the huge mistake and got a 2GB 960 and not the 4GB version and i got a lot of problem thanks to the 2GB Vram.

That means iGPU with max 1GB shared Vram can't display 4K ? Can you explain a little more about your problems.

[clsid]

Windows itself already uses a few hundred MB of GPU memory.

If you use software decoding then only a few frames are buffered in GPU memory. Then 1 GB might be enough. If you use hardware acceleration then much more frames are being buffered, since there can be up to 16 reference frames needed in the decoding process. As an example, here a 4k 10bit video uses 1100MB on top of 200 MB used by Windows 7. But most CPUs are not fast enough for 4k HEVC, so hwaccel is usually needed.

If you use madVR then you need even more memory since that by default uses larger queues/buffers than EVR. Then you can easily go over 2GB.

[CruNcher]

The only time where Nvidias Hybrid Decoder on the GTX 970 so far is more efficient then the 4 Sandy Bridge CPU Cores with ffmpeg is with DivX Hevc 4K Stream complexity @ 8000 Kbps , there it actually can save power :)

Some glitches still persist though but much better efficiency wise especialy when you switch from DXVA Native to CUVID there the Driver Render overhead seems also lower compared to DXVA Native on the CPU with their Hybrid but surely even better manageable with DX 12 and Vulkan in Future Media Player Render Pipelines CUVID already shows no glitches in this example testcase :)

Lower HEVC Complexity Bitstream Playback Stability Test (Nvidia Hybdrid H.265 Decoder R&D)

Nvidia Hybrid (DXVA)

http://i1.sendpic.org/t/7K/7KC18t3CqIVWOShmf9DEYgrGzC3.jpg (http://sendpic.org/view/1/i/hsPUTbmSf6PoDxCkoougEXdw9K3.png)

Nvidia Hybrid (CUVID)

http://i1.sendpic.org/t/jU/jUfcD9Wm5mwcQ7rZHsmm4Ud57aX.jpg (http://sendpic.org/view/1/i/dNNyDlfTjqP4kFNyKj8Hru1ImPo.png)

FFMPEG

http://i1.sendpic.org/t/vO/vOVpYJgoTdDluYmdytOYMcDEuwS.jpg (http://sendpic.org/view/1/i/947B3s8goI0K6I2bxL919BEcGsh.png)

I guess pretty much this was also Nvidias Primary Development goal for the Hybrid Decoder Web Streaming support before the Full Hardware Core was ready ;)

and now they go up to 8K with their IP Core in Tegra and Discrete ;)


And AMD doesn't really seem fully yet to be up to it but they made advances none the less

[P.J]

if you want to display at UHD you should use a card with 4 GB Vram.

2GB is at it's limited when used with UHD. i'm just telling you this because i made the huge mistake and got a 2GB 960 and not the 4GB version and i got a lot of problem thanks to the 2GB Vram.

Would you explain more? you mean madVR?

[chros]

But most CPUs are not fast enough for 4k HEVC, so hwaccel is usually needed.
Yep, my i73630QM can't keep up with >=25fps 4k 10bit hevc (~40Mbps) material (using latest stable LAV).

[CruNcher]

On Mobile Hardware it makes almost always sense to use the fixed function parts instead of hybrid or Software but on edge cases desktop it can be quiet different which path to go (and which in the end is the most beneficial for the users use case) based on the whole system configuration especially in hybrid cases and you need todo smart dynamic decisions based on the system configuration data ;)

[kral2008]

Intel Xeon 1225 V5 :sly: on HP ML10 Gen9 Server and Windows 10 Ent x64

Still stupid results :eek:

Astra.SES.Demo.HEVC.ts
ID : 255 (0xFF)
Menu ID : 10202 (0x27DA)
Format : HEVC
Format/Info : High Efficiency Video Coding
Format profile : Main 10@L5.1@Main
Codec ID : 36
Duration : 2 min
Width : 3 840 pixels
Height : 2 160 pixels
Display aspect ratio : 16:9
Frame rate : 50.000 FPS
Color space : YUV
Chroma subsampling : 4:2:0
Bit depth : 10 bits



Decoder: LAV Video Decoder with DXVA2 Native Enabled:
Decoder Device: HEVC_VLD_Main10
Processor Device: BF752EF6-8CC4-457A-BE1B-08BD1CAEEE9F
Frames: 6400
Time: 184.460 sec
FPS: 34.696 [28-42] fps
CPU Usage: 36 [22-45] %
GPU Usage: -

Decoder: LAV Video Decoder
Decoder Device: -
Processor Device: BF752EF6-8CC4-457A-BE1B-08BD1CAEEE9F
Frames: 6400
Time: 214.116 sec
FPS: 29.890 [27-33] fps
CPU Usage: 89 [13-98] %
GPU Usage: -

Decoder: Microsoft H265 Video Decoder MFT
Decoder Device: HEVC_VLD_Main10
Processor Device: BF752EF6-8CC4-457A-BE1B-08BD1CAEEE9F
Frames: 6397
Time: 184.580 sec
FPS: 34.657 [28-43] fps
CPU Usage: 35 [23-47] %
GPU Usage: -

[wanezhiling]

Xeon 1225 V5 belongs to Skylake family which doesn't support HEVC 10-bit decoding at all

[nsnhd]

Intel Xeon 1225 V5 :sly: on HP ML10 Gen9 Server and Windows 10 Ent x64

Still stupid results :eek:

Astra.SES.Demo.HEVC.ts
ID : 255 (0xFF)
Menu ID : 10202 (0x27DA)
Format : HEVC
Format/Info : High Efficiency Video Coding
Format profile : Main 10@L5.1@Main
Codec ID : 36
Duration : 2 min
Width : 3 840 pixels
Height : 2 160 pixels
Display aspect ratio : 16:9
Frame rate : 50.000 FPS
Color space : YUV
Chroma subsampling : 4:2:0
Bit depth : 10 bits



Decoder: LAV Video Decoder with DXVA2 Native Enabled:
Decoder Device: HEVC_VLD_Main10
Processor Device: BF752EF6-8CC4-457A-BE1B-08BD1CAEEE9F
Frames: 6400
Time: 184.460 sec
FPS: 34.696 [28-42] fps
CPU Usage: 36 [22-45] %
GPU Usage: -

Decoder: LAV Video Decoder
Decoder Device: -
Processor Device: BF752EF6-8CC4-457A-BE1B-08BD1CAEEE9F
Frames: 6400
Time: 214.116 sec
FPS: 29.890 [27-33] fps
CPU Usage: 89 [13-98] %
GPU Usage: -

Decoder: Microsoft H265 Video Decoder MFT
Decoder Device: HEVC_VLD_Main10
Processor Device: BF752EF6-8CC4-457A-BE1B-08BD1CAEEE9F
Frames: 6397
Time: 184.580 sec
FPS: 34.657 [28-43] fps
CPU Usage: 35 [23-47] %
GPU Usage: -

Try CPU decode with MPC-HC Lav 0.68.1 in EVR mode, you may get 50 fps

[kral2008]

Xeon 1225 V5 belongs to Skylake family which doesn't support HEVC 10-bit decoding at all

It supports HEVC_VLD_Main10 as other Skylake family members, but only Hybrid I think.

[huhn]

That means iGPU with max 1GB shared Vram can't display 4K ? Can you explain a little more about your problems.
with widnows you are easily using 300 mb Vram without doing anything.

if i remember correctly 1.4 GB Vram usages was the lowest i have seen with hardware/software UHD decoding on UHD screen and that with everything on min. and other programs can and will use Vram too.

Would you explain more? you mean madVR?

i have to use a GPU buffer of 4 with madVR. madVR doesn't really use more than EVR with default settings. but i'm still pretty much maxed out.

i can re test this maybe next week sorry.

[chros]

with widnows you are easily using 300 mb Vram without doing anything.

if i remember correctly 1.4 GB Vram usages was the lowest i have seen with hardware/software UHD decoding on UHD screen and that with everything on min. and other programs can and will use Vram too.

i have to use a GPU buffer of 4 with madVR. madVR doesn't really use more than EVR with default settings. but i'm still pretty much maxed out.
Thanks for the info, huhn!

[wanezhiling]

It supports HEVC_VLD_Main10 as other Skylake family members, but only Hybrid I think.

'Hybrid' means null :)

[CruNcher]

with widnows you are easily using 300 mb Vram without doing anything.

if i remember correctly 1.4 GB Vram usages was the lowest i have seen with hardware/software UHD decoding on UHD screen and that with everything on min. and other programs can and will use Vram too.



i have to use a GPU buffer of 4 with madVR. madVR doesn't really use more than EVR with default settings. but i'm still pretty much maxed out.

i can re test this maybe next week sorry.

Yes this part of the VRAM is for Rendering the Desktop but you hardly realize this loss and you get a lot of benefits from it and the (LDDM) WDDM DWM Render Model since it was introduced with Longhorn back then in efforts to gain Apples GPU Desktop Efficiency Level.

Though the overall Memory usage up today is dependent on more factors like if you use Variable Refresh or a fixed Refresh Rate 10bit 8 bit hdr or not since Windows 8.1 and 10 there is no real differentiation anymore and it's a fully fixed part integrated and in Windows 10 it drives the whole UWP experience between Windows Devices :)

No 3rd Party Media Player yet uses it fully efficient on Windows 10 and it's many Display changes yet but also drivers are still in a changing stability state ;)

There is no 3rd Party Player yet even existing with a full DX 12 Render Pipeline not even a simple ported one ;)

[P.J]

Don't waste time on 'Hybrid'

with widnows you are easily using 300 mb Vram without doing anything.

if i remember correctly 1.4 GB Vram usages was the lowest i have seen with hardware/software UHD decoding on UHD screen and that with everything on min. and other programs can and will use Vram too.



i have to use a GPU buffer of 4 with madVR. madVR doesn't really use more than EVR with default settings. but i'm still pretty much maxed out.

i can re test this maybe next week sorry.

Then it happens only with madVR not EVR :)

[CruNcher]

Don't waste time on 'Hybrid'



Then it happens only with madVR not EVR :)

No surprise EVR is tuned for System Resource efficiency ontop of DWM not Extreme Quality ;)

[huhn]

UHD playback Vram usages using a UHD source HDR processing is disabled in madVR because EVR doesn't know what HDR is:
lavfilter 0.68.1
mpc-hc 1.7.10.252
clip sony 4K HDR Camp
gpu 960 with a boost clock of 1430 mhz

copyback
madVR GPU queue at 4 1547-1750 mb
EVR CP queue 4 1922-2000 mb

native
madVR GPU queue at 4 present queue 3 1600 mb BLACK SCREEN in full screen.
EVR CP queue 4 1780 mb

a difference of 100 of mb each time are pretty normal.
get more than 2 gb that's all i have to say to this.

the default GPU queue for madVR is 8 which is unusable for UHD with 2 GB Vram. the default GPU queue for EVR is 5 or 6.

GPU usages was ~70% with both EVR and madVR. madVR was at default except queue and FSE mode.

[chros]

get more than 2 gb that's all i have to say to this.
Thanks for the detailed tests!

[leonccyiu]

http://www.notebookcheck.com/Kaby-Lake-Core-i7-7500U-im-Test-Skylake-auf-Steroiden.172422.0.html

Notebookcheck has a DXVA checker screenshot from the Kaby Lake HD650 Graphics

HEVC Main 10 as well as VP9 Profile 0 and 10 bit are now decoded up to 8k.

http://www.notebookcheck.com/fileadmin/Notebooks/MSI/CX72-7QL/dxva_kaby.png

[NikosD]

Nice info for Kaby from Anandtech.

It's all Media update the Kabylake update.
The gain of CPU is the gain of MHz due to the advanced 14+ nm process.
The GPU is the same Gen 9.

But the media capabilities are impressive.

http://www.anandtech.com/print/10610/intel-announces-7th-gen-kaby-lake-14nm-plus-six-notebook-skus-desktop-coming-in-january

[chros]

But the media capabilities are impressive.
Indeed! :)

[NikosD]

The fastest Kabylake processor Core i7 7700K will try to stand against the 8 core/ 16 thread ZEN in the early 2017.

It has no chance at all, it will be eaten for breakfast by ZEN in all apps besides the heavily optimized AVX/ AVX2 ones.

Of course Kabylake has an iGPU inside where ZEN 8 core/ 16 thread will be a CPU only SoC.

But that's a problem of Intel to release 8 core CPUs without iGPUs for mainstream Core i7 and not for the High-End Desktop (HEDT) terribly expensive systems only.

[huhn]

the marketing from intel is terrible in general.

they have mainstream processors without an iGPU but they usually cost extra making them pointless.

so there is absolutely no reason to buy a mainstream CPU without iGPU. you can use the hardware decoder or the iGPU just in case you need it.

AMD ZEN can spare a lot of silicon making them a lot cheaper in production and the TDP will look better on paper.

[NikosD]

Didn't understand a word of what you've written, but OK...Maybe others do.

[Stereodude]

The fastest Kabylake processor Core i7 7700K will try to stand against the 8 core/ 16 thread ZEN in the early 2017.

It has no chance at all, it will be eaten for breakfast by ZEN in all apps besides the heavily optimized AVX/ AVX2 ones.

Of course Kabylake has an iGPU inside where ZEN 8 core/ 16 thread will be a CPU only SoC.

But that's a problem of Intel to release 8 core CPUs without iGPUs for mainstream Core i7 and not for the High-End Desktop (HEDT) terribly expensive systems only.
I know I always enjoy reading your off topic delusional AMD fanboy posts. Please keep it up.

[NikosD]

Just facts, as always.

[Yups]

Zen will lose in 99% of games because it cannot compete with the IPC+clockspeed of Kabylake. Just one example where you are wrong.

[NikosD]

I'm not so sure about games like you.

Because new games using Vulkan and DX12 tend to use more and more cores and threads, it will be an interesting fight with games.

In office applications, production applications, rendering, general purpose benchmarks and Cinebench it will vanish Kabylake for sure.

[Yups]

Doesn't matter really. Over 99% of the games will still use DX11 or even DX9 by next year. And for non gaming apps there is lots of stuff where you don't see improvements over 4 cores. It would be very very sad if a 8 core CPU can't demolish a 4 core CPU in Cinebench. This isn't a fanboy thread for AMD CPUs, you may search for another place to discuss this further.

[P.J]

No more off topic, please @@

http://www.notebookcheck.com/Kaby-Lake-Core-i7-7500U-im-Test-Skylake-auf-Steroiden.172422.0.html

Notebookcheck has a DXVA checker screenshot from the Kaby Lake HD650 Graphics

HEVC Main 10 as well as VP9 Profile 0 and 10 bit are now decoded up to 8k.

http://www.notebookcheck.com/fileadmin/Notebooks/MSI/CX72-7QL/dxva_kaby.png

Saved for Kaby Lake :thanks:

[NikosD]

Polaris DXVA Checker
https://s9.postimg.org/p38ccn03j/dxvachecker_msi_rx470.jpg

[CruNcher]

Most Probably behind one of the uknowns is the VP9 Decoder some time now past since their announcement that it's going to be enabled and yet nothing happened officially ;)

While Intel as well as Nvidia are running, the disadvantage of low resources to work with you could guess.

Overall Polaris DSP really looks like just being in the R&D state of the Previous VPX (Maxwell) and Intel far far ahead of both again (Tegra,Maxwell,Pascal,Polaris).

I really wonder how much sense it really makes to keep that fixed function unit of Nvidias and AMds in a Intel System it makes no real sense efficiency wise especially in the upcoming Kaby Lake Systems it's just pure unneeded overhead on the discrete side and the valuable Wafer space could be used for more 3d and compute logic or to make the grid even denser.


I guess AMD has planned to release/enable VP9 Decoding support officially with the upcoming Polaris 10 Notebook Reviews and supplied Drivers.

[NikosD]

Yes, but is the late Polaris VP9 decoder a fixed-fuction decoder like Maxwell's with 960 card which Nvidia enabled some time later than the initial release of 960 or a hybrid decoder ?

Of course as long as Chrome official releases don't support VP9 HW decoder, all the other uses of VP9 are insignificant.

Only Edge supports VP9 HW decoder.

But the difference between a hybrid and a HW decoder is not insignificant.

[Trevonn]

I really wonder how much sense it really makes to keep that fixed function unit of Nvidias and AMds in a Intel System it makes no real sense efficiency wise especially in the upcoming Kaby Lake Systems it's just pure unneeded overhead on the discrete side and the valuable Wafer space could be used for more 3d and compute logic or to make the grid even denser.

It makes complete sense considering Intel has only just delivered a Hardware decoder for HEVC 10-bit, VP9 when you could have had one from NVIDIA all the way back in January 2015 (VP9 Enabled in December 2015). Plus nobody upgrades their CPU that regularly unlike GPUs because it's a waste of money.

Let's not leave yet another industry to stagnating Intel

[rbej]

I really like all of you guys that after you tell me your off-topic story you call me for writing off-topic.

But you really overcome everyone here by telling me what to say in my own thread and to search somewhere else to write.

It couldn't be more funny.

The truth is that AMD is going to kick for good some @sses and all the fan boys of Nvidia and Intel are desperate and nervous.

Vega will just destroy Pascal for sure, it's another class it will show no mercy.

The same thing is gonna happen to Kabylake.

All of the AAA titles are already DX12 and Vulkan or they are going to get a patch soon.

The rest of the games are simply not interesting.

Most of the apps nowadays are getting a serious boost going hyperthteading, meaning getting 8 threads for 4 real cores.

Just imagine how fast will be 8 real ZEN cores with 16 threads.

Poor Intel...

Poor Nvidia...

2017 will be the year of AMD.

Case closed.

Very funny joke . Today is 1 April??.

You must very "love" AMD, if you realy believe in this.

[ashlar42]

What would currently be the most cost effective solution to build a Windows 10 based HTPC, with 4K and HEVC decoding capabilities? I ask here since you seem to be trying all combinations. Thank you.

[huhn]

the cheapest card is the rx 460. for UHD presentation you should get a GPU with at least 3GB of vram.

[JohnLai]

the cheapest card is the rx 460. for UHD presentation you should get a GPU with at least 3GB of vram.

I disagree with rx 460....the better card would be nvidia gtx 1050 (4gb variant, not yet released) or 1060 3gb.

Don't forget Pascal series support HEVC 8,10 and 12 bits decoding in hardware. Beside, it has been confirmed through nvidia Video Codec SDK that pascal also support VP9 Profile 0 decoding in hardware too.

AMD Polaris only supports 8 and 10 bit HEVC....and there is conflicting info about vp9 support whether it is hybrid or full fixed hardware mode.

EDIT: Oh great, assuming if VCE used in Bristol Ridge is the same as Polaris, then according to http://techreport.com/review/30619/amd-unwraps-its-seventh-generation-desktop-apus-and-am4-platform , VP9 support is limited to 1080p content. Meanwhile Pascal could support up to 4k and 8 k VP9 decoding....., what a bummer....

[huhn]

you should read the question.

and the RC 460 is not the cheapest card that can decode HEVC?

[JohnLai]

you should read the question.

and the RC 460 is not the cheapest card that can decode HEVC?

:p Cheapest but not the best value for video decoding functionality.

[huhn]

so card for double the money that can theoretically decode 12 bit which isn't even part of the DXVA spec is better?

[P.J]

:p Cheapest but not the best value for video decoding functionality.
Maybe good for 4k h.265 @60fps while 960/950 are better.

I would get 1050/1060 for video or waiting for Kaby Lake.

[JohnLai]

so card for double the money that can theoretically decode 12 bit which isn't even part of the DXVA spec is better?

What make you think Microsoft won't update the spec?

Even if Microsoft doesn't update the spec, Nvidia might do it via their CUVID/NVDECODE API. (Seem nvidia does update its cuda decoder)

[huhn]

there is no commercial 12 bit content. even if they are adding it you are unlikely to really need it any time soon.

the last CUVID "update" i know of was just a repack. CUVID has some serious issues on windows 10 on top.

they have a lot todo maybe adding 10 bit is a start...
the problem that only nvidia can do it in theory will slow this down too. commercial product will stay away from it for better compatibility.

[Paul Tronc]

native
madVR GPU queue at 4 present queue 3 1600 mb BLACK SCREEN in full screen.
EVR CP queue 4 1780 mb

a difference of 100 of mb each time are pretty normal.
get more than 2 gb that's all i have to say to this.

the default GPU queue for madVR is 8 which is unusable for UHD with 2 GB Vram. the default GPU queue for EVR is 5 or 6.

GPU usages was ~70% with both EVR and madVR. madVR was at default except queue and FSE mode.

the cheapest card is the rx 460. for UHD presentation you should get a GPU with at least 3GB of vram.

Hello, I just have purchased a 960GTX 2Gb in order to get full hardware 4K HEVC decoding in madvr. Are you saying that it won't work because of lack of memory?

[huhn]

if you are going to use it to display UHD AT UHD resolution.

you could run out of Vram.

in term of madVR you are relatively lucky. you can lower the GPU queue. but deinterlanced content can't be deinterlanced by madVR like this and other issues can show up.

with default madVR settings it will not work.
i'm really regretting buying the 2G version of the 960.

you are totally fine on a 1080p screen (i planned this card for a 1080p screen). just to make that clear.

[Paul Tronc]

you are totally fine on a 1080p screen (i planned this card for a 1080p screen). just to make that clear.

I'm relieved to read this. Actually my brand new projector is 1080p, so I want to downscale UHD content. But it still have to figure out how the video decoding chain works, because I'm using SVP (frc tool triggered via avisynth) together with madvr. I can't say whether SVP is working on 4K resolution or 1080p on my current setup, when I play 4k on my 1080p diffuser. I'd like to get SVP working smoothly on UHD content, together with madvr. I just received the 960/2Gb, paid ~100 bucks (used). Still have to clean all radeon software and switch to this Nvidia card before I can tell how it behaves.

[Paul Tronc]

All right,

My first tests using GTX 960 + HEVC + Madvr 32 + Lav Filter 32 + 1080p diffuser are very positive. That's simple, I could'nt find the decoding limit of my setup. I launched the http://jell.yfish.us/ 400mbps 4k uhd hevc 10bit video, I get a perfect rendering using my standard Madvr settings. The only limitation is the downscaling algorithm, it seems like I have to stay on bicubic. Any other algorithm will cause frame dropping.

I suspect the jellyfish samples to be not fully representative of a high bitrate video, it looks too easy to be true. I tried another 2016p 60fps 50Mbps video, no problem : http://demo-uhd3d.com/fiche.php?cat=uhd&id=90

Next steps, I want to try :

- DXVA checker benchmark
- same tests on my 1440p display
- adding frc (SVP) to get 60fps (I don't understand why SVP/avisynth don't see HEVC files, there must be a wrong setting somewhere)

If you have another suggestion for a stronger stress test...

[v0lt]

If you have another suggestion for a stronger stress test...
Netflix_TunnelFlag_4096x2160_60fps_x265_8bit_700Mbit.mkv (https://yadi.sk/i/swMdlJHMuB6yR)

[huhn]

close i get ~59 FPS in decode mode.

[Paul Tronc]

Netflix_TunnelFlag is not smooth on my 1440p display, I'll check later on my 1080p one. I tried with MPC HC 32, MPC-BE 64, PotPLayer 64, each time with Madvr.

I'm currently encoding a 1080p BluRay to HEVC at 190fps, CPU usage 7% , GPU usage 6%, GPU temp 50°C with the fans off. Pretty impressive.

[CruNcher]

It seems that Nvidias Hybrid H.265 Decoder is more efficient with CUVID Decoding especially on intra frames DXVA Native has the tendency to cause latency spikes (no matter which decoder used) on edge decoding cases i guess it has todo with DXVA Natives Frequency Scaling issues in conjunction with Nvidias Boost Power Management System.

I frames seem to heavy fluctuate in my tests with Nvidias Hybrid Decoder GM204 and a DXVA frequency of around 759 MHz @ 0,862v Real 4K 23.976 Fps

CUVID doesn't show these Latency I frame fluctuations @ 1126 MHz @ 1.025v


NVAPI Hook readouts (MSI Afterburner hooked NVAPI results gained after test)

Though Unwinders code is super nice it itself creates almost 0 fluctuations when polling of Nvapi is kept in sane ranges ;)

Im mainly using as seen MPC-BE Sync Graph which is awesome fine grained for a almost Realtime Graph especially in Analyze Mode 2 when CPU overhead is way lower then with the Full OSD and additional timer polling (NVAPI ?) eliminated :)

@V0lt

Please add a Sync Graph only Analyze Mode without any additional OSD timer running maybe it can further lower the Overhead and it's caused EVR Latency Fluctuations that become visible in the Graph itself increasing the Audio and Video distance from the middle with DWMs tripple buffering :)
same happens with Energy Saving Modes the lines move away when timer precision is lowered and CPU overhead gets higher with lower Frequency :)


PS: About the Intel vs AMD thing Intel had a major advantage with shrinking up to 22nm before AMD could do it thus is the only reason they won the Mobile Space with Baytrail and Cherrytrail adding more CUs each time this advantage is slowly over, but it gave Intel the time to improve the GPU each time and close the gap to AMD and Intels GPU Core being now more feature advanced then all of them including the Video Asic part, where we will see surely another interesting development but i guess not with Kabylake in terms of Intels capability with their Video Decode/Encode time to market updates in the future ;)

At least with 14nm we reach pretty much parity for the first time between virtualy all of them for a longer time period.

Intel = 22nm/14nm/ in the future targeting 11nm but the shrinking advantage will be lower as ever before
Nvidia =28nm/16nm/14nm
AMD = 28nm/14nm

so 14nm is the common denominator on that a lot architecturally will happen finally and no shrinking race advantage anymore for a longer period :)

[aufkrawall]

Who cares about minor fluctuations when you typically have more than a dozen frames queued ahead by decoder and even can have madVR render them ahead?

[Paul Tronc]

It seems that Nvidias Hybrid H.265 Decoder is more efficient with CUVID Decoding especially on intra frames DXVA Native has the tendency to cause latency spikes ...

Indeed : I switched to CUVID, the Netflix_TunnelFlag is almost playable. Globally smooth, only 170 frames dropped for the entire sequence.

[Paul Tronc]

Im mainly using as seen MPC-BE Sync Graph which is awesome fine grained for a almost Realtime Graph especially in Analyze Mode 2 when CPU overhead is way lower then with the Full OSD and additional timer polling (NVAPI ?) eliminated :)

@V0lt

Please add a Sync Graph only Analyze Mode without any additional OSD timer running maybe it can further lower the Overhead and it's caused EVR Latency Fluctuations that become visible in the Graph itself increasing the Audio and Video distance from the middle with DWMs tripple buffering ...

I'll have to dig into this 'Sync Graph', for now I don't understand what it is.

[CruNcher]

It's pretty simple a Graph that shows you the current Render Jitter and this is dependent on many System factors the lower the jitter the smoother the Rendering playback ;)

And the Graph is very sensitive to System Problems and fluctuations in the whole WDDM Playback Chain because it's own Render Overhead isn't really high though it works only in the Render context of D3D9 in MPC-BE for now it's though muich nicer then some static numbers you can't bring into context over time of what you seeing even if it's slightly delayed :)

Push for example the Print Screen button and see if you get a spike ;)

Indeed : I switched to CUVID, the Netflix_TunnelFlag is almost playable. Globally smooth, only 170 frames dropped for the entire sequence.

Not bad if you take into account that you have only 2 GB

Though you wouldn't really expect anything else from VP7/8

Nvidia is very picky about memory see the last AV1 Optimizations they proposed for the De-ringing Filter or their latest AGGA R&D ;)

Nvidias Hybrid Decoder failing not even Broadcast Complexity but high enough to get it performing very unstable ;)

http://i1.sendpic.org/t/cQ/cQIaWZlkF5cxXQEjibEXl65h6Wo.jpg (http://sendpic.org/view/1/i/eb4nMTYvvZQ2XsfTV3F4BbVbFrX.png)

13 sm vs 4 Intel Cores to the rescue ;)

http://i1.sendpic.org/t/pb/pbWkzApGJyjH9CmLG9jP8D3YjcK.jpg (http://sendpic.org/view/1/i/kT8krcaaWAS7gj9Qhau7iM1Mq4T.png)


PS: Did some RX 460,470 480 user tried if Strongene Lentoids Hybrid OpenCL Decoder still works, as each IDCT,MC and PP part is generation individually optimized and chosen at runtime i wonder if it still works @ all on Polaris GCN 4.0 supporting the Decoding or fails or if it's going to fallback to the Tonga\Hawai or some Generic GPU/CPU Path ?

Lentoid can even partly outrun lav video in complex bitstream parts when the CPU starts to starving under heavy load and push out more stable results then but therefore it also uses lot more System Ram and it performs equal in 32 bit as in 64 bit no slowdown like lav video ;)



Fast Scenechange sudden Bitrate Peak test

Lav Video CPU

http://i1.sendpic.org/t/f4/f4Oppp66Wexv9MiIAESTetRV4w3.jpg (http://sendpic.org/view/1/i/gw5BsSf8m04CHQeCGnKR1PK2Yx2.png)


Lentoid CPU

http://i1.sendpic.org/t/t6/t6eICbkKQw85HNV8BkZiuh6MpCq.jpg (http://sendpic.org/view/1/i/x6OZDjn8t6JPWlHQfmRblcLFUch.png)

[Paul Tronc]

I'm back, to share my findings. I'm trying to optimize my CPU and GPU power in order to get the maximum HEVC decoding result.
For now, the best thing I can achieve on a 4K 40Mbps HEVC sample, using a 2600K@4.4 and GTX960, on a 1080p display :

- GPU HEVC decoding : OK.
- Madvr : Jinc and other "rather good" algoritms
- FRC (Frame Rate Conversion) using SVP, from 24p to 60p

The GPU assumes most Madvr + LAV work, while the CPU is used for frame interpolation.

I'm currently tuning the settings to get a rock stable smooth experience.
The 1080p output helps a lot, on my 1440p the rendering time grows quickly.

[wanezhiling]

http://i.imgur.com/CYpKdmp.jpg
http://i.imgur.com/opaTGVd.jpg
Pascal 8K VP9

[NikosD]

And 10bit VP9 ?

[wanezhiling]

Do you have 10-bit VP9 clips?

[NikosD]

No and I haven't found an encoder

[aufkrawall]

Any VP9 decoding still not enabled with Polaris?

[huhn]

here a 3 sec VP9 10 bit file: http://filehorst.de/download.php?file=bDyyChti
i encoded it with VPXENC

edit: i will check out 16.10.1 for polaris VP9 support

[NikosD]

You own a polaris card ?

[huhn]

yes and no VP9 support in 16.10.1.

[NikosD]

It would be useful if you could post H.264 & H.265 (8bit & 10bit) benchmarks of the usual clips of this thread.

Which one by the way ?

[huhn]

RX 480 4Gb.

it's broken or at least i have major problem with it to even get an Constant image have to replace before i can do any proper test.

i will do some test with the card in some of my other PC this weekend to make sure it is really the card. even though it is pretty clearly the card.

in short no performance tests for now. and most likely not any time soon.

[NikosD]

Do you have problems in 2D and 3D besides video ?

[huhn]

i get the problem in madVR and in browser. so it is generally.

nothing special just bad luck.

i could work around it and do some tests but testing this card it self is just more important right now.

trust me i'm very very upset about this.

[NikosD]

Brand ?

[huhn]

XFX.

i don't know why this should matter.

[NikosD]

Out of curiosity.
It's a good brand, bad luck.

Good choice in general the RX 480 card

[CruNcher]

i get the problem in madVR and in browser. so it is generally.

nothing special just bad luck.

i could work around it and do some tests but testing this card it self is just more important right now.

trust me i'm very very upset about this.

Is it only happening on Windows 10 ?

There are lot of reports since the Anniversary Update of Graphical and Stability issues with many IHV Drivers on the indtroduction of WDDM 2.1 ;)

Lot of intermediate Flip issues as well ;)

[NikosD]

XFX.

i don't know why this should matter.
Give a few more € and buy this XFX GTR 8GB RX 480 monster.

1475MHz at 64C max and TDP 133W fully overclocked.

At normal overclock speeds the temp is ~50C and TDP <90W (!)

I think GloFo has started delivering RX 480 chips with good quality.

http://youtube.com/watch?v=zWASNajSdpg

[huhn]

useless to me. i'm not really gaming with it.

[aufkrawall]

Edit: *sorry, wrong thread*

[CruNcher]

useless to me. i'm not really gaming with it.

It's not only the Gaming we rarely use the Power off the GCN Hardware yet for Video based tasks ;)

[NikosD]

Polaris RX 460 2GB/4GB is definitely faster regarding GPU power than 1050 and cheaper - the 2GB model.

For someone using MadVR for example could be a better choice or for a little on-line gaming besides HTPC.

We still don't have clear picture of HEVC decoding capability, but I'm pretty sure Polaris could play Blu-ray UHD.

8K is only for demo/benchmarks right now.

[CruNcher]

Depends on what you want todo NikosD Nvidias VPX is a shared thing which means the more Power the more Streams you could play at once or the Higher Bitrates you can achieve @ 4K ;)

[NikosD]

What do you mean a shared thing ?

Nvidia's VPx it's a fixed-fuction hardware decoder like UVD and QuickSync.

It's independent of the GPU shading power.

Only clock defines its speed for the same architecture.

[NikosD]

You are talking about samples that Pascal also can't play properly, I remember a post of a guy regarding an H.264 SD clip and a Pascal card.

I don't believe it's a hardware problem but rather a driver's issue for both companies.

[clsid]

That clip contained errors, which is why other hardware decoders were also having trouble with it.

[nussman]

Polaris has some problems with h.264 interlaced: https://community.amd.com/thread/203197
Clips with errors are bad for "all" hardware decoders ...

[ashlar42]

If "on paper" specs are confirmed, would it be safe to assume that GTX 1050 will offer every decoding feature needed for the foreseeable future?

[NikosD]

Yes, you can say so.

Only VP9 10bit needs to be confirmed, something that Kabylake from Intel supports for sure.

[ashlar42]

Only VP9 10bit needs to be confirmed, something that Kabylake from Intel supports for sure.Are there any signs pointing to VP9 to move out of YouTube on to other content providers? Partially off-topic, maybe, I don't know. :)

[sneaker_ger]

The VP9 successor AV1 (https://en.wikipedia.org/wiki/AOMedia_Video_1) is supposed to be finalized within 6 months. No hardware support yet. But I believe the GTX 1050 will be good enough to warrant its price for now.

[ashlar42]

The VP9 successor AV1 (https://en.wikipedia.org/wiki/AOMedia_Video_1) is supposed to be finalized within 6 months. No hardware support yet. But I believe the GTX 1050 will be good enough to warrant its price for now.Wow... we're just starting to see a full lineup of HEVC full support from the two major vendors and they're already preparing a new technology? Is there any chance that, by being VP9 capable, the cards will be able to do AV1 as well in hardware?

[nevcairiel]

Is there any chance that, by being VP9 capable, the cards will be able to do AV1 as well in hardware?

Not very likely. But just "finalizing" the codec still puts it years away from wide-spread commercial use, so its nothing to worry about.

HEVC was first approved in 2013, and 3 years later we just have the first consumer content showing up.

To this day, VP9 10-bit is also not to be seen in commercial consumer content, for that matter.

[NikosD]

Is there any chance that, by being VP9 capable, the cards will be able to do AV1 as well in hardware?

It is more or less like expecting from a hardware decoder of H.264/AVC video, to decode in hardware H.265/HEVC video.

[ashlar42]

It is more or less like expecting from a hardware decoder of H.264/AVC video, to decode in hardware H.265/HEVC video.I'm not a coder. So I asked. :)

[NikosD]

It seems that latest Chrome x64 v54 when you write chrome://gpu at the address bar, it says VPx Video Decode: Hardware accelerated for Nvidia and AMD cards.

I don't have an Nvidia card capable of VP9 Hardware acceleration, maybe someone with a GTX 950/960 or Pascal card could try some Youtube VP9 videos.

Also, I don't know if previous versions of Chrome said the same thing.

My Intel system has VP9 acceleration disabled due to a Kabylake VP9 problem with HW acceleration.
You can read more here about Kaby and Chrome:
https://bugs.chromium.org/p/chromium/issues/detail?id=654111

Of course my Haswell iGPU even if VP9 acceleration was enabled, it wouldn't accelerate VP9 codec.

The activation of VP9 acceleration on AMD cards is meaningless because no card is capable of VP9 hardware decode, but it's good that it's there.

So, could anyone with an Nvidia card give a try to the new Chrome x64 and VP9 video from Youtube ?

[Trevonn]

It seems that latest Chrome x64 v54 when you write chrome://gpu at the address bar, it says VPx Video Decode: Hardware accelerated for Nvidia and AMD cards.

I don't have an Nvidia card capable of VP9 Hardware acceleration, maybe someone with a GTX 950/960 or Pascal card could try some Youtube VP9 videos.

Also, I don't know if previous versions of Chrome said the same thing.

My Intel system has VP9 acceleration disabled due to a Kabylake VP9 problem with HW acceleration.
You can read more here about Kaby and Chrome:
https://bugs.chromium.org/p/chromium/issues/detail?id=654111

Of course my Haswell iGPU even if VP9 acceleration was enabled, it wouldn't accelerate VP9 codec.

The activation of VP9 acceleration on AMD cards is meaningless because no card is capable of VP9 hardware decode, but it's good that it's there.

So, could anyone with an Nvidia card give a try to the new Chrome x64 and VP9 video from Youtube ?

It works for me, CPU usage is only a few percent on a 4K VP9 video

[pirlouy]

About the new chrome release (version 54), there's a bug with GIFs which have artefacts at the beginning. Need to disable flag "Hardware-accelerated video decode" to have correct gifs... but then all hardware decoding is disabled (I know, kinda ironic on this thread).

[el Filou]

About the new chrome release (version 54), there's a bug with GIFs which have artefacts at the beginning. Need to disable flag "Hardware-accelerated video decode" to have correct gifs... but then all hardware decoding is disabled (I know, kinda ironic on this thread).
Are you sure it's really GIFs? For example, when Twitter says a video is GIF, it's actually encoded as an AVC MP4 without sound.

Speaking about VP9, I stumbled upon this page from january: http://flatpanelshd.com/news.php?subaction=showfull&id=1452198532 that says YouTube is planning to provide HDR videos "in 2016", which would need to be 10-bit, which would need VP9 Profile 2.
It might be nice then for someone who is looking for a relatively future-proof graphics card to upgrade an older system to make sure their GPU decodes a VP9 Profile 2 bitstream.
I know VP9 still isn't used much outside of YouTube, but YouTube is popular.

[sneaker_ger]

There isn't any PC GPU (card) for VP9 10 bit. Only Intel Kaby Lake.
But AV1 is only a few months ahead. It wouldn't be surprising if Youtube skipped VP9 10 bit and went straight for AV1. Or maybe both.

[NikosD]

Probably AV1 is years ahead for wide adoption, not months.

[sneaker_ger]

Difficult to say. They can roll out the software decoder with Chrome in a matter of days. While AV1 hardware might start spreading only in 2018 or 2019, VP9 10 bit is also barely available. Then again, YouTube can encode both VP9 and AV1 easily as well. HDR videos will only be a tiny fraction of overall videos anyways.

But thinking about it, HDR has its own requirements for OS/driver/display. Do TV SoCs already support VP9 10 bit? If so I agree that VP9 10 bit seems likely.

[clsid]

The HEVC spec was finalized in April 2013. So it took about three years for hardware decoding to become widely available.

[NikosD]

The 1050 Ti is too expensive for HTPC only.

The RX 460 is cheaper even from vanilla 1050.

The main problem here is that both companies AMD and Nvidia have stopped producing cheap entry level cards with low 3D but with full video capabilities, because of the APUs from Intel and AMD.

It's true that all modern CPUs are SoCs (Systems on Chip) with 3 processors (CPU, GPU, Media Processor) that need nothing from a discrete GPU, besides large 3D power for games.

But what about people with no APU or a slow/old iGPU/Media Processor like Sandybridge ?

Do we all have to upgrade to Kabylake or Zen APU in order to get full media capabilities or buy expensive discrete 3D cards just for HTPC ?

[nevcairiel]

The NVIDIA *50 series of cards are not really something I would call expensive. A 750 was available at around 90€, and while the 1050 are still more expensive now, they'll eventually fall again (they have just been released, after all). And AMD has similar offerings (albeit their media block is plagued with problems once again).
But there is just no market for even lower cards, as the CPUs cover that. Cheap products need to sell a lot of stock to be worth producing - and something below a 1050 or RX460 is not going to do that.

[NikosD]

The NVIDIA *50 series of cards are not really something I would call expensive. A 750 was available at around 90€, and while the 1050 are still more expensive now, they'll eventually fall again (they have just been released, after all). And AMD has similar offerings (albeit their media block is plagued with problems once again).
But there is just no market for even lower cards, as the CPUs cover that. Cheap products need to sell a lot of stock to be worth producing - and something below a 1050 or RX460 is not going to do that.

1050 is about 50% more expensive than 90€.

Also 750 Ti 4GB, 950 and 960 are still more expensive than that.

They are about 150€ to 200€.

That's a lot of money for HTPC only, as I said before.


So, your answer didn't reply to my question, because i think it could exist a market from the kind of users I described earlier in my previous post.

[huhn]

the 1050 is not ready for delivery yet but is already at 125 €.
it is very likely that it will drop in price after it gets available in higher numbers.

the HEVC content is very very limited so upgrading a HTPC for "cheap" to get this is simply not important at the moment and a very small market too.

AMD and NVIDIA are creating gaming and professional cards that's there market everything else is an extra.

a cheap card is not going to cut it for UHD anyway the Vram usages is easily over 2 Gb.
a 4Gb card starts at ~125 euro for the rx 460 and the 1050 ti cost about ~150 euro.

i have a 960 for some time now and a UHD screen to and i never really needed the HEVC decoder anyway... it is just a nice to have goodie for now.

[NikosD]

You seem to forget two important things:

Blu-ray UHD and YouTube (or video streaming in general)

And I could add a third one - a lot of Blu-ray movies transcoding from H.264 to H.265.

For the first one you need a HW decoder for HEVC UHD 10bit and for the second a HW decoder for VP9, especially now that the No1 browser, Chrome, has enabled by default the HW acceleration of the default codec - VP9.

Older CPUs and even low cost of new ones, can't handle properly those two codecs.

Those users need to upgrade.

[sneaker_ger]

UHD Blu-ray doesn't work on PCs. Youtube basically only needs VP9 for 4Kp60, there's always AVC to fall back to otherwise. Amazon, Netflix et. al. only use AVC for PC streaming. Actual need for HEVC decoding on PCs is still off.

[huhn]

if they need UHD and the UHD blu ray doesn't work on the PC yet not to mention native HDR.

now you have a very low number of people that can't decode 1080p VP9 with some ancient PCs or atom like PCs. do you really think it is worth to produce a card for these people? getting a new system with a proper CPU is a better investment.

even an i3 is getting bored by youtube VP9 and more important it is not like you have to play the VP9 version the AVC version is going to work splendid. even UHD VP9 works fine on the CPU.

so lets talk about reencodes.
of cause you can do that and you my want to play them.
first the number of HTPC is pretty low. one reason are way cheaper media player. another reason is technically. a media player needs an HDMI cable an HTPC cost more and needs knowledge.

the most important reason for most user to create a HTPC is to game on it too so the low budget cards getting again worthless.

and know it is time to show why hardware decoding isn't important if you aim to play reencodes.
there are a lot of possible ways and good reason to reencode a video in a way that a hardware decoder can't decode them.
if it is just reencoding UHD to FHD 4:4:4 or AVC 10 bit. so if you want to play reencodes you are fast lost with hardware encoders everything else plays on a cheap media player too.

of cause there are some user that would benefit from such a card but the number of people is simply to low.

[NikosD]

@sneaker_ger and huhn

A visit to Intel forums would immediately persuade you that VP9 for a lot Chrome users is a kill.

Really many complaints about that and not for the rare 4K 60fps obviously.

All seems to be as a matter of perspective.

For example, a few posts above you were talking about a yet non-existent codec, AV1, the successor of VP9.

Bluray UHD is much closer to become important than AV1.

And TV/video broadcasting will go to HEVC if it hasn't already gone in some places and services

[huhn]

you can still buy a new PC CPU GPU what ever when the time for HEVC has come.

HEVC broadcast is currently some loops.

if you want to watch UHD right now with an UHD screen but you are missing the GPU to decode the signal i'm sorry but a UHD screen was cheap in the first place.

AMD user don't have VP9 decoding and the world is still turning

[sneaker_ger]

A visit to Intel forums would immediately persuade you that VP9 for a lot Chrome users is a kill.
Yes, it defaults to VP9, unfortunately. Basically buying a card would be useless if it would use AVC instead. There's a Chrome addon for that but unexperienced users have no clue.

For example, a few posts above you were talking about a yet non-existent codec, AV1, the successor of VP9.
Yeah, I'm hypocritical like that. :D But the user specifically asked about future so I deemed it worth mentioning, at least.

And TV/video broadcasting will go to HEVC if it hasn't already gone in some places and services
True. In Germany we already have HEVC for terrestrial but it's only 1080p50 8 bit at low bitrates so not really CPU demanding.

[aufkrawall]

Youtube basically only needs VP9 for 4Kp60, there's always AVC to fall back to otherwise.
You don't want to fall back to YT's H.264 results and also for vids with lower system requirements VP9 hardware decoder might be attractive due to being low on noise and heat, compared to CPU (HTPCs usually don't have an i7).

[CruNcher]

The HEVC spec was finalized in April 2013. So it took about three years for hardware decoding to become widely available.

It took roughly a year for the first GPU Hybrid solutions and not much time after that for the first full Hardware IP Cores adoption rate was much faster then with H.264 overall ;)

Also 750 Ti 4GB, 950 and 960 are still more expensive than that.

They are about 150€ to 200€.

That's a lot of money for HTPC only, as I said before.

Kaby Lake + LGA 1151 motherboard + DDR4 RAM is more expensive than a single GTX 1050 Ti or GTX 1050.

Discrete cards are so easy to upgrade, just plug in and enjoy instead of building a new Kaby Lake PC, installing Windows or OpenELEC etc.

http://www.anandtech.com/show/10610/intel-announces-7th-gen-kaby-lake-14nm-plus-six-notebook-skus-desktop-coming-in-january/3



That's really embarassing on Intel's part, still no HDMI 2.0 support when Nvidia had it since 2014. One can only hope the upcoming desktop Kaby Lake & Z270 chipset has HDMI 2.0b support like Nvidia Pascal GPUs all do.



So, your answer didn't reply to my question, because i think it could exist a market from the kind of users I described earlier in my previous post.

Same Price or even lower Nvidia Shield and you 4K HDR capable ;)

HTPC is nonsense for these low requirement tasks ;)

Also having 2 Media blocks inside the System for AVG user is total nonsense if you could use the discreet die space more efficiently.


The trend is visible already that we gonna see Full Updateable Hardware Cores in the Future on the CPU/APU side, which also is gonna lower adoption time by a very big amount, time to market efficiency will greatly be enhanced.

You can even soon buy very cheap laptops and refunctioning them without any problems to whatever you want modular ;)

[NikosD]

Checking out features regarding video decoding/encoding in HW on discrete cards and prices (Greece) I have a few comments to add:

AMD:

The least expensive card for HW 4K H.264 decoding/encoding was R9 380 ~200€ without HW H.265/VP9 decoding/encoding, before Polaris release.

Now with RX 460 (Polaris 11) at ~120€ you can get 4K HW decoding/encoding of both H.264 & H.265 (8bit & 10bit) and in near (?) future VP9 hybrid decoding with a new driver.

All of the above with a TDP < 75W

For HTPC and video acceleration (HW decoding/encoding) the RX 460 is like a miracle for AMD users, a huge step forward, the best card by far, supporting FreeSync, HDMI 2.0b with HDCP 2.2 and HDR etc (all modern features of HTPC)

For faster 3D you could go to a RX 470, probably the best VFM for 1080p gaming, with all HTPC features like RX 460.

BUT still there is no AMD card <100€ for HW acceleration of HEVC/VP9 or even 4K H.264


Nvidia:

Maxwell GM206 chips (750 v2, 950, 960) support 4K H.264 & H.265 (8bit & 10bit) along with VP9 acceleration in HW.

But 750 v2 is extremely rare and >100 € with low 3D power, 950 costs ~160€ and 960 costs ~200€

Now with Pascal 1050/ 1050 Ti you can get even 8K (useless for now) acceleration (very fast) of HEVC/ VP9 (and 12 bit of HEVC - useless for now) and better VFM than Maxwell in 3D power, with very low TDP but still at ~130€ and ~170€ respectively.

Again, like AMD, there is no Nvidia card <100€ for HW acceleration of HEVC/VP9, the least expensive card starts at ~130€.

My conclusion is that the days of cheap entry level or budget or HTPC or whatever you want to call them cards, are gone.

Those cards had little 3D power, sometimes passive cooler for complete silence, with full video acceleration capabilities and HTPC features at a price of 40€ - 50€

The story of Cruncher with fully update-able hardware cores in the future, sounds like a nice sci-fi movie to me.

[aufkrawall]

I think we still don't know if Polaris' VP9 decoder is hybrid?
If we're lucky, it'd be native. On the other hand, it could simply be broken and never get activated (happened before...).
Hybrid decoder is also very bad in conjuction with madVR if shaders are used for decoding like it was with GM204 and HEVC 8 bit.

[NikosD]

Exactly, your comment is to the point.

From the words used by AMD to describe the VP9 decoder, I think it is hybrid.

[CruNcher]

Makes no sense that it takes so long ergo it's Hybrid i guess we have to wait for VEGA to see the full Hardware Decode it just wasn't ready yet which is a pain again seeing how long Nvidia has it ready now since X1

[NikosD]

Nvidia released GM206 (GTX 960) on January 2015 and enabled the pure HW decoder of VP9 on December 2015 via a driver.

We all thought it would have been hybrid, but it wasn't.

Nvidia just decided to bury a full 4K VP9 HW decoder for almost a year (!)

We really aren't inside the heads of people deciding such things in big companies.

[aufkrawall]

And in the meantime, browsers got VP9 hardware support too (Firefox should follow shortly with version 51).
So it's really about time to have it ready in the drivers.

[NikosD]

It is true that MS released DXVA2 VP9 spec around early October of 2015, but we could/ should know that GM206 had a VP9 HW decoder inside from the beginning.

It is like HEVC 12bit HW decoder of Pascal.

There is no DXVA2 spec from MS for that yet (I don't know if they will ever release it), but Nvidia has already advertised it and we all know that Pascal is capable of HEVC 12bit acceleration.

The question still remains why Nvidia hadn't told anything for a year regarding VP9 HW decoder of GM206.

Of course, it doesn't really matter right now, one year after the activation of VP9 acceleration.

[CruNcher]

Ehh it was known from day 1 that GM206 included it like X1 does, there was no real secret about it.

Though Microsoft was late at the party for support when the whole Linux Mobile Ecosystem already adapted it ;)

And it was not sure when it really would take of and to promote it at all, all these are normal business considerations.

Nvidia allways likes to be prepared very early on especialy with their Linux Mobile side adventure they not do short decissioning phases like Intel for example does and can afford todo so being ready in no time and react ;)

[NikosD]

Do you have a link, presentation, slide of Nvidia advertising VP9 HW decoder of GM206 ?

We didn't even know that GM206 had different video decoder than the rest of the family (GTX 970/980) till we tested the decoder using DXVA2 H.265 files (8bit/10bit)

[CruNcher]

As i told you in the Sync Planing and Cycle completion with X1 it was logical Tegra mostly drives the VPX decisions these days and not only that since Maxwell ;)

And that only GM206 had it is pre planing of Nvidia as well based on the System needs of the user base and how much sense it makes in target range for GTX 970/980 based system it made no sense from the decoding complexity point at all, which saves die space for other things that are more important for that target group ;)

Nvidia very well knows their target groups and is planing in detail on it beforehand to save as much as possible in their end Designs they dont waste any flop or give it for free, they calculate so exactly with their data and make predictions even it's crazy how mostly right they are though they obviously also try to change it to their advantage helping a little.

This is what Roy Taylor at AMD tries to finally brake knowing his ex employer inside out, because he partly developed the most important strategies back then, that brought ATI and then AMD almost to it's knees ;)

[NikosD]

About the new chrome release (version 54), there's a bug with GIFs which have artefacts at the beginning. Need to disable flag "Hardware-accelerated video decode" to have correct gifs... but then all hardware decoding is disabled (I know, kinda ironic on this thread).

The artifacts were produced on Nvidia cards that you probably own and it was a bug of Nvidia drivers.

I can see the below comment in the release notes of latest Nvidia drivers 375.76 that you should probably download and try.


This is GeForce Hot Fix driver version 375.76 that addresses the following:

Occasional flicker on high refresh rate monitors
Artifacts in GIFs

[pirlouy]

Indeed, it was a bug in nVidia drivers 375.57. I also noticed Speakers settings tend to often be reset so I tried 375.63 and .70 without success then I DDuninstall everything and went back to 368.81.
God I hate all brands for their drivers.

