Welcome to Doom9's Forum, THE in-place to be for everyone interested in DVD conversion.

Before you start posting please read the forum rules. By posting to this forum you agree to abide by the rules.

 

Go Back   Doom9's Forum > Video Encoding > VP9 and AV1

Reply
 
Thread Tools Search this Thread Display Modes
Old 3rd August 2022, 00:40   #2441  |  Link
Yups
Registered User
 
Join Date: Sep 2011
Posts: 362
Intel's WORLD FIRST GPU AV1 encoder was worth the hype

https://youtu.be/ctbTTRoqZsM


Very good AV1 result from the Arc A380 in this test.
Yups is offline   Reply With Quote
Old 3rd August 2022, 08:14   #2442  |  Link
ChaosKing
Registered User
 
Join Date: Dec 2005
Location: Germany
Posts: 1,795
Finally the one area the arc gpus are good at ^^"
__________________
AVSRepoGUI // VSRepoGUI - Package Manager for AviSynth // VapourSynth
VapourSynth Portable FATPACK || VapourSynth Database
ChaosKing is offline   Reply With Quote
Old 4th August 2022, 15:46   #2443  |  Link
Mr_Khyron
Member
 
Mr_Khyron's Avatar
 
Join Date: Nov 2002
Posts: 203
SSIM Comparison for Intel Arc A380 QSV
https://rigaya34589.blog.fc2.com/blog-entry-1501.html
Mr_Khyron is offline   Reply With Quote
Old 4th August 2022, 23:08   #2444  |  Link
Yups
Registered User
 
Join Date: Sep 2011
Posts: 362
Quote:
Originally Posted by Mr_Khyron View Post
SSIM Comparison for Intel Arc A380 QSV
https://rigaya34589.blog.fc2.com/blog-entry-1501.html

It's struggling at higher bitrates. Unfortunately he only tested ICQ and didn't check out CBR/VBR/CQP as well as MBBRC and ExtBRC, MB or Ext BRC can sometimes improve the quality.

He says HEVC and H264 encoder on Arc improved over ADL-S iGPU by the way.

It remains to be seen if AV1 QSV can improve over software improvements. oneVPL AV1 support looks rough at the moment.


EposVox made a mistake by not choosing the same gop length in his test, Intel AV1 QSV has a 1s seeking in his samples, means he uses a gop length of 60 (videos are 60 fps). VCE, Intel h264, x264 have a 10s seeking in his encoded videos. As for Intel h264 or h265 QSV a gop increase from 60 to 600 gop should improve the VMAF score by roughly 1 point.

Intel hardware AV1 performs really good at 3500 Kbit relative to the others in this test. It's a blocky mess in motion on NVENC or Intel h264. On higher bitrates Intel AV1 QSV detail preservation is quite poor, that's why it's losing relative to the others at higher bitrates.

Here is a frame example at 3500Kbit

Intel H264 QSV
Nvidia H264 NVENC
AMD H264 VCE
Intel AV1 QSV
x264 very slow
Yups is offline   Reply With Quote
Old 9th August 2022, 18:39   #2445  |  Link
benwaggoner
Moderator
 
Join Date: Jan 2006
Location: Portland, OR
Posts: 4,736
Quote:
Originally Posted by Yups View Post
It's struggling at higher bitrates. Unfortunately he only tested ICQ and didn't check out CBR/VBR/CQP as well as MBBRC and ExtBRC, MB or Ext BRC can sometimes improve the quality.
And average SSIM is not a particularly well subjectively correlated metric, particularly for HDR.

Quote:
It remains to be seen if AV1 QSV can improve over software improvements. oneVPL AV1 support looks rough at the moment.
It's been well over a decade since any HW or even GPU-based encoding quality/efficiency wasn't notably worse than the best available software encoder, and the gap has been increasing as codecs get more complex. With so many modes available, making lots of tight loops with low-latency feedback and early exits is key to performance, and the waterfall processing style of GPU and fixed-function implementations harms quality much more than the theoretical performance edge helps. And a modern CPU with lots of cores and power AVX2+ SIMD can do lots of GPU-style encoding all in the same L3 cache.

HW encoders get used when watts/encode is really constrained, hard realtime is required, or for applications where bitrates aren't particularly constrained, like mezzanine encoding.

Quote:
EposVox made a mistake by not choosing the same gop length in his test, Intel AV1 QSV has a 1s seeking in his samples, means he uses a gop length of 60 (videos are 60 fps). VCE, Intel h264, x264 have a 10s seeking in his encoded videos. As for Intel h264 or h265 QSV a gop increase from 60 to 600 gop should improve the VMAF score by roughly 1 point.
And can vary quite a bit with content type.

Quote:
Intel hardware AV1 performs really good at 3500 Kbit relative to the others in this test. It's a blocky mess in motion on NVENC or Intel h264. On higher bitrates Intel AV1 QSV detail preservation is quite poor, that's why it's losing relative to the others at higher bitrates.
I think we're looking at the graphs differently. I see QSV on 12900 HEVC beat all other contenders by a healthy margin in both 8-bit and 10-bit. It was also the slowest option, so not exactly apples-to-apples. For "quality" all HEVC and most H.264 options beat the Arc AV1; the one exception being 12900 H.264 FF. Some chunk of that is from the GOP duration mismatch, of course.

Or maybe I'm looking at the data weird. It seems ARC HEVC normal beats quality, which I would not expect.
__________________
Ben Waggoner
Principal Video Specialist, Amazon Prime Video

My Compression Book
benwaggoner is offline   Reply With Quote
Old 12th August 2022, 17:24   #2446  |  Link
Yups
Registered User
 
Join Date: Sep 2011
Posts: 362
Quote:
Originally Posted by benwaggoner View Post
I think we're looking at the graphs differently. I see QSV on 12900 HEVC beat all other contenders by a healthy margin in both 8-bit and 10-bit.


You just misread my posting. There is no HEVC comparison in the test made by EposVox. This part is not related to the rigaya test.

https://youtu.be/ctbTTRoqZsM
Yups is offline   Reply With Quote
Old 12th August 2022, 19:41   #2447  |  Link
rwill
Registered User
 
Join Date: Dec 2013
Posts: 341
Quote:
Originally Posted by Yups View Post
Intel's WORLD FIRST GPU AV1 encoder was worth the hype

https://youtu.be/ctbTTRoqZsM


Very good AV1 result from the Arc A380 in this test.
Are we looking at the same graphs ?

If I am looking at the ones at 11:50 I think, well, looks like Intel's GPU is getting beaten by a large margin compared to the Intel software encoder set to realtime mode. Also looks like H.264 cannot keep up at 3500k due to higher bitstream overhead but this normalizes out somewhat at 6000k. Only 3 sample points for a graph are suboptimal too, I'd have used 5+. Also keep in mind, H.264 is 20 years old and 1080p@60hz with 3500k comes out to ~1600k at 24hz for non motion blurred content.

Hardly what I would call 'worth the hype' but I made no flashy YouTube video about yet to deliver my opinion to my followers.
rwill is offline   Reply With Quote
Old 16th August 2022, 16:13   #2448  |  Link
GTPVHD
Registered User
 
Join Date: Mar 2008
Posts: 175
https://www.asrock.com/Graphics-Card...C/index.us.asp
https://www.newegg.com/asrock-arc-a3...82E16814930076

Intel Arc A380 will be available soon in US, hopefully someone can test the AV1 hardware encoder.

Last edited by GTPVHD; 8th October 2023 at 13:56.
GTPVHD is offline   Reply With Quote
Old 24th August 2022, 17:17   #2449  |  Link
GTPVHD
Registered User
 
Join Date: Mar 2008
Posts: 175
https://www.intel.com/content/www/us...ex-series.html
GTPVHD is offline   Reply With Quote
Old 25th August 2022, 21:09   #2450  |  Link
filler56789
SuperVirus
 
filler56789's Avatar
 
Join Date: Jun 2012
Location: Antarctic Japan
Posts: 1,351
I was bored so I built the SVT-AV1 encoder version 1.2.1.

NOTICE: running "svtav1encapp --version" returns

"SVT-AV1 v1.2.0 (release)"

But I tested the .EXE against a Version.avs file and it worked :-|

https://www.mediafire.com/file/h14la...1.2.1.rar/file
filler56789 is offline   Reply With Quote
Old 30th August 2022, 18:37   #2451  |  Link
benwaggoner
Moderator
 
Join Date: Jan 2006
Location: Portland, OR
Posts: 4,736
Quote:
Originally Posted by rwill View Post
Are we looking at the same graphs ?

If I am looking at the ones at 11:50 I think, well, looks like Intel's GPU is getting beaten by a large margin compared to the Intel software encoder set to realtime mode. Also looks like H.264 cannot keep up at 3500k due to higher bitstream overhead but this normalizes out somewhat at 6000k. Only 3 sample points for a graph are suboptimal too, I'd have used 5+.
Yeah, the graph looks impressive, but its utility goes down the more one tries to get applicable information out of it.