[CruNcher]

Thats why i currently avoid Windows 10 entirely as a stable usage system way to much in Flux currently Nvidia has a hard time to get everything of their core stable especially all the hacks and workarounds they did for WDDM 1.1/1.2 and adapting to WDDM 2.0 and now 2.1 and the stability issues have become so heavily high in the last months no fun to be on that OS for now productively especially not on Nvidia Hardware, till everything sorted out and stabilized ;)

And ontop of that they push their own Software stack and create even more problems, because they want to push their Cloud Ecosystem further and need to create new workarounds, im pretty much used to chaos but that's a bit to much :D

[huhn]

one more fix and the decoder is worth testing .

Fixed Issues

H.264 content playback may experience playback issues on internet browsers with hardware acceleration when also running gaming applications or content.

http://support.amd.com/en-us/kb-articles/Pages/Radeon-Software-Crimson-Edition-16.11.4-Release-Notes.aspx

[NikosD]

Right.

But there is also HEVC.

From what you have seen so far, what is the performance of HEVC 8bit/10bit compared to Maxwell GM206/Pascal and Skylake from the results of other users here ?

[CruNcher]

Difference should be at most a pretty small iterative step if any at all between GM206 and GP204 especially on the Encoder side, the biggest impact you really have vs GM204 on the Decoder Performance side

That is massively visible even with all these poor reviews ;)

And in those regards AMD doesn't possition itself so bad with Polaris at all and Vega could be the breakthrough since a long long time if Software parts get ironed out till release :)

[huhn]

RX 480 16.11.4
edit: lavfilter 0.68.1

Beauty_3840x2160_120fps_420_8bit_HEVC_MP4.mp4 (we should really rename this file...)
decode 73 - 77 - 80
playback 69 - 79 - 81
~10% CPU usages from DXVA checker

Life_of_ Pi_draft_Ultra-HD_HDR.mp4
decode 96 - 100 - 107
playback very inconsistent so here a screen: https://abload.de/img/dxvacheckerpsu78.png
~10% CPU usages from DXVA checker


playback has extra GPU usages often 100 % at max clock...
10 bit can be faster than 8 bit and the GPU/CPU usages is strange too.

[NikosD]

Check your driver settings.

You probably have left the default settings which enable various enhancements.

Can you check the same clips that others have already compared in order to have a direct comparison ?

[huhn]

i don't know which files they used.

and video is and was at custom with everything off.
well not a lot of options left anyway https://abload.de/img/crimson2uuo4.png

[aufkrawall]

playback has extra GPU usages often 100 % at max clock...
10 bit can be faster than 8 bit and the GPU/CPU usages is strange too.
I think the GPU usage reported by the AMD driver has been broken since forever with those weird 0%/100% behavior.
Got the same in some games with my R9 390.

[huhn]

well still using a clock speed of 1266 mhz for some reasons

edit: maybe the downscaler is a PS3 script ?

[aufkrawall]

Do you mean that the card doesn't downclock when doing decoding + scaling?
My R9 390 didn't do that either. It was running with full clocks when watching 480p H.264 upscaled to WQHD via DXVA (e.g. in Windows 10 apps), despite of every video filter in the Crimson UI turned off.

[huhn]

the filme & TV app is using 0 GPU usage and 1 CPU usage and the card runs at 608 mhz which is bad but not terrible.

[aufkrawall]

Well, then this got improved, it seems. Could you try WQHD via VSR, if you don't have such a display?

[CruNcher]

A pretty good Sliced HDR 10 Bit Broadcast Sample i propose for to be added to the decoding test

https://abload.de/thumb/decode10bitbt2020i8kbc.png (http://abload.de/image.php?img=decode10bitbt2020i8kbc.png)

[huhn]

i'm at UHD.

playback benchmarking is done at 1280x720.

maybe just a problem with DXVA checker. a potential problem could be if DXVA checker isn't using DXVA for scaling. expect for the CPU usages decode benchmarks are looking fine.

[NikosD]

i don't know which files they used.

and video is and was at custom with everything off.
well not a lot of options left anyway https://abload.de/img/crimson2uuo4.png


1) H.264 -1080p-108Mbps
ftp://helpedia.com/pub/multimedia/x264/testvideos/2012%20-%2001%20-%20QuickSync%20vs%20UVD%202.2%20vs%20VP4/9.Ducks.Take.Off.1080p30fpsRef5-108Mbps.mkv

2) H.264 - 2160p-370Mbps
ftp://helpedia.com/pub/multimedia/x264/testvideos/2160p%20samples/DucksTakeOff_2160p50.x264.CRF24.mkv


3) H.265 8bit -1080p-80Mbps
http://www.imaging-resource.com/PRODS/samsung-nx1/FULLRES/yvid_SAM_0230-ProQuality.MP4

4) H.265 8bit -1080p-80Mbps
http://imaging-resource.com/PRODS/samsung-nx1/FULLRES/yvid_SAM_0235-UHD_Sample1.MP4

5) H.265 10bit - 2160p-140Mbps
http://jell.yfish.us/media/jellyfish-140-mbps-4k-uhd-hevc-10bit.mkv

6) H.265 10bit -2160p-400Mbps
http://jell.yfish.us/media/jellyfish-400-mbps-4k-uhd-hevc-10bit.mkv


Please use latest DXVA Checker v3.14.0 - LAV x64 video 0.68.1.x in DXVA native mode and in decode mode.

Include also playback mode results using the 1280 x 720 resolution

Run each benchmark 3 to 5 times and always exclude the first run, because it reads the file from the disk.

For each file, post the average of your runs in this form min (fps) /avg(fps) /max (fps)

If it's possible check in real-time the GPU clocks because they may have fluctuations, especially in playback mode (due to many reasons)

I think the GPU usage reported by the AMD driver has been broken since forever with those weird 0%/100% behavior.
Got the same in some games with my R9 390.

well still using a clock speed of 1266 mhz for some reasons

edit: maybe the downscaler is a PS3 script ?

I was under the impression that in benchmark playback mode, all of my cards (low power actually) from all three (Intel, AMD, Nvidia) had a GPU usage of 100%.

I'm not 100% sure that this is valid for all cards.

A pretty good Sliced HDR 10 Bit Broadcast Sample i propose for to be added to the decoding test

https://abload.de/thumb/decode10bitbt2020i8kbc.png (http://abload.de/image.php?img=decode10bitbt2020i8kbc.png)

I couldn't figure it out where is the download link.

i'm at UHD.

playback benchmarking is done at 1280x720.

maybe just a problem with DXVA checker. a potential problem could be if DXVA checker isn't using DXVA for scaling. expect for the CPU usages decode benchmarks are looking fine.

I don't think anything has changed in DXVA Checker.

Do you remember a different behavior ?

[huhn]

I don't think anything has changed in DXVA Checker.

Do you remember a different behavior ?

i'm pretty sure my GTX 960 isn't using any GPU and it stays at a low powerstate.

[CruNcher]

I couldn't figure it out where is the download link.

You can find it here

http://demo-uhd3d.com/fiche.php?cat=uhd&id=54

[huhn]

9.Ducks.Take.Off.1080p30fpsRef5-108Mbps.mkv
decode 84 - 97 - 107 CPU 14 % GPU 0 %
playback 86 - 100 - 111 CPU 7% GPU impossible to measure


DucksTakeOff_2160p50.x264.CRF24.mkv
decode 40 - 45 -49 CPU 3% GPU 0%
playback 40 - 45 - 50 CPU 3% GPU - same problem

jellyfish-140-mbps-4k-uhd-hevc-10bit.mkv
decode 45 - 60 - 90 CPU 1-26% GPU 0 %
very high CPU usages at the last part of the video.
playback 54 - 62 - 68 CPU 2% GPU same problem

so weird... i did at least 20 decode tests...

jellyfish-400-mbps-4k-uhd-hevc-10bit.mkv
decode 32 - 41 - 53 CPU 3% GPU 0%
playback 24 - 30 - 40 CPU 2 % GPU 950 mhz 0 %

yvid_SAM_0230-ProQuality.MP4
decode 60 - 69 - 75 CPU 2% GPU 0 %
playback 68 - 70 - 71 CPU 2% GPU 0% 600 mhz

yvid_SAM_0235-UHD_Sample1.MP4
decode 69 - 72 - 75 CPU 3% GPU 0%
playback 71 - 73 - 75 CPU 4% GPU 5% at 600 mhz

39 Mbit BD remux
decode 178 - 188 -195 CPU 3% GPU 0 % 300-450 mhz
playback 182 - 207 - 218 CPU 6% GPU 5% 650-1000 mhz

polaris doesn't like copyback...

1 digit CPU usages is hard to judge the i3 runs at 0.8 -2 GHz most of the time.
i really wonder why playback is often faster than decode...

[NikosD]

Thanks for your tests.

It looks like Polaris is a lot slower than both GM206 and Pascal.
I don't even mention Skylake as it has the fastest media processor of all.

4K H.264 sample is an extreme case, but I didn't expect Polaris to be bandwidth limited so much.

Also Bluray UHD looks like is not going to be decoded in real-time without dropping frames, judging by Jellyfish 140Mbps sample.

Nice try by AMD but Vega needs to have a faster video processor.




polaris doesn't like copyback...



I'm not familiar with any AMD card capable of fast copy-back performance.



i really wonder why playback is often faster than decode...

That's why I told you before to keep an eye on GPU clocks.

I'm pretty sure that in playback mode GPU has faster clocks (maintain faster clocks)

[huhn]

it is 300 - 450 mhz in decode mode as it should be.

[NikosD]

In benchmark decode mode both Intel and I think Nvidia are pushing very hard the GPU and the video processor works in full speed (over 1000MHz)

AMD has a tradition of very low GPU clocks during decoding even in benchmark decode mode.

And this is not normal in benchmark mode

[huhn]

it is well known that maxwell and pascal are using less power in BD playback than polaris.

but that is "easy" to test:

first of all the idle clock of my gtx 960 is 135 mhz and 202.5 vram mhz.

in benchmark mode it goes up to 1430 mhz.

in playback 135 mhz 202.5 mhz vram and 7.6% TDP with 17 % video engine load.

[CruNcher]

In benchmark decode mode both Intel and I think Nvidia are pushing very hard the GPU and the video processor works in full speed (over 1000MHz)

AMD has a tradition of very low GPU clocks during decoding even in benchmark decode mode.

And this is not normal in benchmark mode

Ehh it can be very well bellow 800 mhz without issues for playback and around 8xx mv for 4K H.264 30FPS @ 30 Mbits 8 bit 1.0v for 60 FPS

and it can do both streams at once no problem with 0 spikes at simultaneous playback at 60% of its capacity at 0.600 ms 1304 MHz and 100% capacity at 800 Mhz and 1.2v 1300 rpm

it is content complexity adaptive the DFS scales with the content complexity Nvidias Arbitrator reacts to it especially inloop filtering

This is also why they proposed changes to AV1 ;)

Though with 800 MHz its heavy on the Edge here with that Simultaneous Playback Scenario ;)

800 MHz almost 100% capacity

http://i1.sendpic.org/i/1Y/1Y02ToUNBPcs1bDJ9neXwZoRVet.png

so it's not really a good idea to forcefully restrict it's frequency if the stream complexity suddenly gets higher latency is gonna explode in this example from 0.600 ms to a whoping 30 ms better leave that to Nvidias Arbitrator it's well tuned ;)

1304 Mhz 60% capacity for those 2 Streams

http://i1.sendpic.org/i/qS/qS0uJAeGgH8FRrJXuCBVHq9J4EG.png

@Huhn
You running on I3 ?

[huhn]

yes my HTPC has an passive cooled i3 4130.

[NikosD]

@Cruncher

I'm talking clearly as I wrote above for benchmark mode, not normal realtime playback.

For normal realtime playback, the card can have whatever clock it thinks it should have.

But in benchmark mode it shouldn't have any limits.

Pure unrestricted decoding power in both decoding and playback, I repeat in benchmark mode.

[CruNcher]

Im more interested in real scenarios not benchmarks, benchmarks are a nice indicator but they aren't the whole usage truth.

And as you said AMD still has a copyback defficiency and i think additional using an i3 with that is no good idea at all, and it's technically pretty obvious that intel will win all the time here on performance no matter what.

Also your assumption that Intel has the fastest Media Processor is how you make that assumption wrong, you would get a better picture if you benchmark Tegra X1 vs Cherry Trail for example ;)

[huhn]

well i don't know what intel has to do with this but fine...

[NikosD]

I clearly referred to Skylake's media processor, the top video decoder.

The truth is that Intel will have the fastest video decoder for ever (!) because they are targeting different audiences.

Intel is very much interested in video server community with a requirement of decoding hundreds (!) simultaneous streams in realtime.

They are using Xeons with iGPUs of many EUs and the fastest video processor around, the exact same desktop version actually.

So, Nvidia took a powerful video processor from mobiles/tablets and put it in dGPUs and Intel is interested in realtime decoding in a service environment.

Latest Pascal video decoder has definitely shrinked the gap, but Kabylake in less than two months will increase it again.

AMD has a very limited point of view, just a basic decoding of 1 or 2 streams.

[CruNcher]

You underestimate Nvidia
Also you gonna see AMDs point of view changing rapidly they are next with Intel on the release schedule as you said ;)

[huhn]

AMD has a very limited point of view, just a basic decoding of 1 or 2 streams.

well i see no big problem in that if that would be the case.
but 140 mbit 60 hz HEVC should work and it doesn't...

[CruNcher]

Why should it work it depends on the bitstream complexity overall they haven't the power todo simple 8k complexity yet so nope it doesn't have to work

Based on this all my HEVC test should work fine on the Hybrid GM204 decoder but they fail at certain tools and their complexity ;)

[huhn]

hardware decoder doesn't really care about bitstream complexity and 140 mbit HEVC isn't a lot higher than the UBD spec...

[CruNcher]

Ehh but you testing no conformant speced sample here that wont work and isn't a real use case it is a addition if it works under performing same as it is a addition that Nvidias Decoder can go much Higher and has a relatively big headroom even discrecte.

The most true real use case are WEB VOD (HDR),BD and UHDBD + HDR rest is bonus with a big focus on low latency Screen Game Recording and Decoding.

You don't know why AMD made this decisions so you have to understand them maybe they wanted to have more place for ACE, which makes perfectly sense in many use cases to put the decoder to lower priority for it ;)

Anyway Huhn does that Sliced stream work on your UVD at normal DXVA playback conditions (not benchmarking) ?

http://forum.doom9.org/showthread.php?p=1786639#post1786639

[alfianadjeh]

can i play HEVC 4k 60fps with GTX 1050ti using Full HD monitor HDMI 1.4 ?

i curious because not have 4k monitor with HDMI 2.0 port.

[NikosD]

HDMI 1.4 supports FullHD up to 120 Hz/ fps so if you scale down 4K to Full HD you shouldn't have any problem.

[alfianadjeh]

HDMI 1.4 supports FullHD up to 120 Hz/ fps so if you scale down 4K to Full HD you shouldn't have any problem.

so prefer i buy gtx 1050ti or rx 460 for play HECV 4k 60fps perfectly ?

[NikosD]

I will probably get a Sapphire Nitro+ RX 470 with 8GB VRAM in the next two weeks, so I will tell you for sure what is going on with Polaris HW decoder.

[huhn]

so prefer i buy gtx 1050ti or rx 460 for play HECV 4k 60fps perfectly ?

if you plan on buying an throw away UHD screen than get a card with 4Gb vram.

[CruNcher]

Btw Sonys Camp HDR Demo is also a 8 thread sliced Encode.

Should be also added as a Sliced Encode sample.

Sliced Encodes should be seperated for SMP Decoding tests though.

[alfianadjeh]

I will probably get a Sapphire Nitro+ RX 470 with 8GB VRAM in the next two weeks, so I will tell you for sure what is going on with Polaris HW decoder.

ok, thanks.
i'll waiting your review about Polaris HW Decoder.

[CruNcher]

@huhn

Btw did you tried Strongenes older non Polaris optimized OpenCL Decoder does it still work or does it falls back to CPU only Decoding ?

[kral2008]

if you plan on buying an throw away UHD screen than get a card with 4Gb vram.

Why? 4GB is plenty of VRAM and I think we do not need this amount of memory when playing 4k videos unless playing the games in 4k res.
And, the best card is 1050 or 1050 ti currently. There are good advantages in 1050 over 1060 or AMD new cards, as you may know.

[huhn]

sorry reality looks different.

with my tests they need more than 2GB vram. if you want to make sure playback is smooth get 4GB.

[huhn]

@huhn

Btw did you tried Strongenes older non Polaris optimized OpenCL Decoder does it still work or does it falls back to CPU only Decoding ?

it's 32 bit only can only decode 8 bit. it is slower than my CPU in lavfilter. it isn't using openCL (i don't see GPU usages) for what ever reasons.

and most importantly it crashes when i try to open the filter settings.
the openCL version is from 2014...

[CruNcher]

What i expected for Polaris thx.

[wanezhiling]

http://i.imgur.com/4KtqxA8.png

http://i.imgur.com/q20XPZF.png

The PureVideo VP8 decoder of Pascal GP107 (GTX 1050 & 1050Ti) has already added 10-bit VP9 HW decoding support!!!:p:p:p

[CruNcher]

Btw i wonder actually if the GP107 VPX again has more native decoding features then maybe the GP104 VPX in the Nvidia SDK its mentioned that Specific Features are again only available on Specific Cards only (They don't explicitly state which exactly yet where) and i guess they mean the 1050 GP107 VPX especially once again with that ;)

So they maybe did it again and GP104/106 (1060,1070,1080) user have not the full capabilities of the 1050 GP107 VPX ;)

Which then means we would see those features in Big Size first everywhere with GVxxx again including most probably AV1 support

We should take a closer look at

GP104 (1070/1080)
GP106 (1060)
GP107 (1050)

and if something there is maybe again Hybrid or this time even non existing at all ;)

@wanezhiling

I guess this is what Nvidia stated in the SDK with Specific features only on specific Pascal Versions ;)

the last released VPX get normally the most features Gxx06/Gxx07

and the previous rather Hybrid CUDA based solutions (and this is really interesting if you calculate the user reach and target and overall Performance Metrics of their Hybrid Decoder Research)

I wonder if the Mobile Versions actually derive from it or if it's the same there, though they already balanced some stuff on the 3D side different compared to the Desktop GPU so i wouldn't be surprised if the Mobile support looks different.


I almost was very surprised that it this time it wouldn't be the case but then i read the SDK List and saw those footnotes ;)

[NikosD]

http://i.imgur.com/4KtqxA8.png

http://i.imgur.com/q20XPZF.png

The PureVideo VP8 decoder of Pascal GP107 (GTX 1050 & 1050Ti) has already added 10-bit VP9 HW decoding support!!!:p:p:p
Amazing.

Cruncher is right.

From the figures it looks like Nvidia added a fixed-fuction decoder for VP9 10bit, in GP107 (1050/ 1050 Ti).

Of course Nvidia can surprise us and someone with a 1060 card someday finds out with the release of a new driver that VP9 10bit is enabled for those cards too.

[CruNcher]

Doesn't mean much read the footnotes and it also doesn't explain how the Decoding is actually implemented that's entirely up to Nvidia ;)

* HEVC 10/12 bit encoding SW support coming in Video Codec SDK 8.0
** Max resolution support is limited to selected Pascal chips
*** VP8 decode support is limited to selected Pascal chips

Thats important but they don't say exactly which they keep it in the dark ;)

Even as a Member you cant seem to access

https://developer.nvidia.com/video-encode-decode-gpu-support-matrix

currently ;)

[wanezhiling]

Can someone with GP104 or GP106 install the latest Nvidia 375.95 or later drivers and run the latest DXVA Checker to see if those GPUs support VP9 Profile2?
http://i.imgur.com/m20eY2v.jpg
No luck, same 375.95 driver

[CruNcher]

** Max resolution support is limited to selected Pascal chips
*** VP8 decode support is limited to selected Pascal chips

It doesn't say anything about VP9 10bit Profile2 support, don't need to be snarky and a troll.

You really don't understand how VPX works.

Though 1070/1080 GP104 is not so important GP106 1060 is more important ;)

[NikosD]

It looks like GP107 has the most advanced video decoder right now, even more than Kabylake because i have never seen anywhere that Kaby supports 12bit HEVC which is the only thing missing though.

On the other hand, we have to see also HW encoding capabilities especially for HEVC and of course speed.

My opinion is that Kaby is faster than Pascal in all formats supported by both.

Also desktop Kabylake supports HDCP 2.2, it's not like Kabylake for laptops.

We'll see in less than two months.

[CruNcher]

Of course it will be faster it doesn't have to fight like Pascal for Performance on the PC but wait for the new Pascal Tegra and how it will kill it ;)

Though on Mobile Pascal it could look slighty different also ;)

i want to see the new Kabylake based Atom in Action for 200$ complete system :D


I find it great that Nvidia by the way as mostly the first advises this

Performance represents measured average performance and quality of different classes of videos (camcorder, gaming, screen, synthetic, and telepresence). Performance may vary based on OS and software versions, and motherboard configuration.

[eddman]

Doesn't mean much read the footnotes and it also doesn't explain how the Decoding is actually implemented that's entirely up to Nvidia ;)



Thats important but they don't say exactly which they keep it in the dark ;)

Even as a Member you cant seem to access

https://developer.nvidia.com/video-encode-decode-gpu-support-matrix

currently ;)

I think those footnotes are in regard to GP100 chips which are feature set G.

If GP102, 104, 106 and 107 chips had different hardware accelerating capabilities, nvidia would not have used the same feature set designation, H in this case, for all of them.


http://developer.download.nvidia.com/designworks/video-codec-sdk/secure/7.1/01/NVDEC_VideoDecoder_API_ProgGuide.pdf?autho=1480256815_a4ba43531ca6c92c72e50a50086f8eaf&file=NVDEC_VideoDecoder_API_ProgGuide.pdf

Page 8. GP100 is listed separately from others. It seems the only difference is that GP100 cannot do 8K coding.

http://i.imgur.com/mqGI6Up.png


Tesla P100 devices, which use GP100 chips, are listed as feature set G:

http://us.download.nvidia.com/XFree86/Linux-x86_64/375.20/README/supportedchips.html

http://i.imgur.com/82ZLUqj.png


I think the reason for 1050 cards being VP9 10-bit enabled, is down to drivers. Perhaps nvidia is yet to enable it for other Pascal cards.

Either that, or nvidia made a mistake in feature set naming and GP107 is actually feature set I or something.

[CruNcher]

If GP104, 106 and 107 chips had different hardware accelerating capabilities, nvidia would not have used the same feature set designation, H in this case, for all of them.

This is exactly what suprises me currently but anyway it could support it but maybe not fixed function but via Nvidias CUDA Research Decoder though it's unlikely they would hit the same performance then on AVG Joe Consumer Cards.

I think the reason for 1050 cards being VP9 10-bit enabled, is down to drivers. Perhaps nvidia is yet to enable it for other Pascal cards.

Possible but Review time is long over ;)

Either that, or nvidia made a mistake in feature set naming and GP107 is actually feature set I or something.

Would be a funny Intern error again of course ;)

[eddman]

maybe not fixed function but via Nvidias CUDA Research Decoder

You mean VP9 10-bit? Hmm, yes, maybe it's just hybrid decoding, but even if that's the case, it should be possible for nvidia to enable it on all feature set H cards with future drivers unless, again, they made a mistake and GP107 is a higher feature set chip.

[CruNcher]

yep it's not bad performing on the HEVC side of things at all up of course until a certain complexity ratio and they surely could have improved it even further but they just stooped.

So it would be nice to see how they do with VP9 Decoding on GP104/GP106 with it :)

http://i1.sendpic.org/t/yK/yKIrDmtcdnUzg6RaBHUXd1buSSb.jpg (http://sendpic.org/view/1/i/5uOA8BKxrxKzM8Euj6JiNWFTvRv.png)

It makes me very curious what they reach so far on their entirely own GPU optimized Codec R&D :)

[NikosD]

Please don't forget that GP107 is not like the other Pascal processors.

They are manufactured in different process 14nm, instead of 16nm, by Samsung.

So it is very much possible to have really different fixed-fuction video decoder inside, that Nvidia should mention of course.

[Yups]

I wonder if the decoding/encoding speed/quality differs from bigger Pascal cards, beside VP9 profile2 of course. As a GTX 1080 owner it's a bit disapppointing that the bigger cards are less capable once again, same with Maxwell v2. Do we have DXVA checker screenshots from Polaris and Bristol Bridge with current drivers?

[eddman]

Please don't forget that GP107 is not like the other Pascal processors.

They are manufactured in different process 14nm, instead of 16nm, by Samsung.

So it is very much possible to have really different fixed-fuction video decoder inside, that Nvidia should mention of course.

One has nothing to do with the other. Two chips can be built on entirely different processes and by different manufacturers and yet have the exact same hardware capabilities.

It's all up to the designer of those chips. If nvidia made them to have the same features, then they have the same features.

If turns out that GP107's video capabilities are in fact higher than other Pascal chips, then it was a deliberate design decision by nvidia to upgrade the video block, and not because GP107 is 14nm and the others are not.

So far all Pascal chips, except for GP100, are feature set H, so they SHOULD be the same, but since nvidia sometimes messes things up for no reason, I wouldn't be surprised if GP107 turns out to be feature set H2 or I or something.

We have to wait and see.

[CruNcher]

Im especially interested how it looks on the Mobile Parts now get more and more curius of that because of the weak reviews on such very specific things :D

Can't wait to see NikosD GCN UVD Polaris evaluation there at least i know someone with the capabilities of understanding it does it right even if im not such a fan of a simple synthetic benchmark and we have the winner approach like NikosD follows in it's core and shares unfortunately with most reviewers these days ;)

[NikosD]

One has nothing to do with the other. Two chips can be built on entirely different processes and by different manufacturers and yet have the exact same hardware capabilities.




As a GTX 1080 owner it's a bit disapppointing that the bigger cards are less capable once again, same with Maxwell v2.


It is a very well known tactic since ATI HD 2000 series and probably in older cards too, that smaller GPU chips by manufacturing process (14nm vs 16nm) or by number of shaders (1080 vs 1050) can hold bigger/better video decoders.

It's all about size and space for video decoding inside the GPU chip.

Of course from a commercial point of view, very few people really care about video transcoding features of a card like 1080.

[nevcairiel]

As a GTX 1080 owner it's a bit disapppointing that the bigger cards are less capable once again, same with Maxwell v2

(Assuming its actually a newer decoder in GP107: )

1080/1070 is about half a year older now, so advances can be expected eventually, even if technically still in the same "series".
They could delay these additions until the next full gen, but we wouldn't really benefit from that either, would we.

Its not really about the 1050 being smaller and therefor having more space, its just 6 month newer. The next gen "big" cards will have the same new "bigger" decoder as well.

[CruNcher]

And because of the Higher Shadercount they calculate if it wouldn't be more feasible to use these Shaders for Decoding and stress the CPU therefore a bit more ;)
Nevcariel these decisions are way over a year old what we see is the turn out of simple economic balance and Design decisions about the VPX target and practical goals for specific release schedule it's like partly planing into the future with predicted and existing knowledge of how it will turn out ;)

[eddman]

It is a very well known tactic since ATI HD 2000 series and probably in older cards too, that smaller GPU chips by manufacturing process (14nm vs 16nm) or by number of shaders (1080 vs 1050) can hold bigger/better video decoders.

It's all about size and space for video decoding inside the GPU chip.

Not really. As nevcairiel pointed out, smaller chips usually come, sometimes by a few months, after their bigger counterparts, so it's not surprising that they usually come with better video blocks too.

Manufacturing processes dictate very little when it comes to video engines.

I really doubt that GP107 is different (it's probably a driver feature difference), but I've been wrong about nvidia before, so who knows.

[NikosD]

Its not really about the 1050 being smaller and therefor having more space, its just 6 month newer. The next gen "big" cards will have the same new "bigger" decoder as well.

Not really. As nevcairiel pointed out, smaller chips usually come, sometimes by a few months, after their bigger counterparts, so it's not surprising that they usually come with better video blocks too.


No. It's not that.

Guys, for me it's really obvious that you don't have a clear picture of the subject, lack of knowledge I think.

Since you insist, I have to remind you the first Fermi card of Nvidia that had VP5 decoder was the smallest one - the low entry budget GT520.

Just one month before the release of GT520 the Fermi cards had VP4 inside.

Also, one month AFTER the release of GT520, the 560 Ti card had VP4 decoder.

Release time to market has nothing to do with the new video decoders.

I hope you don't need more proofs.



And because of the Higher Shadercount they calculate if it wouldn't be more feasible to use these Shaders for Decoding and stress the CPU therefore a bit more ;)
Nevcariel these decisions are way over a year old what we see is the turn out of simple economic balance and Design decisions about the VPX target and practical goals for specific release schedule it's like partly planing into the future with predicted and existing knowledge of how it will turn out ;)

Well said, once again.

[huhn]

first problem is these cards are 5 years old and the "new" 560 ti was just a GF110/GF114 these chips where produced already before the 560 ti so they had the same VP.

the GT520 was so slow that the only thing it could do was video decoding.

[NikosD]

I hope some time such completely irrelevant and wrong posts to stop, taking space from the thread and confusing people.

The only purpose besides trolling is to prove me wrong, which is a very very very difficult thing to do :D

[huhn]

you know what happens last time his ego took over him...

[nussman]

I hope some time such completely irrelevant and wrong posts to stop, taking space from the thread and confusing people.

Just stop posting ... :cool:

[NikosD]

Hello world!

We have a troll party here.

May 2011 Nvidia released 560 and November 2011 release 560 ti with 448 cores.

Both AFTER GT520 and both with VP4 and not VP5 decoder

[NikosD]

Hope you guys are no more than 20 or 25 years old.

Because if you are older than that...

[NikosD]

You should have been banned long time ago at least from this thread permanently and probably from the whole forum too.

If it was in my hand I would definitely do it, but lucky you I'm not the admin or even a moderator.

This is the last time ever that I even refer to you.

[eddman]

No. It's not that.

Guys, for me it's really obvious that you don't have a clear picture of the subject, lack of knowledge I think.

Since you insist, I have to remind you the first Fermi card of Nvidia that had VP5 decoder was the smallest one - the low entry budget GT520.

Just one month before the release of GT520 the Fermi cards had VP4 inside.

Also, one month AFTER the release of GT520, the 560 Ti card had VP4 decoder.

Release time to market has nothing to do with the new video decoders.

I hope you don't need more proofs.


IINM, the first GF119 chips were launched on Jan 5, 2011, in the form of 410M and GT 520M.

The first GF114 card, 560 Ti, was launched on Jan 25, 2011.

Yes, GF114 launched 20 days after GF119 and still came with a VP4 engine, but do we know for sure it was only because of die space issues and no other reason?

I remember reading at anandtech that back then nvidia simply would not put better video blocks in mid to high-end cards becasue there was no need for them in the market and to maybe save some money in the process by going with the old design. Video decoding capabilities were focused on low-end cards which were the main go-to cards for HTPC users.

I've been following GPUs since, I think, 2002 and don't remember reading that it was solely because of die space constraints but it's possible I missed that information somewhere.

It's plausible that die space could play a role here, but is it the only player?

If you have a source or something, would you please link it here? It never hurts to read new information.

The only purpose besides trolling is to prove me wrong, which is a very very very difficult thing to do :D

Let's not call the fellow posters trolls, shall we? There is also no reason to bring in ego. I'm looking for truth and facts, not proving others wrong.

[huhn]

and both are old chip with disabled shader parts GF110 and GF100...

[nevcairiel]

I hope you don't need more proofs.

One key point you are ignoring is that its based on the chip - if they re-release an older chip a year later its clearly not going to have more features, and the 560 Ti is the same chip was the 560, and the 560 Ti 448 (the one released in november) is an even older chip (a downclocked GF110)

GF114 was released in January 2011, that includes 560 and 560 Ti.
GF110 was released in December 2010, thats the 560 Ti 448 (which was released much later, in November 2011)

GF117 was released in April, thats the GT520

So even though its only 3 month or so, there is still a clear timeline.
But clearly you prefer calling people trolls.

[eddman]

and both are old chip with disabled shader parts GF110 and GF100...

GF114 and GF119 were new designs; not disabled GF100 and GF110.

One key point you are ignoring is that its based on the chip - if they re-release an older chip a year later its clearly not going to have more features, and the 560 Ti is the same chip was the 560, and the 560 Ti 448 (the one released in november) is an even older chip (a downclocked GF110)

GF114 was released in January 2011, that includes 560 and 560 Ti.
GF110 was released in December 2010, thats the 560 Ti 448 (which was released much later, in November 2011)

GF117 was released in April, thats the GT520

So even though its only 3 month or so, there is still a clear timeline.
But clearly you prefer calling people trolls.

Actually, NikosD is not wrong. GF114 came after GF119 and there are other such examples in other generations, not all of them though.

The point of the debate was that this is done due to die space constraints and limitations. I argued that it's not just that, but mostly a market/cost decision. It is possible to put the same video engine in a big die if there is a need for it.

[NikosD]

Guys please, do me a favor.

Ask a hardware guy to explain you how many YEARS need a new ASIC like a video decoder for R&D and how many million dollars are needed for that and then write your opinions.

Or just read CruNcher's replies who obviously knows about HW.

It's been fun, thank you.

[eddman]

Guys please, do me a favor.

Ask a hardware guy to explain you how many YEARS need a new ASIC like a video decoder for R&D and how many million dollars are needed for that and then write your opinions.

Or just read CruNcher's replies who obviously knows about HW.

It's been fun, thank you.

Implying I don't know anything? Let's not assume things like that about each other. I did not call you anything so I expect the same.

I did mention costs and market in my comment up there. You and CruNcher are very much right about those.

I was just saying that die sizes alone don't dictate everything. It's not a "won't fit on the die" situation.

It's been fun, thank you.

[pirlouy]

Or just read CruNcher's replies who obviously knows about HW.
This guy...
I've been hiding these CruNcher's posts for long time. As far as I remember, the guy was doing very big posts with 0 useful info inside (and useless big signature obviously). Talking about this guy as a reference is the worth you could do.

But even if you're right or wrong, your arrogance is boring to read. I really tried to find a particular sense of humour, but there's definitively not.

[Trevonn]

GTX 1060 also doesn't have VP9 10-bit decode. 375.95

http://i.imgur.com/JbBHzPG.jpg

[CruNcher]

Which becomes economic wise really interesting as if Nvidia locked the requirements for a newer VPX now basically to a very specific Performance range since Maxwell actually.

As the 1050 (Ti) GP107 is the same Performance range as the old 960 GM206 ;)

since everything shifted with the 16/14 nm introduction by 1 chip and +-10 Card number in overall favor of the Energy Efficiency.

I become more and more curious about the actual Mobile Parts VPX status :D

Mobile Part Decisions have very much longer impact on the Avg Consumer user and can't be really changed at all so decisions there have to be taken with greater care for the right time or else you might lose a possible content consumer as your target.

[Trevonn]

The 1050 ti came 4 months after the previous VPX version (June - Oct).

The GTX 960 did the same thing (Sep - Jan).

It's just a natural progression of technology nothing more. NVIDIA starts at the top end first so of course the weaker cards which are released months after the big GPUs will possibly have improved decoding/encoding capabilities

[NikosD]

Some people here really don't understand the concept, while others really trying to make them understand.

[pacuro]

The PureVideo VP8 decoder of Pascal GP107 (GTX 1050 & 1050Ti) has already added 10-bit VP9 HW decoding support!!!:p:p:p

Could you please show us DXVA Checker screenshot with GUID visible or just paste string.

[wanezhiling]

NVIDIA GeForce GTX 1050 Ti
MPEG2and1_VLD: DXVA2/D3D11, SD / HD / FHD / 4K
MPEG2_VLD: DXVA2/D3D11, SD / HD / FHD / 4K
MPEG1_VLD: DXVA2/D3D11, SD / HD / FHD / 4K
VC1_VLD2010: DXVA2/D3D11, SD / HD / FHD
VC1_VLD: DXVA2/D3D11, SD / HD / FHD
32FCFE3F-DE46-4A49-861B-AC71110649D5: DXVA2/D3D11
H264_VLD_Stereo_Progressive_NoFGT: DXVA2/D3D11, SD / HD / FHD / 4K
H264_VLD_Stereo_NoFGT: DXVA2/D3D11, SD / HD / FHD / 4K
H264_VLD_NoFGT: DXVA2/D3D11, SD / HD / FHD / 4K
HEVC_VLD_Main: DXVA2/D3D11, SD / HD / FHD / 4K / 8K
HEVC_VLD_Main10: DXVA2/D3D11, SD / HD / FHD / 4K / 8K
20BB8B0A-97AA-4571-8E99-64E60606C1A6: DXVA2/D3D11
15DF9B21-06C4-47F1-841E-A67C97D7F312: DXVA2/D3D11
MPEG4pt2_VLD_Simple: DXVA2/D3D11, SD / HD / FHD
MPEG4pt2_VLD_AdvSimple_NoGMC: DXVA2/D3D11, SD / HD / FHD
9947EC6F-689B-11DC-A320-0019DBBC4184: DXVA2/D3D11
MJPEG_VLD_PureVideo: DXVA2/D3D11, SD / HD / FHD / 4K
VP9_VLD_Profile0: DXVA2/D3D11, SD / HD / FHD / 4K / 8K
VP9_VLD_10bit_Profile2: DXVA2/D3D11, SD / HD / FHD / 4K / 8K
DDA19DC7-93B5-49F5-A9B3-2BDA28A2CE6E: DXVA2/D3D11
6AFFD11E-1D96-42B1-A215-93A31F09A53D: DXVA2/D3D11
914C84A3-4078-4FA9-984C-E2F262CB5C9C: DXVA2/D3D11
4B2E068F-396A-43DF-B86B-7DC6D2B6726A: DXVA1
312B6A3A-41DE-4A43-9C0F-FB3D5F0DFB9A: DXVA1

[pacuro]

@Wanezhiling
I rather thought about GUID corresponding to VP9_VLD_10bit_Profile2 - try to point over this entry and rectangle with info will appear. This number I am interested in :)

[sneaker_ger]

DXVA_ModeVP9_VLD_10bit_Profile2 Profile
[...]
The associated GUID definition for the corresponding entry in the dxva.h header file is as follows:
// {A4C749EF-6ECF-48AA-8448-50A7A1165FF7}
DEFINE_GUID(DXVA_ModeVP9_VLD_10bit_Profile2,
0xa4c749ef, 0x6ecf, 0x48aa, 0x84, 0x48, 0x50, 0xa7, 0xa1, 0x16, 0x5f,
0xf7);
https://www.microsoft.com/en-us/download/details.aspx?id=49188

[wanezhiling]

http://i.imgur.com/IyNix27.png

[NikosD]

@wanezhiling

If you compare the video decoding speed of your card 1050 Ti with 1060 card and the benchmark results of Trevonn, does it have any performance difference ?

[wanezhiling]

http://jell.yfish.us/media/jellyfish-400-mbps-4k-uhd-hevc-10bit.mkv

I just checked this clip right now

Decode:
GTX 1050ti -59/73/93

Playback:
GTX 1050ti - 60/74/93

While this system is slow, X4 860K, 8g ram*1, no SSD...

I dont think GP107 is different with other Pascal in terms of performance, its just been added VP9 main10 ability by accident....

[NikosD]

Yes, the performance is exactly the same.

It looks like they enabled VP9 10bit only for GP107.

It's more like a commercial/marketing policy of Nvidia rather than a new decoder.

I can't believe they made two video decoders differing only in VP9 10bit decoding.

That would be insane.

But if I was a Nvidia owner of 1060/1070/1080 I would definetely post to their forums to find out what is going on.

[CruNcher]

We need to know what each off the Mobile Parts shows with the newest SDK based Driver.

The decision would be relatively strange indeed if none of the Mobile Parts supports it but surely it's not.

I dont think GP107 is different with other Pascal in terms of performance, its just been added VP9 main10 ability by accident....

The 3.5 GB decision for the GTX 970 wasn't either by accident it was planed also to keep it out of the press to allow a very competitive priced product to enter no one expected to see from Nvidia since the GTX 460.

Nothing happens by accident or you doing the wrong business and have failing strategies.

And i wouldn't be surprised if we see a coming Tegra Pascal supporting everything we talk about maybe a Shield 2 Nvidia needs to update their Android TV Platform it's a very important target and CPU is out of question for the Decoding complexity we are talking about here.

It is so much higher it's crazy and it will get even crazier in the future.

[NikosD]

If there are massive complaints in Nvidia's forums regarding the VP9 10bit addition to GP107 only, then maybe Nvidia could be forced to add it to the other Pascal cards or explain why it doesn't do it.

[CruNcher]

There is a simple mindset behind this and that is lying in the target system requirements in correlatation to the needed CPU Power for a given user target.

Gamer invest heavily into CPU Power and that is what 1080/1070 mainly target.

With this knowledge ahead you can plan wonderful Efficient strategies on your Design and Release Schedule choice and be very efficient with it, that's a very big strength of Nvidia ;)

Also intel is not so dumb as you might think intel releases a feature just in time the market needs it ;)

Nvidia calculates that the lower Cards wont be supported by a that Powerfull CPU and Hardware Decoding more essential and then they have their CUDA (Hybrid supported strategy they can at every time employ) to get at least minimal support (mostly VOD complexity requrements) on even the Gamer Cards ;)

Nvidia does it actually in a very very efficient way of thinking about it to save resources and Hardware space wherever they can to exactly reroute it where needed for their target customers.

Though currently it can be predicted that they mostly gonna to fail vs AMDs new Strategy if they don't adapt fast enough and people start to realize the problems of this approach.

Also as we see a continuously complexity increase and especially faster cycles it becomes questionable if the old ASIC approach still will hold up, thus AMD and Intel already experimenting with a FPGA approach (more release control faster time to markets).

People all the time arguing about coil whine though they don't realize how complex the whole thing really is economic/phsysical/technical and on what for a Edge we stand now.

[NikosD]

This is just an excuse.

If the decoder is exactly the same and you only need a driver to enable VP9 10bit, then the marketing segmentation of the GPUs by the drivers is awful and humiliating for the owners.

Why not a 1060 owner have VP9 10bit acceleration if he has already the hardware inside ?

[aufkrawall]

And the idea that it has been activated by mistake for a single product doesn't seem legitimate to your mind?

I'd rather have better VDPAU support (VP9 & HEVC 10 bit...) than VP9 10 bit for Windows, as long YT uses it only for HDR.

[CruNcher]

It is a possibility of course for a "bang look our customers get that support we are ready for it and now we have a surprise every Pascal supports it" ;)

I don't believe in it seeing all these footnotes though they bring you back to reality ;)

And also it would need the reviewers to redone reviews for bigger impact and yet i see none of this planned press feedback by Nvidia appearing, doesn't mean it not might show up soon ;)

But i would say there might be a chance that the Mobile Pascal parts especially in the lowest target range actually do support it fully in Hardware as well already.

And just about it it might be even that the whole Lineup actually has it inside but Nvidia decides where to make it available through the Driver sure they can do this they have the control but this is heavy speculation without someone proving it ;)

Nvidia has employed a lot of Software Locking into their Product Decisions and Strategies over the years so i wouldn't say this would be even impossible especially as hiding such a capability is a easy if switch but until no one proves that it doesn't exist ;)


If the decoder is exactly the same and you only need a driver to enable VP9 10bit, then the marketing segmentation of the GPUs by the drivers is awful and humiliating for the owners.

If you find that you should be rather scared or fascinated about the future ;)

[eddman]

There are a few possibilities:

1. GP107 uses a newer feature set video block, and nvidia ...
a) made a mistake in their documents and still named it H (instead of, say, H2 or I)
or
b) is repeating a 970-memory-situation and are giving GP107 a better video block without disclosing it to the public.

2. GP107 has the same video block as other Pascal cards and nvidia is ...
a) simply testing its VP9 10-bit functionality with 1050 cards before also enabling it for other cards
or
b) trying to give GP107 based cards an, IMO, unfair edge through drivers and locking other cards out.

Personally, I think and hope it's number 2.a), but they do have a history of doing sneaky things. We won't know for sure until someone asks in their geforce and developer forums or, better yet, a reputable website contacts them and asks for explanation.

I am thinking of maybe buying a 6GB 1060 and wouldn't like to have less video features when nvidia advertises same H video blocks and equal capabilities for all geforce pascal cards.

[NikosD]

OK. That's a good post.

I'm closer to 2b, unless there is pressure from the users.

Go to Nvidia forums!

[aufkrawall]

b) is repeating a 970-memory-situation and are giving GP107 a better video block without disclosing it to the public.

Did Nvidia ever promise VP9 10 bit for the big Pascal GPUs?
I don't think I've ever seen it on any feature slide.

[sneaker_ger]

Nope. That's why everyone is so surprised to have it show up in DXVA checker.

[CruNcher]

There is also another possiblity a very specific one actually can you guess which ;) ?

Okay Okay

how about the

Polaris Mobile vs Pascal Mobile review shootout

that will come sooner or later and should be very close actually and both preparing up ;)

b) trying to give GP107 based cards an, IMO, unfair edge through drivers and locking other cards out.

it would be only unfair if it really exists in the other if it doesn't it would be a perfect valid business decision based on the cards to low shadercount to compensate with CUDA and Systems with to low CPUs using that target

From a Tech Economic Geeks point it would be a ultra efficient decision, that every shareholder would be proud off :D

[NikosD]

We have been warned by Nvidia actually from the previous Maxwell generation.

The first surprise was with GM206 with a new decoder for Maxwell and full fixed-fuction acceleration of HEVC (8bit/10bit).

And the 2nd surprise was again with GM206 (an internal surprise inside the first surprise) that supported VP9 8bit too.

Sneaky Nvidia like eddman accurately said and tricky.

[CruNcher]

That is the gist of it since Maxwell Perf/Watt is crucial in every decision they do.

And that is firiing partly back on terms of stability when it comes to a really optimized multithreaded input/output pipeline.

Nvidia started to extremely bruteforce to get that stability back and the funniest thing about it they made it a trend to see higher clocks as something cool ;)

also the GM206 VP7 was introduced in Tegra X1 (Shield Android TV) first and then later arrived in GM206 and X1 Simultaneously ;)


Im pretty unsure if Nvidia would release another Shield TV now with AV1 so close to finish though ;)

Actually the foundation that we might see VEGAs UVD already with AV1 support might be not bad at all as the first Consumer Chip, though it's a very tight schedule to hit :)

[MisterXDTV]

I have been following this thread for months:

I just have some questions: I just bought a 1050ti that will be my christmas gift this year (haven't opened it yet).

Did I make the right choice for "next-gen" video decoding and eventual UHD Blu-ray playback? Will it support all DRM and copy protection required by windows 10 for streaming 4K on video services?

I read about another codec being released (AV1?). Will it become as important as HEVC and VP9 right away??

[CruNcher]

For now and sometime in the future you are safe ;)

[sneaker_ger]

1050 ti is a very good, if not the best choice at the moment. AV1 will not become as important as HEVC and VP9 right away because no one has any hardware for it right now. It doesn't exist yet. Netflix and Amazon 4K use HEVC, Blu-ray uses HEVC. Even Youtube pushed out VP9 profile 2 and Chromecast Ultra just a few months before the sheduled release of AV1. These companies won't block all their customers out just to push a new video codec. All of them still use H.264 as fall back, for example.
As for DRM: 1050 ti and other Pascal cards are supposed to work with PlayReady 3.0 which is the very latest DRM. We don't yet know why Netflix and Microsoft don't support them yet for 4K on Windows 10. Maybe just a new driver is needed.

So I will say:
- There is no better bang/buck for video decoding than 1050 ti at the moment. If you need a card right now: go for it.
- if you don't need one right now: wait. It isn't working with Netflix 4K yet. UltraHD Blu-ray doesn't work on PC yet. There isn't Amazon 4K on PC yet at all. Pretty much only Youtube 4Kp60 requires it (unless maybe if you have a very fast CPU). No need to rush. (But also: AV1 hardware could be another year or two off. If you are only waiting for AV1 but don't plan on waiting that long: buy the 1050 ti now)

[MisterXDTV]

Thanks for your answers....

In february I also got a Blu-ray drive that can actually read (but not play) UHD Blu-rays

Do you think an updated PowerDVD is all I need for playback, or will I have to wait for AACS 2.0 to be broken?

[sneaker_ger]

Probably "only" need a PowerDVD update.
But:
1.) they haven't announced any ETA yet
2.) they haven't announced prices yet
3.) they haven't announced hardware requirements yet

So again: lots of "ifs" and "whens". Similar for AACS 2.0: no one knows when it will be broken - or if at all. We don't even know whether UltraHD Blu-ray will survive as a format long enough...

[MisterXDTV]

On that subject, I know MadVR already converts HDR10 to SDR. But what about Dolby Vision in the future?

I read that Pascal can/will support 12 bit decoding but I still don't know whether DV is just a "software" thing or needs dedicated hardware not found on Pascal cards

[pacuro]

Update on https://developer.nvidia.com/nvidia-video-codec-sdk

" **** VP9 10/12 bit decode support is limited to select Pascal chips "

[nevcairiel]

Update on https://developer.nvidia.com/nvidia-video-codec-sdk

" **** VP9 10/12 bit decode support is limited to select Pascal chips "

That still doesn't answer any questions however, because it has a similar comment about 8K support, which we know is available on GP104, for example (just not on GP100)

[pacuro]

Still no VP9 10bit hw decoder detected in new 376.09 drivers in GP106. Due to same branch probably so nothing new.

[CruNcher]

As im very interested in Nvidias CUDA Video Research here some more information about their HEVC CUDA Decoder behavior and how it fails with DXVA on EVR-CP very heavily

http://forum.doom9.org/showpost.php?p=1787887&postcount=21387

This is also interesting in General when Benchmarking the Hybrid Decoders :)

Source Nvidia SDK

While Kepler and first generation Maxwell GPUs had one NVENC engine, certain variants of the second generation Maxwell GPUs and Pascal Generation GPUs have two/three NVENC engines physically present on the silicon.

Also they added (updated) footnotes again


* HEVC/VP9 10/12 bit decoding SW support coming in Video Codec SDK 8.0
** Max resolution support is limited to selected Pascal chips
*** VP8 decode support is limited to selected Pascal chips
**** VP9 10/12 bit decode support is limited to select Pascal chips

Support Matrix now available as well we seem now to be in a semi stable state with information available officially :)

[CruNcher]

Nvidia Official Support Matrix Kepler,Maxwell and Pascal

NVDEC (not taking resolution/performance abilities into the matrix @ all)

http://i1.sendpic.org/i/9Y/9YyblcoNV26IRHljJ2Uyss0YBfi.png

NVENC

http://i1.sendpic.org/i/uG/uGW0ZLjYj2Cxa9LQxvilqQnF0qX.png

GP107 not added yet (no pro Quadro/Tesla/Grid product based on it yet official) most probably all ready official that it has 3 encode/decode units and matrix will be updated when pro product arrives ;)

[P.J]

http://www.notebookcheck.net/fileadmin/Notebooks/Acer/Aspire_ES1-332-P91H/dxvachecker.PNG

Intel Apollo Lake's fixed function hardware decoder supports HEVC Main10 8K & VP9 Profile0 4K.

:goodpost:

[CruNcher]

This combination might become a crazy nice testing platform for analyzing Intel and Nvidia in a very similar environment :)

Kabylake + 1050 M

If the rumors are true and the 1050 Mobile will be released (based on GP107) it could have the new VPX like the 1050/TI.

Actually i really like the SOA of the 1050 TI it is a Perf/Watt monster with SOA Decoding capabilities of all for now important solutions (pre AV1) :)

This combined with Kabylake in a Mobile Platfom (yummy yummy)

[CruNcher]

Don't we love Marketing :D

http://cdn.wccftech.com/wp-content/uploads/2016/12/Radeon-Software-Crimson-ReLive-NDA-Only-Confidential-v4-page-043-copy-840x473.jpg

Its definitely VP9 Decoding over AMD's OpenCL accelerated Decoder R&D

http://cdn.wccftech.com/wp-content/uploads/2016/12/file-page10-copy-649x840.jpg

GTX 1060 (GP104) most probably wont do much more on the Encoding side as the GTX 970 (GM204) also does the 30 fps HEVC Limit (at reasonable 15 Mbit/s)

But the GTX 1050 TI GP107 should maybe really reach the 60 FPS Encoding with 3 units :)

So Vega could be also the first Consumer 4K Game 4K 60 FPS Recording :D

[nevcairiel]

That VP9 decode image makes no sense o.o

So Vega could be also the first Consumer 4K Game 4K 60 FPS Recording :D

ShadowPlay can record 4K at 60 FPS, if thats what you are trying to say. That line is rather confused.

[Trevonn]

Finally a free game recorder with hardware HEVC encode support. Your turn NVIDIA

[Trevonn]

That VP9 decode image makes no sense o.o



ShadowPlay can record 4K at 60 FPS, if thats what you are trying to say. That line is rather confused.

I think he means in HEVC

[CruNcher]

GameDVR
Shadowplay

and now ReLive

Massively hurt 3rd Parties as for example products like Action or Bandicam.

We will surely see a lot of these Products dieing in the Future if they don't buildup exciting features not presented by the OS and IHV solutions.

I think he means in HEVC

Nvidia could introduce it in no time they didn't yet which is quiet strange ;)

There would be 0 issues in their Ecosystem doing so up to Shield TV streaming no problem.

And the new Encoder core would even allow for High Quality Lossless HEVC for acquisition

Like you can already do H.264 Lossless for that purpose with Maxwell :)


Which both AMD doesn't support yet.

[Trevonn]

The only decent 3rd party option is OBS I don't think any of other 3rd party options will be missed especially the paid for ones.

The only problems with shadowplay is the lack of granularity in replay/record (Let me input a number!), variable frame rate encoding and the default colour space.

[CruNcher]

variable frame rate recording is no problem it's a feature that AMD neither supports ;)

It will be interesting to see if they push the new AMF with HEVC Encoder support right into GIT on the 8th December so if they sheduled it right on or if they still lose time on mass implementation ;)

@Trevonn

Why should the default colorspace be a problem with the target of Shadowplay being strictly Gamers as a Product and the playout happening to services which only support it :D

[Trevonn]

Regarding CFR, NVIDIA supports it in OBS and it helps with editing in premiere which doesn't like VFR video

I looked into the colour space issue a little more turns out it's not an issue haha nvm.

[aufkrawall]

Like you can already do H.264 Lossless for that purpose with Maxwell :)

Well, I messed around with all available encoder options in OBS with my 1070, and everything except of lossy 4:2:0 H.264 is just broken garbage.

[huhn]

decoding nvidia lossless h264 could be a simple decoding problem.

[aufkrawall]

Does it even support decoding H.264 lossless?

[sneaker_ger]

What is "it"?

[aufkrawall]

DXVA2, I guess.
I haven't payed attention if it was using hardware decoding.

[sneaker_ger]

If it was using software decoding: IIRC LAV H.264 lossless is hacky to account for broken x264 versions. If you upload a sample I can check.

[nevcairiel]

Does it even support decoding H.264 lossless?

NVIDIA hardware can support 4:2:0 8-bit lossless decoding on some GPUs, but its not exposed through any dxva2 decoder I know of.
4:4:4 or 4:2:2 is not supported through any hardware.

The 4:4:4 lossless decoding difference sneaker_ger talks of would result in mild image corruption only, and only with one particular type of coding tool in use, so its not guaranteed to be an issue with every lossless stream.

[sneaker_ger]

Yes, it is unlikely to be used. But also not difficult to check.

[huhn]

i guess the problem is more from the encode. nvidia hardware support the encoding of 4:4:4 lossless and if this stream results in garbage than one can assume nvidia is doing there own thing or OBS can't handle it properly.

[CruNcher]

i guess the problem is more from the encode. nvidia hardware support the encoding of 4:4:4 lossless and if this stream results in garbage than one can assume nvidia is doing there own thing or OBS can't handle it properly.

Im pretty sure BtbN never took High 4:4:4 Predictive Lossless into account really but the Profile works without issues exposed through the SDK and also Decoding works through CUVID at least, also decoding x264 High 4:4:4 Predictive Lossless works without problems.

But it's only really interesting comparing it for archival purposes and they're better solutions like FFV1 or Lagarith.

Though it will be interesting if in the future we can replace it with a VPx or AVx Lossless based option for archival and standardize those :)

[NikosD]

By far better drivers from AMD this year compared to Nvidia which had a lot, lot of problems.

Crimson Relive are extraordinary good, as I read to technical sites/ reviews.

This afternoon I will get a Sapphire Nitro+ RX 470/8GB and test the Polaris HW decoder in the next few days, even in VP9 (hybrid) after Crimson's Relive activation.

Stay tuned ;)

[aufkrawall]

By far better drivers from AMD this year compared to Nvidia which had a lot, lot of problems.

I think I can judge this better than you (owned a R9 390 for almost half of this year and else I've been with Nvidia) and I kinda disagree.

[NikosD]

There were numerous problems with Nvidia drivers and some of them were unusually "brutal" bugs for Nvidia.

For the majority of the users out there, AMD made a far better impression than Nvidia this year regarding drivers.

[huhn]

i running nvidia and AMD system over the past 5 years.

i have a "brutal" bug with AMD for 4 month plus now.

i have to disagree on the driver side too... this was a messy year for AMD driver...

[NikosD]

Nvidia was famous, once upon a time, for its drivers.

Actually Nvidia is a software company more than a HW company.

As the whole Internet says all over the planet, besides people here in this thread :) , this year was one bad or probably the worst year ever for Nvidia regarding drivers.

But, what the Internet and the technical sites know when we have so wise and skilled people around this thread...

[huhn]

nvidia had more flaws than usual that doesn't mean AMD was doing great at all.

there wasn't a major new driver bug that directly affected me using nvidia this whole year not a single one.

but i can instantly recall multiply problems and times where i had to skip AMD drivers.

openGL games like planetary annihilation where broken for month the bug was never added to the known issues.
madVR D3D9 windowed fullscreen is now broken for months on win 10.
and don't let me start about NGU polaris "performance" my R9 270 is most likely out performing my RX 480...
the removal of d3d9 openCL interop from the driver... great times!

and it is nice to read some where on the internet without personal experience.

[NikosD]

The opposite is quite true actually.

Most of the Nvidia users just skipped drivers or were waiting for the next one to fix bugs of the previous that was meant to fix bugs over a previous one.

It was a nasty situation and I'm not talking about video exclusively.

[huhn]

well them tell me your personal experience. it is not hard to find people with driver problems on the internet and there are currently way more nvidia user out there than AMD user so there are more people with problems.

and yes i know there where broken driver that got usually replaced in a couple of days.

i'm not here to tell "everything is all right in the green house" because it clearly isn't.

i mean for how many more years needs the CUVID decoder to die when any openCL program is used at the same time?

the difference is i don't have to skip one driver i have to wait months for a FIX from AMD or no FIX at all and a work around is created.

13 something is most likely still the fastest openCL AMD driver... it is just 3 years old...

[NikosD]

You have a serious issue concentrating on the specific subject we are talking about - 2016 drivers - like all the trolls actually.

So...

[Asmodian]

The opposite is quite true actually.

Most of the Nvidia users just skipped drivers or were waiting for the next one to fix bugs of the previous that was meant to fix bugs over a previous one.

It was a nasty situation and I'm not talking about video exclusively.

I didn't skip any Nvidia drivers, there were a few versions that had minor flickering on the desktop while at 144 Hz but it wasn't a problem at 120Hz and I don't remember any other issues. This flickering has been solved now. :)

I don't have any AMD cards to compare too but calling it a "nasty situation" is exaggerating it a lot, at least in my experiance.

But NGU is oddly slow on my Titan X (pascal) too, being slightly slower than NNEDI3 128. Maybe it is that NNEDI3 is oddly fast on this GPU but I doubt it. :p

[NikosD]

... on my Titan X (pascal) too...



Oh my God!

You threw 1500€ to buy Titan Pascal ?

For how much performance over 1080 ?

Do you also burn money to make heat in the winter ?

AMD needs desperately to release the top version of VEGA and Zen 8c/16t, there is no other to make real profit.

[huhn]

it not hard to find a broken NVIDIA driver: http://www.ghacks.net/2016/11/17/nvidia-375-86-driver-issues/

i mean nikosd tested them all! he must know! i mean this argument is "the internet said so". without links of cause. people that are using hardware in question are simply trolls no question.

[NikosD]

No, not people.

It's actually you and a few others writing here.

And I'm wondering, where are they ?

[huhn]

if you think everyone here is a troll.
and you still have nothing to bring on the table`i wonder what people call that... kind of sad isn't it?

[NikosD]

You make me repeat myself and that it's kind of sad actually.

It's only you and a few others.

I know that you live for commenting me whatever I write wherever I write here in doom9 but after a while it's not funny...

It could be scary actually.

[huhn]

so that how you avoid your own started topic of broken nvdia drivers.
you can block people on this forum just a reminder.

[NikosD]

Thank you for the last one, a good reminder.

[CruNcher]

By far better drivers from AMD this year compared to Nvidia which had a lot, lot of problems.

Crimson Relive are extraordinary good, as I read to technical sites/ reviews.

This afternoon I will get a Sapphire Nitro+ RX 470/8GB and test the Polaris HW decoder in the next few days, even in VP9 (hybrid) after Crimson's Relive activation.

Stay tuned ;)

Nvidia was famous, once upon a time, for its drivers.

Actually Nvidia is a software company more than a HW company.

As the whole Internet says all over the planet, besides people here in this thread :) , this year was one bad or probably the worst year ever for Nvidia regarding drivers.

But, what the Internet and the technical sites know when we have so wise and skilled people around this thread...

Nope you see it very wrong also totally unscientifically at all you don't understand how Nvidias Driver is optimized they only have major problems under Windows 10 nothing else the other WDDM work overly perfect and you can see maybe some problems in rather useless stuff like 3D Vision and G-sync but the core Driver HEART is rock solid overall.

Especially Nvidias Shader Compiler is still the best of the best ;)

And this has very understandable reasons Nvidia invested a lot in going above normal Driver optimization since NT 6 (and they circumvented Microsoft Decisions partly to give users a better overall experience) , but lot of things changed and Windows 10 is a more massive WDDM change that overall AMD was better prepared for also because of different other business related things around a very well known Gaming Console.

In terms of optimization the Arbitrator is very well optimized and tested inside Nvidias driver, they're only very specific situations when it gets problematic and that is when Compute (CUDA) and 3D tasks (OpenGL/D3D/Vulkan) pre emptively hit each other (you need shader power all brute forced trys to get performance will fail, this is how Nvidia scales Gameworks) ;)


Windows 10 is a Driver Minefield in Flux who ever uses that as a work System should be aware that some time will still go by with a lot of problems, there are also a lot of new Security stuff playing a role nobody will really tell you about and problems with different System configurations ;)

Im testing Nvidias Driver codebase on my own and im very picky changing it once im happy with it unless it really gives me a measurable benefit in my workloads.


So if you really want to compare Windows GFX Driver in a correct way please compare them over the whole WDDM timeline since VISTA NT 6 introduction, thank you.


You all speaking about Windows 10 WDDM 2.0/2.1 and don't understand the difference.


i can only lough about the driver issues shown for Nivida here as problematic issues ;)

Really Problematic it was in times of the Arbitrator tuning or introduction of G-sync though some stuff needs certainly adaption to WDDM 2.0/2.1 and new use cases such as Low Latency VR.

VR is also the part where it could get overly problematic for Nvidia old Card user at least pre Paxwell in the future and where AMD will definitely shine at lower cost continuously without user getting headaches, especially in a complete AMD HSA driven Platform like ZEN/VEGA combination in 2017 when the Fusion becomes full reality in the PC mass market after Consoles ;)

Nvidia will heavily need Volta first to be able to recover the loss so Paxwell is a pretty dead investment choice currently.

And Nvidia knows that very very well that's also why their current Paxwell release was the most panic driven ever.



The whole ReLive thing Hype is crazy nothing is really new it's just nice progress for AMDs Driver but it's nothing to go Crazy about other then Marketing if Reviewer really need that kind of advice that AMD is giving them how to test things in that leaked document they should better not review anything anymore ;)

[CruNcher]

Newest Mainconcept SDK Result on Sandy Bridge I5-2400

It survives this Broadcast scene as Cyberlink the only Decoder currently pretty nice resilience decoding result.

their error correction works pretty well in this high bandwith loss situations keeping a very nice smooth motion playback result overall.

http://i1.sendpic.org/t/fV/fVME0iSCTPAFHy8K2Y2wZO1JD34.jpg (http://sendpic.org/view/1/i/ksK0MwQJXxuNsEZLA5kMof9DFuC.png)
http://i1.sendpic.org/t/rw/rwZMs0XshniaaumGCDrsS2d04GC.jpg (http://sendpic.org/view/1/i/cbaknCphv6zOUHWBBxlNSwtqLir.png)

Psy and Decoding Performance wise a great Broadcast Decoding result.

And the non 4K Cameras (Good,Bad Signal results) are still clearly detectable.

Awefull Camera input Signal in the 4K Playout Mix

http://i1.sendpic.org/t/4s/4s4xpCiBu06vwCvqVhuQj5j8PGB.jpg (http://sendpic.org/view/1/i/gVwGOuxtOu7SkrJLxJXLLkOYbTY.png)

[NikosD]

The Nvidia's driver team is clearly in shock, but I don't know the reason.

The whole year almost every release had problems that they were trying to fix in the next release, not always with great success.

That kind of rush has no clear reason and looks amateurish, something that has never happened before for Nvidia.

AMD clearly won the race of 2016 drivers with Crimson.

[CruNcher]

I would agree for Windows 10 only yes but it's everywhere problematic from all points of views for everyone the anniversary update was a bigger core change like a SP back in times and that needed adaption again, especially on the rendering side ;)
Therfore we reached a very low latency OSD Desktop Layer rendering now via Aero since VISTA/WIN7 times :)


Cyberlink does still really well when we get to Low Latency Broadcast Decoding Efficiency :)

Can't wait for their VP9/AV1 results :)

Cyberlink

http://i1.sendpic.org/t/oI/oIcqkyxxIywqY6aJmmd5HQfSNsI.jpg (http://sendpic.org/view/1/i/dDXRHOJF4CAamEm1hlvgzqDsHMN.png)

Mainconcept

http://i1.sendpic.org/t/TN/TNXNWYcTZ0BEHN8R1vomBRptvG.jpg (http://sendpic.org/view/1/i/bYeEGpXQfdv4RFDNn5AsRwTB4CL.png)

I guess you can see the chroma difference very well in the optimized decoding mode ;)

[NikosD]

Hello again.

I used for a few hours a Sapphire Nitro+ RX 470 with 8GB VRAM testing video decoding performance and compatibility issues.

My test system is Core i5-2400 - Win 10 x64 14393.479 - RX 470 - Crimson 16.12.1 (ReLive).

Latest LAV Video 0.68.1-53 (actually I just saw there is LAV Video v0.69 released today, but with no difference than v0.68.1-53) and DXVA Checker v3.14

I also used MPC-HC v1.7.10.269 but with latest LAV Video v0.68.1-53 for playback.

The screenshot of DXVA Checker v3.14 is this:
https://s30.postimg.org/kycts049t/RX_470.png

Now, first of all, as you can see there is no VP9 device decoder yet, so there is no HW acceleration for VP9 using DXVA2 decoders.

Also, Chrome can't use VP9 acceleration for Youtube as of today with a Polaris card.
I don't think we are far away from this day, though.

You can also see that there is DivX (MPEG4-ASP) HW acceleration and MJPEG.
For DivX it's pure fixed-function HW acceleration for MJPEG is GPU based (shaders)

They are not particularly useful nowadays, especially DivX, but it is interesting that AMD provides MFT HW decoders for both, that can be used with DXVA Checker in order to accelerate in HW those codecs.

Regarding H.264/H.265 HW decoding, unfortunately I will not provide specific numbers for specific clips because I found out a very strange behavior of the benchmarking process.

1) The video engine seems to act completely independently from the rest of the chip, regarding monitoring tools.

So, during decode mode benchmarking, the GPU clock most of the times stays at idle clock (!) at 300MHz and the power consumption of the card at ~8W which is the same like idle (!!)

2) Most of the times in decode benchmark mode the performance was lower than in playback benchmark mode.

This is unreasonable and more obvious in codecs like DivX or VC-1 or MPEG-2 where you can have a difference of playback being 3 times faster than decode mode.

Regarding monitoring tools, there is a new (?) GPU utilization parameter exposed by the driver, different than GPU D3D usage, which I haven't seen in Intel iGPUs and it goes from 0% to 100% during benchmarking and playback, fluctuating a lot.

Firstly, I thought that AMD maybe for the first time exposed Video Engine utilization, but this is not the case.

I made a contact with DXVA Checker's developer (actually a lot!) and I'm waiting for an answer if there is a way to benchmark Polaris cards differently, maybe to find another way to put the video engine in a more accurate benchmark mode.

Also, I find out a very strange behaviour for clips that are corner cases for this decoder, like the Ducks 2160p H.264 clip with 375Mbps bandwidth and 50fps, that in realtime playback mode using MPC-HC it could reach ~43 fps dropping frames, but if you repeat the playback procedure could reach 50fps with no dropped frames.

But not all the times.
There is no consistent behavior for those corner cases.

So, after a lot of testing I decided to load my "monster" clip collection of H.264/ H.265 clips with bandwidths up to 1000Mbps and 50fps and these are my results:
(the H.265 clips have been encoded by x265 from uncompressed sources and the H.264 clips by HW H.264 encoder from uncompressed sources)

Decode mode (same or lower than Playback mode) using LAV for H.264 and MS MFT for H.265

2160p H.264

600Mbps at ~30fps

2160p H.265 8bit

600Mbps at ~42fps

2160p H.265 10bit

600Mbps at ~42fps

Please note that for H.264, LAV Video and Microsoft's DirectShow H.264 decoder were very close to performance, but for H.265 8bit/10bit, MS MFT (no DS H.265 decoder at least for my system) was clearly faster than LAV Video.

LAV Video H.265 decoder, after a few runs, could reach the performance of MS MFT but not from the beginning.

Also, it's clear that H.265 decoder is faster than H.264 and HEVC 8bit has no difference than 10bit regarding performance at least on the specific clips.

It is clear from the results above that Polaris HW decoder shouldn't have any problems decoding in realtime Blu-ray UHD.

Regarding compatibility issues, I had no problem at all decoding in HW progressive material of MPEG2, VC-1, WMV3, H.264, H.265 (8bit/10bit).

Also for interlaced content MPEG2, VC-1 were perfect.

The only problem is still for some H.264 interlaced clips, but it will be fixed soon as long as it is mentioned in the drivers to do list.

I will try PotPlayer too and if I find out differences, I will come back on the subject.

Waiting for your comments, after you read carefully this post.

[CruNcher]

Can it decode that TRT4K Euro 2016 Broadcast stream without any latency issues and motion brakeups ?
What about the Channel4K stream and Sony 4K (L)HDR Camp ?

[NikosD]

Which one ?

Do you have a link ?

I don't remember downloading that sample.

[CruNcher]

http://demo-uhd3d.com/fiche.php?cat=uhd&id=54
http://demo-uhd3d.com/fiche.php?cat=uhd&id=143
http://demo-uhd3d.com/fiche.php?cat=uhd&id=144
http://demo-uhd3d.com/fiche.php?cat=uhd&id=146

[NikosD]

From the description I can't see where is the difficulty on decoding those specific clips.

I have started to download the last one, as it is the smallest, but the server is painfully slow.

It gives me only ~140KB/s and unfortunately I have to go.

But I can guarantee that there is no reason of not playing back those clips easily, unless they hide a secret weapon.

[CruNcher]

Good stable Decoding hasn't only todo with how much bitrate i can move around :D

for fixed function decoder the problems lie generally in other parts than that surely Blu-Ray and UHD Blu-Ray should be no problem (and decoder core performance counts only) but Broadcasting is another beast to tame ;)

And AMD from my own point of view has bad credits there in the past in UVD development.

And they where the once that lost back then the whole opportunity to get feedback on Doom9 to begin with and Nvidia/Intel got massive feedback also on their CUDA/OpenCL Decoders from here and improved continuously.

[NikosD]

I have to admit that I'm not easily impressed by SW.

But that ReLive thing is amazing for the easiness and user friendly approach, so i gave it a try and recorded my desktop during real-time playback of the UEFA_Euro_2016_Opening_Ceremony.ts clip.

So, what you see is a real-time HEVC decoded file and at the same time encoded real-time in HEVC!

I couldn't find any issue during playback, but I don't know how you measure latency.
With Jitter ? Sync ?

I used MPC-HC with EVR-CP and statitistics on-screen (Ctrl-J) and below the file.

I have activated all of my monitoring tools at the right of the screen in order to pay attention to those too.

If you find any issues, tell me.

Here it is:
https://www.sendspace.com/file/fkm017

P.S

I used relatively low bitrate for HEVC encoding at 5Mbps and a resolution of 1080p and 60fps.
It's low for that type of HW encoding.

Any issues found out regarding the picture of the encoded file have nothing to do with the real decoded image.

I just wanted to keep the final video small.

Also, I noticed something like tearing in the video, which of course is not present in the real-time decoded file.

The video is just for watching figures, not the quality of the HEVC encoded video ;)

[CruNcher]

Do this please with the Channel 4K sample it pushes the CPU Multithread Decoding to it's maximum EDGE for 4/8 Cores :)

thank you for that realtime insight into the current UVD this kind of video are very rare in such quality also with our comparable system specs pretty nice :)

You would mostly only believe to see this for some heavy 3D computation or encoding but here its sliced 4K 10 bit HDR 60 FPS decoding at the CPUs maximum ;)

http://i1.sendpic.org/t/kB/kBbbRgFpYfROM6nIqH1SOpUSOV0.jpg (http://sendpic.org/view/1/i/eRxASbaKq6NDWBHH2fkQGUUUXUU.png)


Also nice to get a first view on their RePlay Encoder use, looks problematic that intra pulsing isn't so nice :)

Higher bitrate needed if the gop stays that way


Draw time isn't that nice overall but you have the whole timers running and using the MPC-HC fail implementation of the OSD Rendering.

So overall it has to fight heavier :D

[NikosD]

Too easy clip I think.

I captured only 3 and a half minutes from the beginning with some jumps too.

The recording is at 8Mbps this time.

Tell me if you want another part of the clip around 5 minutes max, in order for sendspace to accept it.

https://www.sendspace.com/file/0vgvcu

[CruNcher]

Could you try the same with VSR and MPC-BE

the Clocks looks decent also the RPM :)

I hope though not some Pascal user says now what above 1v my bla bla does 0,8v and 7xx mhz pretty constant which they should achieve with the more powerful VPX core indeed ;)

the Data is pretty comparable to my GTX 970 @ 4K H.264 60 FPS except heat conditions they obviously differ very heavy also RPM

[NikosD]

My native display resolution is only 1680 x 1050.

It's an old display for my backup system.

Using VSR what resolution you want the test to be done ?

[CruNcher]

UHD what else but you really have to higher the RePlay bitrate somehow :)

and don't forget changing to MPC-BE and using the 2nd statistics display mode to reduce overhead you get to much sync offset with all that timers running and overall jitter.

GPU Fan is really impressive :)

i see you almost at the stock clock config for the CPU hardly jumping at all though from the minimum power saving mode

hmm 2 GPU clocks that irritates a little though 2nd seems to react to the complexity try to overlay Afterburner on the EVR CP output this way you also reduce overall timer stress further and you can higher the timer resolution then without to much interception :)

Latency overall feels perfect sound is still pretty much in sync and that from a DXVA Native recording :D

there is only one part in the decoding result im not yet sure about one specific camera movement result but with all this current encoding noise it's rather crazy to make such assumptions anyway, also playing with 50 and recording the DWM surface with 60 is not so optimal.

http://i1.sendpic.org/t/wp/wp4xcyfl8wmAs21PyR5Zz1jdB9n.jpg (http://sendpic.org/view/1/i/pAJdXge8XW3EzFZi4XFtTbTMdpX.png)

Encoder/Decoder evaluation at once remotely :D

Interesting it keeped the 8:5 ratio alive :D

Breite : 1 920 Pixel
Width_Original/String : 1 680 Pixel
Höhe : 1 080 Pixel
Height_Original/String : 1 050 Pixel

That pulsing in the top area that also destroys the video isn't really nice at all but bitrate is seriously way to low for this Low Latency Encoder @ 60 FPS

My guess would be 15-20 MBps at least in that current configuration with the 60 fps target and we could talk about some decent visual result :)

but that 5-6 mbps is way to low for our 60 fps DWM Capture target here in it's current configuration.

Cant wait to see the 8mbps 4K Channel result :)

Saw it nice the Encoder though still cant hold it stable pulsing like crazy as i thought still bitrate way to low though now it goes up to 15 mbps actually :)

Fan is now slowly taking on :) 16xx rpm seems reasonable


Could you run some Vulkan/D12/OpenGL application while recording the playback with VSR at 4K :D ?

But please use MPC-BE on it's 2nd statistics display

So it looks like this

http://i1.sendpic.org/t/aP/aPg9K2qOSA8gT5KLjRg9S7UCenh.jpg (http://sendpic.org/view/1/i/23ktAA4W8r4bonqzDRx5P004HLE.png)

http://www.ozone3d.net/gpu_caps_viewer/

[NikosD]

Also, I find out a very strange behaviour for clips that are corner cases for this decoder, like the Ducks 2160p H.264 clip with 375Mbps bandwidth and 50fps, that in realtime playback mode using MPC-HC it could reach ~43 fps dropping frames, but if you repeat the playback procedure could reach 50fps with no dropped frames.

But not all the times.
There is no consistent behavior for those corner cases.

I will try PotPlayer too and if I find out differences, I will come back on the subject.



OK, there is an update about this.

It seems that most of the decoders have problems with that file and also a bigger problem with this file:
https://www.sendspace.com/file/en6bnw

It's a VFR at 120fps 2160p H.264 clip which both MPC-HC with latest LAV and MPC-BE v1.5.0 can't play fast enough using the Polaris decoder.

MPC-HC (LAV) can reach about ~80fps and MPC-BE it's worse at ~62fps using default EVR-CP renderer.

PotPlayer latest stable x64 1.6.63856 can reach 120fps without any problems using EVR-CP and can play that Ducks clip with 375Mbps with no problem at 50fps steadily.

As always, it's kind of magic what the developers of PotPlayer do and although they use DXVA from ffmpeg (LAV) they have different results.

To the Polaris owners I suggest to test the above 120fps VFR file with various players.



the Clocks looks decent also the RPM :)

the Data is pretty comparable to my GTX 970 @ 4K H.264 60 FPS except heat conditions they obviously differ very heavy also RPM



GPU Fan is really impressive :)

Fan is now slowly taking on :) 16xx rpm seems reasonable


Don't look at the RPM, the monitoring tool can't display 0 RPM which is the default most of the times and just stays at an arbitrary number or the last actual number.
Because Sapphire by default pauses fans if temperature is below 56 Celsius.

The monitoring tool displays real fan speed after >56 C and it's usually <700RPM

I have never seen anything above 900 RPM actually, not even in games (!)

The Nitro series has dual huge fans of 95mm each, so the noise and temperatures are low and can be customized by playing around with the fan speed.

They have an optimized airflow and hot air is pushed out through the vents on the backplate.

Take a look here:
http://sapphirenitro.sapphiretech.com/en/470-8.html#cooling


i see you almost at the stock clock config for the CPU hardly jumping at all though from the minimum power saving mode


Yes, it stays at minimum clock with minimum CPU utilization, like using any other HW accelerator actually :)

Later I'll try MPC-BE with its statistics.

[NikosD]

Really Problematic it was in times of the Arbitrator tuning or introduction of G-sync though some stuff needs certainly adaption to WDDM 2.0/2.1 and new use cases such as Low Latency VR.

VR is also the part where it could get overly problematic for Nvidia old Card user at least pre Paxwell in the future and where AMD will definitely shine at lower cost continuously without user getting headaches, especially in a complete AMD HSA driven Platform like ZEN/VEGA combination in 2017 when the Fusion becomes full reality in the PC mass market after Consoles ;)

Nvidia will heavily need Volta first to be able to recover the loss so Paxwell is a pretty dead investment choice currently.


Things are getting worse for Nvidia.

Microsoft's Win 10 Creators Update on March, introduces new WDDM 2.2

https://s27.postimg.org/m8gc2yzub/WDDM_2_2.jpg

[Yups]

https://community.amd.com/thread/203197

Is Interlaced decoding really broken on Polaris?

[aufkrawall]

In NikosD's world, surely not.

[NikosD]

https://community.amd.com/thread/203197

Is Interlaced decoding really broken on Polaris?

In NikosD's world, surely not.

In my yesterday's post I wrote:



Regarding compatibility issues, I had no problem at all decoding in HW progressive material of MPEG2, VC-1, WMV3, H.264, H.265 (8bit/10bit).

Also for interlaced content MPEG2, VC-1 were perfect.

The only problem is still for some H.264 interlaced clips, but it will be fixed soon as long as it is mentioned in the drivers to do list.


And my last phrase was:


Waiting for your comments, after you read carefully this post.

So, do you have some issues with your life or you are just trolling ?

There is no reason at all to pollute my thread with garbage!

Get a life!

[el Filou]

https://community.amd.com/thread/203197

Is Interlaced decoding really broken on Polaris?
This is just unbelievable. The lack of fixed function VP9 and general lower video decode performance is what made me lose interest in the RX460, but this issue still not being fixed after 6 months is much worse, it's like we're back ten years ago with the obvious video decoding bugs that take months to get taken care of. :(

[NikosD]

90% percent of potential buyers/users of a graphics card, don't even know what interlaced content is and you think someone cares about your "opinion" ?

This is more interesting news:
http://www.guru3d.com/news-story/amd-rx-460-unlocked-with-bios-update-offers-1024-instead-of-896-shaders.html

and of course this one too:
http://www.anandtech.com/show/10905/amd-announces-radeon-instinct-deep-learning-2017/2

VEGA is faster even than Titan X Pascal.

[el Filou]

90% percent of potential buyers/users of a graphics card, don't even know what interlaced content is and you think someone cares about your "opinion" ?
Oh, I'm sorry, I thought this was a video forum, but I seem to have lost my way and somehow ended up in the firmware hacking and machine learning section?

My "opinion" on this is relevant as some people who are interested in a graphics card for the particular task of video playback (which includes a lot of HTPC users) may want to be aware that it cannot reliably decode a decade-old format used in e.g. Blu-ray documentaries or broadcast TV, and this would make the point of how fast it is decoding HEVC (the topic of this thread) a bit moot if some other standard stuff plainly doesn't work, don't you think?
Or do you mean hacking firmware to get 14% more shader power or having a multi-hundred watt GPU being faster at machine learning are more relevant to video playback somehow?

[Yups]

This is just unbelievable. The lack of fixed function VP9 and general lower video decode performance is what made me lose interest in the RX460, but this issue still not being fixed after 6 months is much worse, it's like we're back ten years ago with the obvious video decoding bugs that take months to get taken care of. :(


I couldn't believe after reading numerous times here that AMD drivers were so great this year and Nvidia just garbage. But it seems this is still not fixed. Funy enough, this is called trolling. I wonder what NikosD does here, he is mainly bashing Nvidia with some offtopic nonsense. And he only accepts Nvidia bashing in his thread.

[huhn]

https://community.amd.com/thread/203197

Is Interlaced decoding really broken on Polaris?

i will try it again when and if i get my 480 back.
but there are reports with broken interlaced streams on NVIDIA and AMD and they all come down to stream error which are pretty normal in broadcast.

[VincAlastor]

Please note that for H.264, LAV Video and Microsoft's DirectShow H.264 decoder were very close to performance, but for H.265 8bit/10bit, MS MFT (no DS H.265 decoder at least for my system) was clearly faster than LAV Video.

LAV Video H.265 decoder, after a few runs, could reach the performance of MS MFT but not from the beginning.


Waiting for your comments, after you read carefully this post.

:thanks:

i would like to test MS MFT on a core m 5y71 tablet but i can't find it it directshow filters or Mshevcdec.dll in system. Could you tell us how to use MS MFT, pleae?

[NikosD]

The only way unfortunately is via DXVA Checker.

There are no Video Players (besides built-in Windows) that leverage MFT decoders and I haven't found any other tool besides DXVA Checker to test MFT decoders.

For Polaris owners right now I would suggest definitely PotPlayer with built-in decoders for HW decoding, because MPC-HC and MPC-BE don't look so optimized yet.

Probably the developers of those two popular players should buy a Polaris card some time ;)

[Aleksoid1978]

For Polaris owners right now I would suggest definitely PotPlayer with built-in decoders for HW decoding, because MPC-HC and MPC-BE don't look so optimized yet.

Probably the developers of those two popular players should buy a Polaris card some time ;)

MPC-BE is all be ok with Polaris.

[NikosD]

MPC-BE is all be ok with Polaris.

Not all actually.

As I wrote to my previous post, using MPC-BE v1.5.0 (build 2235) I can't reach 120fps (VFR) with this clip:
https://www.sendspace.com/file/en6bnw

MPC-BE manages to decode around ~62 fps and MPC-HC ~80 fps.

PotPlayer does 120 fps.

[CruNcher]

H.264 resilience MBAFF Decoding is no easy task at all and AMD had allways more problems with Broadcast Decoding since the beginning in UVD so i wouldn't be surprised from my own experience with it.

About Potplayer

lot of care are gone actually especially into Broadcast Decoding after picking up MPC-HC core and restructure and refactoring it

Potplayers DXVA is very nicely integrated into as a whole

Such a stream though is highly not related to consumer reality :D

though MPC-BE surely has no problem to reach 120 fps you can test that easily with the fast forward on DXVA and a fixed function decoder i already did speedups up to 120 fps with DXVA on it fixed function ;)

Not sure how you test it but it could have many reasons but overall it would be way to complex doing cross OS compares so i wont say anything being under Windows 7 still and WDDM 1.1 ;)

i only know that many things in MPC-BE are very nicely and sanely done and i like it for that in normal playback consumer scenarios, it also has rather nice analyzing capabilities especially the sync graph is pretty unique also compared to MPC-HCs because of the better OSD integration that is better layerd without impacting so much on the overall render performance, still waiting for Sync Graph only Statistics Mode though :)


here is some crazy experiment i tried lately with it ;)

http://forum.doom9.org/showpost.php?p=1789431&postcount=21459


Avg consumer doesn't benchmark how fast his player can render things he want's efficiency and a bug free undisturbed Playback (a clean Multithreaded base) experience with OSD (for efficient subtitle display functionality) without braking or impacting the Video Rendering to much :D

He want's performance in scenarios like stream switching lowest delays possible he want's stability in every situation, he wants a easy to understand user interface with fast interaction times and lowest task times, he just want's to enjoy his content as undisturbed as possible.


And he want all that with starting it and playing at best doing no crazy configuration for optimal Performance on his System base.



with 120 FPS you calling not the very AVG (more Gamer then anything else) users but that goes into another future direction of course Performance is important here as well but not yet widely needed for now, it doesn't really surprises me though that the Koreans (Daum) tries to reach that state fast because it also plays a Role for VR Playback Performance and such.

And that is they key of course if you have a 8K Decoder Monster at your back like Nvidia now you can really be more efficient in every case and if Shdaer Power will be able to compensate that is a nice question HEVC was better Designed with this Decoding Scenarios though in mind already VP9 not really and it will be interesting to see AMDs R&D results here on the VP9 OpenCL Decoding part via GCN vs Nvidias HEVC Cuda Decoder.

Of course you not gonna reach the Power Efficiency that way not in the distant future and that anyway cries for the FPGA move first, though more interesting will be if Intel can beat both AMD and Nvidia with Kabylake which im pretty sure they will hands down like always every year kill the Discrete results of both again (and once again their manufacturing leap will be the master key), but more important will be how it looks on the Mobile side of things ;)


I would never really build a HTPC where i make myself entirely dependent on a Discrete solution IP Core no matter from whom ;)

[el Filou]

i will try it again when and if i get my 480 back.
but there are reports with broken interlaced streams on NVIDIA and AMD and they all come down to stream error which are pretty normal in broadcast.
Some broadcast streams may have errors, but the particular sample posted in that forum thread to illustrate the issue on RX series (meteo.ts) decodes and deinterlaces just fine on my old HD 4650 with LAV in DXVA2 Native, and TS Doctor doesn't detect any errors on the video stream, so in this case it's really either the cards and/or the drivers.

[huhn]

not sure if ts doctor is able to detect small decoding problems and it doesn't matter broadcast has to work.

and my r9 270 works fine too.

[nussman]

not sure if ts doctor is able to detect small decoding problems and it doesn't matter broadcast has to work.

and my r9 270 works fine too.
Interlaced .ts files/Streams without errors (ts doctor) works just fine with "every" other hardware (dxva native).
Its clearly an issue with AMD Polaris cards or drivers ...

We need a new thread about media playback @doom9 without this NVIDIA vs AMD nonsense ... :rolleyes:

[NikosD]

There is a certain number of members of this forum that are trully writing non sense in this thread and believe it or not you are one of them.

As I have said in the past, I have to apologise to the normal people reading this thread that I don't have the power to just block those kind of people in order to write their crap in another thread or nowhere.

But please go ahead, open a new thread regarding playback and take with you the army of trolls who are so heavily grabbed by this thread and can't live without writing their crap here.

[Aleksoid1978]

Not all actually.

As I wrote to my previous post, using MPC-BE v1.5.0 (build 2235) I can't reach 120fps (VFR) with this clip:
https://www.sendspace.com/file/en6bnw

MPC-BE manages to decode around ~62 fps and MPC-HC ~80 fps.

PotPlayer does 120 fps.

I understand what's wrong. All because file have VFR. Video decoder can't correct calculate time stamp. That's all. If you make such file but with CBR 120fps - all be perfect.

[NikosD]

The OSD of MPC-HC and MPC-BE and the statistics below the playback, report those numbers which are both over 60 fps during playback.

It's 62 fps for MPC-BE and 80 fps for MPC-HC.

The OSD of PotPlayer reports 120 fps and you can figure out immediately that is faster just by looking at it.

Also, the Ducks 50fps file at 375Mbps is reported as 43fps during playback by MPCs.

PotPlayer of course plays that file at 50fps.

You can test that file which is 50fps < 60fps

[Aleksoid1978]

The OSD of MPC-HC and MPC-BE and the statistics below the playback, report those numbers which are both over 60 fps during playback.

It's 62 fps for MPC-BE and 80 fps for MPC-HC.

The OSD of PotPlayer reports 120 fps and you can figure out immediately that is faster just by looking at it.

Also, the Ducks 50fps file at 375Mbps is reported as 43fps during playback by MPCs.

PotPlayer of course plays that file at 50fps.

You can test that file which is 50fps < 60fps

Need a file with CBR for compare.

P.S. Here http://120hz.net/hypermatrix/120fpsvideo/bf.mp4 1080@120fps, play perfect.

[Aleksoid1978]

I check in DXVAChecker - Microsoft DTV-DVD Video Decoder, Microsoft H264 MFT, LAV Video, MPC Video Decoder.
Decoding - all decoder ~110-120fps.
Playback - all decoder ~130-150fps

So - "bug" not in decoder. Tested LAV/MPC Video decoders in Pot - play perfect. Maybe it's bug in MPC's video renderers, i don't know what's wrong.

[NikosD]

Need a file with CBR for compare.



You make me write the same thing two times or three in a row.

OK, so this is a CFR file at 50fps.

ftp://helpedia.com/pub/multimedia/x264/testvideos/2012%20-%2001%20-%20QuickSync%20vs%20UVD%202.2%20vs%20VP4/9.Ducks.Take.Off.1080p30fpsRef5-108Mbps.mkv

Can you play it flawlessly using MPCs at 50 fps and a Polaris card ?

Because with MPCs I get ~43fps while PotPlayer gives me always 50fps using EVR-CP and Polaris card.

[Aleksoid1978]

I write about it - not decoder issue, possible video-renderer.

[NikosD]

You wrote about a 120 fps VFR and I give you also a 50 fps CFR.

Don't you think it woths to take a deeper look ?

It looks like a more general issue for the MPCs, than just two cases.

Also DXVA Checker doesn't reach those figures you wrote with LAV but lower - haven't tried MPC-BE DS decoder.

How can I recommend MPCs for Polaris owners when PotPlayer is perfect for all my samples and MPCs obviously not ?

[Aleksoid1978]

You wrote about a 120 fps VFR and I give you also a 50 fps CFR.

Don't you think it woths to take a deeper look ?

It looks like a more general issue for the MPCs, than just two cases.

Also DXVA Checker doesn't reach those figures you wrote with LAV but lower - haven't tried MPC-BE DS decoder.

How can I recommend MPCs for Polaris owners when PotPlayer is perfect for all my samples and MPCs obviously not ?

As i say - something wrong with MPC video renderers. This files bad playback also on Nvidia 960.

[Aleksoid1978]

You make me write the same thing two times or three in a row.

OK, so this is a CFR file at 50fps.

ftp://helpedia.com/pub/multimedia/x264/testvideos/2012%20-%2001%20-%20QuickSync%20vs%20UVD%202.2%20vs%20VP4/9.Ducks.Take.Off.1080p30fpsRef5-108Mbps.mkv

Can you play it flawlessly using MPCs at 50 fps and a Polaris card ?

Because with MPCs I get ~43fps while PotPlayer gives me always 50fps using EVR-CP and Polaris card.

1 - link to wrong file(30fps) :)
2 - i found there with 50fps and on my RX460 it's play stable 50fps.

[CruNcher]

Indeed something is strange in Render behavior with that High Framerate VFR on DXVA MPC-BE EVR-CP it fixes itself on exactly 80 FPS over here but resources on VPX are still left enough, that's strange.

Something seems to lock it

Input: LAV Splitter Source

Decoder:

LAV VIDEO avcodec (jitter upto 20 ms) 1 thread

http://i1.sendpic.org/t/kn/knQtndOeNQctAh1f60jmoSjKLpb.jpg (http://sendpic.org/view/1/i/7LgxUCs6S19rbllVLMviVfA9GCS.png)

LAV VIDEO avcodec (jitter upto 20 ms) 2 threads

http://i1.sendpic.org/t/15/15JuLJYnOfD86eBENK4IZKbQVJN.jpg (http://sendpic.org/view/1/i/xEHXVF0ZEMYHN1PouA0N2bBOQR0.png)

LAV VIDEO avcodec (jitter upto 20 ms) 3 threads

http://i1.sendpic.org/t/vu/vuIZI0GlSFFoc3oe0QEOmX0BU9e.jpg (http://sendpic.org/view/1/i/vZFLARRP8hjikLRDMljSSallmmF.png)

LAV VIDEO avcodec (jitter in range of 18 ms) 4 threads

http://i1.sendpic.org/t/qV/qVsBOY6EhCdHxPgRnS044X0NKLv.jpg (http://sendpic.org/view/1/i/cP1VZqwvxeeMNoOAhpkP33YHklr.png)

LAV VIDEO avcodec (jitter in range of 12 ms) 5 threads

http://i1.sendpic.org/t/uZ/uZP8lN4rX879bpxBoi0uGJ1as5T.jpg (http://sendpic.org/view/1/i/rYnCNFJawRlE3GmTMJRuEbJAKoN.png)

LAV VIDEO DXVA Native (jitter 8 ms)

http://i1.sendpic.org/t/5m/5mvmN53Wv7SdyagDj1nyoYyrv0.jpg (http://sendpic.org/view/1/i/mhNIKD024W7x9fD6e6HupSBWYiE.png)

LAV VIDEO DXVA Copy-Back (jitter 8 ms)

http://i1.sendpic.org/t/4a/4awxOS1KiZ75smLZID5Ngc5E7qz.jpg (http://sendpic.org/view/1/i/deX3nL9C2rXVb0hX4ux3gokvFUO.png)

LAV VIDEO CUVID (jitter 8 ms)

http://i1.sendpic.org/t/7n/7nXeVuEJoLgHBBoBxy04y4GywNh.jpg (http://sendpic.org/view/1/i/qtGQAThnFKaAS3O3AcjX66f435T.png)


CUVID has no issues it seems once again with VPX in that case, though jitter still around 8 ms on my system lot of data being pushed on the (copy) subsystem.


Maybe it's really time moving 1 WDDM up ;)


Cyberlinks HAM

Still a really interesting combination implementation, though you really wonder what it needs that additional Shader Power for in the VPX Decoding combination (and obviously you lose Shader PP headroom for the Renderer overall this way) :)

http://i1.sendpic.org/t/uF/uFdXRvB5uQmoeALQboG3ggnrax.jpg (http://sendpic.org/view/1/i/75oviVpy4le6tlGYbtpXysMbNaZ.png)

Overal CUVID is still more efficient using it's mysterious Node 6 Engine

http://i1.sendpic.org/t/r4/r44OH0SJnGoOq3Ip0W2Vw20II7g.jpg (http://sendpic.org/view/1/i/haRhUFOlnnBLGInpKap4Sn1mTf3.png)

[NikosD]

1 - link to wrong file(30fps) :)
2 - i found there with 50fps and on my RX460 it's play stable 50fps.

Sorry, this is the clip:
ftp://helpedia.com/pub/multimedia/x264/testvideos/2160p%20samples/DucksTakeOff_2160p50.x264.CRF24.mkv

It sticks to ~42fps and using DXVA Checker I have the same result ~42fps

[CruNcher]

Overriding LAV VIDEO avcodecs auto threading decisions can improve bitrate spike drooped frame situations significantly :)

in both H.264/H.265 x264/x265 especially sudden encoder overflow situations

http://i1.sendpic.org/t/23/23qZ0HYj5AxO8QmrQpBR36j9zmV.jpg (http://sendpic.org/view/1/i/tTQF1z3b6GjfUVQtVRrmxn0kSCL.png)

Need to retest vs Lentoid :D

[nevcairiel]

Did you increase or reduce the number of threads? Auto uses Core Count * 1.5, which has shown in benchmarks to be the best, but for realtime exceeding the core count is probably not good, so I should maybe reduce that.
(Note that this doesn't affect HW decoding at all)

[Aleksoid1978]

NikosD
Tested on Polaris RX460.
DucksTakeOff_2160p50.x264.CRF24.mkv - MPC-HC/MPC-BE/Pot/Microsoft Video Decoders(DS/MFT) ~30fps
47.Crowd_run@120fps-60Mbps.mkv - as i said have wrong timestamp. I don't know how Pot's splitter handle this file(maybe it's ignore timestamp from file and generate it). Here fixed version(corrected timestamp) - 47.Crowd_run@120fps-60Mbps_fix.mkv (https://yadi.sk/i/74_5v5Kn33X3UL), play perfect ~120fps on MPC-BE/MPC-HC.

[NikosD]

NikosD
Tested on Polaris RX460.
DucksTakeOff_2160p50.x264.CRF24.mkv - MPC-HC/MPC-BE/Pot/Microsoft Video Decoders(DS/MFT) ~30fps


Not here.

30 fps ?

No way.

I get 43fps with MPCs and stable 50fps with PotPlayer.

Can you record your playback with ReLive ?

[Aleksoid1978]

Not here.

30 fps ?

No way.

I get 43fps with MPCs and stable 50fps with PotPlayer.

Can you record your playback with ReLive ?

Why record ?? I see what i see :).
Maybe you have RX470/480 and it's faster then my RX460.

P.S. ~30fps in MPC-BE/MPC-HC with statistics :)
Here records:
MPC-BE (https://yadi.sk/i/YhoyGx9033YuvM)
Pot (https://yadi.sk/i/OuFT-0Cw33Yuwp)

[CruNcher]

Did you increase or reduce the number of threads? Auto uses Core Count * 1.5, which has shown in benchmarks to be the best, but for realtime exceeding the core count is probably not good, so I should maybe reduce that.
(Note that this doesn't affect HW decoding at all)


Im currently swinging between 8 and 12 threads with 4 Physical cores for Edge cases on this System with MPC-BEs OSD Enabled.

The situations we talk about though are very specific when the CPU suddenly jumps from say 25% to above 90% and then in an instant back doiwn

Though DIAVC is still the most impressive it survives these overflows without really any visible issue in the AVC case.

But it's also easy visible why this is the case it never really reaches the 90% at all but stays at 75%, it never really reaches the critical area.

And yeah with 12 threads i can reduce it with Lav Video to 83%

[NikosD]

Why record ?? I see what i see :).
Maybe you have RX470/480 and it's faster then my RX460.

P.S. ~30fps in MPC-BE/MPC-HC with statistics :)
Here records:
MPC-BE (https://yadi.sk/i/YhoyGx9033YuvM)
Pot (https://yadi.sk/i/OuFT-0Cw33Yuwp)

Well, you have to re-examine your sight :D

Your first recording with MPC-BE starts at 50fps then drops to ~40fps and then rises again to ~48fps (!) looking at OSD statistics.

On the other hand, the below playback statistics reveal something more funny, a steady playback at ~60 fps (!)

What is going on ?

Very strange behavior but certainly not 30fps.

Are you sure you recorded the playback of the proper clip ?

Could you open the full statistics of the OSD in your next recording ?

[Aleksoid1978]

With full statistics fps ~25-30fps, the same as Pot.

[CruNcher]

GM204 reaches the 50 FPS as well with MPC-BE here

http://i1.sendpic.org/t/wH/wHrnezK4GTUy76uu74TiLAjbDZb.jpg (http://sendpic.org/view/1/i/bUwBqtk4bMp974Ret4LVwNTMVm9.png)

But statistics mode 2 full statistics is to much timer pressure i rarely use it ;)

And yes Aleksoid still using an older build (dont like what has happened to the Graph scaling)


And on Mobile i really like to avoid the seekbar ;)


Wow that is a pretty bad decoding result ?

was it cached or first play ?

That sample has some very insane I/O peaks up to 288 mb, better caching 1st to avoid possible false results from I/O congestion.

@nevcairiel

Highering the thread count smoothes out this issue as well

http://i1.sendpic.org/t/A6/A6TzCmzn6U6NFjXab6Sliak0rLo.jpg (http://sendpic.org/view/1/i/wf0jW8KnRy8EKNDRZszBb7jGL9h.png)

but still Lentoid is more efficient in that case not 1 peakdown with it.

Absolute undisturbed playback with Lentoid in thiose Ateme Encoder situations

http://i1.sendpic.org/t/mV/mVYvnRdT8OxdjbAhyMYImGahy1R.jpg (http://sendpic.org/view/1/i/2OejQmSS2pNWqRVTu1yzF554XqB.png)

With LAV Video (avcodec) it allways suddenly falls down to 21 fps

The Pattern looks intra frame relatied (scenecuts)

Wow Mainconcepts "Low Latency" Mode in their H.264 Decoder is pretty efficient 0 issues with the x264 encoder overflow best result together with DiAVC so far.

And it is also damn stable each run, but therefore they don't survive the extreme 4K 60 FPS sample so well even with deblocking off they don't seem to be able to maintain the performance needed for that extreme sample, 2 jitter fails then they stabilize and 1 jitter fail again without their deblocking overhead that's pretty strange ;)


Though i get render slowdowns i dont really recognize statistics wise at all but i hear them from the CPU Fan Reaction ;)

actually that's wrong you can see them in the jitter number the graph is not detecting them it's already the difference from 0ms to 1ms i pick up in that sample on its motion as heavy discontinuities (slowdowns).


The Graph seems not to have the resolution to show this 1 ms render difference efficiently.

Also the jitter counter reacts a little later on the slowdown after it passed by.

The Sync Video Renderer has overall also some nice properties the 3 jitter fails become immediately dropped frames and recognized as glitches

and it shows that 1 ms difference as a clear peakdown, perfect :)

This is indeed really cool where Mainconcept in EVR-CP showed the 3 jitter issues of peaks around 16 ms its now dropping frames entirely in those 3 areas with EVR-SYNC and Lav Video survives it with only 1ms peakdowns where on EVR-CP it showed no visible peaks at all in the Sync Graph.

Which in the end means EVR-Sync is even more sensitive and the room for errors like in EVR-CP becomes mega small and you gonna recognize even 1 ms discontinuities from the Decoder with it's Graph and big blow ups will result in dropped frames immediately displayed ;)

[NikosD]

UHD what else but you really have to higher the RePlay bitrate somehow :)

and don't forget changing to MPC-BE and using the 2nd statistics display mode to reduce overhead you get to much sync offset with all that timers running and overall jitter.




Could you run some Vulkan/D12/OpenGL application while recording the playback with VSR at 4K :D ?

But please use MPC-BE on it's 2nd statistics display



Could you try the same with VSR and MPC-BE


OK, I had two problems with VSR:

1) From my native and top resolution of 1680 x 1050, I couldn't go higher than 2560 x 1600.
That's the top resolution supported, no 2160p for my monitor.

2) When I switched to 2560 x 1600 VSR, I couldn't record in that resolution using HEVC.
The decoding frame rate dropped to ~4fps.

It works using AVC at that resolution and HEVC at 1080p.

So, I have two versions of VSR playback.
All of my desktop recordings are at 15Mbps this time.


Normal 1680 x 1050 resolution MPC-BE

1) Full OSD
https://www.sendspace.com/file/493mdg

2) Minimal OSD
https://www.sendspace.com/file/0wp41a


VSR 2560 x 1600 resolution MPC-BE

3) HEVC 1080p minimal OSD
https://www.sendspace.com/file/533rh1

4) AVC 1600p minimal OSD
https://www.sendspace.com/file/4ojrng

[CruNcher]

I will try to prepare the most efficient hevc encoding i can from my desktop this will give a very pretty interesting compare with our pretty close system specs it will be very interesting especially also from the OS Driver point of view between GM204 28nm VPX (NVDEC/NVENC) implementation and Polaris 10 RX 470 14nm UVD/VCE (optimized WDDM 1.1 vs still in optimization phase WDDM 2.0) :)

Obviously though we can't or better shouldn't do this with the H.265 Decoder base GM204 would be very badly limited their on the CUDA Decoder side and that is far far away from Playing that 4K Channel sample and would create to much overhead on the CPU anyways even with simpler complexity levels, also that it is officially not able to play that 4K Channel sample at all ;)

1) From my native and top resolution of 1680 x 1050, I couldn't go higher than 2560 x 1600.
That's the top resolution supported, no 2160p for my monitor.

That VSR is still limited Hardware wise is a pretty heavy let down though it seems 4K is only invested into the Fiji based High cost Platform i guess where it makes more sense.

Though now i wonder if VSR works at least on the RX 480 in 4K like it did previously on the 390x didn't it ?


2) When I switched to 2560 x 1600 VSR, I couldn't record in that resolution using HEVC.
The decoding frame rate dropped to ~4fps.

Very interesting information that you also hardly see in reviews :)

But reviews clearly show that the RX 470 is entirely optimized for the 1080p Edge not even really 2560x1600 overall, so it's not that surprising at all that it seems to brake down especially on 2560x1600 60 FPS HEVC Encoding a 4K output result scaled to that resolution @ 60 FPS at the same time which is heavy on everything, especially that whole thing Discrete in our case over the PCI-E 2 Bus you need some very efficient low latency compression ideas here and very efficient capturing from the OS Driver Level of course.

All in all enough RAW Power with clever ideas and perfect low level optimization not only limited to the Decoder/Encoder IP ;)

PS: The HEVC 2560 Encoding indeed looks very jittery especially in the ballerina scenes but the problem can be many causes all that timers running make me headaches

[baii]

Any info on vp9 acceleration enabled on latest radeon driver? Seems like the only thing tipping htpc user to the red side now other than price.

[huhn]

on they didn't add it and it is very unlikely that it will be a hardware decoder.

[CruNcher]

@NikosD

Could you try the Philips and the Channel 4k sample with EVR-SYNC once with the additional 3rd party hardware counter polling and once without (plain desktop) ?

http://demo-uhd3d.com/fiche.php?cat=uhd&id=135

http://i1.sendpic.org/t/nx/nx3QJ4doQK7WmeZlMvewMJf8ys9.jpg (http://sendpic.org/view/1/i/cKzTXemtZ95aEFo32jnooWGitoX.png)

[Yups]

I already received my i7-7700k which I replaced with my i7-6700k. I did some decoding tests from my 6700k before I switched to the 7700k, comparison between Skylake and Kabylake will follow.


http://fs5.directupload.net/images/161217/h6tn4jdp.png

[CruNcher]

Nice should be stomping AMD and Nvidia into the ground once again :)

Though RayZen might get the same Encoder/Decoder update like Vega.

and with DX12/Vulkan Renderer on the rise it could also get better for Realtime Video stability

[Yups]

i7-6700k 4.0 Ghz, HD 530 1.15 Ghz/i7-7700k 4.2 Ghz, HD 630 1.15 Ghz
Windows 10 x64, lav 0.69, 21.20.16.4551, DDR4-3200, CPU Turbo disabled


DucksTakeOff_2160p50.x264.CRF24.mkv (h264) 134 fps/134 fps
LG_4K_View-the-Feeling.mp4 (HEVC 8 bit) 190 fps/213 fps +12%
bbb-3840x2160-cfg02.mkv (HEVC 8 bit) 173 fps/191 fps +10%
5_crowd_run_2160p50_300Mbps.mkv (HEVC 8 bit) 125 fps/135 fps +8%
4_yvid_SAM_0235-UHD_Sample1.MP4 (HEVC 8 bit) 211 fps/235 fps +11%


H264 decoding speed didn't change, HEVC 8 bit is roughly 10% faster than before. A HEVC 10 bit comparison didn't make sense since Skylake doesn't support 10 bit in hardware. I will check this later with a GTX 1080 comparison.

[CruNcher]

Where is the RyZen Bitstream Su talked about we can get publicly that they used for the Handbrake Transcoding AVC Demo
http://www.amd.com/en-us/innovations/new-horizon

can't find it ?

@yups

even if it's hybrid 10 bit decoding it makes a lot of sense to compare it with that 20x time power reduction 10w to 0.5w advertised ;)

[Yups]

i7-7700k HD 630 1150 Mhz / GTX 1080 1960-2000 Mhz
Rest of the system same as #1609


DucksTakeOff_2160p50.x264.CRF24.mkv (h264) 134 fps/100 fps
jellyfish-250-mbps-4k-uhd-h264.mkv (h264) 141/103 fps

LG_4K_View-the-Feeling.mp4 (HEVC 8 bit) 213/227 fps fps
bbb-3840x2160-cfg02.mkv (HEVC 8 bit) 191/240 fps fps
5_crowd_run_2160p50_300Mbps.mkv (HEVC 8 bit) 135/148 fps fps
4_yvid_SAM_0235-UHD_Sample1.MP4 (HEVC 8 bit) 235/189 fps


10bit_Astra.SES.Demo.HEVC.ts (HEVC 10 bit) 183/238 fps
Eutelsat.Demo.HEVC.ts (HEVC 10 bit) 230/239 fps
10bit_Samsung_UHD_Ride_on_Board.ts (HEVC 10 bit) 208/221 fps
UHD_PQ_Lovely_Swiss.ts (HEVC 10 bit) 205/226 fps
jellyfish-400-mbps-4k-uhd-hevc-10bit.mkv (HEVC 10 bit) 81/83 fps


Out of 9 HEVC videos GTX 1080 wins 8 of them, averaged roughly 10% overall. Both are very fast. Intels HEVC decoding unit is really smart considering that Nvidia requires a 70% higher clock rate to achieve this result with Pascal.

As for H264 Intel is clearly faster, at least in these very high bitrate samples.

Are there any useful VP9 samples available?

[sneaker_ger]

"The World in HDR", VP9 8 bit, 2160p60, 24.9 Mbps:
youtube-dl (https://rg3.github.io/youtube-dl/download.html).exe -f 315 tO01J-M3g0U

"The World in HDR", VP9 10 bit, 2160p60, 18.2 Mbps:
youtube-dl.exe -f 337 tO01J-M3g0U

[CruNcher]

i7-7700k HD 630 1150 Mhz / GTX 1080 1960-2000 Mhz
Rest of the system same as #1609


DucksTakeOff_2160p50.x264.CRF24.mkv (h264) 134 fps/100 fps
jellyfish-250-mbps-4k-uhd-h264.mkv (h264) 141/103 fps

LG_4K_View-the-Feeling.mp4 (HEVC 8 bit) 213/227 fps fps
bbb-3840x2160-cfg02.mkv (HEVC 8 bit) 191/240 fps fps
5_crowd_run_2160p50_300Mbps.mkv (HEVC 8 bit) 135/148 fps fps
4_yvid_SAM_0235-UHD_Sample1.MP4 (HEVC 8 bit) 235/189 fps


10bit_Astra.SES.Demo.HEVC.ts (HEVC 10 bit) 183/238 fps
Eutelsat.Demo.HEVC.ts (HEVC 10 bit) 230/239 fps
10bit_Samsung_UHD_Ride_on_Board.ts (HEVC 10 bit) 208/221 fps
UHD_PQ_Lovely_Swiss.ts (HEVC 10 bit) 205/226 fps
jellyfish-400-mbps-4k-uhd-hevc-10bit.mkv (HEVC 10 bit) 81/83 fps


Out of 9 HEVC videos GTX 1080 wins 8 of them, averaged roughly 10% overall. Both are very fast. Intels HEVC decoding unit is really smart considering that Nvidia requires a 70% higher clock rate to achieve this result with Pascal.

As for H264 Intel is clearly faster, at least in these very high bitrate samples.

Are there any useful VP9 samples available?

Please test the 4K Channel and Sony Camp Demos ;)

And show also the CPU Cores utilization data and active nodes on the HD 630

could you please force Quicksync into Software Decoding mode and check how Kaby-Lake at all performs their with Intels Decoder, and also with Lav Video :)

And does it hold that 1150 MHz entirely also while rendering ?

Some data so far their seems really interesting was this a review test system or did you have other tasks running that could have influenced your measurements ?

You used DXVAChecker ?

Anyway that will get really interesting in the new low power atoms especially vs Tegra P on 4K 60 Decode/Encode :D

[Yups]

Please test the 4K Channel and Sony Camp Demos ;)


Do you have a link?


And show also the CPU Cores utilization data and active nodes on the HD 630

I can do but the CPU utilization is basically zero. I could check however the power draw from the HD 630 with HWInfo.


And does it hold that 1150 MHz entirely also while rendering ?

Yes

Some data so far their seems really interesting was this a review test system or did you have other tasks running that could have influenced your measurements ?

No review system, just my personal computer at home.


You used DXVAChecker ?

DXVAChecker 3.14 was used.

[wanezhiling]

10-bit VP9 clips: https://yadi.sk/d/mADzsqUnzEYfk

[NikosD]

Can you test the "World in HDR" with your 1050 Ti ?

[wanezhiling]

My system is 'broken' again, no decoders are shown in dxva checker....

[NikosD]

Nvidia drivers ? :D

[eddman]

Intels HEVC decoding unit is really smart considering that Nvidia requires a 70% higher clock rate to achieve this result with Pascal.

What's the GPU utilization?

I don't see how clock comparison is relevant when it comes to HW video decoding. IINM, video blocks do not directly rely on GPU's clock speed.

Even if they do, a higher clock speed doesn't necessarily mean it's less efficient. It's just one factor among many.

For example, a stock Radeon 480 runs at 1266 MHz typical boost and a stock GTX 1070 at 1683 MHz, yet the 1070 uses less power in games and is faster too.

[NikosD]

So many mistakes in just one post that I have to reply and because I liked your last post.

What's the GPU utilization?

I don't see how clock comparison is relevant when it comes to HW video decoding. IINM, video blocks do not directly rely on GPU's clock speed.


As I have clearly shown, Intel's IGPU ASIC we call QuickSync and it's responsible for video decoding is directly connected with iGPU clock and its speed is proportional to iGPU's clock.

The overclocking of IGPU is so ridiculously easy - you just increase the multiplier - and free for everyone even the non-K CPUs like Core i7-4790 and Core i3-4170 that I had and have.

I always posted my HD 4600 results at 1.5GHz and now my HD 4400 of Core i3-4170 is at 1.3GHz.

The speed of those Haswells in H.264 HW decoding is proportionally faster than Kaby at 1.15GHz



For example, a stock Radeon 480 runs at 1266 MHz typical boost and a stock GTX 1070 at 1683 MHz, yet the 1070 uses less power in games and is faster too.

All Polaris cards are artificially over-volted probably because AMD wanted them to be overclocking friendly.

My RX 470 had 1.093V and asked for 130W in a particular game.

By under-volting to 0.985 using Wattman - a free and extremely easy tool inside the drivers - the power consumption went down to 85W on the same game and I overlocked the card at the same time from 1260MHz to 1300MHz (!)

Polaris can become extremely power efficient if you want to and AMD just buried them in that matter.

[eddman]

So many mistakes in just one post that I have to reply and because I liked your last post.

As I have clearly shown, Intel's IGPU ASIC we call QuickSync and it's responsible for video decoding is directly connected with iGPU clock and its speed is proportional to iGPU's clock.


I was relying on some old information from my memory that mentioned video blocks ran independently. That's why I wrote "IINM". I was mistaking.

Thanks for the clarification, but what about power efficiency? Pascal runs at a higher clock speed, but how much power does the video block consume?

Perhaps a video playback runtime test is in order? EDIT: on laptops. :D Do you know people in this thread with kaby lake and Pascal laptops?

All Polaris cards are artificially over-volted probably because AMD wanted them to be overclocking friendly.

My RX 470 had 1.093V and asked for 130W in a particular game.

By under-volting to 0.985 using Wattman - a free and extremely easy tool inside the drivers - the power consumption went down to 85W on the same game and I overlocked the card at the same time from 1260MHz to 1300MHz (!)

Polaris can become extremely power efficient if you want to and AMD just buried them in that matter.

How about Pascal cards? Can they be under-volted too, or they become unstable?

EDIT: It seems they can be stably under-volted too.

https://www.youtube.com/watch?v=CVOhBwJZQTI

This is just one example though. I suppose it could behave differently in different games.

[Yups]

What's the GPU utilization?


From both GPUs? Will check later.


I don't see how clock comparison is relevant when it comes to HW video decoding. IINM, video blocks do not directly rely on GPU's clock speed.


There is no separate clock domain for the VPU on Intel GPUs, that's why people always refer to the GPU clock.

LG_4K_View-the-Feeling

HD 630 500 Mhz= 94,5 fps
HD 630 1150 Mhz= 213 fps

230% higher GPU clock results in a 225% faster decoding speed, scales almost lineary. Bandwidth is a small factor, that's why it doesn't scale lineary.



Even if they do, a higher clock speed doesn't necessarily mean it's less efficient. It's just one factor among many.


I'm not talking about power consumption figures, haven't checked this out yet. Clock per clock Intels decoder is much more powerful and faster. Of course the end result matters but If I could decide I always would prefer the bigger unit with lower clocks, it's usually more efficient and could have long term benefits if for some reasons Nvidias next gen clocks are down to Maxwell levels, or even current gen benefits. For example mobile Pascal variants may not clock that high. Intels fastest desktop KBL clocks at 1150 Mhz, mobile variants roughly at 1000 Mhz. Clock dependency is much lower for Intel.

[eddman]

There is no separate clock domain for the VPU on Intel GPUs, that's why people always refer to the GPU clock.

Could you under-clock your 1080 and see how it affects the video performance? If it's been done before in this thread, do you remember a post?

I'm not talking about power consumption figures, haven't checked this out yet. Clock per clock Intels decoder is much more powerful and faster. Of course the end result matters but If I could decide I always would prefer the bigger unit with lower clocks, it's usually more efficient and could have long term benefits if for some reasons Nvidias next gen clocks are down to Maxwell levels, or even current gen benefits. For example mobile Pascal variants may not clock that high. Intels fastest desktop KBL clocks at 1150 Mhz, mobile variants roughly at 1000 Mhz. Clock dependency is much lower for Intel.

That's a good point, although it seems that Pascal's laptop variants aren't much slower than their desktop models.

Obviously video playback power consumption doesn't matter for desktops but it is something to consider for laptops, although I think intel have the edge there anyway, being integrated and whatnot. It'd be nice to have some battery tests while playing videos at their normal frame rates.

[huhn]

it is very unlikely that a dGPU can beat a iGPU in term of power consumption. PCIe transfer the size of the GPU other stuff...

but if you do power consumption comparison stop wasting your time with software TDP/watt measuring tools.

[Yups]

Could you under-clock your 1080 and see how it affects the video performance? If it's been done before in this thread, do you remember a post?



I don't know how to do this. I noticed there is also a GPU Video clock in HWInfo for the GTX 1080. Let's use this video with a similar decoding performance for Kabylake and Pascal:

jellyfish-400-mbps-4k-uhd-hevc-10bit.mkv


Idle GTX 1080
GPU= 253 Mhz
GPU Video= 544 Mhz
GPU Power consumption= ~6W
Total system measured by Energy check 3000= ~50W

Decoding load GTX 1080
GPU= 1999.5 Mhz
GPU Video= 1708.5 Mhz
GPU Power consumption= ~51W
Total system measured by Energy check 3000= ~107W


Idle HD 630 (GTX 1080 is still attached as secondary GPU, means Idle power use from GTX 1080 is included here)
GPU= 150 Mhz
GPU Power consumption= less than 1W
Total system measured by Energy check 3000= ~50W


Decoding load HD 630
GPU= 1150 Mhz
GPU Power consumption= ~3W
Total system measured by Energy check 3000= ~61W


Idle-load delta is 11W for the entire system on my HD630 while on the GTX 1080 the delta is 56W.

[NikosD]

A lot of misunderstandings that I have to clarify


That VSR is still limited Hardware wise is a pretty heavy let down though it seems 4K is only invested into the Fiji based High cost Platform i guess where it makes more sense.

Though now i wonder if VSR works at least on the RX 480 in 4K like it did previously on the 390x didn't it ?



The hardware limitation of VSR it seems that it's not in the card but in the monitor, as I wrote in my previous post.

The first monitor with max 1680 x 1050 resolution went up to 2560 x 1600.
But my 2nd monitor with max 1920 x 1080 resolution went up to 3840 x 2160 using my RX 470

So, no problem using VSR


Very interesting information that you also hardly see in reviews :)

But reviews clearly show that the RX 470 is entirely optimized for the 1080p Edge not even really 2560x1600 overall, so it's not that surprising at all that it seems to brake down especially on 2560x1600 60 FPS HEVC Encoding a 4K output result scaled to that resolution @ 60 FPS at the same time which is heavy on everything, especially that whole thing Discrete in our case over the PCI-E 2 Bus you need some very efficient low latency compression ideas here and very efficient capturing from the OS Driver Level of course.

All in all enough RAW Power with clever ideas and perfect low level optimization not only limited to the Decoder/Encoder IP ;)


I have to clarify that when I started desktop recording of 2560 x 1600, I stopped it immediately when I saw the decoding of the clip at 4fps.

So, it's not the recorded video at 4fps but the decoding of the file when HEVC transcoding is used at VSR 2560 x 1600.

I'll check with my 2nd monitor at 3840 x 2160 the framerate of the recorded clip and post again.


PS: The HEVC 2560 Encoding indeed looks very jittery especially in the ballerina scenes but the problem can be many causes all that timers running make me headaches

There is no HEVC 2560 encoding, only AVC 2560 encoding.

The HEVC encoding of VSR 2560 is at 1080.

@NikosD

Could you try the Philips and the Channel 4k sample with EVR-SYNC once with the additional 3rd party hardware counter polling and once without (plain desktop) ?

http://demo-uhd3d.com/fiche.php?cat=uhd&id=135


I will do it on my 2nd monitor at 1920 x 1080, but since the card has gone to its owner, I don't know when I'll do the test.

What do you mean EVR-SYNC by the way ?

[eddman]

I don't know how to do this. I noticed there is also a GPU Video clock in HWInfo for the GTX 1080. Let's use this video with a similar decoding performance for Kabylake and Pascal:


Thanks for the test. You mean that when you under-clock the 1080, its video clock speed doesn't change?

Yea, comparing desktop power consumption isn't really useful, since there is an entire card that needs to be powered vs. an integrated solution.

I did know that there is a difference but didn't expect it to be this big.

What was the GPUs and also video engines loads? Could you also play the video at its native frame rate and compare that?

[CruNcher]

I don't know how to do this. I noticed there is also a GPU Video clock in HWInfo for the GTX 1080. Let's use this video with a similar decoding performance for Kabylake and Pascal:

jellyfish-400-mbps-4k-uhd-hevc-10bit.mkv


Idle GTX 1080
GPU= 253 Mhz
GPU Video= 544 Mhz
GPU Power consumption= ~6W
Total system measured by Energy check 3000= ~50W

Decoding load GTX 1080
GPU= 1999.5 Mhz
GPU Video= 1708.5 Mhz
GPU Power consumption= ~51W
Total system measured by Energy check 3000= ~107W


Idle HD 630 (GTX 1080 is still attached as secondary GPU, means Idle power use from GTX 1080 is included here)
GPU= 150 Mhz
GPU Power consumption= less than 1W
Total system measured by Energy check 3000= ~50W


Decoding load HD 630
GPU= 1150 Mhz
GPU Power consumption= ~3W
Total system measured by Energy check 3000= ~61W


Idle-load delta is 11W for the entire system on my HD630 while on the GTX 1080 the delta is 56W.

So we have a Decoding load on the Card Subsystem of 56W for 4K 400 mbps ?
Not sure if those measurements though are that reliable with a power wall plug measurement but obviously vs the dgpu efficiency no chance but as huhn said this was pretty clear from the beginning,

Though on P1 we only have 256 shaders that will lower power out under full load also a lot same for the atom version of the HD 630 it will have fewer EUs and the P1 will have the same Core Decoding state as the 1050 ti VPX.

They overall gonna calculate reaching those 60 FPS.


So it would be nicer if measurements are not done in Benchmark Mode but Realistic Playback Scenario (with 4K Render overhead) also then 400 mbps is insane out of avg consumer reality

Not 1 TV Manufacture releases a Demo at such a overall Bitrate complexity level , it's a nice test file surely but has 0 todo with reality, avg consumer reality starts with UHD Blu-Ray, Broadcast and Web not with jellyfish 400 mbps ;)

Many avg consumer will be already absolute happy being able to transcode in Realtime 3x realtime is surely nice but not really the mass market target, lowering power consumption to the maximum at those 60 FPS and one 4K Output is the next big goal :)


Avg Consumer (Sony, Microsoft, AMD, corporate Console target)
Gamers (Avg Consumer with Premium targets, also partly establishes even in the Console world) <- heavily exploded in prices and insane products over the last years a bubble to blow (mostly putting a led on something and selling it for a big uptake)
Pro Summer
Professionals

Gamers can also be devided in the sub target group of overcliockers and a new established target group "Streamers".

A lot of misunderstandings that I have to clarify



The hardware limitation of VSR it seems that it's not in the card but in the monitor, as I wrote in my previous post.

The first monitor with max 1680 x 1050 resolution went up to 2560 x 1600.
But my 2nd monitor with max 1920 x 1080 resolution went up to 3840 x 2160 using my RX 470

So, no problem using VSR



I have to clarify that when I started desktop recording of 2560 x 1600, I stopped it immediately when I saw the decoding of the clip at 4fps.

So, it's not the recorded video at 4fps but the decoding of the file when HEVC transcoding is used at VSR 2560 x 1600.

I'll check with my 2nd monitor at 3840 x 2160 the framerate of the recorded clip and post again.



There is no HEVC 2560 encoding, only AVC 2560 encoding.

The HEVC encoding of VSR 2560 is at 1080.



I will do it on my 2nd monitor at 1920 x 1080, but since the card has gone to its owner, I don't know when I'll do the test.

What do you mean EVR-SYNC by the way ?

yes that rescaled 1080p output it shows playback jitter that isn't recognized by the graph really EVR-SYNC is less forgiving in that it will either droop the frames and depending in how high and long the duration of the jitter that can become very critical upto displaying nothing at all or cause a peakdown if the jitter can be compensated fast enough as you see in the FPS difference for that 24 fps philips sample alternating up and down up and down on the 60 Hz refresh :)

you might see those jitter as a 1 ms difference but it's latency before display is bad the Graph is overall more reliable EVR-CP/EVR-SYNC recognizing these issues and here EVR-SYNC espeicialy as it will be less forgiving and show immediately a peakdown for every 1 ms discontinuity :)

Though in your case i guess the frames where dropped on the recording side not on the playback side at all like you also seem to recognize the 4fps render result being entirely recording efficiency related when scaling up to 2560 and recording it in native res with the HEVC Encoder at the 59 hz DWM output but as i said even the 1080p recording at that 15 mbps now shows issues but with that hwinfo and the graph display gadget all firing so many timers it could be very normal. also depending on the power management state, its Windows after all.

Though we shouldn't forget that it's Sandy Bridge and i think since Skylake intel took over it from Microsoft and could also improve counter reading overhead when frequency scaling or even core parking which was a big issue for Realtime 3D causing timer issues, i guess though you had core parking disabled ?.


I will show you where exactly i see issues @ playback so others might be able to confirm or i find a problem on my side ;)

We talking about

MPC-BE_Channel4K_VSR_to_1080_HEVC.mp4

just to be sure :)

i can see 3 very problematic areas that i cant pinpoint yet as playback issues on my side @ 59 Hz (actually not as timer issues, therfore its to stable repeating each run).

1) Choppy Head Motion

http://i1.sendpic.org/t/3L/3LYQki8MgeTQWkQBF9KxkWEuqve.jpg (http://sendpic.org/view/1/i/7Lh9O6BogLdZWy1q58Ume1I87sQ.png)

2) Choppy Air Throw

http://i1.sendpic.org/t/ry/ryJWALTaMpH6x6abxSohmVwYXOs.jpg (http://sendpic.org/view/1/i/dwR37dkPs4C6PMe7XC0Iv0kckU4.png)

3) Choppy Rotation Movement (not 100% sure about this though)

http://i1.sendpic.org/t/yy/yyMLnncysFcBNmqnpIHxgaqPYuq.jpg (http://sendpic.org/view/1/i/wo7bOWBH14NbL1gg6IVpTVhHNNp.png)

Dont be surprised it's nearest neighbour i want to be absolutely sure that everything is as low overhead as possible (internally).

The glitch count was a test between CUVID and avcodec first to run with Lav Video (avcodec) 2nd with Lav Video (CUVID) on the CUDA Decoder.

[Yups]

So it would be nicer if measurements are not done in Benchmark Mode but Realistic Playback Scenario (with 4K Render overhead) also then 400 mbps is insane out of avg consumer reality



Power consumption is much higher even in playback with MPC. Difference depends on the video, power consumption with HD 630 sits between 55-60W during playback with my sample videos. GTX 1080 has bigger fluctuations, 85W or even 100W in some 4K videos, smaller videos can be below these figures.

[huhn]

+50 watt is pretty bad...

[eddman]

Power consumption is much higher even in playback with MPC. Difference depends on the video, power consumption with HD 630 sits between 55-60W during playback with my sample videos. GTX 1080 has bigger fluctuations, 85W or even 100W in some 4K videos, smaller videos can be below these figures.

Really want to see its GPU utilization numbers. Theoretically, it should be very low.

If it consumes that much even for consumer level HEVC videos, then it sure is odd.

According to TPU's review, a stock 1080 consumes about 7W while playing a blu-ray, in a scene that goes up to 40 Mbps.

https://www.techpowerup.com/reviews/NVIDIA/GeForce_GTX_1080/24.html

If it's efficient while playing h.264, then why can't it do the same for HEVC? Did nvidia get HEVC decoding wrong?

Could you test one or two h.264 videos and measure the power draw?

[CruNcher]

Power consumption is much higher even in playback with MPC. Difference depends on the video, power consumption with HD 630 sits between 55-60W during playback with my sample videos. GTX 1080 has bigger fluctuations, 85W or even 100W in some 4K videos, smaller videos can be below these figures.

This was for DXVA Native zero copy on plain EVR ?

Anyway such Wall Plug measurements wont bring us far it's to fast switching we need higher resolution like only the most advanced reviewer have measuring at those.

Please show the Node usage of the GTX 1080 for LG View the Feeling and Crowd run

[nevcairiel]

I don't think comparing power usage of a huge high-performance GTX 1080 with a tiny iGPU is really a fair comparison in any sense (or even very meaningful). Just powering the GPU to run on medium or high clocks without load will use way more power then even the iGPU at full load.
It generally also depends what kind of player and video renderer you use, as that can have a huge impact on which power level the GPU runs in.

[CruNcher]

Sure more efficient would be a compare especialy with the 1050 which VPX will also land in the P1

At the End there will stay x86 ATOM Celeron vs ARM P1 ;)

Nintendo Switch could become a nice Media Player ;)

Or we might see a new revision of the Android Shield and Nvidias Streaming service update to VP9 or H.265, which still could be deployed backward compatible for both and theoretically a deal with Nintendo could also be arranged vs Sonys Streaming service :)

Though im not sure if Nvidia would get the Deal to get Nintendo Games on their Cloud Platform for PC Gamers (Inside the Geforce Experience Ecosystem) or even their Shield users technically obviously absolutely possible but i guess Nintendo wont do it for their "Exclusive titles" ;)

But Nintendow Switch surely will get a GameStream App to the PC that's as clear as kloßbrühe, interesting question will Nvidia supply it and slowly try to also reach AMD users with it ;)

[eddman]

It generally also depends what kind of player and video renderer you use, as that can have a huge impact on which power level the GPU runs in.

I was thinking the same and searching the web for some info. If Pascal's video engine sits in its own clock domain, then the GPU itself (shaders, etc.) should not have to run at full 3D clocks, if you're only playing a video with full HW decoding.

The card alone consuming 30-50W power just for HW decoding doesn't seem right.

Sure more efficient would be a compare especialy with the 1050 which VPX will also land in the P1


So far all Pascal chips (except for GP100) have the same feature set H video engines, unless nvidia made a mistake in its documents. I really doubt 1050 has a different video block. The VP9 profile 3 support that we are seeing is probably due to drivers, just like the missing VP8 support for GP102 based Quadro and Tesla. Don't know about Titan X P, since there is no info on nvidia's website.

[CruNcher]

I was thinking the same and searching the web for some info. If Pascal's video engine sits in its own clock domain, then the GPU itself (shaders, etc.) should not have to run at full 3D clocks, if you're only playing a video with full HW decoding.

The card alone consuming 30-50W power just for HW decoding doesn't seem right.



So far all Pascal chips (except for GP100) have the same feature set H video engines, unless nvidia made a mistake in its documents. I really doubt 1050 has a different video block. The VP9 profile 3 support that we are seeing is probably due to drivers, just like the missing VP8 support for GP102 (Titan X).


Feature Set aren't worth much footnotes are, though i agree with you the power measurements look insane so far.

Because it wont work at all efficient in the Mobile Pascal Platforms and they would be empty in no time ;)

As im not able to measure such high resolutions i myself prefer measuring heat and fan speed and actual noise levels instead and their correlation ;)

As the PWM Systems are very fast reacting lower then the OS level and reading those performance counters doesn't influence so heavily at least via NVAPI if you don't over pressure the pooling ;)

VPX was in the beginning a 400 MHz clocked domain by now it should have roughly reached 500 MHz

Btw i really like to compare Nvidias SOA with Cheap under 100$ Chinese ARM SOC IP solutions ;)

[eddman]

Feature Set aren't worth much footnotes are

Please use punctuation marks. Reading your comments are rather hard.

Yes, the footnotes say "limited to select Pascal chips", but we don't know which chips, and even then, it might just be a driver limitation.

I think it'll all become clear once Video Codec SDK 8.0 is released. If it turns out that GP107 actually has a better video block, then I'd be somewhat pissed at nvidia for misleading us.

[Yups]

Really want to see its GPU utilization numbers. Theoretically, it should be very low.

If it consumes that much even for consumer level HEVC videos, then it sure is odd.

According to TPU's review, a stock 1080 consumes about 7W while playing a blu-ray, in a scene that goes up to 40 Mbps.

https://www.techpowerup.com/reviews/NVIDIA/GeForce_GTX_1080/24.html

If it's efficient while playing h.264, then why can't it do the same for HEVC? Did nvidia get HEVC decoding wrong?

Could you test one or two h.264 videos and measure the power draw?


I can see this low power draw in several videos in my collection, also several with an exorbitant high power draw. As far as I can see it really comes from the Video Engine because GPU load stays very low. HWinfo log is messy unfortunately and GPUz doesn't display/log some things like GPU Video clock or GPU Power (TDP only).

[CruNcher]

you could lower your overhead using Process Hacker and it's Node View for looking up the domain states utilization

What for a GTX 1080 you have exactly ?

Are you running in a GPU Mixed config or is the Nvidia Card the Primary (intel igpu bios disabled) ?

Your Power Management setup CPU/GPU ?

OS i guess Windows 10 ?

Which samples give you the highest power draw highest VPU utilization ?

LGE View the Feeling should be pretty darn low same for NikosD AMD VCE recordings, the DIVX Profile samples also very low.

Dont forget the 400 mbps and have you not seen samples also gonna suck power from other subsystems ;)

If you want to be sure only 1 part is fixed consuming and avoid extra latency move the entire sample into the Ram (Ramdrive) or on a SSD always prefer the ramdrive though.

[NikosD]

"The World in HDR", VP9 8 bit, 2160p60, 24.9 Mbps:
youtube-dl (https://rg3.github.io/youtube-dl/download.html).exe -f 315 tO01J-M3g0U

"The World in HDR", VP9 10 bit, 2160p60, 18.2 Mbps:
youtube-dl.exe -f 337 tO01J-M3g0U
Very nice way to download VP9 10bit from YouTube and probably the only way if you have a non HDR monitor like me and the majority of people.

Because when Youtube sees a non HDR monitor automatically offers VP9 8bit only streams, so it's impossible to download the 10bit streams from Youtube directly.

[Yups]

LG_4K_View-the-Feeling power draw is low after 5-10 seconds of playback. In contrast UHD_PQ_Lovely_Swiss is way higher.

[mariush]

I'm using this Firefox extension : https://addons.mozilla.org/en-US/firefox/addon/youtube-video-and-audio-dow/?src=api
Adds an orange button under the videos, when clicked it overlays all the video and audio options over the video and allows you to select specific formats.

The version I downloaded 2160p 60fps is about 460 MB (~458 MB + 2 mb opus track automatically selected by extension to mux with the video)

Is it the HRD one or the 8bit one?

How would the html 5 code of the player detect that you have a HDR (or non-HDR) monitor installed, is there even some way for the browser to retrieve that information and give html players or javascript code on some page acccess to that information?

--

I should get an XFX RX 470 in a few days, waiting for the courier to drop it off.
If anyone will be interested by then, I can post power consumption numbers as I have a kill-a-watt type of device and a decent power supply (Seasonic X-650, 90+ efficiency), so the measurements should be pretty accurate.

Could be interesting as I have two monitors with different resolutions (1920x1200 for primary, 1920x1080 for secondary) and in the past from what I heard, there were problems in drivers like increased power consumption at idle when two monitors with mismatched frequencies or resolutions were used.

[NikosD]

The one with that size is the 8 bit.

Use the youtube downloader app to download the 10bit version.

I don't know how they recognize the monitor.

About RX 470, don't forget to undervolt it and measure again.

[sneaker_ger]

I don't know how they recognize the monitor.
Pretty sure they don't. The client is responsible for advertising its codec support (or only accessing the codecs it supports - however their JS logic works). I doubt any PC browser will do that at the moment. They expect you to buy a Chromecast Ultra and connect it to a HDR10-compatible HDMI display or use a SmartTV with the feature built-in.

[P.J]

We need some benchmark charts :rolleyes:

[CruNcher]

Can
RX 470
GTX 1080
Intel Kabylake


Playback this without any issues ?

https://mega.nz/#!ilgxWaJA!sjqwfIWuCknm4o_mQibD_xRHFV7yliiVP0-jSCLv05I

[Trevonn]

Can
RX 470
GTX 1080
Intel Kabylake


Playback this without any issues ?

https://mega.nz/#!ilgxWaJA!sjqwfIWuCknm4o_mQibD_xRHFV7yliiVP0-jSCLv05I

The video is longer than the metadata says (22 secs), there's some small corruption within the 1st second of playback but after that the video seems to plays fine. This is on a GTX 1060 with MPC-HC. I don't think it was encoded 100% correctly

[Aleksoid1978]

File encoded incorrectly - video resolution change some seconds after start.

[baii]

Have a rx 460 in htpc, but vp9 decode is no where to be found with latest driver:confused:

Do the 1060/70/80 does hardware vp9?

[huhn]

even some maxwell cards have VP9 so yes they do.

[mariush]

I had some spare time to download some of the videos mentioned in the last few pages on this thread and put them on one of my dedicated servers. Here's download link (FTP, anonymous) : HIT ME (ftp://helpedia.com/pub/multimedia/testvideos/)

The anonymous transfers may be limited to 1 MB/s per IP but resume is supported and from memory I think i capped the ftp server at 3 simultaneous connections per IP ... anyway, point is you can feel free to download them as long as you won't complain they're not blazing fast downloads.

I've also set up an upload ftp account, in case someone here wants to submit more test files. I'll keep it active for a couple of days or something like that.

Host : helpedia.com User : testvideos Password : upload

Everything uploaded will show up in the _uploads folder and if the files are good i'll move them in main folder eventually.

The account supports creating folders and appending to files (to resume uploads if your connection is crap) but no deleting (to prevent people from deleting other users' videos if more than one people care to upload). You can pm me if you screwed something and want it deleted.
So the idea is that if someone has an interesting test file or finds a good one, i wouldn't mind getting a copy uploaded there.. or have someone upload it there and link to it from here, instead of sites like mega which sometimes ask for registration or have loads of ads on them.

// There's about 60 GB of free disk space on the drive with that folder but i can make more room if need be.

[CruNcher]

File encoded incorrectly - video resolution change some seconds after start.

I don't think so nothing states that it is out of spec changing anything at all at encoding runtime and the decoder could accept it.

Many of the Sat test streams change all kind of properties from one sample to the next.

It would be mainly the fault of the demulitplexer, decoder and renderer if they can't do it in a efficient manner without braking and dynamically adapting to the different stream properties signaled appropriately at runtime.

[nevcairiel]

Many of the Sat test streams change all kind of properties from one sample to the next.

Test streams do that, but actual consumer broadcasts typically do not, because its not a seamless transition. All sorts of things need to re-init to change properties like frame size or bitdepth, which is just going to disrupt playback and possibly glitch a bit.

In this case all you get is a bit of lagging/hiccuping, which is common in such a case because generally the decoder and the renderer need to finish with the old sequence before they can start the new one, so any pre-buffering goes out the window - and you get a hiccup.

[Yups]

The World in HDR_VP9

This 10 bit VP9 video is demanding, even with hardware decoding with my GTX 1080 there is a CPU load of ~20% on my i7-7700k at 4.2 Ghz. I don't see such CPU load numbers with HEVC videos.

[nevcairiel]

The World in HDR_VP9

This 10 bit VP9 video is demanding, even with hardware decoding with my GTX 1080 there is a CPU load of ~20% on my i7-7700k at 4.2 Ghz. I don't see such CPU load numbers with HEVC videos.

The GTX 1080 doesn't have 10-bit VP9 hardware decoding, and if you are decoding on the Intel GPU and transfering it to the 1080 then you have a bit of load just from that.

[CruNcher]

Could be the Cuda Decoder, as i said it would also make a lot of sense for the higher shader count cards to save the asic logic (or just do it when the asic is not ready yet the cuda decoder gives faster target hit times) this is actually a fascinating strategy, not new at all though ;)

Now if we get the same result for the GP107 and it shows lower utilization overall it would be pretty confirmed ;)

@yups

Which actuall Nodes get utilized on it ?

Btw 20% CPU is also what i see on the HEVC Cuda Decoder on it's peak possible decoding complexity on GM204 at 1266 on a i5-2400 though

and i also correlated the overall decoding power vs that 4 cores on software alone (avcodec) the Decoder is from that point weaker overall then the CPU alone but obviously more power efficient then the old CPU generation at certain complexity stages (use cases) alone ;)

[Yups]

The GTX 1080 doesn't have 10-bit VP9 hardware decoding, and if you are decoding on the Intel GPU and transfering it to the 1080 then you have a bit of load just from that.


Does it have a Hybrid decoder? It should have because disabling hardware decoding results in over 90% CPU load.


CruNcher, what nodes?

[nevcairiel]

Does it have a Hybrid decoder? It should have because disabling hardware decoding results in over 90% CPU load.


Not with current drivers. (unless its hidden somwhere where I have not found it yet)

Which decoder/hardware decoding mode are you using? LAV Video should be able to tell you which GPU its running on, for example.

[CruNcher]

Does it have a Hybrid decoder? It should have because disabling hardware decoding results in over 90% CPU load.


CruNcher, what nodes?

GPU Nodes from the Windows NT 6 API (Longhorn (Vista) upward)

http://www.geeks3d.com/20111205/process-hacker-2-25-gpu-monitoring/

You can see all of them either via Process Hacker or Process Explorer.

https://msdn.microsoft.com/en-us/library/windows/hardware/dn265449(v=vs.85).aspx

http://www.ozone3d.net/public/jegx/201107/process_explorer_gtx460_gt240_furmark.jpg
http://www.ozone3d.net/public/jegx/201112/process-hacker-2-gpu-usage.jpg

[Yups]

Not with current drivers. (unless its hidden somwhere where I have not found it yet)

Which decoder/hardware decoding mode are you using? LAV Video should be able to tell you which GPU its running on, for example.


There is no iGPU involved, looks like CPU decoding with lav is much stronger than I expected.

[CruNcher]

This LG View the Feeling sample makes me headache why is it named as .mp4 and in reality a transport stream.

What is also really interesting a big bunch of all samples that are currently floating around are actually from Atemes Titan Encoder Core then we have Fraunhofers/HHI Encoder and some others that are rather hard to identify :)

Samsung absolutely exclusively using Atemes Titan

LG pretty much also the latest LG Chess which is the absolute top new version of the HDR Encoder ATEME Titan KFE 3.7.1 (4.7.1.0).

I wonder if LG View the Feeling was done by LG @ all the way it floats around in that x265 version ;)

also makes no sense why LG should use x265 1.3+206-199e8f2e0d54abd1 • Compiling by snayper [x265.ru]

looks like a user transcode but from which original LG input source ?


@Yups

It is nice to see some results of it to correlate to from your Kaby-Lake 20% is nice :)

Not undisturbed realtime Playback Capable @ 60 FPS (lot of jitter parts where cpu dies) with that complexity level here on the i5-2400

http://i1.sendpic.org/t/vx/vxY2hkbjDLkCyHFQt5F46xdZB9p.jpg (http://sendpic.org/view/1/i/7hQpVigALfJnTJjaYVdsbo1bb9K.png)

50 FPS might work though :D

could you limit the core count to 2 and let us see the result ? :)

@nevcairiel

i think you don't really want to expose decoding tricks (cheating) via the lav video gui to its users ?

[NikosD]

By using the registry hack from this thread https://community.amd.com/message/2768536 you can force the use of VP9 OpenCL acceleration as the registry key says: (change the value from 0 to 1)

HKEY_LOCAL_MACHINE\SYSTEM\ControlSet001\Control\Class\{4d36e968-e325-11ce-bfc1-08002be10318}\0000\UMD\DXVA\VForceOCLVP9


The DXVA Checker says:
https://s28.postimg.org/u4yxc6w59/RX_470_VP9v3.jpg

LAV doesn't seem to support VP9 acceleration using this hack but PotPlayer drops the CPU utilization of Sandy Core i5-2400 to ~30% even in the most demanding VP9 4K 60fps clips.

The problem is the framerate is very low 7 fps to 13 fps and the GPU utilization/clocks not very high.

Probably AMD has a reason of not activating yet the VP9 HW decoder.

Let's hope that when they do it, it will work OK even as a clearly GPU and not HW based decoder.

[NikosD]

New 16.12.2 driver fixed:


Chromium may fail to utilize hybrid decode for VP9 content.


but it's not working according to the reports.

I'll give it a try.

[NikosD]

After a clean install the registry "hack" disappeared (that's a good sign that the installation of the new driver was really clean) but unfortunately, still VP9 acceleration is not working for the latest Chrome v55 x64 even using Crimson 16.12.2.

The "World in HDR" gives a 90+ % CPU utilization of Sandy Core i5-2400.

[sneaker_ger]

Maybe it just sucks? Open a second tab "chrome://media-internals/" while playing a Youtube VP9 video. Click the blob and look for "video_decoder". "VpxVideoDecoder" is software AFAIK.

[NikosD]

Using chrome://gpu it says hardware accelerated.

It was the first thing I checked.

[sneaker_ger]

But is chrome://gpu showing the current status of the specific video or only showing the general availability of accelerated VP9 decoding?

[NikosD]

The general availability...It means that hw acceleration is present and ready to be used (and compatible)

Now, why it's not used is another question.

It's a bug probably or not ready for release.

[Yups]

VP9 decoding doesn't work either with Chrome 55 for Kabylake, but it works with Chrome 56 Beta. Edge works out of the box, Firefox doesn't support it in version 50. LAV also doesn't support it. Works out of the box with MPC-BE, no idea what decoder they are using.

[sneaker_ger]

The general availability...It means that hw acceleration is present and ready to be used (and compatible)

Now, why it's not used is another question.

It's a bug probably or not ready for release.
So what does "video_decoder" say in the respective chrome://media-internals/ blob? You still didn't say.

[NikosD]

The card is not mine.

I do some tests when I and the owner have some time.

I'll try again.

[NikosD]

Maybe it just sucks? Open a second tab "chrome://media-internals/" while playing a Youtube VP9 video. Click the blob and look for "video_decoder". "VpxVideoDecoder" is software AFAIK.

Yes, it says "VpxVideoDecoder" while using AVC codec says "GpuVideoDecoder"

It looks like SW mode for VP9.

BTW, Edge gives me always the H.264 stream, not VP9.

[NikosD]

GPU Nodes from the Windows NT 6 API (Longhorn (Vista) upward)

http://www.geeks3d.com/20111205/process-hacker-2-25-gpu-monitoring/

You can see all of them either via Process Hacker or Process Explorer.



No GPU module and some others I see in your post. Only CPU, Memory, I/O.

I don't have a discrete card, only Core i3-4170 and the built-in iGPU HD 4400.

I run the portable version.

[huhn]

So what does "video_decoder" say in the respective chrome://media-internals/ blob? You still didn't say.

GpuVideoDecoder:
https://abload.de/img/vp9amdgzu79.png

[NikosD]

From the above link of the AMD forum someone says that this is the solution for fluid playback of VP9 4K 60fps using Chrome and Youtube:

chrome.exe --enable-accelerated-vpx-decode=2

I'll try it tomorrow though.

[huhn]

works fine. with the advanced parameter and bad without but both are using GPU decoding.
UHD 60 hz VP9 is doable with it.

[CruNcher]

Is the Rendering mainly working over ANGLE via Chrome so wrapped DX11 ?

They had allways a real benefit with it even supporting down to XP with 3D Acceleration with it when Mozilla failed ;)

it will be interesting when slowly Vulkan takes over the Rendering overall :D

How does the overall utilization looks like via Chrome currently on the World in HDR sample with Crimson driver on the RX ??? ?

and are the 60 FPS also holdable in a window running within Chrome (not fullscreen) scaled to a slightly lower resolution or the entire Render surfface as a background video playing for example ?

[huhn]

i can't properly test full screen because webbrowser are still unable to deal properly with high DPI screen.

i can't watch the HDR version and external video are not hardware decoded for what ever reason.

GPU usage is unreadable on the RX 480 like always. it jumps between 0- 100% like crazy

[CruNcher]

Most probalby only over mediasource for DRM Reasons maybe over the Protected Media Path which might add some more overhead currently, most probably AMD will only release it if they absolutely sure it works reliable encrypted over the trusted zone.
On diffferent capable setups so far only Intel has the only reliable working solution via their AES stack and Netflix enabled that for Kaby-lake Windows 10 but only on EDGE so far.

Especialy on the discrete parts latency problematic getting everything right and if we ad Hybrid to it even more problematic ;)

[huhn]

my TV can't do it anyway.

Vp9 profile2 isn't listed anyway.

and the h264 blocking issues are removed from the known issue.

[CruNcher]

and the h264 blocking issues are removed from the known issue.

Nice can someone confirm it works ?

[huhn]

i can confirm it doesn't work.

and i can't find the issue here: http://support.amd.com/en-us/kb-articles/Pages/Radeon-Software-Crimson-ReLive-Edition-16.12.2-Release-Notes.aspx

[Aleksoid1978]

H.264 interlaced MBAFF still picture corrupted with DXVA.

P.S. RX460 - force registry key "VForceOCLVP9" to "1" don't work for me. Windows 10 ENT x64, 16.2.2 Crimson.

[CruNcher]

Hmm so maybe a mistake removing it ?

Even the current cheapest Kaby Lake should reach reatime like YUPS I7 with marginal higher Utilization for the World in HDR

also

i5-7400
i5-7600

should look well in the current CPU Decoder state already :)

The cheapest Kaby Lake vs ZEN @ same price will be very interesting :)

[NikosD]

BTW, Edge gives me always the H.264 stream, not VP9.

In order for Edge to use the VP9 stream you have to do these steps:

1.Open Edge and type "about:flags" in url bar.
2.Find "Enable VP9 video format" and set it to "Always On".
3.Restart Edge then open the online videos you want.

[huhn]

H.264 interlaced MBAFF still picture corrupted with DXVA.

P.S. RX460 - force registry key "VForceOCLVP9" to "1" don't work for me. Windows 10 ENT x64, 16.2.2 Crimson.

maybe a performance reasons a 480 is about 2.5-3 times as fast as a 460.

[NikosD]

H.264 interlaced MBAFF still picture corrupted with DXVA.

P.S. RX460 - force registry key "VForceOCLVP9" to "1" don't work for me. Windows 10 ENT x64, 16.2.2 Crimson.

The first one is true, AMD hasn't said yet that H.264 interlaced bug is fixed.

We have exactly the same platform: Windows 10 x64 Ent - Crimson 16.12.2

By using the registry hack and your player (MPC-BE) I can see that DXVA is used according to the stats (Ctrl-J) and the low CPU utilisation, but the performance is awful.

The card works at highest state 7 (1300MHz/8000MHz) with a GPU utilisation of 80% - 96% and it barely reaches ~10fps for 4K VP9 clips (8bit)

Exactly the same situation with PotPlayer.

On the other hand, even without the registry hack, if you run chrome with the above parameters you can watch 4K30fps in window mode with <30% CPU utilisation of Core i5-2400.

Edge, on the other hand, when you force the use of VP9 stream with or without the registry hack, fails to use hybrid acceleration - it's always SW.

Also, if you do the registry hack you can run chrome without the extra parameters and still have GPU (hybrid) decoding.

In all cases that Chrome works with hybrid acceleration, media internals say "GpuVideoDecoder"

As a last comment I would say to give AMD a little more time to see if they can give general availability of hybrid VP9 acceleration to all browsers and Media Players with proper speed.

Chrome seems to work, but it's the only one so far and not out of the box.

[baii]

Can confirm the chrome thingy work on rx 460 and this is on 16.12.1.

I remember pot player internal dxva decoder can auto do vp9 at low framerate in an older driver.

So in all, no youtube + madvr as of yet for AMD.

[CruNcher]

It's a pity we don't see Ateme in any Encoder Compare anymore they really rock the boat hard on HEVC even more so as they did already on AVC :)

They also continuously improving their PSY Core, i love their very unique psy work results they improved them a lot since the AVC Encoder and their complexity balancing is awesome even in the early core, hard contender and all that in a very rough Realtime environment :)

http://i1.sendpic.org/t/dX/dXYCQ1FZxLtfQdHx8oH1QvMbPmz.jpg (http://sendpic.org/view/1/i/gvLe9jrnhT67ejpifsevMIZkVqf.png)

Their 10 bit path Strobing Volumetric results are just awesome to watch in motion, Broadcast (Realtime) limited awesome work.

http://i1.sendpic.org/t/lN/lNCQpe06bkYEr24CBOhvf9gTiio.jpg (http://sendpic.org/view/1/i/vipN8qiUymWDaVCeWa62w16wmSa.png)

http://i1.sendpic.org/t/yb/ybRgJUVK8uxFGoKQOCtifWFH1mw.jpg (http://sendpic.org/view/1/i/tIVNb6pl487oldrOgsACQHcWUsk.png)

But as the Borg said once upon a time "Resistance is Futile" ;)

[NikosD]

P.S. RX460 - force registry key "VForceOCLVP9" to "1" don't work for me. Windows 10 ENT x64, 16.2.2 Crimson.

Try this one:
chrome.exe --ignore-gpu-blacklist --enable-accelerated-vpx-decode=2

[Aleksoid1978]

Try this one:
chrome.exe --ignore-gpu-blacklist --enable-accelerated-vpx-decode=2

I meen using DXVA for VP9. Chrome don't interesting for me :)

[huhn]

DXVA doesn't work for me too.

[MisterXDTV]

Just bought my GTX1050ti but MPC-HC can't hardware decode files with HVC1 codec ID works fine with HEVC ID

What am I doing wrong??? what's the difference between HVC1 and HEVC???

[sneaker_ger]

Which decoder are you using? Are you using LAV Video? Use "DXVA2 (copy-back)". If it still doesn't work post MediaInfo and sample.

[MisterXDTV]

Which decoder are you using? Are you using LAV Video? Use "DXVA2 (copy-back)". If it still doesn't work post MediaInfo and sample.

I resolved remuxing the mp4 file to mkv.... :o:o

The codec ID of the mp4 file was not recognized by LAV Video

[NikosD]

The codec ID of the mp4 file was not recognized by LAV Video

Why don't you post the sample to LAV's video thread in order for the developer to check it out ?

[CruNcher]

@MisterXDTV

im not yet stumbled above that fourcc at all but

fourcc.c

search

A("h265"),

and on the end add

A("hvc1"),

this should then let ffmpeg recognize it as supported and let it through.

@All

Lentoid partly still performs better then lav video, especially when it gets really rough on the CPU Multithreading Edge on 4 Cores really nice for a 2015 Decoder with only 1 major decoding issue :)

http://i1.sendpic.org/t/2X/2XpDnqIyRYx5DxDUN3WsIhqE6IJ.jpg (http://sendpic.org/view/1/i/vJFQd9s4pmgvfSsiFxgbFSWytKN.png)


currently torturing 160W (250W 3D Engine) stable out of my I5-2400 system (complete i/O everything for video playback) with all 4 cores under maximum load while rendering on EVR CP so i should be pretty maxed out at the 95W tdp :D

the LG Chess Demo should be added.


Idle state (after caching, cycle overhead mostly Firefox related, though causes no visible CPU breakdowns)

http://i1.sendpic.org/t/uM/uM6JWTFOj44dkDAiZ1BfUjzzMsA.jpg (http://sendpic.org/view/1/i/y5f4FiShIwdZkvcyh1RSKmPyZGe.png)

Playback Result (Decoding/Rendering)

http://i1.sendpic.org/t/uy/uy2ONC28q4NfaBANa2r82hw4SyV.jpg (http://sendpic.org/view/1/i/kbOlJXnoiYyTezTj0OXLKzykIhp.png)

@Yups

How does this look on your Kaby-Lake currently with Lav Video (avcodec) and compared to the Quicksync Decoder (Via DXVA and via Quicksync) ?

[MisterXDTV]

Why don't you post the sample to LAV's video thread in order for the developer to check it out ?

It's the Sony 4K HDR Demo here:

https://mega.nz/#!0VMWhRoD!6hR2-lRoYKqO1Dpf7ylHNna6z1MIuNO4ByySxUIp-E4

Check it out



General
Complete name : ok_Sony_4K_HDR_Camp.mp4
Format : MPEG-4
Format profile : Base Media
Codec ID : isom (isom)
File size : 1.12 GiB
Duration : 2 min 7 s
Overall bit rate mode : Variable
Overall bit rate : 75.8 Mb/s
Encoded date : UTC 2016-02-03 08:01:30
Tagged date : UTC 2016-02-03 08:01:30

Video
ID : 1
Format : HEVC
Format/Info : High Efficiency Video Coding
Format profile : Main 10@L5.1@High
Codec ID : hvc1
Codec ID/Info : High Efficiency Video Coding
Duration : 2 min 7 s
Bit rate : 75.6 Mb/s
Maximum bit rate : 123 Mb/s
Width : 3 840 pixels
Height : 2 160 pixels
Display aspect ratio : 16:9
Frame rate mode : Constant
Frame rate : 59.940 (60000/1001) FPS
Color space : YUV
Chroma subsampling : 4:2:0 (Type 2)
Bit depth : 10 bits
Bits/(Pixel*Frame) : 0.152
Stream size : 1.12 GiB (100%)
Encoded date : UTC 2016-02-03 07:59:49
Tagged date : UTC 2016-02-03 08:01:32
Color range : Limited
Color primaries : BT.2020
Transfer characteristics : SMPTE ST 2084
Matrix coefficients : BT.2020 non-constant
Mastering display color primar : R: x=1.000000 y=1.000000, G: x=1.000000 y=1.000000, B: x=1.000000 y=1.000000, White point: x=1.000000 y=1.000000
Mastering display luminance : min: 0.1000 cd/m2, max: 0.5000 cd/m2

Audio
ID : 2
Format : AAC
Format/Info : Advanced Audio Codec
Format profile : LC
Codec ID : 40
Duration : 2 min 7 s
Source duration : 2 min 7 s
Bit rate mode : Variable
Bit rate : 192 kb/s
Maximum bit rate : 197 kb/s
Channel(s) : 2 channels
Channel positions : Front: L R
Sampling rate : 48.0 kHz
Frame rate : 46.875 FPS (1024 spf)
Compression mode : Lossy
Stream size : 2.91 MiB (0%)
Source stream size : 2.91 MiB (0%)
Language : English
Encoded date : UTC 2016-02-03 07:59:49
Tagged date : UTC 2016-02-03 08:01:32

[nevcairiel]

That file plays just fine here, hardware decoding and all (in both native and cb modes), must be something on your end.

[CruNcher]

Ahhh so it was already added jesus actually mostly the Sony are hvc1 i wonder these is the encoder Lentoid 2015 has problems with it seems
if Channel 4K matches that it gets interesting.

Codec-ID : 36

huh


Sonys colors Demo has the same problematic with Lentoid all use Slices it seems

Codec-ID : hvc1

hmm which encoder is this ?

Now i wonder if all these Codec ID 36 where written on the stream capturing part from some hardware device.

But yeah Lav Videos Parser/Decoder doesn't really care about these ids and other parser let them through as well so the ID seems not the only issue.

also older Intel Platforms have Problems with these sliced streams

https://communities.intel.com/message/433628



What i find really crazy

http://xhevc.com/en/hevc/decoder/download.jsp

they say no HEVC 10 bit decoding but lot of HEVC streams shown as 10 bit work only these sliced Sony Demos and other Japanese Test streams make real problems compared to lav video (openHEVC)

[MisterXDTV]

That file plays just fine here, hardware decoding and all (in both native and cb modes), must be something on your end.

As i said, remuxing to MKV container changed the ID to V_MPEGH/ISO/HEVC and that alone fixed the problem. It took me 5 seconds..

The LG Chess was an MP4 with the proper hevc ID and worked fine directly without remuxing

[CruNcher]

No its no .mp4 lot of these demos have wrong file extensions but in general that shouldn't be really a major problem
Most of these Demos are Satellite Captures or coming from Device Demos that where on the Devices themselves for Showroom Demo purposes.
Some are actual x265 retranscodes from uknown source.

Some could be also legit x265 tests in cases of LG but LG widely uses ATEMEs Titan Encoder same as Samsung

[MisterXDTV]

No its no .mp4 lot of these demos have wrong file extensions but in general that shouldn't be really a major problem
Most of these Demos are Satellite Captures or coming from Device Demos that where on the Devices themselves for Showroom Demo purposes.
Some are actual x265 retranscodes from uknown source.

Some could be also legit x265 tests

Yeah, I couldn't believe it at first, the remuxing was a wild guess by me but I didn't believe it could work. Usually when a file doesn't play right remuxing to another container does nothing to fix the problem. In this case it did but I don't know why.... :D

[CruNcher]

Im not surprised by anything currently looking at my own oddities in behavior outside of lav video ;)

but nevcairiel is on Nvidia and it would be really crazy if his older VPX has not these issues and the one on the Driver (Firmware) side of the 1050 Ti has those, nothing impossible but highly unlikely.

did you tried a direct playback chain (Graphedit) maybe something gets in the Dshow Chain and interferes somehow ?

and is the Channel 4K stream http://demo-uhd3d.com/fiche.php?cat=uhd&id=54 with Codec ID: 36 also failing for you on the 1050 TI not remuxed ?


Found some new camera Hardware Encoder sample with a very interesting resolution signaling :)

https://www.youtube.com/watch?v=AqlTk4uAcTI

[MisterXDTV]

Im not surprised by anything currently looking at my own oddities in behavior outside of lav video ;)

but nevcairiel is on Nvidia and it would be really crazy if his older VPX has not these issues and the one on the Driver (Firmware) side of the 1050 Ti has those, nothing impossible but highly unlikely.

did you tried a direct playback chain (Graphedit) maybe something gets in the Dshow Chain and interferes somehow ?

and is the Channel 4K stream http://demo-uhd3d.com/fiche.php?cat=uhd&id=54 with Codec ID: 36 also failing for you on the 1050 TI not remuxed ?


Found some new camera Hardware Encoder sample with a very interesting resolution signaling :)

https://www.youtube.com/watch?v=AqlTk4uAcTI

I will download that stream but that server is really slow, it will take a while

this is my graph edit for that Sony file, looks fine to me

https://abload.de/img/grapheditssq8e.jpg

EDIT: I have the same problem with this Exodus demo clip (hev1 ID)

http://files.hdrsamples.com/downloads/hdr/Exodus_UHD_HDR_Exodus_draft.mp4

Again remuxing to mkv fixes the problem and DXVA2 copy-back works perfectly

[CruNcher]

I will download that stream but that server is really slow, it will take a while

this is my graph edit for that Sony file, looks fine to me

https://abload.de/img/grapheditssq8e.jpg

EDIT: I have the same problem with this Exodus demo clip (hev1 ID)

http://files.hdrsamples.com/downloads/hdr/Exodus_UHD_HDR_Exodus_draft.mp4

Again remuxing to mkv fixes the problem and DXVA2 copy-back works perfectly

this is really interesting slowly i guess we can pinpoint your issue it to certain HDR10 bitstreams i wonder if the fourcc actually in the standard for HDR10 is actually hev1/hvc1.

that maybe can explain why Codec ID: 36 maybe gets not correctly detected by some logic and handled differently then it should and in your case maybe it's the same by using the normal Main10 fourcc something gets ignored by the Hardware parser and then it works

Which in the end might be meaning though that you will lose HDR on the Renderer side and need to manually set it up to get the correct wide color gamut and luminance response ?.

Though in this case you should really refer to MadVR instead especialy on pre Windows 10 Aniversary ;)


https://www.youtube.com/watch?v=fHa10EbyKTk

Apollo Lake playing 4K 120Mbps HEVC Main10 video.

Looks great so far but in reality way to expensive if your target is just efficient avc/hevc/vp9 4K 60 decoding at lowest power possible ;)

For that price you get a Entry 40 4K TV with well performing MediaTek Platform with Fake HDR ;)

Though at 250 entry point it becomes very interesting and the following HDMI sticks as well :D

[famasfilms]

I have an i5-6600 and GTX 970.

I use MPC-HC, lavfilters and madVR, but trying to play 10bit HEVC files results in dropped frames.

I thought this PC would handle everything 4k but apparently the 970 series doesn't have hardware HEVC decoding.

Is there alternative software I can use or is the only answer upgrading my gfx card (pointless as I am not a PC gamer) or downgrading to a 960?

[CruNcher]

I have an i5-6600 and GTX 970.

I use MPC-HC, lavfilters and madVR, but trying to play 10bit HEVC files results in dropped frames.

I thought this PC would handle everything 4k but apparently the 970 series doesn't have hardware HEVC decoding.

Is there alternative software I can use or is the only answer upgrading my gfx card (pointless as I am not a PC gamer) or downgrading to a 960?

On Nvidias side you have 3 choices

960/1050TI/1060

neither is really optimal and you should chose wisely based on your heaviest workflows.

The GTX 970 CUDA HEVC Decoder is limited it copes with Medium 8 bit Complexity rather well but if the complexity is 1 step above it falls apart fast and there is the problem that the PWM works not so stable at all over Windows DXVA and you better refer to CUVID instead.

a good balance is 1060 if you could survive VP9 10 Bit Decoding maybe not becoming available or being as in your current situation for H.265 on GTX 970 Cuda Driven and so overall limited to the Shader array performance.

the 1060 is virtually exactly the highest Performing GTX 970 but updated towards lower power consumption and no additional latency impact due to to low ram with a more punchy VPX heart overall the GTX 970 was Designed for 4K 30 FPS 3D so is the GTX 1060 both can drive 60 FPS for Video without any problem and low processing 3D like the Windows GUI has itself no issues running 60 FPS stable also optimized 3D results have no Problem running 4K 60 stable as long as they not heavily use the shader array for Compute tasks at the same time for example, the GTX 1060 is a bit more efficient here then the GTX 970 which can help on latency when both use cases crash into each other heavily. :)

If you mix Compute, 3D and Video Decoding/Encoding at 4K 60 at the same time you need some punch for Realtime results and i advise to think about the 1070 in that case with more headroom in the Consumer space, don't invest though then in the most expensive that's total nonsense get the cheapest with balanced results.

All my tests on GM204 also showed that still the efficiency of Nvidias Mpeg-2/4 (ASP) Decoder Core is not good @ all and it can be better for efficiency purposes to not use them @ all but instead use Lav Video (avcodec) at least for Playback Decoding (DVD,Broadcast and Blu-Ray, Xvid, DivX) i would advise to prefer avcodec, this could save a lot of Energy for most System configurations especialy when you use a overclocked GTX 970.

Happy new Year everybody.

[CruNcher]

You can already see from the size just for 4K playback it is inefficient

http://www.intel.com/content/www/us/en/nuc/nuc-kit-nuc7i5bnk.html

But maybe it can press some usefull results out of MadVRs processing chain

That thing becomes though a interesting contender to Nvidias Shield TV slowly :)

Maximum System output seems to be around 65W that's a lot if you can't use it efficiently

i would go more for that Apollo lake @ cheap :D


N4200
J4205
J3455

but you should keep an eye on the USB3 transfer speeds and forget USB2 devices at least 1 USB3 port ;)



I hate Youtube reviewers these bunch of corrupted idiots.

Advising people to get overpriced Chinese Mini PCs for so much more then a complete Notebook from a well known Brand with the same hardware how dumb must you be.

[NikosD]

We need RyZen ASAP to escape from another Intel's fraud, the most miserable and useless "upgrade" ever for the desktop PC (besides media capabilities)

The shameless guys of Intel released a new CPU with the same IPC, with exact same performance clock for clock with the previous generation.

AMD, please help us!

[JohnLai]

We need RyZen ASAP to escape from another Intel's fraud, the most miserable and useless "upgrade" ever for the desktop PC (besides media capabilities)

The shameless guys of Intel released a new CPU with the same IPC, with exact same performance clock for clock with the previous generation.

AMD, please help us!

Down with INTEL CPU, thumb up with intel IGPU features.....

[CruNcher]

Sorry NikosD but you dont seem to understand Chip Design and Balancing.

And that you have to make decisions where to invest your shrink advantage.

Or you endup with Chips overheating at maximum gate load ;)

Apollo lake is very impressive for what it packs inside and to the outside for a non RISC Chip.

https://www.youtube.com/watch?v=HjXbAPXf_Vo

And as i told you ARM feared only Intel would reach their GFX capabilities fast and by now they overrun them the ARM CEOs worst nightmare became true with Apollo lake ;)


http://i1.sendpic.org/t/lh/lhE3x7ro7Q3UuLmx2hRMoX49m1S.jpg (http://sendpic.org/view/1/i/bBBAEXHivMwmT3OcNwJ0QFMUnpO.png)

[NikosD]

What I understand for sure is that Intel is so afraid of RyZen that was forced to update Skylake with an overclocked Skylake naming it Kabylake, just to have something to compete with the architecturally more advanced RyZen - in terms of 8C/16T design with less than 100W TDP very close to i7 7700K TDP actually.

So, AMD managed to produce a CPU for desktop that is competing with HEDT CPUs of Intel which sells them in the price range of more than 1000$ and a TDP of 140W.

I'm waiting for the official release of VEGA and RyZen to start real celebration for AMD products climbing up to the roof of both CPU and GPU performance

[CruNcher]

AMD still lacks a lot on UVD and VCE

Quicksync is much better overall integrated and reaches astonishing results since Haswell already :)

If AMD cant integrate as efficient using GCN as the foundation it will look bad and they will fail at low power again.

The real advances happen at low power not where Rayzen currently is targeting for.

6W SDP 15W TDP thats where AMD always had problems maintaining Performance and if you lose their you lose everywhere ;)

Also AMD never wanted to catch Intel @ all they knew they never could they just wanted to position as 2nd and survive.

With RyZen they will at least manage that again and hopefully the successor to Carizzo will also show that.

[NikosD]

What happened to your arguments regarding Chip Design and Balancing ?

Went back to UVD and VCE ?

Tell me more about the fraud of HEDT chips of Intel.

[CruNcher]

What happened to your arguments regarding Chip Design and Balancing ?

Went back to UVD and VCE ?

Tell me more about the fraud of HEDT chips of Intel.

Never had a problem with those issues always made sane decisions on intels side with the chip capabilities and tried to find a balance so i found the prices off intel on the i-5 always ok compared to AMDs Power Hungry HEDT with less performance ;)

i7 plays no role for me its unbalanced ;)

and AMD has not to win against i-7 but i-5 if they really want the crown with RyZen you should understand what ZEN really stands for, it doesn't stand for Gamer owning as a marketing word but initially it stands for "balance" ;)

comparing vs i-7 is absolutely not what ZEN stands for and i found it a big marketing gag ;)

in my perspective i-7 lineup could die tomorrow wouldn't miss it ;)

[oddball]

Someone suggested I post decoding results of LG's HDR Chess demo here.

First is MPC-BE with LAV filters using DX Copyback and EVR renderer.

http://i.imgur.com/IPjPrs7.jpg

Second is with MadVR. I cannot seem to get it under control GPU usage wise using MadVR.

http://i.imgur.com/4IqPHJB.jpg

This is using an AMD Radeon RX 480 overclocked to 1400/2200Mhz and a Intel i7 4770K @ 4.2Ghz with 16GB DDR3 @ 2200Mhz.

[CruNcher]

4% or lower with DXVA Native ?

was this EVR Custom or EVR Native and what playback application is this ?


The MadVR overhead is really crazy double the consumption something is going very wrong somewhere.

Are you sure there is none of the OpenCL/Directcompute Processing running ?

Very strange that extreme overhead if here only nearest neighbor is active.

What is shown in the Rendering Statistics CTRL+J ?

[tahaa7]

I've been following this thread, and it's been a fascinating and very informative read.

A quick question (from a relative noob): I'm buying a new laptop soon, and let's say I buy one with a Kaby Lake CPU and without a dedicated GPU. Would I be better off buying a CPU with Iris Plus or, say, 7700HQ (with Intel HD 630), in terms of multimedia performance (HEVC 4K 10-bit decoding, VP9 decoding etc.)? I see many of these U-series CPUs have higher prices than quad-core H-series ones, so I wonder why exactly that is. Is it because of graphics performance, or something else?

[MisterXDTV]

Do you think it's theoretically possible that Pascal cards could get their HDMI port upgraded to 2.1 via firmware upgrade? They decode 8K after all...

[huhn]

no not everything the connecter can't do 48 mbit.

some of the HDMI 2.1 stuff could be added. they did this with kepler before.

[aufkrawall]

I'm fiddling around with Linux a bit atm, so I could give VDPAU decoding a try.
Unlike DXVA2 native, it doesn't suffer by extremely blurry chroma, quality seems to be the same as software decoding with mpv and opengl-hq profile (has deband active, didn't pay attention to banding).
Unfortunately, Nvidia doesn't yet support VP9 and HEVC 10 bit decoding via VDPAU, which is a real bummer.

[CruNcher]

You should look into Wayland it's awesome for Video Desktop Performance, sadly not much supports it yet natively :)

[aufkrawall]

Dunno if it is possible yet with Nvidia.
I'm without any compositor since everyone for X11 doesn't seem to work perfectly.

[CruNcher]

It is most supporting Distro currently seems Red Hats Fedora (Workstation), some really nice results there on latency and usability i hope it wont fail it's a big step since compiz and beryl to a real nice user experience on every device :)

https://www.youtube.com/watch?v=F6PFjoYuml0
https://www.youtube.com/watch?v=Ux-WCpNvRFM
https://www.youtube.com/watch?v=s8XiEyGVKcI
https://www.youtube.com/watch?v=ETSsex3M_Oc

https://vimeo.com/3220996

[aufkrawall]

But only with open-source drivers, and you can totally forget about Nouveau.
And I also hate Gnome, waiting for KWin Wayland session that works with proprietary Nvidia driver...

[NikosD]

I've been following this thread, and it's been a fascinating and very informative read.

A quick question (from a relative noob): I'm buying a new laptop soon, and let's say I buy one with a Kaby Lake CPU and without a dedicated GPU. Would I be better off buying a CPU with Iris Plus or, say, 7700HQ (with Intel HD 630), in terms of multimedia performance (HEVC 4K 10-bit decoding, VP9 decoding etc.)? I see many of these U-series CPUs have higher prices than quad-core H-series ones, so I wonder why exactly that is. Is it because of graphics performance, or something else?

The performance of HW decoding alone, is only dependant from the clock for the same generation chips e.g Kabylake.

But for scaling, the low number of EUs could be a problem e.g when you try to decode a 4K clip on 720p resolution.

The weak Intel iGPUs have such problems.

If you are interested in HW encoding too, then more EUs means faster HW encoding.

And of course more EUs means faster 3D and GPGPU (computing) performance.

Products like Iris (Pro or Plus) have embedded memory which helps not only the GPU but the CPU too and for certain tasks can be very helpful but not that much in general CPU tasks.

For example Broadwell Core i7 5775C is faster than any desktop Core i7 even Kabylake in some compression tests.

Personally I wouldn't mind that much for embedded ram and I would wait some months for RyZen for sure.

We don't know the multimedia features of VEGA that RyZen APUs will have inside.

[CruNcher]

The performance of HW decoding alone, is only dependant from the clock for the same generation chips e.g Kabylake.

But for scaling, the low number of EUs could be a problem e.g when you try to decode a 4K clip on 720p resolution.

The weak Intel iGPUs have such problems.

If you are interested in HW encoding too, then more EUs means faster HW encoding.

And of course more EUs means faster 3D and GPGPU (computing) performance.

Products like Iris (Pro or Plus) have embedded memory which helps not only the GPU but the CPU too and for certain tasks can be very helpful but not that much in general CPU tasks.

For example Broadwell Core i7 5775C is faster than any desktop Core i7 even Kabylake in some compression tests.

Personally I wouldn't mind that much for embedded ram and I would wait some months for RyZen for sure.

We don't know the multimedia features of VEGA that RyZen APUs will have inside.

Waiiiiit
What
Intel HD Graphics has the best Hardware Scaler since GEN2 way more advanced then AMD and Nvidia back then and partly even still today.

Intels Hardware Scaler is Lanczos4 fully Hardware implemented, since GEN2 when Nvidia and AMD where still pushing around with Bicubic/Billinear by default.

And i think even today not much changed their and Intel still has the better Hardware Scaling and the Performance effects of it are almost non existent natively.

Nvidias and AMDs approach go more over the shaders if you want higher quality then Billinear/Bicubic.

Intel doesn't neeed to touch the EUs for Scaling really at all and that overhead doesn't exist with better quality because Intels Hardware scaler is also Adaptive.

[NikosD]

It was about time to reveal one more secret of Intel iGPUs, although I have posted that info in Doom9 a few times.

Almost all of the problems mentioned in Intel forums and even by PJ (IIRC) there with the low EUs APUs of Intel trying to playback 4K files, I have concluded that come from the scaler.

I had myself a Pentium Haswell G3420 and although it had the same QuickSync ASIC like the bigger models, it had definitely a lot of problems using playback mode of 4K clips scaled down to lower resolutions.

If you try to decode those files and scale them in lower resolutions you will see that GPU usage goes to almost 100%.

With all pre/post filtering off, there is no other excuse for the huge GPU utilisation of pure HW decoding.

The number of EUs play a significant role in scaled playback of 4K files even when they are decoded in HW.

My experience from Pentium G3420 says so.

[CruNcher]

It was about time to reveal one more secret of Intel iGPUs, although I have posted that info in Doom9 a few times.

Almost all of the problems mentioned in Intel forums and even by PJ (IIRC) there with the low EUs APUs of Intel trying to playback 4K files, I have concluded that come from the scaler.

I had myself a Pentium Haswell G3420 and although it had the same QuickSync ASIC like the bigger models, it had definitely a lot of problems using playback mode of 4K clips scaled down to lower resolutions.

If you try to decode those files and scale them in lower resolutions you will see that GPU usage goes to almost 100%.

With all pre/post filtering off, there is no other excuse for the huge GPU utilisation of pure HW decoding.

The number of EUs play a significant role in scaled playback of 4K files even when they are decoded in HW.

My experience from Pentium G3420 says so.

I have a Bay Trail system and i test a lot on 4K AVC Decoding efficiency on it and it's just awesome that it can easily beat very advanced ARM Soc IP that is not able to playback 4K 60 FPS at all ;)

Scaling doesn't play a part here Zero copy efficiency does more then anything else and certain things can be problematic under Windows in that regards, this shows up at that 4W target extremely.

The Problem with Sonys Color Demo has different reasons.

Overhead of EVR-CP should also not be underestimated as well especially not for MPC-HC ;)

And i can't really wait to get my fingers on N4200 to be true :D

And well see if it's really impossible getting it efficient working without official Windows 7/8/8.1 support

[NikosD]

Try your internal iGPU of Sandybridge

[CruNcher]

That doesn't support 4K and i currently put it in sleep mode and optimized everything that way not sure what will happen if i wake it up to be honest, but i gonna do that some day again.

Else you would see in a lot of my Screenshots the HD Graphics Framebuffer as well ;)

[MisterXDTV]

no not everything the connecter can't do 48 mbit.

some of the HDMI 2.1 stuff could be added. they did this with kepler before.

I think almost everything HDMI 2.1 has to offer can be added except new resolutions/frame rates. (as 8K/120fps)

So dynamic metadata HDR, eARC and Variable refresh rate DON'T require the new 48G cable. They only need the current 18G cable....

Basically current HDMI 2.0b can become "2.0c" or something like that.....

[tahaa7]

OK, so it seems you guys are suggesting that Iris iGPU would be a better choice. However, in the case of mobile Kaby Lake processors, there is a trade-off in terms of CPU cores: you have a choice of either getting a dual-core CPU with Iris graphics, or a quad-core CPU with Intel HD 630 graphics. How much does that influence final performance (in terms of HEVC 10 playback in MPC-HC, madVR, etc.)?

(5775C chip you mentioned @NikosD, is also a quad-core chip).

[CruNcher]

They will be no major differences only difference might be overall cpu utilization changing but all will playback up to 4K 60 at least without problems.

So it doesn't really matter only in terms of overall efficiency their are big differences with N4200 the most efficient it seems so far thus used in entry laptops :)

Then you have those hardembeded ITX barebones then comes the NUC and then Desktop.

All have very different targets and so the chips are laid out for these :)

But all of them will pretty certainly reach the 4K 60 10 bit Quicksync Decoding goal.

So if only Decoding is your goal obviously you go with the most power efficient ;)

that currently is N4200

http://ark.intel.com/products/95592/Intel-Pentium-Processor-N4200-2M-Cache-up-to-2_5-GHz


You surely will also see a lot of cheap China Mini PCs based on it but most of the time their wont be worth it vs a Laptop platform which is exactly the same with additional display ;)

Though the price is overall heavy as such in one or another way you get highly optimized ARM based OctaCore HDR solutions much much cheaper that can also do 4K 60 and if you have a UHDTV of the newest Generation you need nothing overall they all come pre included with such High Efficient Playback and on some you can directly install other Playback solutions that will work out of the Box ;)

And they will still overun the Intel Kaby Lake Barebones in Price/Value/Efficiency and in terms of Price vs N4200 Kaby Lake is no match and much to expensive just to be used for that usage scenario.

Though if you want to use some very Specific things like Netflix HD/4K you better of with the Intel solution.

The Barebones are the only semi efficient solution Price/Value/Efficiency wise but still Efficiency lacks.

https://www.youtube.com/watch?v=F_eEJk6lWlc

is pretty efficient :)

[tahaa7]

They will be no major differences only difference might be overall cpu utilization changing but all will playback up to 4K 60 at least without problems.

So it doesn't really matter only in terms of overall efficiency their are big differences with N4200 the most efficient it seems so far thus used in entry laptops :)

Then you have those hardembeded ITX barebones then comes the NUC and then Desktop.

All have very different targets and so the chips are laid out for these :)

But all of them will pretty certainly reach the 4K 60 10 bit Quicksync Decoding goal.

So if only Decoding is your goal obviously you go with the most power efficient ;)

that currently is N4200

http://ark.intel.com/products/95592/Intel-Pentium-Processor-N4200-2M-Cache-up-to-2_5-GHz


You surely will also see a lot of cheap China Mini PCs based on it but most of the time their wont be worth it vs a Laptop platform which is exactly the same with additional display ;)

Though the price is overall heavy as such in one or another way you get highly optimized ARM based OctaCore HDR solutions much much cheaper that can also do 4K 60 and if you have a UHDTV of the newest Generation you need nothing overall they all come pre included with such High Efficient Playback and on some you can directly install other Playback solutions that will work out of the Box ;)

And they will still overun the Intel Kaby Lake Barebones in Price/Value/Efficiency and in terms of Price vs N4200 Kaby Lake is no match and much to expensive just to be used for that usage scenario.

Though if you want to use some very Specific things like Netflix HD/4K you better of with the Intel solution.

The Barebones are the only semi efficient solution Price/Value/Efficiency wise but still Efficiency lacks.

https://www.youtube.com/watch?v=F_eEJk6lWlc

is pretty efficient :)

OK, thank you very much. Well, I will be buying a laptop that will be my main computer, but also occasionally be used as a HTPC. If the prices are right, I will certainly be looking to get one with a discrete GPU, and/or a quad-core CPU.

[Yups]

If you try to decode those files and scale them in lower resolutions you will see that GPU usage goes to almost 100%.

With all pre/post filtering off, there is no other excuse for the huge GPU utilisation of pure HW decoding.


I have never seen such a big GPU load on a Skylake and Kabylake GT2 nor do I noticed shader performance issues on a pure hardware decode, except for Hybrid decoding. As for encoding it is very important because of the Hybrid mode, even more for HEVC. But even with the huge shader performance advantage AMD has no chance regarding speed and quality with their current hardware. They have to catch Nvidia @Pascal as well. Personally I'm not interested in RyZen because of the lack of iGPU. Desktop APU seems to come only in H1 2018, which is probably a 4C SKU. I'm more interested in Coffe lake 6C in this case.

[CruNcher]

That is strange

https://youtu.be/y5EqYQ-6cEI?t=929

The Sony Camp HDR sample definitely fails

doesn't fit to what we saw in Tralaks jellyfish test at all

Interestingly it gets color graded as well not more into the slog range displayed it was recorded in looks like somekind of HDR to SDR being done :)

https://youtu.be/RxPxW8WaPks?t=489

[nevcairiel]

The Sony clip lacks proper HDR metadata, different players handle it differently. Some notice that its metadata is broken and assume defaults, some use the broken data and everything plays back all super saturated.

[CruNcher]

Hmm it looks more like the metadata get interpreted from the Windows 10 Parser/Decoder or even Quicksync side and a conversion takes place thus the colors stay not dull but get saturated like some lut applied

And the Kodi result on S912 is the pure one that resambles more the native slog space of the camera without any conversion going on or lut applied which is virtualy the most case players do it based on ffmpeg i saw so far what gets outputed by the Windows 10 Decoder (Player) here looks very different to anything else.

But the Performance is unusable in that N4200 result anyways no matter what goes on on the HDR side here .

Far away from what we saw here

https://www.youtube.com/watch?v=fHa10EbyKTk

Maybe a Driver issue

All these reviewers concentrate to much on jelyfish it's ridiculous

https://youtu.be/Fk5i3fvmIJc?t=354

Ok much better then this one this is really nahh 24 2000 kbps unbelievable that's a real challenge geez a Youtuber tests 24 @ 2000 kbps 4K :D

https://youtu.be/WvWX0MTo7WA?t=685


Redskull at least makes usable reviews

https://www.youtube.com/watch?v=y5EqYQ-6cEI

Also ok (but limited to jellyfish)

https://www.youtube.com/watch?v=Fk5i3fvmIJc

seems Chrome still runs the libvpx Decoder

Limits himself to DX11 tests on a Windows 10 system with a GPU Designed for DX12 ouch ;)

Cinebench 15 seems to be the most interesting Score to compare the CPU Efficiency with virtually every reviewer uses it in one way.

it indicates that my i-5 2400 is ~2x more efficient then the N4200 in that Laptop configuration which is when we compare the maximum output difference very impressive :)

from 2nd 22nm max 95W ->14nm 7th Generation max 6W it seems CPU Performance in direction of Tegra X1 i guess good comparable

i guess no bad compare point on the 3D side ;)

https://www.youtube.com/watch?v=qm16p4qYjF4

of course P1 will kill it inside the Nintendo NX but it's allways nice to see what Intel runs for ;)

Anyway cant wait to get my N4200 Laptop and take that thing apart in all Details :)

Also interesting because AVX is missing in N4200 :D

oha Nvidia has allready prepared his answer ;)

https://www.youtube.com/watch?v=AK9hpNb68xY

a Shrinx of X1 with i guess the newest VPX thats also in GP107

looks super sleek and sexy :D

Intels preparation of full module updates at least on the SOC side ;)

https://www.youtube.com/watch?v=VWtSBSRSg3s


This will make Video Hardware Codec upgrading very easy for TVs very useful, i saw this future coming :)

Hoho now they prepare to attack MediaTek :D

New Video Codec comes out your TV doesn't support no problem just replace the Module in the back (others surely will follow the idea including MediaTek).

This is so great so much Waste will be reduced and you don't have to buy a TV each Video Codec cycle but just replace the Module, that's the future :)

You can also upgrade obviously advancements in Post Processing this way too (more advanced Motion interpolation ect, instead of X amount of models that only differentiate by that) :)

So instead you need to buy a new TV for say 800$ you upgrade it for 200$

https://www.youtube.com/watch?v=aMeVl4PLKaQ

http://ark.intel.com/products/95449/Intel-Core-m3-7Y30-Processor-4M-Cache-2_60-GHz-
http://ark.intel.com/products/95592/Intel-Pentium-Processor-N4200-2M-Cache-up-to-2_5-GHz

very efficient and includes AVX compared to N4200


They Balancing Instructions vs Cores and GPU here to get higher efficiency :D

And strangely the Resolution is better specified then with N4200

http://www.cpubenchmark.net/compare.php?cmp%5B%5D=2864&cmp%5B%5D=2886&cmp%5B%5D=2875


Nice :)

Compared vs my Desktop

http://www.cpubenchmark.net/cpu.php?cpu=Intel+Core+i5-2400+%40+3.10GHz&id=793


I wonder if that Sony Clip Performance result of the N4200 could have todo with what YUPS said about the Clock Domain and how it works in combination

https://forum.doom9.org/showpost.php?p=1790164&postcount=1622

Also maybe the missing AVX plays into it especialy in that form factor but ok it's actively cooled and shouldn't really fluctuate on the clocks so much at all but unfortunately we never gonna see such indepth tests from any Reviewer in the World ;)


AMD is on the right track

http://www.cpubenchmark.net/compare.php?cmp%5B%5D=2861&cmp%5B%5D=793&cmp%5B%5D=2889

RyZen will be a exciting Value opportunity for End Users of all target groups and paired with Vega it could even do Hybrid Workflows more efficient then we ever seen before :)

https://www.youtube.com/watch?v=lGuFVc8AFXk

https://youtu.be/FwcUMZLvjYw?t=1458

[Yups]

That is strange

https://youtu.be/y5EqYQ-6cEI?t=929

The Sony Camp HDR sample definitely fails

doesn't fit to what we saw in Tralaks jellyfish test at all

Interestingly it gets color graded as well not more into the slog range displayed it was recorded in looks like somekind of HDR to SDR being done :)

https://youtu.be/RxPxW8WaPks?t=489


Why is this strange? His player clearly doesn't support HW decode.

[CruNcher]

Ehh thats the Default Windows Player it should automatically go into DXVA and you can see that it does else that Astra Sample would never play so smooth, pretty impossible on that CPU :D

Does the Sony 4K HDR Clip get accelerated for you via the Default Films & TV App Player and do you see a color grading taking place like in the Review ?

http://demo-uhd3d.com/fiche.php?cat=uhd&id=144

[JohnLai]

Anyone have Kaby Lake HEVC 10bit encode sample with Target Usage 1 (TU1)?

[edigee]

Sorry ,I don't have the time to read the whole topic here. Which discrete video card is better for HEVC decoding? GTX 960 or GTX 1050(or the TI version)? Or a similar AMD product ,similar price range.
Thanks?

[huhn]

the 1050 can decode HEVC up to 12 bit(no DXVA spec for 12 bit) the 960 "only" up to 10 bit. the 1050 can do easily 120 FPS stable with quite a some headroom.
on top of it the 1050/1050 ti supports VP9 10 bit all other Dcards didn't support this yet.

the AMD GPUs can "only" decode HEVC in hardware no VP9 and can at best do 60 FPS HEVC.

if you want to play UHD at UHD make sure you have more than 2 GB Vram.

[P.J]

Is 1050Ti 4GB better than 1060 6GB for 4K/8K?

[huhn]

GPU specs are relatively unimportant.
if you want a VP9 10 bit decoder yes.

and for 8K wrong year...

[nsnhd]

And a GPU for HDR display is invisible now which many are looking for, including me.

[NikosD]

Sorry ,I don't have the time to read the whole topic here. Which discrete video card is better for HEVC decoding? GTX 960 or GTX 1050(or the TI version)? Or a similar AMD product ,similar price range.
Thanks?

After several tests with my RX 470, I can say that AMD followed once again a rather conservative way, covering most of the cases of HW decoding.

So, with Polaris video decoder you can be sure that every consumer disc (DVD, Bluray, Ultra HD Bluray) can be decoded in HW in terms of HW video decoding (I'm not referring to the other requirements of the discs)

More specifically UVD (Polaris HW decoder) can decode 100Mbps bitrate of : (FPS values are on average)

2160p H.264 ~ 100-120 fps

2160p H.265 ~ 70-115 fps

2160p H.265 10bit ~70-90 fps

I've chosen 100Mbps on purpose, as it is the maximum bitrate for video stream alone that Ultra HD Bluray is allowed to have (and put that benchmark to H.264 too)

There is no video stream of any disc that Polaris can't decode and with minimum fps always above 60fps.

With that philosophy in mind, you can expect that Polaris doesn't support 12bit H.265, 8K (both are useless right now) and only supports VP9 8bit in hybrid mode (not by default enabled in the drivers yet)

Overall, Polaris UVD decoder tries not to miss HW decoding support of critical video formats (consumer discs) than to support everything there is out there.

[edigee]

Thank you very much for all the info. It seems that 1050Ti is a winner in terms of price and decoding quality and features.

[baii]

AMD can do VP9 in chrome(hybrid or not, it works), which is sort of the main purpose for VP9.

I read people are saying hdr in new hdr games is actually working? at least on nvidia 10xx series.

[ibius]

Yeah, 1050 Ti is the king in TDP/performance class right now, I have one myself.
But if you're gonna use it for gaming or madVR, I would suggest getting one with an 6-Pin PCIe connector, otherwise you're just leaving performance on the table.
I bought ASUS Expedition first and it hit it's power limit on many occasions during my playback tests with madVR, so I exchanged it.
All of these slot-only cards are hardware/bios limited and don't go near 75W, they hit their power limit at 70W and even though with GPU boost 3.0 they can go up to 1720 MHz boost clock (no manual/factory overclock), these clocks are just not sustainable on a 70W TDP.

[P.J]

the 1050 can decode HEVC up to 12 bit(no DXVA spec for 12 bit)

Only 1050/1050 Ti?

GPU specs are relatively unimportant.
if you want a VP9 10 bit decoder yes.

and for 8K wrong year...

VP9 10 bit, useful?

But Pascal can decode 8K :rolleyes:

[huhn]

all pascal GPU can decode up to 12 bit HEVC.
don't forget you can't even use it with DXVA.

VP9 10 bit, useful?
youtube is using VP9 10 bit right now rarely but it is. so unlike 12 bit HEVC it is useful.

But Pascal can decode 8K
and how do you output it?

and about hybrid decoding...

using 2 displays at the same time is a common setup and loosing FPS because you are watch a video is just terrible design...

[el Filou]

I suppose many of you will have seen this, but if not: Palit has released a fully passive 1050 Ti: http://www.palit.com/palit/vgapro.php?id=2823&lang=en

This could be an interesting upgrade for people like me who want to unburden their weak CPU of HEVC and VP9 playback, or who want madVR and silence but find Intel's iGPUs not powerful enough, all this while not having to void your graphics card's warranty by installing a 3rd-party cooler yourself.

I'm buying this as soon as it's available in a shop that delivers in my country, and will report on temperatures and any power issues when installed in my old HTPC with a big case and one single slow fan.

For reference, on Amazon France Palit's passive 750 Ti was 40 euros more expensive than the cheapest 750 Ti so we may be looking at something around 190 euros, but for me it's worth it.

[huhn]

https://www.computerbase.de/2017-02/palit-geforce-gtx-1050-ti-kalmx-test/3/

well good luck.
the heatsink is simply to small to passive cool this low TDP card.

i can get this card for below 180 in germany and hardware is relative expensive in germany.

and you know you can always start a new thread they are answered rapidly here on the forum.

[nevcairiel]

Passive GPUs generally rely on some basic air movement in the case, preferably straight over the GPU. A fully passive system requires special careful design and appropriate components, not something you can just assemble from off-the-shelf things usually.
Personally I prefer GPUs with good semi-passive coolers, they stay silent most of the time and even when active the good ones are usually inaudible from a decent sitting distance (and a noise-dampened case).

[wanezhiling]

DXVAChecker Version 3.15.0 (2017/01/28)
Changed naming rule of vendor original device
Added "VP9_VLD_Intel" as an alternative name of Decoder Device
Added support for Intel iGPU in Video Acceleration Setting
Changed Video Acceleration Setting UI
Fixed Video Acceleration Setting issue under specific conditions
Added support to enable/disable DXVA for Microsoft MPEG Video Decoder MFT and Microsoft WebM MF VP8 Decoder Transform
Fixed wrong data of D3D11_GetDecoderBuffer and D3D11_SubmitDecoderBuffers event in trace log
Discontinued support for WindowsXP
Minor changes and fixes

FYI

[clsid]

Latest AMD driver changelog mentions a fix for the H.264 corruption after seek bug.

[v0lt]

Arri65_from_raw.mp4 (http://www.dmitry.com/tests/ARRI/) (HEVC 10-bit 6560x3100x23,976 263Mbit)

[NikosD]

Arri65_from_raw.mp4 (http://www.dmitry.com/tests/ARRI/) (HEVC 10-bit 6560x3100x23,976 263Mbit)

Impressive clip.

Using my Core i5-2400 and Decode benchmark of DXVA Checker (LAV Video 0.69), I got only 9fps.

You need a Ryzen 7 CPU (available from March 2nd) or even better a HW HEVC decoder like Pascal, Skylake*, Kabylake supporting 8K resolution in order to decode it in real-time.

*Edit:
It's a 10bit HEVC clip, so only Nvidia Pascal and Intel Kabylake can decode that clip in HW.

[CruNcher]

Arri65_from_raw.mp4 (http://www.dmitry.com/tests/ARRI/) (HEVC 10-bit 6560x3100x23,976 263Mbit)

Geez i had to scratch together that Space somewhere with all the encoding/decoding tests running

Thx this is directly from ARRI RAW->SDR Grade to HEVC ?

And the Scene came from the Alexa 6.5K Film mode, which Lens was used ?

yeah a beast I/O peaks can be also very high :)

Though lot of noise, hmm encoder wasn't x265 but Reference 10 bit Encoder ?

The Alexa Mounted on the bottom of a Helicopter facing straight downwards ?

But that Horizontal Pan is so perfectly precise really a Helicopter with some Gyro Stabilization ?

http://i1.sendpic.org/t/ac/ac94qZD9essYEnzY3hn4ABDkUyF.jpg (http://sendpic.org/view/1/i/72wcSF3ljSAhmh9u3VaLJH4Z5gO.png)
http://i1.sendpic.org/t/vH/vHRxRDXi0CRqfKMIIUEmvB8OawW.jpg (http://sendpic.org/view/1/i/syFepfLorohAP2NAdHcn4Qe1tK5.png)

[v0lt]

Thx this is directly from ARRI RAW->SDR Grade to HEVC ?

And the Scene came from the Alexa 6.5K Film mode, which Lens was used ?
I do not know the details. Information from the official page:ALEXA 65 Footage
All three ALEXA 65 clips were shot with project rate 23.98 in ARRIRAW Open Gate format. One clip is in sync speed the other two in 60 fps high speed. The used lenses are Prime 28, 50 and 80mm.

[NikosD]

Does anyone have the slightest idea why an 8 core RyZen 7 has 4.8 billion transistors which is the same as Broadwell-EX with 15 cores (!) ?

To give you an idea, biggest Broadwell-E processor, the 10 core Core i7-6950K, has "only" 3.2 billion transistors.

What is inside RyZen 7 occupying so many transistors ?

[huhn]

512kb level 2 cache per core intel is using 256 kb per core , maybe more IO most of the south bridge is on the CPU now and http://techreport.com/news/22100/amd-corrects-muffed-bulldozer-transistor-count

not that it matter but we will know fast enough

[CruNcher]

Dont forget the Trustzone powered by ARM ;)

http://www.amd.com/en-us/press-releases/Pages/amd-strengthens-security-2012jun13.aspx

The Hypervisor is a ARM Cortex Ax which will be tightly integrated most probably rather similar to how Intels ME works in combination with their AES 256 Intstructions ;)

http://hackaday.com/2016/11/28/neutralizing-intels-management-engine/

In the End this will allow you to play the 4K content Bus Encrypted via DXVAs Protected Media Path and controlled by the Trustzone in Realtime for integrity.

http://i1.sendpic.org/t/3y/3yFvzt0Fl1NBpAoyiOt5Pk13Bhq.jpg (http://sendpic.org/view/1/i/7mWXtMvUv8C5KOWMbVAozOMH1vm.png)

Which gonna mean this time we have fought content piracy successful and brought it finally successful to the PC it was about time as soon as everyone switched to Windows 10 and coming OSes we can pull the switch and transfer all the content finally and MSE is integrated as well in the chain, also we can finally start to introduce all this new exciting monetarization ways via tighter OS Content Control a Dream comes true so many years of work finally become reality :)

Now that every Hardware is ready it will strengthen our tumbling Economies and loses over the coming years finally reduced and the Web will get cleaner as well for AVG users in the transition Process most Online Pirates and Real Criminals wont even realize what hit them, sometime it needs force to make avg users understand what's good for them :)

[el Filou]

Which gonna mean this time we have fought content piracy successful and brought it finally successful to the PC it was about time as soon as everyone switched to Windows 10 and coming OSes we can pull the switch and transfer all the content finally and MSE is integrated as well in the chain, also we can finally start to introduce all this new exciting monetarization ways via tighter OS Content Control a Dream comes true so many years of work finally become reality :)

Now that every Hardware is ready it will strengthen our tumbling Economies and loses over the coming years finally reduced and the Web will get cleaner as well for AVG users in the transition Process most Online Pirates and Real Criminals wont even realize what hit them, sometime it needs force to make avg users understand what's good for them :)
It's hard to know when you're serious or not from your posts, but this time I have to hope you're joking.
Like every DRM technology before it, it will eventually be either cracked or bypassed and won't stop piracy, all while bringing additional costs to consumers, producers and every intermediate party.
Every DRM scheme is a disease and this is the HIV of DRM.

[CruNcher]

That's your opinion but those who implement it think different it's bringing order to the chaos in their view of it and it will have a impact on society in several ways good/bad, of course the economical impact in the future will be huge and so will be it's misuse therefore we have to fight which rules are rules Society can agree upon like we have found rules for Law and Order in Real Life but it needs to be incorporated directly in IT Systems and that is happening right now :)

Some respect the Law and Order system some don't and you get punished and jailed if you disrespect it and the Society it's Values it's Protecting being it Real Property or Intellectual or your Life itself sooner or later IT Systems will completely fall under this Judgement in Realtime you can't avoid it with the rising complexity and issues coming from things like IOT or Augmentation not if your Societies System is broken in itself and you need to drive it on the way it is.

Society needs Protection on this Layer DRM/DRS will ensure it like the real Law and Order System ensures what's your property and what's not and what happens if you go against the Rules you gonna get Punished how that Punishments will look and these Future Courtrooms is still to be decided.

The biggest Problem is that currently the decisions about it are purely driven by Corporate Economical interests and that needs to stop and the overall Process becoming more transparent.

[jkauff]

I would be in favor of an end to piracy if it was accompanied by an end to price-gouging. The popular movies and TV shows most people want to own have recouped their cost many, many times over by the time they're released on disc. Disc manufacturing is a trivial cost to major studios (not to mention the cost of streaming, which is basically nothing). Blu-ray movies should cost half what they do, and the gouging will get even worse with 4K.

[kypec]

I would be in favor of an end to piracy if it was accompanied by an end to price-gouging. ...
My words exactly. Content owners and providers should understand that restricting online access to only certain regions is wrong. I have started to use and pay monthly Netflix account in my country last year when it became available here in Slovakia but it still irritates me that there is almost never a TV show or movie that I want to watch. Most of the content is either available only to U.S. customers or not available at all, not even the movies that were released like 3+ years ago. I mean, what's the point of paying online streaming service when there is no interesting content to watch. :sly:

[NikosD]

It's been around 2.5 years since my first post in this thread, evaluating SW and Hybrid HEVC decoders of that time, since no pure fixed-function HW HEVC was available.

The samples that I had chosen back then, they were all low bitrate in order to be feasible for SW and hybrid decoders to decode them in reasonable framerate.

Now that all big three (AMD, Intel, Nvidia) provide pure HW HEVC decoders (8bit & 10bit) it's time to gather a new collection of samples that fit better to HW video decoding and post results for all three.

You can find a link to download each clip right below it or you can find them altogether in this host: (thanks to mariush)

ftp://helpedia.com/pub/multimedia/testvideos/HEVC_HW_decoders_clips/ (Max 2 downloads/user, Max 1MB/s)


All tests were done using DXVA Checker v3.x, LAV Video 0.6x in benchmark decode and benchmark playback mode (downscaled to 1280 x720) using DXVA native


8bit HEVC 4K/8K samples

Cards/iGPUs tested:

nVidia Turing (up to 8K)
Nvidia Pascal (up to 8K)
Nvidia GM206 (3rd gen Maxwell) (up to 4K)
Intel Kabylake (up to 8K)
Intel Skylake (up to 8K)
AMD Polaris (up to 4K)


1) LG_4K_View-the-Feeling (2160p@29.970fps ~25Mbps)
http://demo-uhd3d.com/fiche.php?cat=uhd&id=101


Decode:


GTX 1660 - 338/357/370

GTX 1060 - 217/223/227

GTX 1050 Ti 187/193/204

GTX 1050 172/180/194

Kabylake HD 620 156/170/195

Kabylake HD 610 138/165/194

GTX 960 137/145/163

RX 470 97/105/111



Playback:


GTX 1660 - 340/358/368

GTX 1060 - 217/223/227

GTX 1050 Ti 195/199/204

GTX 1050 183/190/201

Skylake HD 530 146/168/202

Kabylake HD 620 146/160/187

GTX 960 141/145/165

Kabylake HD 610 97/123/147

RX 470 101/107/113




2) Samsung 4K@30fps-80Mbps file -0235
http://imaging-resource.com/PRODS/samsung-nx1/FULLRES/yvid_SAM_0235-UHD_Sample1.MP4



Decode:


GTX 1660 - 289/295/307

Kabylake HD 620 181/188/194

GTX 1060 - 183/185/189

Kabylake HD 610 134/173/183

GTX 1050 Ti 158/162/164

GTX 1050 149/155/162

GTX 960 119/120/122

RX 470 96/99/106


Playback:


GTX 1660 - 293/302/308

Skylake HD 530 174/190/201

GTX 1060 - 185/186/190

Kabylake HD 620 172/177/180

GTX 1050 Ti 156/161/165

GTX 1050 148/154/158

Kabylake HD 610 130/140/151

GTX 960 121/123/126

RX 470 95/98/102


3) Samsung 4K@30fps-80Mbps file -0230
http://www.imaging-resource.com/PRODS/samsung-nx1/FULLRES/yvid_SAM_0230-ProQuality.MP4


Decode:


GTX 1660 - 248/254/263

GTX 1060 - 175/176/179

Skylake HD 530 160/164/168

Kabylake HD 620 153/156/158

Kabylake HD 610 149/153/158

GTX 1050 Ti 148/153/157

GTX 1050 137/141/145

GTX 960 114/115/117

RX 470 94/96/99


Playback:


GTX 1660 - 257/258/262

GTX 1060 - 175/176/180

Skylake HD 530 147/155/160

GTX 1050 Ti 148/154/158

Kabylake HD 620 148/149/152

GTX 1050 138/143/148

Kabylake HD 610 107/122/128

GTX 960 116/116/121

RX 470 95/98/102


4) Crowd_run 4K@50fps -300Mbps
https://mega.nz/#!RVNy0BhB!DpsJlf5n_TJV1lptFGH4mAZblTA9-vT0P12VikS3xko


Decode:


GTX 1660 - 176/180/180

GTX 1060 - 142/145/147

GTX 1050 Ti 113/125/128

Skylake HD 530 112/120/123

Kabylake HD 620 101/110/113

Kabylake HD 610 97/109/113

GTX 1050 92/104/109

GTX 960 83/93/97

RX 470 60/70/77


Playback:


GTX 1660 - 176/182/182

GTX 1060 - 139/145/147

GTX 1050 Ti 115/125/128

Skylake HD 530 102/114/119

Kabylake HD 620 96/106/109

GTX 1050 94/106/114

GTX 960 83/94/98

Kabylake HD 610 72/92/98

RX 470 60/72/79


5) 1st Bonus clip:
Netflix_TunnelFlag_4096x2160_60fps_x265_8bit_700Mbit
https://yadi.sk/i/swMdlJHMuB6yR


Decode:


GTX 1660 - 96/109/127

GTX 1060 - 81/91/101

GTX 1050 Ti 66/80/93

Kabylake HD 610 64/80/96

Kabylake HD 620 62/76/88

GTX 1050 49/64/73

GTX 960 48/59/70

RX 470 33/44/51


Playback:


GTX 1660 - 96/109/127

GTX 1060 - 81/91/101

GTX 1050 Ti 66/80/92

Kabylake HD 620 62/75/85

Kabylake HD 610 57/72/87

GTX 1050 49/64/74

GTX 960 47/59/71

RX 470 34/46/55


6) 2nd Bonus clip (8K):
8K_Ex_sample- 7680x4320@24fps-53Mbps
http://art11111.com/image/video/8K_Ex_sample.mp4


Decode:


GTX 1660 - 95/105/108

GTX 1060 - 55/59/62

GTX 1050 46/52/55

Kabylake HD 620 42/46/48

Kabylake HD 610 41/46/49



Playback:


GTX 1660 - 95/105/108

GTX 1060 - 55/59/62

GTX 1050 45/52/54

Kabylake HD 620 41/44/47

Kabylake HD 610 36/41/43



10bit HEVC 4K/8K samples

Cards/iGPUs tested:

nVidia Turing (up to 8K)
Nvidia Pascal (up to 8K)
Nvidia GM206 (3rd gen Maxwell) (up to 4K)
Intel Kabylake (up to 8K)
AMD Polaris (up to 4K)


1.UHD_PQ_Lovely_Swiss-51Mbps@60fps
http://demo-uhd3d.com/files/uhd4k/Demo_Samsung_2014_-_Lovely_Swiss.zip

Decode:


GTX 1660 - 310/340/365

GTX 1060 - 215/222/242

GTX 1050 171/179/197

GTX 1050 Ti 159/179/211

GTX 960 112/124/153

Kabylake HD 610 74/115/152

RX 470 73/87/108



Playback:


GTX 1660 - 305/335/360

GTX 1060 - 213/221/237

GTX 1050 185/191/205

GTX 1050 Ti 171/190/217

GTX 960 111/122/155

RX 470 80/94/116

Kabylake HD 610 52/76/87



2.Crowd_run_2160p50_Main10_100Mbps
https://www.sendspace.com/file/giy7br


Decode:


GTX 1660 - 292/298/298

GTX 1060 - 205/208/209

GTX 1050 Ti 175/179/181

GTX 1050 161/164/165

Kabylake HD 620 155/157/158

Kabylake HD 610 112/118/121

GTX 960 108/109/113

RX 470 86/89/98 (CPU usage ~20% on quad core)


Playback:


GTX 1660 - 290/294/294

GTX 1060 - 204/207/208

GTX 1050 Ti 173/176/177

GTX 1050 162/166/169

Kabylake HD 620 137/139/141

GTX 960 108/112/115

RX 470 94/96/98

Kabylake HD 610 68/75/79



3.Jellyfish-140-mbps-4k-uhd-hevc-10bit
http://jell.yfish.us/media/jellyfish-140-mbps-4k-uhd-hevc-10bit.mkv


Decode:


GTX 1660 - 177/208/256

GTX 1060 - 151/169/191

GTX 1050 Ti 130/146/165

Kabylake HD 620 117/133/151

GTX 1050 112/128/147

Kabylake HD 610 94/110/124

GTX 960 94/103/114

RX 470 67/80/111


Playback:


GTX 1660 - 178/210/255

GTX 1060 - 151/169/191

GTX 1050 Ti 131/147/165

GTX 1050 114/131/150

Kabylake HD 620 111/123/139

GTX 960 92/102/113

RX 470 74/85/96

Kabylake HD 610 65/77/83



4.Jellyfish-400-mbps-4k-uhd-hevc-10bit
http://jell.yfish.us/media/jellyfish-400-mbps-4k-uhd-hevc-10bit.mkv


Decode:


GTX 1660 - 86/105/130

GTX 1060 - 67/84/109

GTX 1050 Ti 60/74/96

GTX 1050 53//83

Kabylake HD 620 54/67/89

Kabylake HD 610 53/65/83

GTX 960 38/55/71

RX 470 32/40/52


Playback:


GTX 1660 - 85/104/129

GTX 1060 - 67/84/109

GTX 1050 Ti 60/75/96

Kabylake HD 620 53/66/85

GTX 1050 42/55/74

GTX 960 44/55/72

Kabylake HD 610 44/55/68

RX 470 32/41/53



5) 1st Bonus clip (7K):
5.Arri65_from_raw-6560x3100-263Mbps@24fps
http://www.dmitry.com/tests/ARRI/Arri65_from_raw.mp4


Decode:


GTX 1660 - 103/103/106

GTX 1060 - 74/76/77

GTX 1050 Ti 63/68/72

Kabylake HD 620 54/56/58

GTX 1050 52/55/59

Kabylake HD 610 40/47/50



Playback:


GTX 1660 - 100/103/107

GTX 1060 - 74/76/77

GTX 1050 Ti 64/68/72

GTX 1050 52/55/59

Kabylake HD 620 48/51/53

Kabylake HD 610 30/35/37



6) 2nd Bonus clip (8K):
MBC UHD Music Core GFriend - Rough (4in1) 8k@60fps-80Mbps
https://mega.nz/#!8hcmwDoK!UwfMhPQvtiJw9SkZpijJijxyC4k45iqrNTwMVHjUZ6Y


Decode:


GTX 1660 - 98/103/107

GTX 1060 - 59/60/63

GTX 1050 Ti 53/56/58

GTX 1050 52/54/56

Kabylake HD 620 43/48/56

Kabylake HD 610 31/35/41


Playback:


GTX 1660 - 96/101/104

GTX 1060 - 59/60/63

GTX 1050 Ti 54/55/56

GTX 1050 52/54/55

Kabylake HD 620 38/41/47

Kabylake HD 610 24/25/29


Comments:

1) It is now clear from the benchmark results, that Intel has changed its strategy to bring the fastest HW video decoder and targets the most efficient.

When Sandybridge back on 2011 introduced for the first time QuickSync fixed-function decoder/encoder, it was the fastest ASIC of that time with a clock around 1300 MHz, when AMD and Nvidia had clocks around ~500 MHz.

But Intel in every generation manages to increase performance and at the same time lowers the clock, so Kabylake's iGPU has exactly the same clock as Skylake (1150 MHz) which is lower than Sandy, 6 years ago.

On the other hand, Nvidia with Pascal cards managed to reach ~1700 MHz which is almost 50% faster clock than Kabylake and reached in raw terms the decoding speed of Intel, but not so efficiently in order to achieve that clock.

Turing cards go even faster at ~1800 MHz for Video Clock and for HEVC seem to be in a different category especially for 8K clips which seem to be a lot more optimized than Pascal.

But 8K VP9 is almost the same for Turing and Pascal, like 4K H.264 too.


2) AMD RX 470 has obviously a different philosophy, covering 60fps only at "normal" bitrate clips up to 80Mbps which should cover a very large amount of real world clips out there.


3) Turing GTX 1660 has a video clock of 1800 - 1815 MHz, while GTX 1050 Ti runs at 1645 MHz "GPU Video Clock" and GTX 1060 at 1708.5 MHz.
GTX 1050 runs at 1582 - 1607 MHz
The GPU Video Clock difference is 4% for Pascal cards but the real-world video decoding performance is ~15% faster for 1060 in most of the clips (besides the huge bitrate clips in which they are a little closer)


4) Nvidia GTX 960 has a "GPU Video Clock" of 1202.3 MHz, so a 1050 Ti is ~37% faster clock for clock.

It seems that the HEVC 8bit decoder follows exactly the clock difference, so the decoder should be actually the same.

But regarding HEVC 10bit decoder, the first clips show an almost double difference than 8bit decoder favoring, of course, the Pascal chip.

The huge bitrate 10bit clip (400Mbps) has exactly 37% difference though, like the clock difference.


5) Skylake HD 530 and Kabylake HD 630, both run at the same speed of 1150 MHz and Kabylake HD 620/610 run at 1050 MHz.
The results of pure decoding speed of Kabylake and Skylake look similar on 8bit clips, they could have the exact same HEVC 8bit decoder (only difference in MHz)
The same goes to 610/620 which could/ should have the same speed on both 8bit & 10bit HEVC clips and the only difference should be the playback scaled down to 720p speed due to a lot more units of 620.
Different drivers and different memory configuration (single channel/ dual channel) could affect performance on iGPUs, too.

[loekf]

Time permitting, will try to put my newly installed GTX 1050TI through the test.

I was looking for these numbers. What happened to AMD's Polaris UVD ? Looks like it has much lower processing power and has a "all-is-ok-if-I-output-60fps" philosophy.

[NikosD]

I will add tomorrow 10bit HEVC clips too, so just wait for another day ;)

AMD was always conservative to media performance.

They don't care for Media Servers and multiple stream decoding, so they are just to the point for one stream performance.

[loekf]

I will add tomorrow 10bit HEVC clips too, so just wait for another day ;)

AMD was always conservative to media performance.

They don't care for Media Servers and multiple stream decoding, so they are just to the point for one stream performance.

BTW, this link is not correct:

http://imaging-resource.com/PRODS/samsung-nx1/FULLRES/yvid_SAM_0235-UHD_Sample1.MP4

[NikosD]

BTW, this link is not correct:

http://imaging-resource.com/PRODS/samsung-nx1/FULLRES/yvid_SAM_0235-UHD_Sample1.MP4

When I click on this link, it immediately starts downloading the file.

[loekf]

When I click on this link, it immediately starts downloading the file.

Weird, here at work it works. At home I got a 404 error at that Samsung site (or missing page).

Anyways, lemme download LAV tonight and do some testing.

[loekf]

Okay here we go:

Tools: DXVA checker rev. 3.15, LAV Filters 0.69.0

System: Intel Core i5 7400 (Kabylake), Nvidia GTX1050 TI
Decode benchmarks settings: 64-bit
Playback benchmarks settings: 64-bit, downscaled to 1280 x 720

For all results: 5 runs, usually first run is worst. Took the last run.

1. LG_4K_View-the-Feeling.mp4

Decode: 187 - 193 - 204
Playback: 195 - 199 - 204

2. yvid_SAM_0235-UHD_Sample1.mp4

Decode: 158 - 162 - 164
Playback: 156 - 161 - 165

3. yvid_SAM_0230-ProQuality.mp4

Decode: 148 - 153 - 157
Playback: 148 - 154 - 158

4. crowd_run_2160p50_300Mbps.mkv

Decode: 113 - 125 - 128
Playback: 115 - 125 - 128

5. Netflix_TunnelFlag_4096x2160_60fps_x265_8bit_700Mbit.mkv

Decode: 66 - 80 - 93
Playback: 68 - 80 - 92

BTW, your link to test case 3 is really wrong. Issue is the wrong file name. Should be:

http://www.imaging-resource.com/PRODS/samsung-nx1/FULLRES/yvid_SAM_0230-ProQuality.MP4

[NikosD]

Okay here we go:

Tools: DXVA checker rev. 3.15, LAV Filters 0.69.0

System: Intel Core i5 7400 (Kabylake), Nvidia GTX1050 TI
Decode benchmarks settings: 64-bit
Playback benchmarks settings: 64-bit, downscaled to 1280 x 720

For all results: 5 runs, usually first run is worst. Took the last run.



Your results are perfect, containing all min/avg/max values.

We usually abandon the first run and the best run (if it's too high) and take the average of the rest.

Because maybe your last run has too high or different than average results.

Also, what GPU did you test ?

Kaby or 1050 Ti ? I think 1050 Ti, but at what clocks ? (core/ memory)

Can you run the exact same tests on both GPUs ? (Kaby and 1050 Ti)

I will upload in a moment the 10bit samples, so you can run the test on them too.

Thanks!

Update:
Your results do not much with GTX 1060.
It has a good difference.
Clocks maybe ?


BTW, your link to test case 3 is really wrong. Issue is the wrong file name. Should be:

http://www.imaging-resource.com/PRODS/samsung-nx1/FULLRES/yvid_SAM_0230-ProQuality.MP4

Right.
I updated the link.
Thanks!

[loekf]

Also, what GPU did you test ?


Only the GTX 1050 TI inside my HTPC. The iGPU of the Kabylake CPU is disabled, guess automatically via the BIOS. To test the Kabylake iGPU guess I would have to take the GPU out and install proper Intel drivers.

The 1050 TI is a MSI with 1455 MHz clock (7GHz memory). So compared to a 1060 that's roughly 15% lower. The 1060 usually also has a 8 GHz vs 7 GHz clock. Looks like the difference between the two is smaller than you would expect based on clock speeds.

I will try to make time for another round of testing.

[NikosD]

The 1050 TI is a MSI with 1455 MHz clock (7GHz memory). So compared to a 1060 that's roughly 15% lower. The 1060 usually also has a 8 GHz vs 7 GHz clock. Looks like the difference between the two is smaller than you would expect based on clock speeds.

I will try to make time for another round of testing.

Use a monitoring tool like GPU-Z to see the GPU clocks when you run the tests but write down the results without GPU-Z running.

I posted the new five HEVC 10bit clips.

[NikosD]

I will try to make time for another round of testing.

I have updated the previous post with your 8 bit results.

Waiting for new feedback regarding the 10 bit clips.

[Trevonn]

nvidia cards have a video clock. It can be seen in HWINFO so decoding speed is probably based on that

[loekf]

nvidia cards have a video clock. It can be seen in HWINFO so decoding speed is probably based on that

I had a discussion on another forum about that, so why there's a difference between a 1060 and a 1050. You could say that the video decoder block on Pascal cards (VP8) is a fixed function block, so likely should have fixed clock regardless of (what's usualy quoted) the clock the shader blocks are running on. Memory clock speed and bus width could still make a difference.

EDIT:

http://docs.nvidia.com/gameworks/content/gameworkslibrary/coresdk/nvapi/group__gpuclock.html#ga3ee5350b6ca7dce4aa076c25c8f5d66d

Nvidia GPUs have indeed four clocks, reported via the driver:

NVAPI_GPU_PUBLIC_CLOCK_GRAPHICS
NVAPI_GPU_PUBLIC_CLOCK_MEMORY
NVAPI_GPU_PUBLIC_CLOCK_PROCESSOR
NVAPI_GPU_PUBLIC_CLOCK_VIDEO

SIV can also read it out (also free):

http://rh-software.com/siv_gtx.png

[NikosD]

You could say that the video decoder block on Pascal cards (VP8) is a fixed function block, so likely should have fixed clock regardless of (what's usualy quoted) the clock the shader blocks are running on. Memory clock speed and bus width could still make a difference.

That's not true in general.

For example the Intel's iGPUs share exactly the same clock for EUs (3D execution units) with the video fixed-function block (QuickSync)

AMD cards seem to have a different clock for UVD than shaders, not exposed in any monitoring tool and the same across all Polaris cards.

But UVD especially in Polaris cards is like a black box, actually.

Now regarding Nvidia I think the truth is somewhere in the middle, meaning that it could have a different clock for VP8 but not the same across the cards.

A faster 3D clock for 1080 or 1060 than 1050 Ti, could mean a faster VP8 clock for those cards too.

So different clock than shaders for video, but not fixed across different cards but rather a fractional clock.

[NikosD]

SIV can also read it out (also free):


What SIV gives you for video clock when running DXVA Checker in benchmark mode ?

Latest version is 5.16, you have an older one.

I would suggest to verify the number with GPU-Z and HWiNFO.

[baii]

A little quick test on the 10bit clips with my 1060 3gb. My stock clock result is consistent higher than what is listed, and overclock does increase the fps. I didn't exactly check for video clocks though.
Also tested one of the 8bit clip on rx460 and the result is almost the same compare to the 470 result.

[loekf]

Quick test on the clocks on my 1050 TI. Regardless of the codec and resolution you see the video clock spike from 540 MHz idling to 1645 MHz immediately in HWinfo. Memory clock and GPU clocks go to their max as well.

So whether you doing SD x264, HD x264 or 4K HEVC, all result into the same clocks.

Could someone with a 1060, 1070 or 1080 checks the clocks ? As mentioned in my case, the video clock is 540 MHz idling (just desktop) increasing to 1645 MHz when decoding video. Clocks are stable during playback. SIV reports little higher clocks (570 MHz idle, 1835 MHz max).

I expected more intelligent clock throttling from Nvidia, but guess those are the base clocks. I'm sure they doing stuff like clock gating statically (HW block by block, so turn off what's not needed) and dynamically (directly on logic gates) to drive down power consumption.

[NikosD]

1645 MHZ is a reasonable clock for a modern video decoder.

During decoding benchmarking (not playback benchmarking) the clocks are going up and down ?

It should be max all the time during benchmarking (decode or playback)

What is the stable clock during playback ?

[loekf]

1645 MHZ is a reasonable clock for a modern video decoder.

During decoding benchmarking (not playback benchmarking) the clocks are going up and down ?

It should be max all the time during benchmarking (decode or playback)

What is the stable clock during playback ?

Stable as a rock .. so you see it climb to 1645 MHz and stay there until it finishes. As I mentioned resolution, bit rate, type of codec doesn't seem to matter. The clock doesn't change at all. Same for the other clocks.

[NikosD]

OK. ..And the other clocks during benchmarking ?

I mean GPU core and GPU memory.

Is it time to write down your 10bit results ? ;)

[ibius]

Quick test on the clocks on my 1050 TI.
As mentioned in my case, the video clock is 540 MHz idling (just desktop) increasing to 1645 MHz when decoding video. Clocks are stable during playback.

I expected more intelligent clock throttling from Nvidia, but guess those are the base clocks. I'm sure they doing stuff like clock gating statically (HW block by block, so turn off what's not needed) and dynamically (directly on logic gates) to drive down power consumption.

What kind of clock throttling do you expect during benchmarking?

I also have a 1050 Ti and during regular video playback all clocks fluctuate, as expected.

[loekf]

What kind of clock throttling do you expect during benchmarking?

I also have a 1050 Ti and during regular video playback all clocks fluctuate, as expected.

Throttling as in a different clocks depending on how many resources are used for video decoding. However, I now realise the decoder is pushed for maximum fps, so max resources, so max frequency.

Guess if you use VLC or ffmpeg you see more dynamics.

I will do the 10-bit stuff tonight.

[loekf]

My 10-bits results (MSI GTX 1050 TI 4GB):

1. UHD_PQ_Lovely_Swiss.ts

Decode: 159-179-211
Playback: 171-190-217

2. 2.Crowd_run_2160p50_Main10_100Mbps.mkv

Decode: 175-179-181
Playback: 173-176-177

3. jellyfish-140-mbps-4k-uhd-hevc-10bit.mkv

decode 130-146-165
playback 131-147-165

4. jellyfish-400-mbps-4k-uhd-hevc-10bit.mkv

Decode: 60-74-96
Playback: 60-75-96

5. Arri65_from_raw.mp4 (not 8K, but more like 6560 x 3100 pixels ?!)

Decode: 63-68-72
Playback: 64-68-72

Note: all playback benchmarks have output downscaled to 720p

Also looked at clocks during video playback using VLC. VLC 2.2.4 has some bug with DXVA2 (crashed). 3.0 beta worked fine and indeed during SD or 1080p x264 files (BD rip) the video clock "hoovers" around 540 - 570 Mhz. CPU load is not more than 4%. GPU and memory clock stay low as well (didn't write down the numbers).

[CruNcher]

1645 MHZ is a reasonable clock for a modern video decoder.

During decoding benchmarking (not playback benchmarking) the clocks are going up and down ?

It should be max all the time during benchmarking (decode or playback)

What is the stable clock during playback ?

Are you kidding for a Asic that's crazy never ever that's the VP8 Asic Frequency.

400-600 MHz sounds reasonable but never ever 1600, though in Nvidias Case we have to think indeed more about a MultiCore Structure.

And the overall Performance seems to depend on the core Count or rather a Virtualized Splitted core count.

http://i1.sendpic.org/i/bP/bPW89uDcWkoLwoDorkuSi10Zg0V.png


2) AMD RX 470 has obviously a different philosophy, covering 60fps only at "normal" bitrate clips up to 80Mbps which should cover a very large amount of real world clips out there.

I see that different i see it in the amount of Streams the ASIC is capable decoding at once the Performance can be either used to get a higher bitrate or higher resolution split.

That Nvidia can decode such high bitrates is only because their target was 8K so you either decode 2x4K(UHD) at reasonable bitrate or 1x8k at lower bitrate/FPS (half) then what is possible at 4K(UHD).

The same you could decode mostly hundreds of SD at once depending on your complexity ;)

You should more think of it like 0-100% Utilization if you dont go over the 100% Utilization you can do whatever you want, it doesn't matter what you playback or how much bitrate the highest complexity possible is the overall Performance ;)

I can easily playback 4X UHD AVC 30Mbit 30 fps Streams at once on GM204 and im not reaching the 100% ;)

4x30 = 120 Mbits @ 30 FPS

So in theory GM204 could playback 8K AVC at 60 mbits 30 fps stable ;)

http://oi66.tinypic.com/w9te09.jpg

Almost exactly the same result as if i Decode 2x the same complexity @ 60 FPS ;)

2x30 = 60 Mbits @ 60 FPS

http://oi64.tinypic.com/16krlo7.jpg

3x30 = 120 Mbits @ 60 FPS brakes apart and over utilizes the ASIC

http://oi64.tinypic.com/140cay1.jpg

[NikosD]

My 10-bits results (MSI GTX 1050 TI 4GB):


Thank you very much for your time and effort.

Since the 1st bonus clip is closer to 7K, I added a 2nd one in my 10bit HEVC list which now has 6 clips.

The 2nd bonus clip is 8K but huge (~2GB)

You can download it and test it from here:
https://mega.nz/#!8hcmwDoK!UwfMhPQvtiJw9SkZpijJijxyC4k45iqrNTwMVHjUZ6Y

My 10-bits results (MSI GTX 1050 TI
Also looked at clocks during video playback using VLC. VLC 2.2.4 has some bug with DXVA2 (crashed). 3.0 beta worked fine and indeed during SD or 1080p x264 files (BD rip) the video clock "hoovers" around 540 - 570 Mhz. CPU load is not more than 4%. GPU and memory clock stay low as well (didn't write down the numbers).

Can you use MPC-HC or MPC-BE or PotPlayer and try again ?

I would definitely like to know your GPU Video clock along with GPU core/ memory clocks but using more familiar (and better) video players during playback.

Using GPU-Z you could see the Video Engine utilisation and you could try some heavy (bitrate), high framerate clips like Netflixx 4K HEVC 8bit@60fps-700Mbps in order to see the clocks rising.

As a last favor, if you have more time and you are in a good mood, I would like to add your Kabylake's results.

I saw that it's using HD 630 iGPU at 1.0GHz.

Thanks in advance!

[NikosD]

Are you kidding for a Asic that's crazy never ever that's the VP8 Asic Frequency.

400-600 MHz sounds reasonable but never ever 1600, though in Nvidias Case we have to think indeed more about a MultiCore Structure.


I'm never kidding actually talking about HW evaluation/ benchmarks.

When Intel introduced for the first time in 2011 a highly integrated full fixed-function ASIC (QuickSync) for video decoding/encoding inside Sandybridge's (2nd generation Core) iGPU, the reason that destroyed the competition of ATI/Nvidia was mainly (among other things) it's clock ~ 1150 MHz, when ATI/Nvidia had half (or less) clock.

I was able and I'm still able to overclock this ASIC inside my Core i5-2400 at around 2000MHz even today.

Do you believe that Nvidia Pascal with those crazy GPU clocks ~2000MHz, is holding the video engine clock to a max benchmark mode (decode or playback) at ~400-600 MHz ?

Is it even possible to hit Kabylake's (7th generation Core) video decoding performance when the iGPU HD 630 of Kaby has a video clock of 1150MHz, with your Pascal's estimation of half clock (~400-600)?

Now I think you are kidding me and that was a good joke, I have to admit it.

[CruNcher]

Still you are wrong even with Decoding HEVC the Main Core has not that of a high Frequency as you suggest here or rather believe.

[loekf]

Thank you very much for your time and effort.

Since the 1st bonus clip is closer to 7K, I added a 2nd one in my 10bit HEVC list which now has 6 clips.

The 2nd bonus clip is 8K but huge (~2GB)

You can download it and test it from here:
https://mega.nz/#!8hcmwDoK!UwfMhPQvtiJw9SkZpijJijxyC4k45iqrNTwMVHjUZ6Y


Checked the 8K stream:

Decode: 53-56-58
Playback: 54-55-56

(using Nvidia's latest drivers, 378.78)

Also had a look at clocks again:

for a BD-rip, 1080p H.264, 9.2 Mbps video + 1.5Mbps DTS (bitstreaming) you see:

645 MHz for the GPU clock
570 Mhz for the GPU video clock
202.5 Mhz for the GPU memory clock

(clocks are stable, doesn't matter whether you have a quiet scene or a scene full of action and CGI)

suppose you take a 4K, HEVC file, 140 Mbps:

1835 MHz for the GPU clock
1645 Mhz for the video clock
1752 MHz for the memory clock

(using MPC-HE, x64, clocks traced using HWInfo)

Kabylake still pending.

[NikosD]

Checked the 8K stream:

Decode: 53-56-58
Playback: 54-55-56

(using Nvidia's latest drivers, 378.78)


Thanks, I will update my post.
It seems that we managed to find a file that Pascal 1050 Ti can't decode in real time, although it's close.

Also had a look at clocks again:

suppose you take a 4K, HEVC file, 140 Mbps:

1835 MHz for the GPU clock
1645 Mhz for the video clock
1752 MHz for the memory clock

It makes sense, that file needs maximum clocks, so I consider the 1645 MHz as benchmark mode clock (for playback and decode mode)

Next time see the video decoding utilisation.
Nvidia is the only company providing this info and is useful because you can see the margins of video decoding performance.
Use GPU-Z for that.

Kabylake still pending.

That would be extremely useful in order to compare it with Pascal.

[CruNcher]

Samsungs UHD Highlight Demo of

https://www.youtube.com/watch?v=pnuj1svkH8w

is slowly leaking through it seems ;)

[loekf]

Next time see the video decoding utilisation.
Nvidia is the only company providing this info and is useful because you can see the margins of video decoding performance.
Use GPU-Z for that.


Actually DXVA checker also reports the GPU (and CPU) load. It's usually always > 90% and breathing a little bit (a few %).

I guess the video decoder (nvdec or purevideo) utilization scales with clock speed. Not sure how it is calculated. The Nvidia SDK docs show little information, e.g.:

http://developer.download.nvidia.com/compute/cuda/5_5/rel/nvml/nvml.5.319.43.pdf

In other words, the video clock scales up or down based on what you "throw" at it, so the driver and GPU HW will always scale the clocks such that the load fits and the clocks are as low as possible (engineering hat on: make them such that you can process @ 90% + 10% margin).

So.. ideally the load should be around 90%.

[NikosD]

Actually DXVA checker also reports the GPU (and CPU) load. It's usually always > 90% and breathing a little bit (a few %).

So.. ideally the load should be around 90%.

What I'm talking about has absolutely nothing to do with GPU and CPU load.

Those two should be <5% or even ~0% during decode benchmark.

But for playback benchmark mode, yes the GPU utilisation should be close to 100%

DXVA checker for Nvidia cards should show a 3rd utilisation of Video decoder.

That is what I ask.

You can play the HEVC 10bit 140Mbps clips from inside DXVA Checker using normal playback (not benchmark) and check that 3rd parameter.

Otherwise just use GPU-Z sensors tab that report video utilisation.

[mixus]

Note :
4K 120 fps = 8K 30 fps
4K 240 fps = 8K 60 fps


8K - 30 fps = OK
4K - 180 fps = OK
8K - 60 fps = FAIL

If You Want 8K 60 Fps Wait Nvidia Volta or AMD Navi


Result
http://s6.dosya.tc/server8/mznlzn/Result.rar.html

[NikosD]

Note :
4K 120 fps = 8K 30 fps
4K 240 fps = 8K 60 fps


8K - 30 fps = OK
4K - 180 fps = OK
8K - 60 fps = FAIL

If You Want 8K 60 Fps Wait Nvidia Volta or AMD Navi


Result
http://s6.dosya.tc/server8/mznlzn/Result.rar.html

Hey,

I can see you have a Pascal GTX 1060.

What is your clocks (GPU/Memory/Video clock) during benchmarks and is there any difference in performance between your card and 1050 Ti that I have already posted ?

If performance difference is big, maybe you could post your results in my samples collection in order to put them in that comparison table.

Also, are there any links for those clips that you have provided screenshots only ?

Now, as a last point, I don't know about Volta but AMD's Navi is very difficult to include a 8K60fps decoder, simply because AMD is not interested in any non-consumer resolution, video codec, bitrate etc.

Polaris UVD has included HEVC decoding/ encoding after years of the HEVC release just because BluRay Ultra HD is close and has started to move a little.

We have to wait enough for AMD to give us 8K60fps decoder.

[mixus]

Hey,

I can see you have a Pascal GTX 1060.

What is your clocks (GPU/Memory/Video clock) during benchmarks and is there any difference in performance between your card and 1050 Ti that I have already posted ?

If performance difference is big, maybe you could post your results in my samples collection in order to put them in that comparison table.

Also, are there any links for those clips that you have provided screenshots only ?

Now, as a last point, I don't know about Volta but AMD's Navi is very difficult to include a 8K60fps decoder, simply because AMD is not interested in any non-consumer resolution, video codec, bitrate etc.

Polaris UVD has included HEVC decoding/ encoding after years of the HEVC release just because BluRay Ultra HD is close and has started to move a little.

We have to wait enough for AMD to give us 8K60fps decoder.

Yeah man, I have GTX 1060.

I think AMD and Nvidia will add to the next architecture.

I prepared video for you. please download and look.

http://s6.dosya.tc/server8/285lne/GTX_1060.rar.html

[NikosD]

Yeah man, I have GTX 1060.

I think AMD and Nvidia will add to the next architecture.

I prepared video for you. please download and look.

http://s6.dosya.tc/server8/285lne/GTX_1060.rar.html

Nice video, thank you for your time and effort.

I'm afraid you didn't answer a few questions.

Are those videos in public or private ?

Are there any links for 8K videos ?

I saw that you have also a HD 530 iGPU and you are familiar with DXVA Checker.

Why don't you test my samples collection with both your GPUs ? (1060 and HD 530)
If you have time and you are in a mood of course.

Now, the GPU core clock of your Pascal 1060 is huge - 2GHz - but GPU-Z doesn't provide the GPU Video clock.

You have to use HWiNFO in order to find out the GPU Video Clock during benchmarking.

One last thing.

I didn't see LAV video in your codecs.
Don't you use it ?

[NikosD]

Actually DXVA checker also reports the GPU (and CPU) load. It's usually always > 90% and breathing a little bit (a few %).


Now that I saw this video http://s6.dosya.tc/server8/285lne/GTX_1060.rar.html the monitoring parameter I was talking about regarding DXVA Checker is called VE usage (Video Engine usage) and it's right next to GPU/VE usage ;)

[aufkrawall]

The shown usage might be as mislading as the shown GPU usage as long as the GPU isn't running with full clocks.
Means if it only runs with ~1GHz but shows 50% VE/VPU usage, the latter value might be much lower for the same task if the GPU runs with a higher frequency.
So it's not a good indicator for the actual performance.

Btw: Dunno if this has already been mentioned here, but I've been told that all Pascal GPUs except of GP104 are able to decode VP9 10/12 bit.

[sneaker_ger]

Told by whom? GTX 1060 is GP106 and in tests it didn't show VP9 10 bit. (Haven't seen any Titan X or GTX 1080 Ti tests). And I don't remember anyone saying VP9 12 bit is supported on any Pascal card.

[aufkrawall]

Told by somebody close to Nvidia:
https://www.forum-3dcenter.org/vbulletin/showpost.php?p=11298988&postcount=36
You can question how well communication between Nvidia and their internal testers is, but unless you have a better bet...

[huhn]

just take this source:
https://developer.nvidia.com/nvidia-video-codec-sdk

some pascal card can do 12 bit VP9.

[sneaker_ger]

I see. Interesting.

[nevcairiel]

Yet 1060 (GP106) cards do not support it as of the most recent drivers, while we've heard that a 1050 does. So it would seem either some implementation isn't quite there yet (which would be odd if it shows up and works on a 1050), or the information isn't fully accurate.

[mixus]

Nice video, thank you for your time and effort.

I'm afraid you didn't answer a few questions.

Are those videos in public or private ?

Are there any links for 8K videos ?

I saw that you have also a HD 530 iGPU and you are familiar with DXVA Checker.

Why don't you test my samples collection with both your GPUs ? (1060 and HD 530)
If you have time and you are in a mood of course.

Now, the GPU core clock of your Pascal 1060 is huge - 2GHz - but GPU-Z doesn't provide the GPU Video clock.

You have to use HWiNFO in order to find out the GPU Video Clock during benchmarking.

One last thing.

I didn't see LAV video in your codecs.
Don't you use it ?


8K Video from www.youtube.com/user/martinheck1

8K 29.970 fps - Youtube MP4 66 Mbps.
8K 60 fps - Youtube 30 fps VP9 15 Mbps. (Converted 60 fps with http://www.spirton.com/convert-videos-to-60fps/) + Hybrid "www.selur.de"
4K 60 fps - http://demo-uhd3d.com/fiche.php?cat=uhd&id=152 (Converted 60 fps to 180 fps)

Sources Converted to HEVC Main10. (used Hybrid "www.selur.de")


I Don't know Both usage. Sorry ( HD 530 + Nvidia )

I checked with HWiNFO64. GPU Video Clock 1700 MHz. http://sendpic.org/view/1/i/dUPrpylf2BURgwk8zZwvJKui7IZ.png

I don't love MPC-HC. MPC-BE is better than MPC-HC.

Note: My upload speed 0.65 mbps. i don't make upload :(

I Know lav filters :)

[GTPVHD]

Unless someone with a GP102/GP104/GP106 running the latest 378.78 or later drivers & DXVAChecker posts a screenshot showing VP9_VLD_10bit_Profile2 like the link below.

http://forum.doom9.org/showthread.php?p=1787451#post1787451

I would find it hard to believe those GPUs suddenly picked up VP9 Profile2 support.

[NikosD]

Thanks for the info.


I Don't know Both usage. Sorry ( HD 530 + Nvidia )


What do you mean ?
You just plug a cable to HD 530 output and your monitor and run the tests selecting HD 530 from DXVA Checker and then Pascal 1060 :)


I checked with HWiNFO64. GPU Video Clock 1700 MHz. http://sendpic.org/view/1/i/dUPrpylf2BURgwk8zZwvJKui7IZ.png


Thanks.

[huhn]

Unless someone with a GP102/GP104/GP106 running the latest 378.78 or later drivers & DXVAChecker posts a screenshot showing VP9_VLD_10bit_Profile2 like the link below.

http://forum.doom9.org/showthread.php?p=1787451#post1787451

I would find it hard to believe those GPUs suddenly picked up VP9 Profile2 support.

only newer cards like the GP107 have this decoder this is no secret. maybe the 1080 ti will get it but it is still a GP102 so i don't think so.

* HEVC/VP9 10/12 bit decoding SW support coming in Video Codec SDK 8.0
** Max resolution support is limited to selected Pascal chips
*** VP8 decode support is limited to selected Pascal chips
**** VP9 10/12 bit decode support is limited to select Pascal chips

[loekf]

only newer cards like the GP107 have this decoder this is no secret. maybe the 1080 ti will get it but it is still a GP102 so i don't think so.

AFAIK, the release notes for the latest driver show:

http://us.download.nvidia.com/Windows/378.78/378.78-win10-win8-win7-desktop-release-notes.pdf

New Features
* Added support for the following products:
GeForce GTX 1080 Ti
GeForce GTX 1050 notebook
GeForce GTX 1050 Ti notebook
* Video SDK 8.0
High-bit-depth (10/12-bit) decoding (VP9/HEVC)
OpenGL input surface support for encoder
Weighted Prediction
H.264 ME-only mode enhancements

All Pascal based cards have the same video decoder, so GP102 - GP107. Difference is in clock speeds, #shaders, ROPs and memory size & speed.

I have never seen files/streams in VP9 format, I know Youtube serves them if you use Chrome/Edge/FF in some cases.

Found this one:

https://www.vcodex.com/news/a-nice-example-of-4k-video-encoded-using-vp9/

It's 4K, but ~13 Mbps, not very challenging... should be ...

[nevcairiel]

All Pascal based cards have the same video decoder, so GP102 - GP107.


They do not. Not sure how you draw that conclusion from the release notes, it doesn't mention anything about hardware support of those features either way.
The only semi-official word we received was that GP104 does not have VP9 10-bit support, but it was also claimed that GP106 does, which is either still disabled in current drivers, or wrong information - in which case only the GP107 in the 1050 has VP9 10/12-bit support.

[loekf]

They do not. Not sure how you draw that conclusion from the release notes, it doesn't mention anything about hardware support of those features either way.
The only semi-official word we received was that GP104 does not have VP9 10-bit support, but it was also claimed that GP106 does, which is either still disabled in current drivers, or wrong information - in which case only the GP107 in the 1050 has VP9 10/12-bit support.

Yeah, read the stuff on either videohelp or hardocp. Still puzzling why a GP107 would do 10-bits VP9 and a more expensive sibling like GP104 or GP106 would not.

Maybe it's tapeout related. GP107 got introduced after GP104 and GP106, so perhaps GP107 has video IP with a higher revision number (bug fixes).

EDIT: when I go this Chinese site DXVA checker reports VP9 profile 0 decode for GTX 1060 (= GP106):

https://www.mobile01.com/topicdetail.php?f=298&t=4866089

https://attach.mobile01.com/attach/201607/mobile01-3988607dd710f26cad9bbd49cf1b5bf0.jpg

And GTX1070, via https://forums.geforce.com/default/topic/989353/geforce-1000-series/cannot-get-hevc-x265-to-work-under-any-circumstances-could-i-have-a-hardware-defect-/

http://i.imgur.com/dTweKQE.jpg

and GTX1080:

http://www.notebookcheck.net/fileadmin/Notebooks/Sonstiges/Grafikkarten/NVidia/GTX1000/dxvachecker.png

So:

GP106 (GTX 1060), GP107 (GTX1050): support VP9
GP104 (GTX1070 and GTX1080): should be, think just rumours. Makes no sense if GTX1070 can do VP9, but GTX1080 can not.
GP102 (Titan etc, GTX1080 TI): got introduced later, should be

[wanezhiling]

when I go this Chinese site DXVA checker reports VP9 profile 0 decode for GTX 1060 (= GP106):
That's 8-bit VP9, not 10-bit.


Look at GP107 here: http://forum.doom9.org/showthread.php?p=1787451#post1787451

[CruNcher]

Yeah, read the stuff on either videohelp or hardocp. Still puzzling why a GP107 would do 10-bits VP9 and a more expensive sibling like GP104 or GP106 would not.

Maybe it's tapeout related. GP107 got introduced after GP104 and GP106, so perhaps GP107 has video IP with a higher revision number (bug fixes).

EDIT: when I go this Chinese site DXVA checker reports VP9 profile 0 decode for GTX 1060 (= GP106):

https://www.mobile01.com/topicdetail.php?f=298&t=4866089

https://attach.mobile01.com/attach/201607/mobile01-3988607dd710f26cad9bbd49cf1b5bf0.jpg

And GTX1070, via https://forums.geforce.com/default/topic/989353/geforce-1000-series/cannot-get-hevc-x265-to-work-under-any-circumstances-could-i-have-a-hardware-defect-/

http://i.imgur.com/dTweKQE.jpg

and GTX1080:

http://www.notebookcheck.net/fileadmin/Notebooks/Sonstiges/Grafikkarten/NVidia/GTX1000/dxvachecker.png

So:

GP106 (GTX 1060), GP107 (GTX1050): support VP9
GP104 (GTX1070 and GTX1080): should be, think just rumours. Makes no sense if GTX1070 can do VP9, but GTX1080 can not.
GP102 (Titan etc, GTX1080 TI): got introduced later, should be

Market, Target, Design choices anyways Nvidias was allways very clever where and when they support something at the right time to market ;)

Also what endsup in the PC had no priority yet because of the Security Architecture missing, thus why exposing of features was always first in Tegra where the Security architecture is fully actively in Place.

See the new Andorid Shield Maxwell shrink i bet with you it has the full GP107 VPx features ;)

In general Nvidias strategy is always logical based on the overall Target Systems capabilities most of the time.

So the more powerfull VPX in the low end GPU consumer hardware tries to compensate for a weaker overall System ;)

And then they make a decission sometime if it makes sense for the target to bring some features on the higher end via the Shader capabilities (invest into additional Software development and R&D) (first then these come mainly for the Professional GPUs, with a lot of Cuda Power).

So Nvidia might still expose Hybrid 10 bit support especially on Titan X, 1080 TI and GP102, not sure though how much sense it really would make, so close before Volta with the complete overhaul.

Also VP9 has not that support internaly they fully working on AV1.

[NikosD]

Yeah, read the stuff on either videohelp or hardocp. Still puzzling why a GP107 would do 10-bits VP9 and a more expensive sibling like GP104 or GP106 would not.

Maybe it's tapeout related. GP107 got introduced after GP104 and GP106, so perhaps GP107 has video IP with a higher revision number (bug fixes).

EDIT: when I go this Chinese site DXVA checker reports VP9 profile 0 decode for GTX 1060 (= GP106):

GP106 (GTX 1060), GP107 (GTX1050): support VP9
GP104 (GTX1070 and GTX1080): should be, think just rumours. Makes no sense if GTX1070 can do VP9, but GTX1080 can not.
GP102 (Titan etc, GTX1080 TI): got introduced later, should be

I really like guys who like to search for sources, googling in various sites, read Nvidia's (AMD's, Intel's) docs etc.

I'm one of them actually.

But I have to tell you sincerely, that the greatest source of almost every publicly available info regarding those issues is here and sometimes months before anyone else write about it ;)

You must read carefully and read more of this thread in order to get replies and solve mysteries regarding SW/HW video decoding.

And obviously I don't mean only myself, I just happened to start this thread.

[loekf]

You must read carefully and read more of this thread in order to get replies and solve mysteries regarding SW/HW video decoding.

And obviously I don't mean only myself, I just happened to start this thread.

Yeah, saw that too. Was too lazy to read through all 90+ pages. Appearently it was already known that Nvidia is doing strange things with their video IP (put a more advanced decoder in your later introduced low-end product). As someone also mentioned, high-end GPUs are targetting gamers, who are mostly interested in graphics' fps, not HEVC fps...

[mixus]

Thanks for the info.



What do you mean ?
You just plug a cable to HD 530 output and your monitor and run the tests selecting HD 530 from DXVA Checker and then Pascal 1060 :)



Thanks.

Hi bro.

HD 530 is very bad gpu. Because cannot play 4K main10 HEVC 30p or 60p.

[mixus]

I Installed Lastest Driver. 378.78

http://i.hizliresim.com/4PbBap.png

Note : VLC 3 now playable HEVC 4K and 8K.

http://i.hizliresim.com/m31YXP.png

http://i.hizliresim.com/j8Vkqj.png

[NikosD]

Hi bro.

HD 530 is very bad gpu. Because cannot play 4K main10 HEVC 30p or 60p.
It doesn't have 10bit HEVC decoder.

But you could test 8bit HEVC clips only.

[mixus]

It doesn't have 10bit HEVC decoder.

But you could test 8bit HEVC clips only.

For you :


INTEL HD 530 - HEVC 8 Bit

HEVC 4K MAIN 30 FPS 24 mbps --- Clip Name : LG_4K_OLED_Wonders
Decoder: Microsoft H265 Video Decoder MFT
Decoder Device: HEVC_VLD_Main
Frames: 7110
Time: 26,657 sec
FPS: 266,724 [235-301] fps
CPU Usage: 1 [0-3] %
GPU Usage: -

HEVC 4K MAIN 60 FPS 41 mbps --- Clip Name : Samsung_SUHD_Sensational_Picture
Decoder: Microsoft H265 Video Decoder MFT
Decoder Device: HEVC_VLD_Main
Frames: 7110
Time: 27,857 sec
FPS: 255,228 [131-290] fps
CPU Usage: 1 [0-5] %
GPU Usage: -

HEVC 8K MAIN 30 FPS 140 mbps --- Clip Name : NEW ZEALAND ASCENDING "Youtube"
Decoder: Microsoft H265 Video Decoder MFT
Decoder Device: HEVC_VLD_Main
Frames: 3620
Time: 54,951 sec
FPS: 65,877 [59-77] fps
CPU Usage: 1 [0-3] %
GPU Usage: -


Note:
MPC-HC + LAV Filters 0.69 (Copy-Back and Native)
Not played 8K Main 30p

Extra:
http://i.hizliresim.com/695AzP.png

[NikosD]

For you :

Extra:
http://i.hizliresim.com/695AzP.png

Oh man.

You really like doing things your own way!

Thanks for all this info, but we have here some specifics clips that I told you to test, using LAV video actually.

There is a collection of 8bit HEVC and 10bit HEVC clips in order for all of us to test the same clips.

Nevermind, thanks.

You made me smile :D

[mixus]

Oh man.

You really like doing things your own way!

Thanks for all this info, but we have here some specifics clips that I told you to test, using LAV video actually.

There is a collection of 8bit HEVC and 10bit HEVC clips in order for all of us to test the same clips.

Nevermind, thanks.

You made me smile :D

:D


I5 6600K 4.4 GHz - VP9 8K Profile 0 15 mbps
http://i.hizliresim.com/1LMzY1.png

I5 6600K 4.4 GHz - H264 8K 66 mbps
http://i.hizliresim.com/Ly19pJ.png

INTEL HD 530 - 4K 120 fps HEVC Main 8 Bit 223 Mbps
http://i.hizliresim.com/r3ZPQB.png

INTEL HD 530 - 4K 120 fps HEVC Main 8 Bit 223 Mbps
http://i.hizliresim.com/7qMRkm.png

INTEL HD 530 - 4K 120 fps HEVC Main 8 Bit 223 Mbps
http://i.hizliresim.com/P0PXqv.png


Final

HD 530 Optimized HEVC 4K Main 60 Fps and AVC 4K 60 Fps.

Not Ready Main10.

[mixus]

GTX 1060 6 GB - 4K 120 fps HEVC Main 8 Bit 223 Mbps
http://beta.hizliresim.com/ldvnvb

[P.J]

EVGA GeForce GTX 960 SuperSC ACX 2.0+ 2GB (GPU Video Clock: 1202.3 MHz)
Win 10 x64 1607, Forceware 378.78, LAV x64 0.69.0.14

1) LG_4K_View-the-Feeling (2160p@29.970fps ~25Mbps)

Decode: 137/145/163
Playback: 141/145/165

2) Samsung 4K@30fps-80Mbps file -0235

Decode: 119/120/122
Playback: 121/123/126

3) Samsung 4K@30fps-80Mbps file -0230

Decode: 114/115/117
Playback: 116/116/121

4) Crowd_run 4K@50fps -300Mbps

Decode: 83/93/97
Playback: 83/94/98

5) 1st Bonus clip:
Netflix_TunnelFlag_4096x2160_60fps_x265_8bit_700Mbit

Decode: 48/59/70
Playback: 47/59/71

1.UHD_PQ_Lovely_Swiss-51Mbps@60fps

Decode: 112/124/153
Playback: 111/122/155

2.Crowd_run_2160p50_Main10_100Mbps

Decode: 108/109/113
Playback: 108/112/115

3.Jellyfish-140-mbps-4k-uhd-hevc-10bit

Decode: 94/103/114
Playback: 92/102/113

4.Jellyfish-400-mbps-4k-uhd-hevc-10bit

Decode: 38/55/71
Playback: 44/55/72

[NikosD]

Perfect !

Thank you.

[Kadano]

Hi all,

I recently bought an Intel G4560 because of its amazing performance per price ratio. However, the one potential problem I see with it is the integrated graphics HD 610, which seems to be half of a HD 630 used in all Kaby Lake CPUs that are sold for 85€ or more.

Is it known whether the HD 610 is also chopped down in encoding / decoding ability or is the performance only reduced for 3D / gaming applications?
If it is not known, I'd like to help by running tests on the G4560.

[NikosD]

l
Is it known whether the HD 610 is also chopped down in encoding / decoding ability or is the performance only reduced for 3D / gaming applications?
If it is not known, I'd like to help by running tests on the G4560.

The 3D power of Intel's IGPU plays a significant role in HW encoding, because Intel uses hybrid encoding and not pure fixed function (actually it has a choice of pure fixed function but it's crippled)

But in HW pure decoding only the clock really matters *

In playback benchmark mode and realtime playback, 3D power plays a role in scaling playback and also using EVR CP or even heavier renderers, like MadVR.

Anyway, I'm personally more than interested in testing your IGPU in all video samples using both decode and playback mode.

If you do it, post your results here or send me a PM.

* There are cases that the higher GPU models have more HW units than the lower but I think it only matters for encoding not decoding (they are encoders actually)

[Kadano]

Anyway, I'm personally more than interested in testing your IGPU in all video samples using both decode and playback mode.

If you do it, post your results here or send me a PM.
How do I run decode / playback benchmark with LAV Filters? Do I need to set some flag in PotPlayer / mpc-hc? I've looked through your posts but I can't find any information on how to prepare the testing settings.

[Kadano]

How do I set up the testing environment? I've looked through your post, but wasn't able to find any details.

In DXVA Checker > "DirectShow/MediaFoundation Decoder", do I set the Target to "DS + MF (32bit)" or something else? Also, after "Re-Check (Connect to Renderer)" on LAV Video Decoder, the text turns red. Does this mean there is something wrong?

Also, do I use 32bit or 64bit for playback?

[NikosD]

OK here we go:

Use LAV Video x64 one of the latest versions along with DXVAChecker v3.15 and you select "DS + MF 64bit" for all benchmark modes (decode & playback)

For playback your selection should be 1280 x 720.

The LAV video turning red means success.
It successfully connected to the renderer (EVR) using DXVA2.

Run each of the 12 files at least 4 times for each mode and write down the average of min/avg/max, usually excluding the first run because it loads the file.

Be extra careful to disable everything from the Intel's control panel, because Intel by default enables a few options for "better" video quality.

Be sure to disable everything.

[CruNcher]

This is really interesting

http://images.anandtech.com/doci/11229/cropped_evrcp_575px.png
http://images.anandtech.com/doci/11229/cropped_kodi_575px.png


Though i see absolutely no information about Anandtechs Testing Methodology here on the EVR-CP side it is MPC-HC though Fullscreen Ex or Fullscreen Windowed on Win 10

for UHD it is definitely a master difference at those 12W @ peak between glitch free playback and not on Kodi with a peak GPU load of ~5% at ~7W

[wanezhiling]

No doubt about it, vanilla EVR is always the best&only choice for weak platforms.

[CruNcher]

yeah also comes close on my own cherry trail results but there the overhead also came from the gui rendering and timers of MPC-BE, especially the seekbar was critical even with EVR.

So i wonder if Anandtechs result is with any gui rendering or without

also in case of MadVR im not sure didn't test it on low power but with the options all on lowest and DXVA all the time it might be ok but in general then still the CPU overhead is higher anyways then even EVR-CP.

[clsid]

Why is the GPU usage so high? Did they perhaps use PS2.0 bicubic resize? DXVA scaling would be interesting is MPC-HC ever gets it.

[pirlouy]

No doubt about it, vanilla EVR is always the best&only choice for weak platforms.
Maybe this has already been asked but why madVR which seems to be well coded can compete with EVR if you you use basic filters (bilinear) and you enable all "trade quality for performance" settings ?

[huhn]

madVR is nothing a modern iGPU can't handle.

adding DXVA scaling to EVR sounds pretty pointless to me. DXVA scaling is too limited it only works with a low number of pixel formats.

[NikosD]

Why is the GPU usage so high? Did they perhaps use PS2.0 bicubic resize? DXVA scaling would be interesting is MPC-HC ever gets it.
Kadano, a user with iGPU HD 610 just informed me that if you don't disable the default "Noise reduction" from the control panel you get 70fps.

If you disable it, you get 122 fps.

Anandtech could have easily forgotten the default settings on, something that I always suggest to disable.

[pirlouy]

madVR is nothing a modern iGPU can't handle.But why ? If you use dxva scaling and all "trade quality" settings, madVR should be at the same level, no ?

[CruNcher]

Kadano, a user with iGPU HD 610 just informed me that if you don't disable the default "Noise reduction" from the control panel you get 70fps.

If you disable it, you get 122 fps.

Anandtech could have easily forgotten the default settings on, something that I always suggest to disable.

Nope if you read also previous htpc targeted platform tests you will see they are aware of the Driver Settings Influence on the Shader Overhead and recommend to disable any post processing on the Driver side especially with MadVR.

But why ? If you use dxva scaling and all "trade quality" settings, madVR should be at the same level, no ?

You should think so but MadVR has also has by default no matter which settings you chose a higher CPU overhead and consumes more CPU cycles which can get critical as well @ 6W-12W overall CPU/GPU target ;)

[nevcairiel]

But why ? If you use dxva scaling and all "trade quality" settings, madVR should be at the same level, no ?

huhn used double negative, meaning that even iGPUs can handle madVR if you stick to modest settings.

[vlado08]

Intel NUC7i3BNH
CPU: i3-7100U 2.4Ghz
GPU: Intel HD 620
RAM: 16GB DDR4-2132MHz - dual channel
Windows 10 x64 1607 os build 14393.969
Graphics driver 21.20.16.4590
LAV Video x64 0.69.0
DXVA Checker 3.15.0


Power Options - High Performance
(GPU @ 1000Mhz during benchmark)
video enhancement - disable; monitor@1080/59p

HEVC 8bit:

1) LG_4K_View-the-Feeling (2160p@29.970fps ~25Mbps)

Decode: 156/170/195 CPU 3[0-6]
Playback: 146/160/187 CPU 5[3-10]

2) Samsung 4K@30fps-80Mbps file -0235

Decode: 181/188/194 CPU 3[1-5]
Playback: 172/177/180 CPU 4[2-7]

3) Samsung 4K@30fps-80Mbps file -0230

Decode: 153/156/158 CPU 3[1-4]
Playback: 148/149/152 CPU 5[3-7]

4) Crowd_run 4K@50fps -300Mbps

Decode: 101/110/113 CPU 3[1-6]
Playback: 96/106/109 CPU 4[1-7]

5) 1st Bonus clip:
Netflix_TunnelFlag_4096x2160_60fps_x265_8bit_700Mbit

Decode: 62/76/88 CPU 2[1-7]
Playback: 62/75/85 CPU 4[2-8]

6) 2nd Bonus clip:
8k_ex_sample 8k@24fps - 53Mbps

Decode: 42/46/48 CPU 2[0-5]
Playback: 41/44/47 CPU 3[2-5]

HEVC 10bit:

2.Crowd_run_2160p50_Main10_100Mbps

Decode: 155/157/158 CPU 4[3-6]
Playback: 137/139/141 CPU 6[5-6]

3.Jellyfish-140-mbps-4k-uhd-hevc-10bit

Decode: 117/133/151 CPU 4[2-11]
Playback: 111/123/139 CPU 6[4-11]

4.Jellyfish-400-mbps-4k-uhd-hevc-10bit

Decode: 54/67/89 CPU 4[1-9]
Playback: 53/66/85 CPU 5[2-13]

5.1st Bonus clip:
Arri65_from_raw 6560x3100@23.976fps 263Mbps

Decode: 54/56/58 CPU 3[1-7]
Playback: 48/51/53 CPU 6[4-10]

6.2nd Bonus clip:
MBC UHD Music Core GFriend - Rough (4in1) 8k
7680x4320@59.94fps 82.4Mbps

Decode: 43/48/56 CPU 3[0-8]
Playback: 38/41/47 CPU 4[0-10]

[vlado08]

Intel NUC7i3BNH
CPU: i3-7100U 2.4Ghz
GPU: Intel HD 620
RAM: 16GB DDR4-2132MHz - dual channel
Windows 10 x64 1607 os build 14393.969
Graphics driver 21.20.16.4590
LAV Video x64 0.69.0
DXVA Checker 3.15.0


Power Options - High Performance
(GPU @ 1000Mhz during benchmark)
video enhancement - disable; monitor@1080/59p

HEVC 8bit:

1) LG_4K_View-the-Feeling (2160p@29.970fps ~25Mbps)

Decode: 156/170/195 CPU 3[0-6]
Playback: 146/160/187 CPU 5[3-10]

2) Samsung 4K@30fps-80Mbps file -0235

Decode: 181/188/194 CPU 3[1-5]
Playback: 172/177/180 CPU 4[2-7]

3) Samsung 4K@30fps-80Mbps file -0230

Decode: 153/156/158 CPU 3[1-4]
Playback: 148/149/152 CPU 5[3-7]

4) Crowd_run 4K@50fps -300Mbps

Decode: 101/110/113 CPU 3[1-6]
Playback: 96/106/109 CPU 4[1-7]

5) 1st Bonus clip:
Netflix_TunnelFlag_4096x2160_60fps_x265_8bit_700Mbit

Decode: 62/76/88 CPU 2[1-7]
Playback: 62/75/85 CPU 4[2-8]

6) 2nd Bonus clip:
8k_ex_sample 8k@24fps - 53Mbps

Decode: 42/46/48 CPU 2[0-5]
Playback: 41/44/47 CPU 3[2-5]

HEVC 10bit:

2.Crowd_run_2160p50_Main10_100Mbps

Decode: 155/157/158 CPU 4[3-6]
Playback: 137/139/141 CPU 6[5-6]

3.Jellyfish-140-mbps-4k-uhd-hevc-10bit

Decode: 117/133/151 CPU 4[2-11]
Playback: 111/123/139 CPU 6[4-11]

4.Jellyfish-400-mbps-4k-uhd-hevc-10bit

Decode: 54/67/89 CPU 4[1-9]
Playback: 53/66/85 CPU 5[2-13]

5.1st Bonus clip:
Arri65_from_raw 6560x3100@23.976fps 263Mbps

Decode: 54/56/58 CPU 3[1-7]
Playback: 48/51/53 CPU 6[4-10]

6.2nd Bonus clip:
MBC UHD Music Core GFriend - Rough (4in1) 8k
7680x4320@59.94fps 82.4Mbps

Decode: 43/48/56 CPU 3[0-8]
Playback: 38/41/47 CPU 4[0-10]

[Kadano]

I tested Intel Pentium G4560 / HD 610. No dedicated GPU attached. 1x 8GB DDR3 RAM (1600MHz) used.

8bit HEVC

1) LG_4K_View-the-Feeling (2160p@29.970fps ~25Mbps)

Decode: 165 fps [138-194]
Playback [1280x720]: 123 fps [97-147]

2) Samsung 4K@30fps-80Mbps file -0235

Decode: 173 fps [134-183]
Playback [1280x720]: 140 fps [130-151]

3) Samsung 4K@30fps-80Mbps file -0230

Decode: 153 fps [149-158]
Playback [1280x720]: 122 fps [107-128]

4) Crowd_run 4K@50fps -300Mbps

Decode: 109 fps [97-113]
Playback [1280x720]: 92 fps [72-98]

5) 1st Bonus clip:
Netflix_TunnelFlag_4096x2160_60fps_x265_8bit_700Mbit

Decode: 80 fps [64-96]
Playback [1280x720]: 72 fps [57-87]

6) 2nd Bonus clip (8K):
8K_Ex_sample- 7680x4320@24fps-53Mbps

Decode: 46 fps [41-49]
Playback [1280x720]: 41 fps [36-43]


10bit HEVC 4K/8K samples

1) 1.UHD_PQ_Lovely_Swiss-51Mbps@60fps

Decode: 115 fps [74-152]
Playback [1280x720]: 76 fps [52-87]

2.Crowd_run_2160p50_Main10_100Mbps

Decode: 118 fps [112-121]
Playback [1280x720]: 75 fps [68-79]

3.Jellyfish-140-mbps-4k-uhd-hevc-10bit

Decode: 110 fps [94-124]
Playback [1280x720]: 77 fps [65-83]

4.Jellyfish-400-mbps-4k-uhd-hevc-10bit

Decode: 65 fps [53-83]
Playback [1280x720]: 55 fps [44-68]

5) 1st Bonus clip (7K):
5.Arri65_from_raw-6560x3100-263Mbps@24fps

Decode: 47 fps [40-50]
Playback [1280x720]: 35 fps [30-37]

6) 2nd Bonus clip (8K):
MBC UHD Music Core GFriend - Rough (4in1) 8k@60fps-80Mbps

Decode: 35 fps [31-41]
Playback [1280x720]: 25 fps [24-29]



Detailed results: https://docs.google.com/spreadsheets/d/1p_z3pLO_zFvcB0A57WXDILUwOjRFWbcETYG8SOIURxI/

[NikosD]

I tested Intel Pentium G4560 / HD 610. No dedicated GPU attached. 1x 8GB DDR3 RAM (1600MHz) used.


That's perfect thank you.

But where are the above 4K results ?

I think that HD 610 supports up to 8K acceleration.

Also, can you confirm using GPU-Z or HWINFO that GPU clock is 1050MHz (or something else ?)

[NikosD]

I tested Intel Pentium G4560 / HD 610. No dedicated GPU attached. 1x 8GB DDR3 RAM (1600MHz) used.


Now that I put your data on the table with the rest of the HW video decoders, they look more interesting and helpful than what I was expecting to be.

Yes, I was expecting a drop in playback performance due to only 12 EUs since HD 610 is just a GT1 version, but even pure decoding performance is lower than I was personally expecting.

Its decoding performance is better for 8bit clips than 10bit, but even in the worst case scenario of 10bit playback performance, it's still faster than Polaris UVD, but most of the times it's slower than everything else (in worst case scenario)

I have deleted the HD 630 results because they were incomplete and I'm waiting for your results on the rest 4 clips.

[Lollerke]

Now that I put your data on the table with the rest of the HW video decoders, they look more interesting and helpful than what I was expecting to be.

Yes, I was expecting a drop in playback performance due to only 12 EUs since HD 610 is just a GT1 version, but even pure decoding performance is lower than I was personally expecting.

Its decoding performance is better for 8bit clips than 10bit, but even in the worst case scenario of 10bit playback performance, it's still faster than Polaris UVD, but most of the times it's slower than everything else (in worst case scenario)

I have deleted the HD 630 results because they were incomplete and I'm waiting for your results on the rest 4 clips.

Maybe because the RAM is single channel and runs only at 1600MHz? With integrated GPU's fast dual channel RAMs are essential for gaming performance. In this case two 2400MHz DDR4 sticks in dual channel would more than double the memory bandwidth.

[NikosD]

RAM and general CPU performance has nothing to do (or shouldn't have) with HW decoding, especially in DXVA native mode.

It must be a weaker ASIC (less MFX units)

[Kadano]

That's perfect thank you.

But where are the above 4K results ?

I think that HD 610 supports up to 8K acceleration.

8bit HEVC

5) 1st Bonus clip:
Netflix_TunnelFlag_4096x2160_60fps_x265_8bit_700Mbit

Decode: 79 fps [64-96]
Playback [1280x720]: 72 fps [57-87]

6) 2nd Bonus clip (8K):
8K_Ex_sample- 7680x4320@24fps-53Mbps
[DS64] LAV Video Decoder 0.69.0.18-git Unsupported

10bit HEVC 4K/8K samples

5) 1st Bonus clip (7K):
5.Arri65_from_raw-6560x3100-263Mbps@24fps
[DS64] LAV Video Decoder 0.69.0.18-git Unsupported

6) 2nd Bonus clip (8K):
MBC UHD Music Core GFriend - Rough (4in1) 8k@60fps-80Mbps

Decode: 35 fps [31-41]
Playback [1280x720]: 25 fps [24-29]


Also, can you confirm using GPU-Z or HWINFO that GPU clock is 1050MHz (or something else ?)
GPU clock is 1050 MHz during DXVA usage, according to GPU-Z. Memory clock 800 MHz.

A different question, some anime encoders started using HEVC 12-bit. On G4560, software HEVC 12-bit 720p decoding is a bit stressing, with SVP also running. GTX 1000 series and Kaby Lake should be able to decode 12-bit HEVC, but it seems LAV Filters don't support them. Is it possible to set up 12-bit encodes to bypass LAV Filters and be decoded by Microsoft decoders (which do support 12-bit, according to DXVA Checker)?

[nevcairiel]

For iGPUs RAM still has some impact, as it needs to read and write the video frames in there - the system ram also doubles as video ram, so even if its not copied around (ie. by DXVA Copy-Back), it can matter. If it really does - well, one could benchmark some iGPU with different RAM configurations, i guess.

[NikosD]

Interesting post.

What do you mean "the system ram also doubles as video ram" ?

What is the difference between iGPU and dGPU for HW decoding using DXVA native ?

[Kadano]

Nvidia released 381.65 WHQL drivers today, new driver branch that supports Win10 Creators Update & WDDM 2.2, can Pascal users update to the latest driver and run DXVA Checker and post screenshots here?

http://www.geforce.com/whats-new/articles/quake-champions-closed-beta-game-ready-driver

Old: http://i.imgur.com/Q8zRdEP.jpg

New: http://i.imgur.com/YgZV3vJ.jpg

Seems nothing has improved.

[pacuro]

There were nothing to improve with GP107 as it has VP9.2 hardware decoder since the begining.
Unfortunatelly nothing improved with GP106, which is most expected I guess.

http://static.frazpc.pl/board/2017/04/a9a2f3e5be60fab9.jpg

[vlado08]

Day or two ago made 1 post but it didn't appear?
Repeated it again and it didn't appear?
I'm trying now again.

[vlado08]

Intel NUC7i3BNH
CPU: i3-7100U 2.4Ghz
GPU: Intel HD 620
RAM: 16GB DDR4-2132MHz - dual channel
Windows 10 x64 1607 os build 14393.969
Graphics driver 21.20.16.4590
LAV Video x64 0.69.0
DXVA Checker 3.15.0


Power Options - High Performance
(GPU @ 1000Mhz during benchmark)
video enhancement - disable; monitor@1080/59p

HEVC 8bit:

1) LG_4K_View-the-Feeling (2160p@29.970fps ~25Mbps)

Decode: 156/170/195 CPU 3[0-6]
Playback: 146/160/187 CPU 5[3-10]

2) Samsung 4K@30fps-80Mbps file -0235

Decode: 181/188/194 CPU 3[1-5]
Playback: 172/177/180 CPU 4[2-7]

3) Samsung 4K@30fps-80Mbps file -0230

Decode: 153/156/158 CPU 3[1-4]
Playback: 148/149/152 CPU 5[3-7]

4) Crowd_run 4K@50fps -300Mbps

Decode: 101/110/113 CPU 3[1-6]
Playback: 96/106/109 CPU 4[1-7]

5) 1st Bonus clip:
Netflix_TunnelFlag_4096x2160_60fps_x265_8bit_700Mbit

Decode: 62/76/88 CPU 2[1-7]
Playback: 62/75/85 CPU 4[2-8]

6) 2nd Bonus clip:
8k_ex_sample 8k@24fps - 53Mbps

Decode: 42/46/48 CPU 2[0-5]
Playback: 41/44/47 CPU 3[2-5]

HEVC 10bit:

2.Crowd_run_2160p50_Main10_100Mbps

Decode: 155/157/158 CPU 4[3-6]
Playback: 137/139/141 CPU 6[5-6]

3.Jellyfish-140-mbps-4k-uhd-hevc-10bit

Decode: 117/133/151 CPU 4[2-11]
Playback: 111/123/139 CPU 6[4-11]

4.Jellyfish-400-mbps-4k-uhd-hevc-10bit

Decode: 54/67/89 CPU 4[1-9]
Playback: 53/66/85 CPU 5[2-13]

5.1st Bonus clip:
Arri65_from_raw 6560x3100@23.976fps 263Mbps

Decode: 54/56/58 CPU 3[1-7]
Playback: 48/51/53 CPU 6[4-10]

6.2nd Bonus clip:
MBC UHD Music Core GFriend - Rough (4in1) 8k
7680x4320@59.94fps 82.4Mbps

Decode: 43/48/56 CPU 3[0-8]
Playback: 38/41/47 CPU 4[0-10]

[vlado08]

when I post my testing they didn't appear. There was short message when I made the post which I couldn't read.

[NikosD]

Day or two ago made 1 post but it didn't appear?
Repeated it again and it didn't appear?
I'm trying now again.

when I post my testing they didn't appear. There was short message when I made the post which I couldn't read.

@mods
I received the same complain from another user that he was trying to post a simple text without images or attachments and he got a message that he must wait for approval.

Why?

The approval didn't come and he didn't show his message posted.

Is there some kind of restriction for new users in the forum with a small number of posts regarding their ability to post new messages ?

[NikosD]

For iGPUs RAM still has some impact, as it needs to read and write the video frames in there - the system ram also doubles as video ram, so even if its not copied around (ie. by DXVA Copy-Back), it can matter. If it really does - well, one could benchmark some iGPU with different RAM configurations, i guess.

Interesting post.

What do you mean "the system ram also doubles as video ram" ?

What is the difference between iGPU and dGPU for HW decoding using DXVA native ?

I replied you with the above post but you didn't reply.

What is the difference regarding HW decoding using DXVA native between iGPU and dGPU ?

In the meantime, Intel wants to reply to me (and you) with that new KabyLake-G:
http://wccftech.com/intel-kaby-lake-g-hbm2-gpu-multi-die/

A new Kabylake, if the rumor is true, with multi-die design.
One die a 4 core CPU with iGPU GT2 and the other die a dGPU with HBM2 memory as VRAM connected with a PCIe 3.0 x8.

A mystery around dGPU.
Is it Intel or AMD ?

[nevcairiel]

Excuse me for being out yesterday evening. :p

The difference is that a dGPU has dedicated fast video memory (ie. GDDR5 or what ever yours has), which is much faster then the normal system RAM.
An iGPU does not have this, it uses your normal system RAM as video memory. Even pure DXVA Native decoding needs to read and write video frames (ie. read reference frames, write decoded frames), so it is very well possible that the speed of the RAM can have an impact on decoding performance. How much (or if at all) would only be possible to actually be tested with different RAM configurations.

Regarding KabyLake-G, what I wonder is whos media features get used. I suppose the media features of the Intel GPU are still available, much like on an Optimus laptop? Who knows.

[NikosD]

Maybe because the RAM is single channel and runs only at 1600MHz? With integrated GPU's fast dual channel RAMs are essential for gaming performance. In this case two 2400MHz DDR4 sticks in dual channel would more than double the memory bandwidth.
Based on the above reply of nevcairiel, I realized that I had forgotten that system RAM is used as both system RAM and video RAM for iGPU.

Yes, a single channel DDR3 1600 could be a small (?) bottleneck even for pure HW DXVA native acceleration.

Only a benchmark with both configurations (single channel/dual channel) could clarify this.

[Kadano]

I tested Nvidia GTX 1050. CPU: Intel G4560. 1x 8GB DDR3 RAM (1600MHz) used. 3 samples each time, first sample dropped.

8bit HEVC

1) LG_4K_View-the-Feeling (2160p@29.970fps ~25Mbps)

Decode: 180 fps [172-194]
Playback [1280x720]: 190 fps [183-201]
GPU Video clock during decoding/playback: 1607 MHz

2) Samsung 4K@30fps-80Mbps file -0235

Decode: 155 fps [149-162]
Playback [1280x720]: 154 fps [148-158]
GPU Video clock during decoding/playback: 1595 MHz

3) Samsung 4K@30fps-80Mbps file -0230

Decode: 141 fps [137-145]
Playback [1280x720]: 143 fps [138-148]
GPU Video clock during decoding/playback: 1595 MHz

4) Crowd_run 4K@50fps -300Mbps

Decode: 104 fps [92-109]
Playback [1280x720]: 106 fps [94-114]
GPU Video clock during decoding/playback: 1595 MHz

5) 1st Bonus clip:
Netflix_TunnelFlag_4096x2160_60fps_x265_8bit_700Mbit

Decode: 64 fps [49-73]
Playback [1280x720]: 64 fps [49-74]
GPU Video clock during decoding/playback: 1595 MHz

6) 2nd Bonus clip (8K):
8K_Ex_sample- 7680x4320@24fps-53Mbps

Decode: 52 fps [46-55]
Playback [1280x720]: 52 fps [45-54]
GPU Video clock during decoding/playback: 1595 MHz


10bit HEVC 4K/8K samples

1) 1.UHD_PQ_Lovely_Swiss-51Mbps@60fps

Decode: 179 fps [171-197]
Playback [1280x720]: 191 fps [185-205]
GPU Video clock during decoding/playback: 1595 MHz

2.Crowd_run_2160p50_Main10_100Mbps

Decode: 164 fps [161-165]
Playback [1280x720]: 166 fps [162-169]
GPU Video clock during decoding/playback: 1595 MHz

3.Jellyfish-140-mbps-4k-uhd-hevc-10bit

Decode: 128 fps [112-147]
Playback [1280x720]: 131 fps [114-150]
GPU Video clock during decoding/playback: 1595 MHz

4.Jellyfish-400-mbps-4k-uhd-hevc-10bit

Decode: fps [53-83]
Playback [1280x720]: 55 fps [42-74]
GPU Video clock during decoding/playback: 1582 MHz

5) 1st Bonus clip (7K):
5.Arri65_from_raw-6560x3100-263Mbps@24fps

Decode: 55 fps [52-59]
Playback [1280x720]: 55 fps [52-59]
GPU Video clock during decoding/playback: 1582 MHz

6) 2nd Bonus clip (8K):
MBC UHD Music Core GFriend - Rough (4in1) 8k@60fps-80Mbps

Decode: 54 fps [52-56]
Playback [1280x720]: 54 fps [52-55]
GPU Video clock during decoding/playback: 1607 MHz

[NikosD]

Perfect, thanks!

Most of the times your GPU Video clock is the same 1595 MHZ but a few times dropped to 1582 MHZ and for the last clip raided to 1607 MHZ

Weird.

Could it be inside the margin of error of the monitoring tool ?

Is it set to monitor every 1sec or more ?

Was the GPU Video clock steady each time for each clip during the whole process ?

[MrScientist]

I tested my four HEVC hardware encoders. I know you already have many of the results, but this was mostly for my information. For the 1050, I ran the tests three times. For the rest, I ran the tests four times and I threw out the first result. The first group of results is the decode benchmark, the second group is the playback benchmark. The first result is Nvidia 1050, second is AMD 460, third is Intel 610 (in G3930), fourth is Intel "HD Graphics" (5xx? in x5-Z8300).

1.LG_4K_View-the-Feeling.mp4

193-194-194
95-94-95
181-181-181
67-68-68

195-197-197
108-108-108
166-169-170
53-52-52

2.yvid_SAM_0235-UHD_Sample1.MP4

165-165-165
99-98-99
199-200-200
79-78-79

159-165-164
101-102-101
173-183-182
60-60-60

3.yvid_SAM_0230-ProQuality.MP4

155-155-155
96-95-95
165-164-166
62-63-63

148-156-156
96-96-97
157-157-157
53-52-853

4.crowd_run_2160p50_300Mbps.mkv

127-127-127
67-67-67
117-118-117
45-45-45

124-127-127
71-70-71
114-115-115
40-40-41

5.Netflix_TunnelFlag_4096x2160_60fps_x265_8bit_700Mbit.mkv

81-81-81
41-41-41
82-82-82
40-40-41

78-80-80
39-39-39
81-81-81
33-33-33

6.8K_Ex_sample.mp4

51-54-53
41-41-41
49-49-49
20-20-20

51-53-53
39-39-39
47-47-47
17-17-17

1.UHD_PQ_Lovely_Swiss.ts

195-197-191
73-74-73
177-178-178
ND

192-192-192
91-91-90
155-155-155
ND

2.Crowd_run_2160p50_Main10_100Mbps

163-177-176
78-76-76
167-166-166
ND

174-181-181
93-93-93
144-145-144
ND

3.jellyfish-140-mbps-4k-uhd-hevc-10bit.mkv

141-147-148
76-76-76
142-142-143
ND

143-148-147
84-84-84
129-130-130
ND

4.jellyfish-400-mbps-4k-uhd-hevc-10bit.mkv

74-74-74
41-41-40
71-70-71
ND

73-74-74
40-40-41
70-69-69
ND

5.Arri65_from_raw.mp4

68-68-68
ND
28-28-28
ND

60-68-68
ND
28-28-28
ND

6.MBC UHD Music Core GFriend - Rough (4in1) 8k.mkv

54-55-55
ND
51-51-51
ND

54-54-54
ND
36-36-36
ND

[NikosD]

Awesome!

Thank you for your time and effort.

[NikosD]

The first result is Nvidia 1050, second is AMD 460, third is Intel 610 (in G3930), fourth is Intel "HD Graphics" (5xx? in x5-Z8300).


Can you tell me a little more of your systems configuration ?

OS, drivers and especially for G3930 and Atom x5-Z8300 your ram configuration.
(Memory speed and single channel/dual channel)

[MrScientist]

Can you tell me a little more of your systems configuration ?

OS, drivers and especially for G3930 and Atom x5-Z8300 your ram configuration.
(Memory speed and single channel/dual channel)
Sure, sorry I didn't do that earlier.
1. Nvidia 1050, 378.92 drivers, AMD 9590, 32GB dual channel 1333MHz DDR3, Win 10 Pro
2. AMD 460, 17.4.1 drivers, AMD E1-2100, 8GB single channel 1600MHz DDR3, Win 10 Pro
3. Intel 610, 21.20.16.4534 drivers, Intel G3930, 8GB dual channel 2133MHz DDR4, Win 10 Pro
4. Intel "HD Graphics", Intel x5-Z8300, 2GB single channel DDR3, Win 10 Home (This is a tablet. I'll get the drivers and RAM speed later, it's currently not responding.)

[NikosD]

Sure, sorry I didn't do that earlier.
1. Nvidia 1050, 378.92 drivers, AMD 9590, 32GB dual channel 1333MHz DDR3, Win 10 Pro
2. AMD 460, 17.4.1 drivers, AMD E1-2100, 8GB single channel 1600MHz DDR3, Win 10 Pro
3. Intel 610, 21.20.16.4534 drivers, Intel G3930, 8GB dual channel 2133MHz DDR4, Win 10 Pro
4. Intel "HD Graphics", Intel x5-Z8300, 2GB single channel DDR3, Win 10 Home (This is a tablet. I'll get the drivers and RAM speed later, it's currently not responding.)

Your results are freaking fast.

I compare your RX 460 with my RX 470 and 460 is way faster than mine.

Your HD 610 and GTX 1050 are way faster than HD 610 and GTX 1050 of Kadano.

So, what's your secret ?

Do you cherry pick the results based on the maximum figures ?

Do you have a completely clean installation with no apps running in the background ?

One last request:
Can you post also your GPU Video Clock for HD 610, GTX 1050 and RX 460 ?

Thanks!

[MrScientist]

Your results are freaking fast.

I compare your RX 460 with my RX 470 and 460 is way faster than mine.

Your HD 610 and GTX 1050 are way faster than HD 610 and GTX 1050 of Kadano.

So, what's your secret ?

Do you cherry pick the results based on the maximum figures ?

Do you have a completely clean installation with no apps running in the background ?

One last request:
Can you post also your GPU Video Clock for HD 610, GTX 1050 and RX 460 ?

Thanks!

Hi,
I didn't cherry pick, I did the tests with either three or four passes as I described. I followed the test procedure I found here, https://www.youtube.com/watch?v=RhZCtE1P40w. Did I follow the correct protocol? I used 64bit for both benchmarks with "well-used" operating systems.

How do I report GPU video clock? On the benchmark results, it shows "Video Engine Usage." Should I do the tests again with GPU-Z running?

[NikosD]

Hi,
I didn't cherry pick, I did the tests with either three or four passes as I described. I followed the test procedure I found here, https://www.youtube.com/watch?v=RhZCtE1P40w. Did I follow the correct protocol? I used 64bit for both benchmarks with "well-used" operating systems.

How do I report GPU video clock? On the benchmark results, it shows "Video Engine Usage." Should I do the tests again with GPU-Z running?

Yes, I read that you run the tests 3 or 4 times but maybe you didn't get the average value but the best values of those runs.

The procedure in video is fine, as long as you selected DXVA native using LAV Video x64.

What version of LAV did you test ?

The "GPU Video Clock" is the "GPU clock" for Intel reported by GPU-Z.

The "GPU Video Clock" is the "GPU Video Clock" for Nvidia reported by HWiNFO

The "GPU Video Clock" is unknown for AMD, but just post GPU clock/ GPU Memory reported by GPU-Z.

I advise you and everyone else, against to run the tests with monitoring tools and benchmark procedure running at the same time.

So, please don't run at the same time tests and monitoring tools.

Run the GPU-Z and HWiNFO without reporting benchmark values of HW decoding.

I just need the GPU Video Clock from these monitoring tools while you run the benchmark in decode/playback mode.

[Lollerke]

Your results are freaking fast.

I compare your RX 460 with my RX 470 and 460 is way faster than mine.

Your HD 610 and GTX 1050 are way faster than HD 610 and GTX 1050 of Kadano.

So, what's your secret ?

Do you cherry pick the results based on the maximum figures ?

Do you have a completely clean installation with no apps running in the background ?

One last request:
Can you post also your GPU Video Clock for HD 610, GTX 1050 and RX 460 ?

Thanks!

I think it's fast because he has dual channel DDR4 2133MHz RAM for the HD610 :) Don't know about the other VGA's tho, but GPU clock shouldnt affect the decoding speed, because all of these cards decode videos purely by the video decoding engine and there's no need for the GPU's graphics processing power.

[NikosD]

Hi,
I didn't cherry pick, I did the tests with either three or four passes as I described.

I've just installed Creators Update (clean install) and latest drivers 17.4.1

I re-ran the benchmark for the first clip and I saw a HUGE performance gain ~40%.

I'll post my results later for all the clips.

Yes, your 460 figures seem reasonable right now.

Do you have creators update ?

[CruNcher]

I think it's fast because he has dual channel DDR4 2133MHz RAM for the HD610 :) Don't know about the other VGA's tho, but GPU clock shouldnt affect the decoding speed, because all of these cards decode videos purely by the video decoding engine and there's no need for the GPU's graphics processing power.

Unless you don't activate the internal PP Engines which can run over the shaders from within the driver directly or do Hybrid Compute Based Decoding/Encoding (OpenCL/DirectCompute/Cuda) that's correct.

[NikosD]

4.Jellyfish-400-mbps-4k-uhd-hevc-10bit

Decode: fps [53-83]
Playback [1280x720]: 55 fps [42-74]
GPU Video clock during decoding/playback: 1582 MHz



This one is missing the Decode Average value.

[NikosD]

Hi,
I didn't cherry pick, I did the tests with either three or four passes as I described.

I've just started posting your results on the general table and I realized that most of the times the min - avg - max values are all equal.

Also, many times the first value (min) is just one frame bigger than middle value (average)

How is it possible ?

The clips are not decoded that fast in order to have such issues.

I think your results are rounded by you in a little weird way !

Can you take a look of your results ?

[NikosD]

I have updated my RX 470 results in the general table, after the new OS+drivers.

I ran the tests after the clean installation of Creators Update + 17.4.1 drivers without any monitoring tools running or other background apps.

I re-ran the tests with all of my monitoring tools running and after finishing some basic apps installation, including background apps and there were no difference than the clean installation.

I haven't seen before such a performance gain regarding HW decoding.

The decode and playback performance of H.265 is more about 40% - 50% higher than before.

Probably the new drivers pushed the ASIC clock higher, I can't think of anything else (I don't believe it's the Creators Update)

P.S

There is also a significant gain ~25% for H.264 decoding

[P.J]

http://www.anandtech.com/show/11266/cyberlink-launches-powerdvd-17-with-uhd-bluray-and-vr-hmd-support

=)

[NikosD]

The TDP for the new RX 500 AMD cards has gone up in general, but for video playback is lowered down significantly due to a new intermediate memory state which seems to affect a lot the power consumption of the whole card due to the power hungry GDDR5 memory type.

So, instead of going from 300MHz (idle) to 2000MHz (full) during video playback it goes to 1000MHz and the power consumption is significantly lower than RX 400 series and even Pascal 1060 for high bitrate H.264 & HEVC clips.

Read more here:
http://www.anandtech.com/show/11278/amd-radeon-rx-580-rx-570-review/2

[huhn]

how they are managing to get 93-96 watt with video playback and 74 watt idle comes down is beyond me. the bad choice of PSU has something to do with it but 74 idle is just...

but yes the 1060 with the current driver is using a very high mem clock. it is usually 202.5 to 405 mhz for hardware decoding.

[pacuro]

Do we benchmark also VP9 decoding and playback in this thread? DXVA works well with 8K60 VP9 8bit 66Mbps clip well on my GTX1060 + last nightly LAV Filters, but I am not able to run DXVA Checker benchmark due to some unidentified error (empty with yellow exclamation mark)

http://static.frazpc.pl/board/2017/04/d2c705df778b4eb1.jpg

[NikosD]

...but I am not able to run DXVA Checker benchmark due to some unidentified error (empty with yellow exclamation mark)


Did you try latest 3.15.2 ?

DXVA checker had a few issues lately.

[pacuro]

Yes, 3.15.2.
Previous 3.15.0 ends with the same empty window with yellow "!"

[NikosD]

Yes, 3.15.2.
Previous 3.15.0 ends with the same empty window with yellow "!"

DXVA Checker v3.15.3 is out and fixed the problem I reported to the developer.

If the latest version doesn't fix your issue, maybe you should report that to the developer.

[sneaker_ger]

Latest LAV x64 nightlies are up to 10% faster for software decoding of HEVC compared to older versions here (Win 7x64, i5-2500K, Ducks Take Off sample). This is the biggest, if not the only, speedup I've seen to HEVC software decoding since 2014.

[clsid]

29-10-2013 | LAV 0.58.2-97 x64 | 68,6 fps
30-10-2013 | LAV 0.58.2-104 x64 | 131,2 fps
03-08-2014 | LAV 0.62.0-8 x64 | 231,0 fps
03-08-2014 | LAV 0.62.0-9 x64 | 280,3 fps
19-01-2015 | LAV 0.63.0-41 x64 | 346,5 fps
13-07-2016 | LAV 0.68.1-31 x64 | 370,4 fps
23-04-2017 | LAV 0.69.0-39 x64 | 381,7 fps

TearsOfSteal 1080p sample

[sneaker_ger]

Interesting. I didn't notice a significant speed change between 2015 and 2016. Guess I'll have to test again later.

[NikosD]

29-10-2013 | LAV 0.58.2-97 x64 | 68,6 fps
30-10-2013 | LAV 0.58.2-104 x64 | 131,2 fps
03-08-2014 | LAV 0.62.0-8 x64 | 231,0 fps
03-08-2014 | LAV 0.62.0-9 x64 | 280,3 fps
19-01-2015 | LAV 0.63.0-41 x64 | 346,5 fps
13-07-2016 | LAV 0.68.1-31 x64 | 370,4 fps
23-04-2017 | LAV 0.69.0-39 x64 | 381,7 fps

TearsOfSteal 1080p sample

Your CPU ?

Have you noticed significant speed-up gains due to specific SIMD optimizations ?

A 5.5x speed-up in 3.5 years is impressive.

[NikosD]

Latest LAV x64 nightlies are up to 10% faster for software decoding of HEVC compared to older versions here (Win 7x64, i5-2500K, Ducks Take Off sample). This is the biggest, if not the only, speedup I've seen to HEVC software decoding since 2014.

29-10-2013 |
13-07-2016 | LAV 0.68.1-31 x64 | 370,4 fps
23-04-2017 | LAV 0.69.0-39 x64 | 381,7 fps

TearsOfSteal 1080p sample

No change here (Core i3-4170, Win 10 x64 Creators Update) going from 0.69.24 to 0.69.39 on Ducks Sample (0%)

But a small regression on 10bit HEVC Gravity sample.

At least for me latest nightly is a little slower.

[sneaker_ger]

29-10-2013 | LAV 0.58.2-97 x64 | 68,6 fps
30-10-2013 | LAV 0.58.2-104 x64 | 131,2 fps
03-08-2014 | LAV 0.62.0-8 x64 | 231,0 fps
03-08-2014 | LAV 0.62.0-9 x64 | 280,3 fps
19-01-2015 | LAV 0.63.0-41 x64 | 346,5 fps
13-07-2016 | LAV 0.68.1-31 x64 | 370,4 fps
23-04-2017 | LAV 0.69.0-39 x64 | 381,7 fps

TearsOfSteal 1080p sample

TearsOfSteal 1080p on i5-2500K
19-01-2015 | LAV 0.63.0-41 x64 | 289.5 fps
13-07-2016 | LAV 0.68.1-31 x64 | 293.9 fps
23-04-2017 | LAV 0.69.0-39 x64 | 294.9 fps

Guess I misremembered. Nothing to see. Sorry for the confusion.

[nevcairiel]

The speed ups are likely going to be bigger the lower the bitrate is, which means more importance is being placed on actual image processing instead of CABAC bitstream reading (which is by far the biggest bottleneck in the decoder, and unfortunately very hard to optimize, and practically impossible to use SIMD on).
With multi-threaded decoding a 100% accurate and reproducable benchmark seems to be very hard to achieve though, there is always quite a bit of variance, so differences under 1% often just disappear in noise.

[clsid]

There wasn't any speedup with the 4k version of the same video, while the 720p version had even bigger gains than the 1080p result above. The 720p one is decoded with 700+ fps. So it seems to me that your recent optimizations work good for reducing the overhead per decoded frame. Unfortunately probably not very noticeable at real-time speeds.

[GTPVHD]

https://videocardz.com/69009/nvidia-pascal-gp108-300-graphics-processor-pictured

GT 1030 (GP108 GPU) coming soon, smallest Pascal GPU with same hardware decoders as GP107.

[nsnhd]

https://videocardz.com/69009/nvidia-pascal-gp108-300-graphics-processor-pictured

GT 1030 (GP108 GPU) coming soon, smallest Pascal GPU with same hardware decoders as GP107.
Hopefully it will add 4K HDR to old systems.

[huhn]

the is no reason to remove it but but it is not going to work everywhere thanks to the most likely to low Vram.

[NikosD]

Hopefully it will add 4K HDR to old systems.
If you are interested in 4K HDR, then AMD RX 550 is an alternative choice as it in the same price range with GT 1030 and has 4GB VRAM option for sure.

Also, it is available worldwide as we speak.

[nsnhd]

If you are interested in 4K HDR, then AMD RX 550 is an alternative choice as it in the same price range with GT 1030 and has 4GB VRAM option for sure.

Also, it is available worldwide as we speak.

Is the situation of VP9 hardware decoder improved by AMD RX cards ?

[NikosD]

VP9 is not hardware accelerated by AMD cards in pure mode, only hybrid and only in YouTube.

But, there is no VP9 usage outside YouTube so I don't think it's a major issue.

[NikosD]

Nvidia and Netflix just added another reason to avoid 1050 and <4GB VRAM cards.

In the preview support of Netflix 4K streaming service, besides some weird requirements like Windows Insider OS (!) they added a restriction of at least 3GB VRAM.

From anandtech.com

In any case, for the moment NVIDIA is calling this a preview, presumably due to the Insider OS requirements.
Despite this qualifier, 4K Netflix support should fully work on all Pascal GPUs with 3GB or more of VRAM, though this means that 2GB GTX 1050 cards are oddly excluded.


Update:
It seems that you can manually install 381.74 drivers from Nvidia on Win 10 Creators Update and do the trick.
You don't actually need Windows Insider.

[huhn]

is this about an Netflix application or web browser?

because 4K youtube with 2GB Vram is totally fine but a proper video player will need quite a lot of Vram and 2 GB can make a huge problem.

like before if you want 4K video playback get a card with at least 3 GB Vram better 4.

[wanezhiling]

Off-topic, I wanna buy a 4K(5K/8K) HDR monitor, any good choices? :)

[huhn]

the number of HDR monitors is close to zero.

[nevcairiel]

Off-topic, I wanna buy a 4K(5K/8K) HDR monitor, any good choices? :)

Acer Predator X27 is coming out later this year, 4K, 144Hz, HDR, AdobeRGB and NVIDIA GSYNC. Probably quite expensive. :p

[baii]

Eizo and dell just announce their s, but you may as well just buy a tv for any entertainment related experience.

[nsnhd]

is this about an Netflix application or web browser?

because 4K youtube with 2GB Vram is totally fine but a proper video player will need quite a lot of Vram and 2 GB can make a huge problem.

like before if you want 4K video playback get a card with at least 3 GB Vram better 4.

And what's about memory bandwidth 128-bits (RX550) vs 64-bits (GT1030), does that have impact on fps assuming these cards have 4GB vram ?

[huhn]

i don't know sorry.
if it is 64 bit with GDDR5 it is maybe fast enough but i don't believe in a 1030 with 4Gb Vram there is no 1050 with 4 GB on the market right now.

[el Filou]

GeForce 1030 is official: http://www.geforce.com/hardware/desktop-gpus/geforce-gt-1030/specifications

Will be interesting to see if the 2 GB of RAM and/or limited bandwidth are a limitation with video outside of madVR, because 30W TDP is interesting.

[nevcairiel]

With a very simple renderer like EVR its probably going to be fine.

[leonccyiu]

GeForce 1030 is official: http://www.geforce.com/hardware/desktop-gpus/geforce-gt-1030/specifications

Will be interesting to see if the 2 GB of RAM and/or limited bandwidth are a limitation with video outside of madVR, because 30W TDP is interesting.

It's a huge shame there isn't a 3GB version that works with netflix in 4k. Spec's wise, it's exactly half a 1050Ti and sells for half a 1050Ti, I think it could probably do 1080p23 > 2160p with NGU low in MADVR

[NikosD]

The main issue with those kind of cards, like GT 1030 and plain GTX 1050, is that there is no 4GB version.

Due to 2GB VRAM only, those cards could definitely have problems displaying the 4K resolution.

On the other hand, a much more capable GPU like AMD RX 550 has more 3D power, 4GB VRAM, costing around the same price like GT 1030 with a very low TDP too.

[huhn]

polaris cards are very slow for NGU so a 550 is most likely slower than a 1030 for NGU and ignoring Vram.
the pathetic 48 GB/s could be a real problem.

to be honest the 1030 looks terrible.
TDP is not that important i wouldn't be shocked if a 1050 has the same power consumption when playing back videos.

about 4K playback with 2GB Vram using EVR-CP this was fine with DXVA copyback and unplayable with DXVA native using a 2GB 960. Vram was at his limited.
a lot has changed since a used that card.

[NikosD]

RX 550 has 512 cores and 128bit memory bus.

It should be faster than GT 1030 in all processing cases.

[nevcairiel]

A 4GB RX550 is around ~30€ more expensive around here though, which at that level is almost 30%. But yeah the 1030 is not something I would ever buy or recommend to anyone in any case.
For another 20-30€ over the RX550 (ie. starting at ~135€), you can get a 4GB 1050Ti already, which is probably as low as I would go.

[el Filou]

It's a huge shame there isn't a 3GB version that works with netflix in 4k. Spec's wise, it's exactly half a 1050Ti and sells for half a 1050Ti, I think it could probably do 1080p23 > 2160p with NGU low in MADVR
Maybe they'll update the drivers to make the lower-end cards compatible with Netflix later?
My 1050 Ti when decoding YouTube 4K60 in Chrome and outputting 1080 doesn't even use 500 MB (and this includes around 200 MB for Windows and the browser). When outputting actual 4K that might go up a few 100s MB, but I don't see why just the DRM support would need more than 1 GB for itself and the whole couldn't run on 2 GB.
Has someone already tested Netflix 4K using the Insider Preview NVIDIA driver and measured how much VRAM is actually used?

@NikosD: the RX 550 is indeed a better all-round card (though I doubt it beats a 1030 in pure shader power by much, only the memory is much faster), but the 4 GB ones start around 100 USD. We'll have to see how low the OEMs can price their GF 1030s.

[huhn]

that's why: https://abload.de/img/thatswhyo1uge.png (EVR-CP)

and yes webbrowser video player use a lot less Vram and there is a reason picture quality is pathetic...

[el Filou]

that's why: https://abload.de/img/thatswhyo1uge.png (EVR-CP)

and yes webbrowser video player use a lot less Vram and there is a reason picture quality is pathetic...
Yes, I remember that from your tests.
"pathetic" is pushing it a bit, don't you think? :)

Anyway there is no way to play Netflix using MPC-HC ;), I was just wondering if 2 GB would be fine with the Netflix app or in the browser but I can't find any public test results yet. I guess the fact the driver is still in beta explains it.

[huhn]

Yes, I remember that from your tests.
"pathetic" is pushing it a bit, don't you think? :)

most browser video renderer today can't deal with windows scaling and they do everything as fast as possible.

i still think think pathetic is a good word for the default web browser renderer.
and it is pretty obvious what a web browser renderer is doing to get the lowest possible Vram usages it's not like EVR is wasteful or something like that.

mpc-hc may not support it but jriver say they support it: https://www.jriver.com/video.html
i don't know how it works in jriver.

[NikosD]

It looks like GT 1030 has a different (slower) Video Decoder than the rest of the Pascal family.

Tested in Linux and H.264 - 720p/1080p, it looks like having a GTX 950 video decoder.

Unless that test is not about pure decoding, but rather playback performance, so the weak GPU shrinks the pure video decoder's performance.

http://www.phoronix.com/scan.php?page=article&item=nvidia-geforce-1030&num=10

[nevcairiel]

Might be limited by its low memory bandwidth, doubtful they would put an older chip in there - but someone with DXVAChecker could figure that out.

[NikosD]

After Windows Creators Update, the old Chrome trick via the --enable switch to force enable VP9 hardware acceleration on Polaris RX 400/500 cards doesn't properly work.

Does anyone have any idea how to enable the VP9 acceleration of Chrome/YouTube on Polaris cards ?

I want access to VP9 4K60fps content, so no other browser is suitable for that.

[mrcorbo]

After Windows Creators Update, the old Chrome trick via the --enable switch to force enable VP9 hardware acceleration on Polaris RX 400/500 cards doesn't properly work.

Does anyone have any idea how to enable the VP9 acceleration of Chrome/YouTube on Polaris cards ?

I want access to VP9 4K60fps content, so no other browser is suitable for that.

Seems to work in Edge on my GTX 1060. Tested up to 8k VP9 w/ 10% CPU usage and no dropped frames. Chrome can't even do this, actually, at least not on my system. Same video pegs my CPU and drops frames like crazy.

[NikosD]

Nvidia is different and it has a pure video VP9 decoder.

And I don't think Edge can access all YouTube streams.

Some of them, like 4K60fps are VP9 only and Chrome only (I think)

[sneaker_ger]

Edge has VP9 but I don't know if it's compatible with AMD.
https://www.reddit.com/r/Windows10/comments/5ojx3j/microsoft_edge_cant_play_4k_videos_on_youtube/

[NikosD]

Besides Chrome-only streams of YouTube, I want to avoid Edge for one more reason.

I haven't found out yet a way to move Edge's cache from my SSD (C:) to HDD (D:)

Is it possible ?

[mrcorbo]

Nvidia is different and it has a pure video VP9 decoder.

And I don't think Edge can access all YouTube streams.

Some of them, like 4K60fps are VP9 only and Chrome only (I think)

According to "Stats for Nerds", all the streams I tried were VP9.

[P.J]

Wonder why youtube used H.264 for 8K :stupid:

http://www.mysterybox.us/blog/2017/2/21/delivering-8k-using-avch264

[huhn]

it's VP9 too...

[NikosD]

The new Windows 10 Insider Preview Build 16226 adds a GPU utilization Tab in Windows 10 task manager.
The Performance tab shows GPU utilization information for each separate GPU component (such as 3D and Video encode/decode), as well as graphics memory usage stats.

The Details tab shows you GPU utilization info for each process.
Please note that this feature is still under construction and you will find issues and bugs.

https://s3.postimg.org/uc75eqh77/untitled-1.png

[edwdevel]

Hello, this is my first post here on Doom9. I just purchased a
gt 1030 for the ancient 8 yr old Intel icore duo at work, just
to benchmark this board representing the least expensive nvidia
solution for HEVC decoding. My PC (Dell Dimension 780) is a
stand-in for the millions of old systems that now cannot decode
HEVC.


Benchmarks: Nvidia Gt 1030 @1227Mhz/Intel E8400 2core3.0Ghz
LAV 0.70.0 DXVA Checker 3.15.3 Nvidia driver 382.53

All benchmarks are playback only downscaled to 1920x1080

-----------------------------------------------------------
NikosD benchmark clips (see March 1, 2017 forum entry)

Clip: 4k 8bit Frames Cpu% Gpu%
LG View the Feeling 89-109-139 2-7-25 13-17-62
LG View the Feeling 139-198-205 6-12-23 23-26-31 Nvidia 384.76
Sams 4K 8bit -235 107-118-122 3-8-20 14-17-53
Sams 4K 8bit -230 97-112-115 4-9-18 15-16-17
Crowd Run 4K300m50f 61-89-98 5-12-26 12-12-14
Tunnelflag700m60f 41-76-93 4-15-24 3-15-21 Nvidia 384.76


Clip: 4k 10bit Frames Cpu% Gpu%
Lovely Swiss 66-116-143 1-8-29 5-20-60
Lovely Swiss 124-185-208 2-12-30 29-33-42 Nvidia 384.76
Jellyfish 140m30f 64-88-103 4-11-22 11-15-23

Clip: 7k 10bit Frames Cpu% Gpu%
Arri65 7K263m24f 30-44-50 3-12-21 14-20-25
8Kexsamp 40-52-55 4-9-23 16-20-26 Nvidia 384.76



-----------------------------------------------------------
Other clips I think should be benchmarked:

Clip: 4k 8bit Frames Cpu% Gpu%
Tears of Steel 24f 123-165-177 1-8-15 26-35-97

Clip: 4k 10bit Frames Cpu% Gpu%
Exodus Trailer 24f 113-132-130 1-10-23 18-25-31
LifeofPi Trail 24f 97-115-123 1-8-21 17-22-30
LifeofPi Trail 24f 121-171-191 5-11-20 27-32-40 Nvidia 384.76
ColorsofJourney60f 106-128-139 2-6-18 17-22-30
ColorsofJourney60f 132-191-207 2-11-23 31-35-45 Nvidia 384.76
Sony Camp 60f 92-109-134 0-7-21 15-20-27
Sony Camp 60f 130-177-210 2-10-31 25-32-43 Nvidia 384.76


About the choice of benchmark clips:

Although NikosD's choice of clips are illuminating, I feel
they don't represent the real-world videos and movies
that people are actually going to view. "Lovely Swiss", for
example, while very pretty to watch, is really just a slideshow
of individual images with virtually no movement or panning at all,
making it a perfect foil for interframe compression. Almost all
real video/film content, on the other hand, is basically all
movement, due both to camera and subject.

Thus, I included some clips (all easily available on the web) that
I think are more representative of content that people will
actually be exposed to now and in the future. "Exodus" and
"Life of Pi" Trailers represent professionally directed and encoded
UHD HDR 24fps content, and "Colors of Journey" and "Sony Camp" are
4K 60fps video that represent non-movie but again professionally directed
and encoded 10 bit content. "Tears of Steel" is an interesting
8-bit professional 4K short movie.


Observations:

The GT 1030 can play all of the videos above perfectly and with only
about 20% CPU utilization on an ancient 8 year old 2 core 3Ghz
wolfdale PC. This lowest-end (so far) Pascal GPU furthermore
has the lowest TDP in the 10x series, which means that no old
PC system has to worry that their old power supply can support the
graphics card. The 2 GB video memory seems perfectly adequate to play
the most demanding 4k videos, while *simultaneously* downscaling to
whatever resolution the current PC monitor has. This seems to me
to demonstrate that this GPU is ideally suited to the millions of old
PC systems, and all of the HTPC systems, that would like to upgrade
their systems to allow playback of what I think will be the universal
video encoding standard. Unlike some other posters in this forum, I
would heartily recommend this GPU. At $69 (ZOTAC GT 1030) I don't
think the card has any drawbacks at all.

Plus, now even Google Earth can draw its' 3D buildings and terrain
on my old system with some alacrity. :)

I have nothing to do with Nvidia or Zotac, I'm just reporting
benchmarks and observations.

[NikosD]

Hello, this is my first post here on Doom9.

Benchmarks: Nvidia Gt 1030 @1227Mhz/Intel E8400 2core3.0Ghz
LAV 0.70.0 DXVA Checker 3.15.3

All benchmarks are playback only downscaled to 1920x1080


Very interesting first post :)

We usually post both pure decode benchmark and playback benchmarks down scaled to 1280x720.

Also, can you post your DXVA Checker screenshot ?

It would also be useful to post your video clock not GPU clock during playback benchmarking.

What is your OS ?

[nsnhd]

Observations:

The GT 1030 can play all of the videos above perfectly and with only
about 20% CPU utilization on an ancient 8 year old 2 core 3Ghz
wolfdale PC. This lowest-end (so far) Pascal GPU furthermore
has the lowest TDP in the 10x series, which means that no old
PC system has to worry that their old power supply can support the
graphics card. The 2 GB video memory seems perfectly adequate to play
the most demanding 4k videos, while *simultaneously* downscaling to
whatever resolution the current PC monitor has. This seems to me
to demonstrate that this GPU is ideally suited to the millions of old
PC systems, and all of the HTPC systems, that would like to upgrade
their systems to allow playback of what I think will be the universal
video encoding standard. Unlike some other posters in this forum, I
would heartily recommend this GPU. At $69 (ZOTAC GT 1030) I don't
think the card has any drawbacks at all.

Plus, now even Google Earth can draw its' 3D buildings and terrain
on my old system with some alacrity. :)

I have nothing to do with Nvidia or Zotac, I'm just reporting
benchmarks and observations.

But benchmarks on DXVA Checker don't guaranty smooth play back with common apps like madVR in reality, especially by HDR output.

[NikosD]

Benchmarks: Nvidia Gt 1030 @1227Mhz/Intel E8400 2core3.0Ghz
LAV 0.70.0 DXVA Checker 3.15.3

All benchmarks are playback only downscaled to 1920x1080

-----------------------------------------------------------
NikosD benchmark clips (see March 1, 2017 forum entry)

Clip: 4k 8bit Frames Cpu% Gpu%
LG View the Feeling 89-109-139 2-7-25 13-17-62
Sams 4K 8bit -235 107-118-122 3-8-20 14-17-53
Sams 4K 8bit -230 97-112-115 4-9-18 15-16-17
Crowd Run 4K300m50f 61-89-98 5-12-26 12-12-14

Clip: 4k 10bit Frames Cpu% Gpu%
Lovely Swiss 66-116-143 1-8-29 5-20-60
Jellyfish 140m30f 64-88-103 4-11-22 11-15-23

Clip: 7k 10bit Frames Cpu% Gpu%
Arri65 7K263m24f 30-44-50 3-12-21 14-20-25

-----------------------------------------------------------
Other clips I think should be benchmarked:

Clip: 4k 8bit Frames Cpu% Gpu%
Tears of Steel 24f 123-165-177 1-8-15 26-35-97

Clip: 4k 10bit Frames Cpu% Gpu%
Exodus Trailer 24f 113-132-130 1-10-23 18-25-31
LifeofPi Trail 24f 97-115-123 1-8-21 17-22-30
ColorsofJourney60f 106-128-139 2-6-18 17-22-30
Sony Camp 60f 92-109-134 0-7-21 15-20-27


Two more favors:

1) I don't see any 8K clips - 8bit or 10bit

Can you test some 8K clips from my samples collection ?

2) If you have a fast Internet line (e.g VDSL) can you test Youtube 4K60fps VP9 clips using Chrome ?

I would like you to test them in a standard Chrome window-mode (not full screen) and see the stats for any frame drop.

thanks!

[edwdevel]

I am currently under 1030 driver level 382.53, but will install the latest version later. Here is the current DXVA checker screen (see attachment).

NikosD, I did test the arri 7K clip, as it could be downloaded quickly, but the others are very large and the servers refused fast downloading. The "girls" video takes a long time to benchmark and this cannot be done at work :)

I am able to view 8k chrome you tube videos (from forum discussion here last month) but the actual bit rate is too high for the internet connection here at work. Post a link to your 4k VP9 videos and I'll try to view them.

[edwdevel]

Oh- I am under Windows 7 SP1.

Also, I don't know how to the "video" clock as opposed to the graphics clock for the GT 1030. Is this part of DXVA checker?

[NikosD]

Oh- I am under Windows 7 SP1.

Also, I don't know how to the "video" clock as opposed to the graphics clock for the GT 1030. Is this part of DXVA checker?
You have to use a tool called HWiNFO and look for "GPU Video Clock" in the monitoring parameters.

For VP9 4K60fps content in YouTube, just run Chrome and put in the search field of YouTube "4K60" and you will see a list of 4K60FPS videos.

It doesn't really matter which one you choose.

Just see the CPU usage while playing and frame drop in "Stats for nerds" (right click on the clip)

It will also say if the clip is 60fps and in VP9 format

The image of DXVA Checker is not visible.
The best way is to copy paste a direct URL of some free "upload image" service using the insert image button of Doom9.

[v0lt]

@edwdevel
Use external photohostings. For example: sendpic.org, jpegshare.net.

Please test Sony_4K_HDR_Camp.mp4 (http://demo-uhd3d.com/fiche.php?cat=uhd&id=144) in MPC-BE (https://yadi.sk/d/hlf1lfC8mKU58/!_last) and make a screenshot with statistics enabled (Ctrl+J) after 15 or more seconds of playback (so that the chip's frequency is stabilized).

[edwdevel]

http://jpegshare.net/d2/3a/d23a630cf160163c5057946f6df5056c.jpg.html

[edwdevel]

Here is an MPC-BE screenshot of sony camp:

http://jpegshare.net/37/88/3788aa4cd0ee38811ba7290c84f9e884.jpg.html

[NikosD]

NikosD, I did test the arri 7K clip, as it could be downloaded quickly, but the others are very large and the servers refused fast downloading. The "girls" video takes a long time to benchmark and this cannot be done at work :)

Post a link to your 4k VP9 videos and I'll try to view them.

This is a link with all of my samples listed in a server with a max download of 1MB/s
ftp://helpedia.com/pub/multimedia/testvideos/HEVC_HW_decoders_clips/

Take a look here:
ftp://helpedia.com/pub/multimedia/testvideos/HEVC_HW_decoders_clips/8K_Ex_sample.mp4

It's a 8K sample with only 133MB footprint on your disk.

I would like also you to test this 4K clip with huge 700Mbps bandwidth.
It's only 787MB, take a look here:
ftp://helpedia.com/pub/multimedia/testvideos/HEVC_HW_decoders_clips/Netflix_TunnelFlag_4096x2160_60fps_x265_8bit_700Mbit.mkv

OK, here is a link of YouTube with 4K60fps in VP9 codec:
https://www.youtube.com/watch?v=ppy_gjBp71E

You have to manually select 4K60 fps from the "Settings"->"Quality" -> 2160p60 (4K)

Now, regarding your tests and your DXVA Checker screenshot, it seems that you have the full-featured video decoder of Pascal 1050/1050 Ti chips regarding features but the decoding speed of 1030 is between AMD RX 470 and Nvidia 960.

Something like that: (regarding decoding speed)
RX 470 < GT 1030 < GTX 960

Your 2GB VRAM is a no go for Netflix 4K streaming, which requires >3GB VRAM

Also, as others have already mentioned here, using madVR with 4K content requires >2GB VRAM, especially for the clips you mentioned to test like "Exodus" and "Life of Pi" clips which are HDR.

Did you find the "GPU Video Clock" ?

[v0lt]

Here is an MPC-BE screenshot of sony camp:

http://jpegshare.net/37/88/3788aa4cd0ee38811ba7290c84f9e884.jpg.html
Thank you.
The GT 1030 is enough to play 4k HEVC HDR 60fps video. :)

[huhn]

why should there be a problem at FHD?

[mrcorbo]

Seems to work in Edge on my GTX 1060. Tested up to 8k VP9 w/ 10% CPU usage and no dropped frames. Chrome can't even do this, actually, at least not on my system. Same video pegs my CPU and drops frames like crazy.

Wanted to update this to say that YouTube HDR clips are also playing in Edge in HDR if you turn on HDR mode in Windows (and you have an HDR TV/monitor, obviously).

In "Stats for Nerds", Codec is reported as VP9.2 and color is reported as smpte2084(PQ)/bt2020.

This as of Windows 10 Insider Build 16232. Not sure if it works on any earlier version.

Notably, on my 1060 this seems to be using hybrid acceleration and at 60fps drops frames like crazy. It can mange 24fps (at least) smoothly, though.

[NikosD]

Notably, on my 1060 this seems to be using hybrid acceleration and at 60fps drops frames like crazy. It can mange 24fps (at least) smoothly, though.

Only 1030/1050/1050 Ti cards have fixed-function HW decoder for 10bit VP9, used by YouTube for HDR clips.

And Kabylake from Intel of course.

I think 1060 doesn't even have a hybrid 10bit VP9 decoder.
It must be CPU only.

[jmonier]

Only 1030/1050/1050 Ti cards have fixed-function HW decoder for 10bit VP9, used by YouTube for HDR clips.

And Kabylake from Intel of course.

I think 1060 doesn't even have a hybrid 10bit VP9 decoder.
It must be CPU only.

The 1060 DOES have a HW decoder.

[NikosD]

The 1060 DOES have a HW decoder.
We are talking about VP9 10bit.

Are you following the conversation ?

[v0lt]

@jmonier
GTX 1050 (https://forum.doom9.org/showthread.php?p=1802947#post1802947) vs GTX 1060 (https://forum.doom9.org/showthread.php?p=1802971#post1802971)
The GTX 1060 does not have a VP9_VLD_10bit_Profile2, but it does exist in the GTX 1050. Need or do not need this feature, this is another question.

[NikosD]

Latest Nvidia drivers are out - 384.76 - and it seems that Nvidia kept its promise, eventually.

They added support of DX12 on Fermi cards, although it's just support not optimized drivers for DX12.

https://s4.postimg.org/xzc6qs72l/GTX-580-_Ge_Force-384.76-_DX11-vs-_DX12.jpg

[NikosD]

OK, it seems that 384.76/384.80 drivers of Nvidia, enabled one more undocumented feature - besides DX12 support on Fermi - and that is Netflix UHD support on Pascal cards.

If someone has a 1030/ 1050 which have <3GB VRAM could try a Netflix UHD account and tell us if it works.

You need Netflix Windows Store (UWP platform) app or Edge browser, Windows 10 and HDCP 2.2 4K monitor.

http://techreport.com/news/32195/nvidia-384-76-drivers-quietly-enable-uhd-netflix-on-pascal

https://www.techpowerup.com/234882/nvidia-geforce-384-series-driver-removes-need-for-new-cpus-for-4k-netflix

[NikosD]

It is the first time I see Video Decoding usage of Intel's iGPU outside of Intel's own tools (GPA monitor)

https://s9.postimg.org/cg72m9bin/small_gpu_monitor.jpg

[P.J]

It is the first time I see Video Decoding usage of Intel's iGPU outside of Intel's own tools (GPA monitor)

I can't find anything in GPA monitor :confused:

[NikosD]

Do you have GPA monitor installed ?

It's a special tool from Intel, it's not in the drivers.

You have to read the manual.

[P.J]

Do you have GPA monitor installed ?

It's a special tool from Intel, it's not in the drivers.

You have to read the manual.

Yea but it doesn't show video engine load, only fps.

[aufkrawall]

I tried out using the Skylake IGP for decoding with rendering and signal output via dGPU, and it worked like a charme on Linux with mpv and vaapi-copy (mpv btw. worked great in general with the IGP on Linux).
On Windows, we are still limited to the old QuickSync decoder in LAV Filters which doesn't support HEVC 10 bit and VP9, right? So Kabylake users with dGPU are cut off their best video functionality...

[NikosD]

I think that DXVA copy-back can do that using recent versions of LAV filters, but nevcairiel should reply for sure.

[nevcairiel]

You can do that if the GPU is active. It needs a connected screen for that.

[aufkrawall]

I think that DXVA copy-back can do that using recent versions of LAV filters, but nevcairiel should reply for sure.
Only D3D11VA-CB could do that, but nev stated it'd be really ugly for him to implement.
mpv has D3D11VA-CB support, but it's slow and it seems to be lacking support for using different devices.

You can do that if the GPU is active. It needs a connected screen for that.
It wouldn't be enough to simply add a second cable that connects the display with the IGP and dGPU at the same time, right? If I'm not mistaken, there needs to be an actual signal send to the display?

[huhn]

even a fake display is enough so there is no need to send a signal.

far from perfect but better than nothing.

[aufkrawall]

It always annoyed me to the death that the mouse cursor could leave the real display and got lost on the fake one.

[pankov]

aufkrawall,
I think it was since the Creators Update that if you position your fake display in the bottom right corner of a real one so they have only one corner touching (one vertex to be precise) the mouse won't be able to pass to the fake one.

---------------
| |
| R |
| |
---------------
---------------
| |
| F |
| |
---------------

[huhn]

this works but this doesn't change that it is not practical for the AVG user.

[edwdevel]

It's a 8K sample with only 133MB footprint on your disk.

I would like also you to test this 4K clip with huge 700Mbps bandwidth.

OK, here is a link of YouTube with 4K60fps in VP9 codec:
https://www.youtube.com/watch?v=ppy_gjBp71E

Your 2GB VRAM is a no go for Netflix 4K streaming, which requires >3GB VRAM

Also, as others have already mentioned here, using madVR with 4K content requires >2GB VRAM, especially for the clips you mentioned to test like "Exodus" and "Life of Pi" clips which are HDR.

Did you find the "GPU Video Clock" ?

I've edited my 29th June 2017, 18:52 forum entry with the "tunnelflag" 700Mbs clip and the "8Kexsamp" 8K sample.

I've also done some benchmarking on the vp9 youtube clips, and have done more investigation on different platforms using the vp9 codec on youtube. To get to the point with the gtx 1030, I am working with my old 2core3.0Ghz wolfdale and Windows7, and it seems that the GPU is not being used for decoding under Chrome, as the cpu is maxed out at 1080p vp9, although there is some GPU activity (23%), and there are many dropped frames at 1080p. Furthermore, it seems that the coding efficiency is not too good for vp9, as even 1080p seems to require around 50Mbs to run without frame drops. Here at work with the 1030 and around 50Mbs max internet (if that) I must drop down to 720p before I have no dropped frames (this is with your provided clip).

Contrast the above with my home system, which is a sandybridge 4core3.4Ghz system with an Nvidea GT 970. My internet speed at home is 80Mbs+, and 1080p is flawless although the cpu is running high (50%). Even 4K can probably be decoded by the cpu, but the bitrate required is <100Mbs for the vp9 clip, so I run out of internet bandwidth.

The bottom line is that it is youtube and chrome that dictate the efficiency of the download. Google should immediately convert the download of all of it's content to HEVC, and jettison vp9. Then my poor old PC at work with its good-enough gt 1030 can view all of these clips at a reasonable internet bit-rate. :)

I think it's clear that 2G is enough to decode *and* downsize basically any 4k clip. If netflix needs more than this it has its' own reasons, but certainly technically the 1030 can do the job.

I've also done some madvr investigation, and believe 2G is still adequate as far as memory is concerned, but madvr is a completely different animal as far as renderers go and basically it acts like a 3D game that needs as much GPU power as you can afford. Still, I can run both the move clips (exodus and lifeofpi) using madvr (via MPC-BE) with the GT 1030 perfectly, but clearly the 1030 is overloaded at 60fps with, for instance, the "Sony Camp" clip. I'll comment more on what I've discovered in response to a different reply to my June29 comment.

[el Filou]

I've also done some benchmarking on the vp9 youtube clips, and have done more investigation on different platforms using the vp9 codec on youtube. To get to the point with the gtx 1030, I am working with my old 2core3.0Ghz wolfdale and Windows7, and it seems that the GPU is not being used for decoding under Chrome, as the cpu is maxed out at 1080p vp9, although there is some GPU activity (23%), and there are many dropped frames at 1080p. Furthermore, it seems that the coding efficiency is not too good for vp9, as even 1080p seems to require around 50Mbs to run without frame drops. Here at work with the 1030 and around 50Mbs max internet (if that) I must drop down to 720p before I have no dropped frames (this is with your provided clip).
It looks like Chrome wants/needs (maybe a dev can clear that out?) at least Windows 10 1607 or later to enable VP9 acceleration: https://codereview.chromium.org/2182263002/

On my own 2-core Wolfdale under Windows 10, YouTube VP9 is perfectly smooth in 1080 with absolutely no frame drops whatsoever with a 30 Mbps connection. 2160 is a bit too much for my connection, but if I buffer for long enough it will also play without any frame drops.

[huhn]

2G for madVR should be totally save for 1080p output.

madVR uses an advanced HDR-SDR conversation which is maybe to much for a 1030 even at 1080p but a 1030 should be able to do 60 FPS with the correct settings.

and why should google change from VP9 too HEVC your card should be able to hardware decode it in chrome and they need to pay for HEVC unlike VP9.

[aufkrawall]

aufkrawall,
I think it was since the Creators Update that if you position your fake display in the bottom right corner of a real one so they have only one corner touching (one vertex to be precise) the mouse won't be able to pass to the fake one.

---------------
| |
| R |
| |
---------------
---------------
| |
| F |
| |
---------------
Thx, good to know. Let's just hope this is by design and not actually a bug.

[NikosD]

To get to the point with the gtx 1030, I am working with my old 2core3.0Ghz wolfdale and Windows7, and it seems that the GPU is not being used for decoding under Chrome, as the cpu is maxed out at 1080p vp9, although there is some GPU activity (23%), and there are many dropped frames at 1080p.

Your OS is too old, at least for Microsoft and probably Google, you need to upgrade to Win 10 Creators Update in order to HW accelerate VP9 8bit/10 bit.

Using Windows 10 you could try even Edge - the Win 10 only browser of Microsoft - which probably can HW accelerate VP9 too.

Furthermore, it seems that the coding efficiency is not too good for vp9, as even 1080p seems to require around 50Mbs to run without frame drops. Here at work with the 1030 and around 50Mbs max internet (if that) I must drop down to 720p before I have no dropped frames (this is with your provided clip).

Contrast the above with my home system, which is a sandybridge 4core3.4Ghz system with an Nvidea GT 970. My internet speed at home is 80Mbs+, and 1080p is flawless although the cpu is running high (50%). Even 4K can probably be decoded by the cpu, but the bitrate required is <100Mbs for the vp9 clip, so I run out of internet bandwidth.

The bottom line is that it is youtube and chrome that dictate the efficiency of the download. Google should immediately convert the download of all of it's content to HEVC, and jettison vp9. Then my poor old PC at work with its good-enough gt 1030 can view all of these clips at a reasonable internet bit-rate. :)


You are totally wrong here.

I have a VDSL2 (FTTC) true 50Mbps line and I can easily keep up with the 4K60fps stream of the clip provided above by me.

For 1080p the required bandwidth is below <20Mbps.

Probably there is something wrong with your line - maybe it's not real 80Mbps.

I can't see something wrong with Google servers or VP9, it must be your side.


I think it's clear that 2G is enough to decode *and* downsize basically any 4k clip. If netflix needs more than this it has its' own reasons, but certainly technically the 1030 can do the job.


I don't own a 4K monitor, it's only 1080p, but from what I have read here the problem is when you try to decode and display 4K clips on 4K native resolution without downsizing - this is where you need more than 2G VRAM - and probably the same goes for Netflix.

[NikosD]

I've edited my 29th June 2017, 18:52 forum entry with the "tunnelflag" 700Mbs clip and the "8Kexsamp" 8K sample.


Your two new results, place 1030 above 960 and closer to 1050 and even a little above for "tunnel" clip.

Did you change something (driver, setting) ?

Maybe you could run again some of your previous tests with your new configuration or it is something coincidental.

[edwdevel]

Your two new results, place 1030 above 960 and closer to 1050 and even a little above for "tunnel" clip.

Did you change something (driver, setting) ?

Maybe you could run again some of your previous tests with your new configuration or it is something coincidental.

Hmm. It looks like the latest Nvidia drivers are far better than the previous ones- I've edited again my my 29th June 2017, 18:52 forum entry with new benchmarks for the 384.76 drivers.

I think that is the only change to the June29 setup.

The new benchmarks seem 50% faster. I think everyone with a Pascal benchmark should re-do them with 384.76 to ensure there is not something strange in my setup :)

[edwdevel]

It looks like Chrome wants/needs (maybe a dev can clear that out?) at least Windows 10 1607 or later to enable VP9 acceleration: https://codereview.chromium.org/2182263002/

On my own 2-core Wolfdale under Windows 10, YouTube VP9 is perfectly smooth in 1080 with absolutely no frame drops whatsoever with a 30 Mbps connection. 2160 is a bit too much for my connection, but if I buffer for long enough it will also play without any frame drops.

Yes, I'm sure you're right about Chrome needing Windows10 for hardware VP9 decoding. I wonder if it has something to do with the DirectX version? At any rate, if Google wanted, it means they could make it work in Windows7 as well, it's just a matter of desire.

As far as how youtube does VP9 streaming is concerned, it is difficult to know just what the actual bitrate of the clip is from the "stats for nerds" that youtube provides. Apparently youtube just buffers the desired video at the maximum bit rate the user can support and then dribbles it out as needed. But obviously, if the average bit rate at the desired resolution (say 1080p or 4k) cannot be sustained, then the video will drop frames no matter how fast the cpu or gpu are. Furthermore, as I have discovered and is also obvious, different clips have different required bit rates. We should keep our discussion to the clip NikosD provided- https://www.youtube.com/watch?v=ppy_gjBp71E, which happens to be a very difficult 4k video requiring a high bit rate. Other VP9 4K videos obviously have a much lower bit rate requirement.

To compare actual bit-rate requirements for different videos, we need VP9 4K 10 bit downloadable content to actually calculate the bit-rates. I haven't seen any such videos to test, but I haven't also looked very hard :)



r

[NikosD]

Yes, I'm sure you're right about Chrome needing Windows10 for hardware VP9 decoding.

As far as how youtube does VP9 streaming is concerned, it is difficult to know just what the actual bitrate of the clip is from the "stats for nerds" that youtube provides. Apparently youtube just buffers the desired video at the maximum bit rate the user can support and then dribbles it out as needed. But obviously, if the average bit rate at the desired resolution (say 1080p or 4k) cannot be sustained, then the video will drop frames no matter how fast the cpu or gpu are.


Did you see my previous reply regarding those issues ?

[edwdevel]

2G for madVR should be totally save for 1080p output.

madVR uses an advanced HDR-SDR conversation which is maybe to much for a 1030 even at 1080p but a 1030 should be able to do 60 FPS with the correct settings.

and why should google change from VP9 too HEVC your card should be able to hardware decode it in chrome and they need to pay for HEVC unlike VP9.

For 1080p madvr, the 2G memory of the GT 1030 is perfectly adequate (at least for the clips I have tested). It is 4k 60fps HEVC that is questionable.

We need HEVC for the simple reason that it is more efficient, so that *I can view the same video at the same quality* for half the bit-rate, which is critical because for most people internet bandwidth is finite. I know Google doesn't care about this, but all of its' users do. :)

[sneaker_ger]

To compare actual bit-rate requirements for different videos, we need VP9 4K 10 bit downloadable content to actually calculate the bit-rates. I haven't seen any such videos to test, but I haven't also looked very hard :)
Look no further:
https://www.youtube.com/watch?v=tO01J-M3g0U (download using youtube-dl)

[huhn]

We need HEVC for the simple reason that it is more efficient, so that *I can view the same video at the same quality* for half the bit-rate, which is critical because for most people internet bandwidth is finite. I know Google doesn't care about this, but all of its' users do. :)

HEVC isn't even remotely close to 2x compression efficiency compared to VP9.

[edwdevel]

Look no further:
https://www.youtube.com/watch?v=tO01J-M3g0U (download using youtube-dl)

Have you actually downloaded this file? I'm having trouble getting it: I keep getting the H264 720*1280 version. Please download it yourself, ensure that it is in fact the 4K VP9 version, play it using MPC-BE or DXVA checker, and then let me know how you did it :).

[nevcairiel]

We need HEVC for the simple reason that it is more efficient, so that *I can view the same video at the same quality* for half the bit-rate, which is critical because for most people internet bandwidth is finite. I know Google doesn't care about this, but all of its' users do. :)

HEVC and VP9 are in a similar efficiency range, you won't get any significant savings from using HEVC over VP9 - if any at all.

[sneaker_ger]

Have you actually downloaded this file?
It is not one file. Youtube provides 41 different streams for this video. You can get a list via
youtube-dl -F https://www.youtube.com/watch?v=tO01J-M3g0U

Then download e.g stream #337 (4kp60 VP9 10 bit HDR):
youtube-dl -f 337 https://www.youtube.com/watch?v=tO01J-M3g0U

https://rg3.github.io/youtube-dl/

[stax76]

I'm also using youtube-dl, great tool, here is a great Firefox extension which can download it too:

https://addons.mozilla.org/de/firefox/addon/youtube-video-and-audio-dow/?src=api

[NikosD]

Hmm. It looks like the latest Nvidia drivers are far better than the previous ones- I've edited again my my 29th June 2017, 18:52 forum entry with new benchmarks for the 384.76 drivers.

I think that is the only change to the June29 setup.

The new benchmarks seem 50% faster. I think everyone with a Pascal benchmark should re-do them with 384.76 to ensure there is not something strange in my setup :)

Now that the driver is tuned enough and your results look close to 1050/1050 Ti, I suggest you to download everything from here:
ftp://helpedia.com/pub/multimedia/testvideos/HEVC_HW_decoders_clips/

and test them with latest LAV x64 filters and DXVA Checker in both decode and playback mode of 1280x720 scaling in order to put your results here in the table:
https://forum.doom9.org/showthread.php?p=1799099#post1799099

[clsid]

What is the current state of decoding on AMD GPUs with recent drivers? Does HEVC work properly now? Have the interlaced video problems been fixed?

[NikosD]

You can take a look at my links regarding performance of HW decoding on both H.264 & H.265.

It is more than fast enough for anything meaningful e.g Blu-ray UHD (not corner cases like 8K)

HEVC never had a problem using Polaris card, although it was a little bit on the edge regarding performance.

But not anymore, new drivers solved that "problem".

Now, regarding H.264 interlaced content, my last test using latest drivers for AMD, Nvidia, Intel and latest LAV x64 filters gave me a little surprise.

AMD has now exactly the same behavior like Nvidia and Intel.

So, H.264 interlaced is definitely fixed too.

[clsid]

Good news. Thanks. Since which driver version has the performance increased? 17.4 or later?

[NikosD]

Yes, around there 17.4.x

[edwdevel]

HEVC and VP9 are in a similar efficiency range, you won't get any significant savings from using HEVC over VP9 - if any at all.

Please understand I am not interested at all in proving HEVC is better than VP9 or vice versa. I was under the impression, from basic scientific consensus, that in encoding efficiency HEVC produces about half the bit rate of VP9 for videos at the 4k level. Just google "HEVC vs VP9 efficiency" for many papers that show this. Here is one:

http://iphome.hhi.de/marpe/download/Performance_HEVC_VP9_X264_PCS_2013_preprint.pdf

But, as I've been shown a 4k video that is encoded under both HEVC and VP9, I will undertake to look at both and report back :)

[huhn]

it's from 2013 and the used encoder plays a major if not the most important role for compression.

what so ever in term of compression potential VP9 and HEVC are not far apart.

[nevcairiel]

Please understand I am not interested at all in proving HEVC is better than VP9 or vice versa. I was under the impression, from basic scientific consensus, that in encoding efficiency HEVC produces about half the bit rate of VP9 for videos at the 4k level. Just google "HEVC vs VP9 efficiency" for many papers that show this. Here is one:

http://iphome.hhi.de/marpe/download/Performance_HEVC_VP9_X264_PCS_2013_preprint.pdf

That paper uses PSNR to measure video. That might be "scientific", but its not any way to actually measure video quality accurately (which in itself is not something thats really "solved", but PSNR is one of the worse things to use). Early HEVC encoders achieved good PSNR by creating blury video, which was very obvious to any human viewer, but rated quite well in PSNR tests.

[littleD]

dxva output with newest intel drivers

[NikosD]

Well, it seems that I managed to come up to some conclusions regarding VP9 acceleration (hybrid or pure fixed-function) of the three well known browsers (Chrome, Firefox, Edge) using Youtube on different hardware.

First of all VP9 acceleration is supported by all three nowadays and it definitely works on recent hardware supporting fixed-function VP9 acceleration like Kabylake for Intel and 950/960 for Nvidia, along with all Nvidia Pascal cards of course.

Some browsers could have problems with 8K VP9 acceleration though, but that's a very new and extremely high resolution option of Youtube streams, added lately.

The other fact is that for Chrome and Chromium based browsers you need a very recent Win OS, which is Windows 10 RS1 (Anniversary Update with a release on August 2016) or later according to this:

https://groups.google.com/a/chromium.org/forum/#!topic/chromium-reviews/qSePjavyJsU

The first official stable Chrome version supporting VP9 acceleration is v54, released after Anniversary Update on November 2016.

Now, Edge as we know is a Windows 10-only browser so obviously you need a Windows 10 OS and Edge enabled VP9 acceleration in Windows Insider builds before Anniversary Update, but the first stable Edge version supporting VP9 acceleration is inside Windows 10 RS1 (Anniversary Update) according to this:

https://blogs.windows.com/msedgedev/2016/04/18/webm-vp9-and-opus-support-in-microsoft-edge/#aPHsQSA3aIvOpy38.97

I don't know when exactly Firefox enabled in stable releases the VP9 acceleration and what are its Win OS requirements, but it seems that using Windows 10 and latest versions like v55, it works.

VP9 hybrid acceleration was firstly supported by Intel iGPUs like Broadwell and Skylake and AMD added that functionality officially on December 2016 inside Crimson 16.12.1 drivers only for Chrome:

VP9 Decode Acceleration: 4K 60Hz GPU-Accelerated Video Streaming enabled on supported Google™ Chrome web browsers.

Requires supported Chrome™ web browser versions with Hardware Acceleration enabled.
Compatible with AMD Radeon™ GCN and Radeon RX 400 series enabled products on Windows® 7/8.1/10.


It says Windows 7/8.1 but as I told you that is wrong according to my sources.

Also, as I read, it's an OpenCL based solution not only for RX 400/RX 500 series but some previous generations, too.

The first driver didn't enable the VP9 hybrid acceleration by default and the second driver 16.12.2, fixed that according to AMD.

Fixed Issues

Chromium may fail to utilize hybrid decode for VP9 content.


The truth is that Chrome never enabled by default the VP9 hybrid acceleration for Polaris cards.

You have to force enable it using this switch:

chrome.exe --enable-accelerated-vpx-decode=2

The reason is probably low performance.

But for Intel and CPUs prior to Kabylake, things are even worse.

All of the VP9 hybrid acceleration capable CPUs, like Broadwell and Skylake, are not allowed to use VP9 hardware decoding of Chrome according to this patch:

https://groups.google.com/a/chromium.org/forum/#!topic/chromium-reviews/BghP-1eqyhg

I have no idea regarding Edge and Firefox support of Intel's hybrid VP9 acceleration.

For Polaris cards, the actual use of that somewhat crippled VP9 hybrid decoder, lasted only 4 months between the first driver support and Windows 10 Creators Update which broke VP9 acceleration.

So, for that reason, I have opened a thread in AMD forums for anyone interested to read and contribute to the common effort of an actual fix - if possible - of VP9 hybrid acceleration.

You can read and write your comments here:
https://community.amd.com/thread/218278

I have a Core i5 2400 right now with a RX 470/8GB card and a real 50Mbps VDSL2 (FTTC) line, testing 4K60 fps clips and I think AMD should have kept its promise of offering VP9 4K60fps support using Polaris cards.

[peca89]

Hello,

I'm following this forum for a month now and I didn't find definitive answer if a new GT 1030 would be adequate upgrade for my years old HTPC. It will be connected directly to 4K HDR TV using HDMI.


Can it upscale 1080p60 and output 2160p60 using EVR? Can it do it using madVR?
Can it decode AVC/HEVC/VP9 2160p60 SDR and output 2160p60 SDR? EVR or madVR?
Can it decode AVC/HEVC/VP9 2160p60 HDR and just pass it through to a HDR compatible TV, therefore skipping resource-hungry 4K scaling and HDR<->SDR conversion?

[NikosD]

All these are good questions, but I'm afraid GT 1030 is not so popular so the one and only owner appeared here has declared that he doesn't own a 4K display.

We must find a combination of 1030 using a 4K display, which could be rare considering low GPU power of the card and only 2GB VRAM.

[jmonier]

Based on my limited experience with a GTX1060 3gb, I would say there's no chance. In all these cases, memory usage was well above 2 gb and seemed to be pushing the limits for both memory and performance. I'm going to a GTX1070 as a result.

[v0lt]

@NikosD
2 GB is enough to play 4k video. Of course, you can specially tweak the settings, but this is not an indication that the video card is bad.

[NikosD]

The problem could be the output of 4K not the decoding and then downscaling to 1080p.

I don't have a 4K display, but seems strange that Nvidia has set the requirement to >3GB for 4K Netflix playback, if 2GB could be enough.

[huhn]

while it is possible to playback UHD videos at UHD with 2GB Vram you are really pushing it.

i was using a 960 with 2Gb for sometime doing this just opening a web browser was pushing it over the limited for me.

cutting corners by using 8 bit buffers like mpc-BE may help in term of Vram usages but that doesn't mean everyone is unable to see the banding...

[NikosD]

Good news!
It seems that latest Chrome Canary v62 (x64) works like a charm using VP9 hybrid acceleration of a Polaris card and Creators Update!

Firstly, like plain Chrome v59 (x64) it recognizes out of the box VP9 Youtube content but it doesn't enable by default the hybrid VP9 acceleration.

BUT if you force enable VP9 acceleration using the --enable-accelerated-vpx-decode=2 switch, you see no stuttering in the video decoding and it has a minimal frame dropping of just 35 frames during the whole duration of the very difficult 4K60 fps Youtube clip posted above.

The GPU load works exactly like Chrome, but without stuttering and you see the GPU and memory clocks going to the highest levels.
The CPU usage is around ~50% on average.

So, it seems that AMD's VP9 hybrid acceleration works, at least on one Chromium based browser and force enabled.

Of course Canary has some strange behavior as an experimenting browser, so we definitely need a more stable overall browser like Chrome to implement correctly the VP9 hybrid acceleration.

[wanezhiling]

https://techreport.com/news/31627/amd-publishes-patches-for-vega-support-on-linux
The UVD 7.0 decoder and Video Coding Engine 4.0 encoder are reported to be included in the upcoming Vega based GPUs.
Hope AMD could catch GM206 now..

[v0lt]

@NikosD
Netflix wrote a complete nonsense. "GTX 1050 (not Ti) with 3 GB of video memory".

[NikosD]

No. It was not Netflix.

I told you what Nvidia wrote in their requirements.

>3GB VRAM

[el Filou]

All these are good questions, but I'm afraid GT 1030 is not so popular so the one and only owner appeared here has declared that he doesn't own a 4K display.

We must find a combination of 1030 using a 4K display, which could be rare considering low GPU power of the card and only 2GB VRAM.
Spec-wise, a 1030 is pretty much half a 1050 Ti in every way: half the quantity of VRAM, half the number of transistors and half the number of compute cores, and nearly half (43%) the memory bandwidth.
So in theory you just need someone with a 1050 Ti and a 4K TV (sorry, 1080p for me), and ask them to watch with GPU-Z if GPU usage, VRAM usage, and memory controller usage never go above 50% (or let's say 40-45% to keep a safe margin) and if it doesn't then it should be fine for a 1030, right?

[huhn]

you don't need a 4k display to test 4k "output" at least not with nvidia.

the problem is the Vram and nothing else. if you do a proper render chain with proper buffer you will run out of Vram or close to it.

[peca89]

Thanks for your replies. I must admit I do not really understand why do you think (or why it is) 4k so demanding. If you accept the logic that performance and ability to play video in realtime can be linearly scaled like some people here mentioned, can you explain why my ages old Radeon 5670 with only 512MB of VRAM and only 400 MHz of clock during playback is perfectly capable of chewing 1080p at 60 Hz. It stays below 250 MB of memory usage and below 50% of GPU usage using Win 10, MPC-HC, EVR and LAV video with DXVA native. 4k is "only" 4 times more pixels than 1080p and 1030 has 4x more VRAM, 3x GPU clock and much faster shaders.

Main question: Why does 4k need >2GB of VRAM when 1080p needs 10 times less?

[el Filou]

Thanks for your replies. I must admit I do not really understand why do you think (or why it is) 4k so demanding. If you accept the logic that performance and ability to play video in realtime can be linearly scaled like some people here mentioned, can you explain why my ages old Radeon 5670 with only 512MB of VRAM and only 400 MHz of clock during playback is perfectly capable of chewing 1080p at 60 Hz. It stays below 250 MB of memory usage and below 50% of GPU usage using Win 10, MPC-HC, EVR and LAV video with DXVA native. 4k is "only" 4 times more pixels than 1080p and 1030 has 4x more VRAM, 3x GPU clock and much faster shaders.

Main question: Why does 4k need >2GB of VRAM when 1080p needs 10 times less?
You are right, and 2 GB is well enough for "basic" 4K video playback (Edit: except for Netflix, if you believe NVIDIA, but they've never explained the exact reason), it's just that quality is a highly subjective question.
A lot of people on this forum are videophiles, and if you want to take full advantage of the processing power of a PC GPU to enhance the rendering quality and smoothness of your video playback, then you'll have to use things like a long buffer queue, internal processing using high precision floating point formats etc. All of this combined can increase VRAM usage quite a lot.

Here are some numbers on my 1050 Ti, playing HEVC Main10 2160p60 140 Mbps (i.e. Blu-ray UHD level) with DXVA Native, and outputting to 1080p:

MPC-HC EVR: 705 MB
MediaPortal EVR-CP: 1110 MB
MPC-HC madVR: 1483 MB
MPC-HC EVR-CP 8 buffers: 1765 MB (I can't understand why it uses more than madVR)
MediaPortal madVR: 1802 MB

Using DXVA CopyBack instead of Native adds 370 MB to those numbers.
Outputting to 2160p 10 bit instead of 1080p would add a few 100 MBs again.

[v0lt]

MPC-HC EVR-CP 8 buffers: 1765 MB (I can't understand why it uses more than madVR)
Why 8 buffers? What is the format of these buffers?

[el Filou]

Why 8 buffers? What is the format of these buffers?
The renderer setting is just called "EVR Buffers", the default is 5 but I found that increasing that value a bit prevents skipped frames in rare conditions.
I guess it's a setting equivalent to madVR's "GPU queue size".

[huhn]

@el Filou

you are aware that at UHD output the render queue and the output frame are at least 4X as big which is more than 100 mb...

like i said before you can test it using nvidia DSR. and again UHD output with 2 GB is possible but you are really pushing it.

[el Filou]

@el Filou you are aware that at UHD output the render queue and the output frame are at least 4X as bigYes I am.

which is more than 100 mb...I know, that's why I wrote "a few 100 MBs", but I should have written "a few hundreds".

like i said before you can test it using nvidia DSR. and again UHD output with 2 GB is possible but you are really pushing it.Thanks for the clarification, I didn't understand how it could be done when you said we could test 4K output.
I've tried DSR and here is the VRAM usage I find in 2160p, with DXVA Native:

EVR: 1105 MB
EVR-CP, render queue 5: 1775 MB
madVR (my custom settings): 2480 MB :o
madVR shortest render queue: 1988 MB.

So EVR is fine, MPC-HC's EVR-CP is still alright but approaching the limit.

[huhn]

windows high DPI settings can take up some Vram just opening a web browser can make it harder too.

adding needed 10 bit processing and output is going to make it even harder.
subtitles and other things too.

i made madVR work with 2 GB too but i will never ever recommend someone to do the same so do yourself a favor and get 4 Gb Vram or more.

[v0lt]

Never use 32-bit Floating Point textures (in MPC-HC, this is called Full Floating Point Processing). This leads to a meaningless consumption of video memory.

If you really need a floating point for intermediate results, then use 16-bit Floating Point textures (in MPC-HC, this is called Half Floating Point Processing). This is quite enough.

I'll advise you to see the video settings and playback statistics in MPC-BE. There it is more clear.

[edwdevel]

Hello,

I'm following this forum for a month now and I didn't find definitive answer if a new GT 1030 would be adequate upgrade for my years old HTPC. It will be connected directly to 4K HDR TV using HDMI.


Can it upscale 1080p60 and output 2160p60 using EVR? Can it do it using madVR?
Can it decode AVC/HEVC/VP9 2160p60 SDR and output 2160p60 SDR? EVR or madVR?
Can it decode AVC/HEVC/VP9 2160p60 HDR and just pass it through to a HDR compatible TV, therefore skipping resource-hungry 4K scaling and HDR<->SDR conversion?


The short answer to your questions is: if you want to use madvr with a 4k TV,
you will probably not be happy with a GT 1030. :)

The long answers to your questions are:

First, I am the poster here who benchmarked several clips using a GT 1030 and a
wolfdale 2core 3Ghz and a 1600x900 monitor. Since I posted the benchmarks,
my old FHD LCD TV died, and has since been replaced with a brand new Sony XBR900E UHD TV.
Unfortunately, I don't have a HTPC (nor will I), so I still won't be able to
definitively tell you that a 1030 can drive a 4k TV. However, I have been doing
some research with madvr vs. EVR using the 1030, and I believe that some reasonably
confident conclusions can be made about the performance of the GPU with 4k output.

Several posters here have shown that the 1030 with its' 2GB memory *should* be able
to drive a 4k TV. Also, is it obvious that there is nothing special about the 1030 in
Nvidia 10 series line of 1050, 1060, etc. when it comes to 4k output? Nvidia has not
come out with a bulletin stating that, sorry, our 1030 just cannot drive 4k output :).
I have done playback benchmarks using DXVA Checker with native 4k resolution and EVR
with various 4k clips. In "Colors of Journey" I got over 100fps, with low cpu and 100%
GPU (this may be normal for playback benchmarks). It is illogical to think that
the GPU scanning electronics cannot drive an HDMI output at 4K from a video memory
resident frame. Thus, I think we can confidently state that the 1030 can display 4k60fps
clips using HEVC and VP9 and EVR.

Madvr, though, is a completly different story. I benchmarked a 4k24fps clip "Exodus"
downsizing to monitor size (about 1080p) with MPC-BE and madvr, and basically
got perfect beautiful HDR output, with the cpu at 10% and the gpu at 90%- without DXVA
downsizing. As soon as DXVA downsizing is set on, the GPU goes to 20% (though the
colors are all wrong- I reported this as a bug on the madvr forum here). This
demonstrates the real issue with madvr: it basically is an image processor using D3D
GPU function for even simple downsizing and is completely dependant on the performance
of the GPU board. My 1030 cannot play 4k60fps clips with madvr bilinear downsizing
to 1080p- not because of the 2G memory, but simply because the 1030, which is the
weakest GPU in the 10 series, cannot keep up with madvr's demands. I can say, though,
both "Exodus" and "Life of Pi" 4k24fps can play perfectly- downsized to 1080p.

So, if you want to use madvr to do image processing such as downscaling, upscaling,
sharping or basically anything, you can forget about using the 1030. You should hang
out in the madvr forum here, where people complain about $600 GTX 1080 boards having
performance problems. :)

But why do you even need an HTPC if you are going to a modern 4k TV? I've been enjoying
my new android wi-fi state of the art full array dimming UHD TV for the past few days,
which can play all of my 4k clips using wi-fi DLNA perfectly, not to mention all 4k
youtube clips and, yes, Netflix UHD, directly over wi-fi connected to the router (though
I did have problems with the old wi-fi speed, and have since upgraded both my router and
internet connection- but that is another story)

Regards,
edwdevel

[NikosD]

Major performance drop using latest 17.7.2 drivers.

The GPU clocks have been stabilized and using pure decoding mode, the memory clock is always at min state (300 MHz) and the GPU clock is at min state (300 MHz) around 90% of the whole decoding process time, going up to 466 MHz occasionally.

Of course pure video decoding mode has nothing to do with GPU clocks, but memory clock always at min state (300 MHz) could be a problem.

AMD has probably put the internal video decoder clocks lower (they are invisible to the user)

The result is a pure decoding speed less than the <17.4 drivers, a lot less to be more exact. Huge drop in performance.

Now, during playback benchmark mode we are on the opposite situation, regarding GPU/ memory clocks.

GPU clock and Memory clock are at max speed, but still the decoding speed is less than previous drivers, but better than <17.4 drivers.

I also got a micro-stuttering issue while moving cursor on the desktop, it seems that the cursor pauses just for a sec with no reason, like some older situations when incompatible drivers have been installed on the system or some other kind of interrupt caused that little pause.

17.7.2 is a major update, but a probably a little unpolished.

I don't know if the decision of slowing down internal video clocks of the video decoder (invisible to the user) in pure decode mode is permanent for AMD.

[el Filou]

Major performance drop using latest 17.7.2 drivers.

The GPU clocks have been stabilized and using pure decoding mode, the memory clock is always at min state (300 MHz) and the GPU clock is at min state (300 MHz) around 90% of the whole decoding process time, going up to 466 MHz occasionally.

AMD has probably put the internal video decoder clocks lower (they are invisible to the user)Did they slow it down so much that you can't play 2160p60 smoothly?

[NikosD]

In playback mode the decoding speed is:

Drivers earlier than 17.4 < 17.7.2 < Drivers after 17.4.x up to 17.7.1

So, you can play 2160p60 smoothly, but previous drivers from 17.4.x up to 17.7.1, were definitely faster.

[el Filou]

Maybe they introduced a bug while trying to optimise power profile?
I understand that in this new computing landscape where laptops and other mobile devices represent the mass market, you'd want the lowest possible power consumption while still providing real-time playback, but they should provide an advanced option to put the video decoder to highest performance state for desktop parts.

[aufkrawall]

I've followed Phoronix news quite thoroughly during the last months, and I think there's still not a single sign of AMD supporting VP9 with Vega on Linux. :(
Nvidia's cuda decoding isn't great either since it has more often problems with certain formats (e.g. HEVC 10 bit HDR), but at least it works well for VP9 YT videos (and without quality degradation, unlike shit APIs by Microsoft).

[nevcairiel]

it works well for VP9 YT videos (and without quality degradation, unlike shit APIs by Microsoft).

Microsofts decoding APIs are perfectly bit-exact decoding. Try not to spread false rumors.

[NikosD]

Least expensive RX VEGA 56 will cost around 400$, so I don't think a lot of people would buy it for HTPC or watching YouTube videos.

Only video quality freaks using madVR could find that card interesting for video playback, along with 3D gaming of course.

But I think the plans of AMD involve a desktop APU release on early 2018, along with some medium level cards based on VEGA replacing Polaris cards, so yes, in a longer term it's important for VEGA core to support VP9 hardware decoding.

[aufkrawall]

Microsofts decoding APIs are perfectly bit-exact decoding.

You likely know the issues very well, no need to be extra-smart.

[nevcairiel]

You likely know the issues very well, no need to be extra-smart.

In which case you likely do not, otherwise you would know better.

There are absolutely no issues to get a perfect bit-exact image out of the hardware decoder using DXVA. FFmpeg has regression tests that test the bit-exactness of decoding, you are free to run those with DXVA decoding to confirm this fact.

I can only guess you are referring to the DXVA-native "issue" that madVR has (and some other players following a similar implementation).
This is not related to decoding, the decoder gives you a perfect bit-exact image back, it only gets degraded through the process of how these renderers convert it into a texture - which is a D3D9 limitation (lack of 4:2:0 textures). You could use DXVA with D3D11 to avoid this (ffmpeg calls this "d3d11va"), or even map the decoded surface to OpenGL - or the easy alternative, albeit with a performance hit (although in my experience on NVIDIA the impact is barely measurable), use Copy-Back.

If you build against Direct3D, using D3D11 is clearly the best option, because you should be using that anyway to get NV12/4:2:0 support for your textures, so you don't have to cheat the chroma upsampling shaders.

NVDEC decoding (formerly CUVID) isn't any different. Either you use copy-back or you map the CUDA memory as a OpenGL surface, or even a D3D surface - although if you use D3D9, the same problems would manifest themself.

With any decoding API one of the main problems is how to get the image from the decoder format (typically a simple GPU memory buffer) into a format for display (typically a texture), but this is generally more limited by the display API you are using, not the decoder. In this case above, D3D9 (not DXVA) is just limited, but we have a modern replacement already.

[aufkrawall]

Which renderer does not show quality issues in shape of chroma blur and additional banding on Nvidia DXVA2/d3d11va without copyback?
madVR, MPDN and mpv do. I also gave Microsoft's Windows 10 video app a try some time ago, looked bad too (might have to recheck this, but not expecting anything).
That there is no quality degradation by just outputting the frame to memory is a totally different topic which I wasn't talking about. Read more carefully perhaps?
And mpv also has the issue with d3d11va and DX11 ANGLE renderer.

The issues do not occur with native cuda (no copyback) on both Windows and Linux and neither with VDPAU (no copyback) on Linux.
It's only bad with Microsoft APIs involved, such coincidence.
(The sames goes btw. for GPU compositing in browsers.)

[nevcairiel]

Its perfectly possible to do this with D3D11 (which madVR or MPDN cannot do right now, since they don't have a decoder yet that outputs D3D11 surfaces - in fact does MPDN support DXVA-native at all even?), or even OpenGL (in theory, I haven't personally tested the DXVA -> OpenGL interop)
I don't know what mpv does in d3d11va zero-copy mode, but considering its primarily linux-focused, I wouldn't expect this to be a fully optimized path.

I wouldn't expect the Microsoft Video App to behave high-quality under any definition, it just tasks the GPU to convert the video to RGB, it has no interest in getting a bit-exact copy of the decoded video. The same thing browsers do.


That there is no quality degradation by just outputting the frame to memory is a totally different topic which I wasn't talking about. Read more carefully perhaps?


You didn't actually specify any topic you were talking about, just generically referenced "decoding".

Since you seem to have tested that quite extensively, I would be interested in a sample that shows these differences quite obviously for future reference.

[NikosD]

VEGA supports "Trusted memory zone" for Ultra HD Blu-ray and Netflix 4K:

http://www.tomshardware.com/news/amd-radeon-rx-vega-64-specs-availability,35112.html

[aufkrawall]

Since you seem to have tested that quite extensively, I would be interested in a sample that shows these differences quite obviously for future reference.
It's true that, unlike direct DXVA2, direct d3d11va doesn't show quality degradation with 8 bit input and the Windows 10 video app can also play HEVC 10 bit video without banding (just rechecked it).
However, mpv is limited to rounding down 10 to 8 bit with d3d11va, so probably the decoder can't be told to dither down to 8 bit and angle can't deal with 10 bit input.
Quite stupid limitation, not a problem with cuda or vdpau (and I guess neither vaapi).

[wanezhiling]

Nothing special, just add what Apollo Lake lack

[NikosD]

Another CPU launch, another fraud from Intel.

Coffee Lake is incompatible with Kabylake's chipset, but Intel had said that it will be compatible.

All the happy owners of 100 and 200 series are moving to Ryzen, the best overall mainstream processor ever.

Nobody with a 6C/12T CPU is interested in iGPU of Coffee.

Not even the biggest hard core Intel fan boys like you are going to but that useless and expensive CPU when Ryzen 7 has 8C/16T and is cheaper.

[huhn]

to bad that most people that doesn't really need a PC right now are avoiding to buy one.

[nevcairiel]

Another CPU launch, another fraud from Intel.

Coffee Lake is incompatible with Kabylake's chipset, but Intel had said that it will be compatible.

I don't remember them saying that anywhere officially, and none of the articles talking about the Coffee Lake and 200-series incompatibility have mentioned anything of that sort, got any source for that?

Not really interested in buying a Coffee Lake either way, though, but rumors have a 4c/8t i3 showing up, which might be good value (first quad-core i3, if its true)

[NikosD]

LGA 1151 is LGA 1151 whether it's Skylake, Kabylake or Coffee Lake.

The entire planet thought that Intel will eventually change its notorious mind, but still they keep going on the same way.

6C/12T for mainstream desktop and i3 with 4C/8T (if true) are the consequences of Ryzen competition.

[huhn]

that's not how a chipset works...

[peca89]

I've given up waiting for relevant information to show up on somewhere on the internet so I decided to buy the GT 1030 and test it myself. Let me first answer my own questions:


Can it upscale 1080p60 and output 2160p60 using EVR? Yes, absolutely!
Can it do it using madVR? Did not test properly yet.
Can it decode AVC/HEVC/VP9 2160p60 SDR and output 2160p60 SDR? Yes, absolutely! No problem at all.
EVR or madVR? Tried only EVR so far
Can it decode AVC/HEVC/VP9 2160p60 HDR and just pass it through to a HDR compatible TV, therefore skipping resource-hungry 4K scaling and HDR<->SDR conversion? I managed to set up madVR to trigger HDR mode on TV. However, colors are completely wrong in fullscreen. I need more time figuring out how madVR works


Everything I threw to the card (up to 4K res), it managed to decode and render properly. I've tried most of the HEVC test clips mentioned in this topic and everything works fine except higher than 4K resolutions. VRAM usage at 4K 60fps 10bit HEVC output at 4K@60Hz is up to 1500MB when using EVR.

If anyone has an idea how to set up madVR to do basically nothing and just pass HDR through, please write here. I know that's not what madVR was made for, but it is probably only thing possible using low end video card. Also, I can do some more tests if someone is interested.

[huhn]

don't use DXVA native with madVR.
set the GPU queue to 6 and pray.
don't use FSE with HDR for know it's kind of buggy in the nvidia GPU driver.

give the windows 10 HDR API a try with 10 bit windowed.

it is not possible to do basically nothing to an decoded image using madVR or EVR.

try either DXVA chroma scaling or bilinear both are absolutely terrible...

and make sure no other program like a web browser is running in the background.

[jmonier]

For me, the few HDR samples that are available (Life of Pi, Exodus, etc.) work perfectly with FSE under Win 8.1 and the latest Nvidia driver.

[nsnhd]

Can it decode AVC/HEVC/VP9 2160p60 HDR and just pass it through to a HDR compatible TV, therefore skipping resource-hungry 4K scaling and HDR<->SDR conversion?

Also, I can do some more tests if someone is interested.

For those who don't own HDR displays, HDR ->SDR conversion is crucial. Could you pls test that to confirm the GT 1030 capability ? Thanks

[peca89]

Thanks for the tips. I need to reinstall Windows to get Creators Update to get Windows HDR API because I'm currently on LTSB 2016...

HDR->SDR conversion is mixed success. The best I could get is successful 4K conversion at 24fps, but only if I set "preserve hue in low quality" in madVR settings. Life of Pi sample plays fine. All of 60fps 4K HDR clips I have (e.g. Sony camp) play as a slideshow. 1080p HDR samples at 60fps in VP9.2 (downloaded from Youtube) play fine.

Since madVR requires copy-back (thanks huhn) it's no surprise that 4K 60fps does not work. Even SDR 4K 60fps using EVR does not work if copy-back is used (memory gets maxed out).

But hey, 24fps is fine for movies! :)

[huhn]

madVR works with DXVA native but with native it is using more Vram. i guess the decode queue is in the Vram too now.

[NikosD]

Intel hasn't unveiled a new architecture since August 2015, which was Skylake.
It's been already two years.

The "new" Coffee Lake CPUs are actually clones of AMD's concept of multi-core.

So, probably yes, Core i3 8300 could be a 4C/8T chip, matching all previous generations Core i7s from 2011 Sandybridge at least, regarding number of cores/threads.

Digging in Intel's internals, it seems they face a serious problem of not being able to produce desktop powered/ performance chips using 10nm process - only low-powered.

Intel has a broken 10nm process right now.

With Ryzen/Threadripper/EPYC chips alive and kicking, things couldn't be worse for Intel this period.

Let's see their funny reactions, like Coffee Lake for example.

[el Filou]

Digging in Intel's internals, it seems they face a serious problem of not being able to produce desktop powered/ performance chips using 10nm process - only low-powered.

Intel has a broken 10nm process right now.
Who else has tried to produce 4 GHz+ high power chips in 10 nm? Samsung and TSMC have only produced mobile chips in 10 nm until now, and Samsung '10 nm' is barely finer than Intel's 14 nm, so Intel has quite a margin of development IMO.
Let's wait and see how (AMD) - sorry, GloFo - manages 10 nm too.

[NikosD]

GloFo is moving directly to 7nm for Zen 2 and Navi GPU.

[nevcairiel]

Comparing the marketing process node names is impossible at the current time. Companies have been using different references to measure them, which leads to some smaller numbers that don't really indicate an advantage. Intel may be playing catchup on their lineup, but their processes are still quite stable.
As el Filou already said, Samsung 10nm isn't really smaller then Intel 14nm, its just a different reference of measurement. The same can be applied to TSMC as well. Don't know the details about GloFo off-hand.