Quote:
Also keep in mind, H.264 is 20 years old and 1080p@60hz with 3500k comes out to ~1600k at 24hz for non motion blurred content.
Bitrate has a less-than-linear increase with frame rate, same as with frame size. With a higher frame rate less happens between frames, so individual frame predictions are more accurate. Also, any visual defect is visible for less time and so less noticeable. Also, as IDR placement is generally in seconds, not frames, higher frame rates mean a lower IDR percentage, which also improves efficiency.

Motion blur is a whole other matter. 60p stuff tends to have a 1/60th of a second shutter at the slowest, and can be much, much faster for daylight shoots. 24p, except for cell animation, almost always used a 1/48th of a second shutter. Generally more motion blur is helpful for encoding, although there are complexities that sometimes confound that.

My rule of thumb is that doubling the frame rate requires a 20-40% increase in bitrate, content dependent. If it's the same content (like encoding a 60p source at 60p and 30p), it's on the lower end of the range.
__________________
Ben Waggoner
Principal Video Specialist, Amazon Prime Video

My Compression Book
benwaggoner is offline   Reply With Quote
Old 8th September 2022, 20:12   #2452  |  Link
rwill
Registered User
 
Join Date: Dec 2013
Posts: 341
Quote:
Originally Posted by benwaggoner View Post
Motion blur is a whole other matter. 60p stuff tends to have a 1/60th of a second shutter at the slowest, and can be much, much faster for daylight shoots. 24p, except for cell animation, almost always used a 1/48th of a second shutter. Generally more motion blur is helpful for encoding, although there are complexities that sometimes confound that.

My rule of thumb is that doubling the frame rate requires a 20-40% increase in bitrate, content dependent. If it's the same content (like encoding a 60p source at 60p and 30p), it's on the lower end of the range.
Oops I did not see the reply.

I agree for normal shot content but the R/D graph I meant was for captured video game content. So most likely no motion blur + hard edges/contrast + small detail. My guess is that it requires a ~80% increase in rate to keep the same quality when doubling the frame rate. The only thing I can see it saving is motion vector length really.
rwill is offline   Reply With Quote
Old 10th September 2022, 11:21   #2453  |  Link
LigH
German doom9/Gleitz SuMo
 
LigH's Avatar
 
Join Date: Oct 2001
Location: Germany, rural Altmark
Posts: 6,746
New uploads: (MSYS2; MinGW32 / MinGW64: GCC 12.2.0)

AOM v3.4.0-397-gaeee77c48

rav1e 0.5.1_g25f6c0f Clang 14.0.6

dav1d 1.0.0-61-g5247319

SVT-AV1 v1.2.1-17-g470e1823

avif 0.10.1_6624e76
dav1d [dec]:1.0.0-61-g5247319, aom [enc/dec]:3.4.0-397-gaeee77c48, rav1e [enc]:0.5.0 (p20220906-3-g25f6c0f)
__________________

New German Gleitz board
MediaFire: x264 | x265 | VPx | AOM | Xvid
LigH is offline   Reply With Quote
Old 16th September 2022, 18:06   #2454  |  Link
benwaggoner
Moderator
 
Join Date: Jan 2006
Location: Portland, OR
Posts: 4,736
Quote:
Originally Posted by rwill View Post
Oops I did not see the reply.

I agree for normal shot content but the R/D graph I meant was for captured video game content. So most likely no motion blur + hard edges/contrast + small detail. My guess is that it requires a ~80% increase in rate to keep the same quality when doubling the frame rate. The only thing I can see it saving is motion vector length really.
That said, there can be a lot of static elements on-screen in a game, depending on the time. The HUD and GUI, for example. Also, since any temporal errors last only half as long when frame rate doubles, one can get away with somewhat higher QPs.
__________________
Ben Waggoner
Principal Video Specialist, Amazon Prime Video

My Compression Book
benwaggoner is offline   Reply With Quote
Old 16th September 2022, 18:09   #2455  |  Link
benwaggoner
Moderator
 
Join Date: Jan 2006
Location: Portland, OR
Posts: 4,736
I was a little surprised by how little discussion of AV1 there was at IBC this year. Support was mentioned here and there, but I certainly saw a lot more VVC demos, and discussion of real-world VVC products.

8K talk was also quite muted, and VR/AR wasn't as prominent either. Sustainability was talked about way more than ever before, as Europe faces a winter energy crisis due to Russia this winter.
__________________
Ben Waggoner
Principal Video Specialist, Amazon Prime Video

My Compression Book
benwaggoner is offline   Reply With Quote
Old 20th September 2022, 16:32   #2456  |  Link
GTPVHD
Registered User
 
Join Date: Mar 2008
Posts: 175
https://nvidianews.nvidia.com/news/n...-rtx-40-series

Quote:
Dual NVIDIA Encoders (NVENC) cut export times by up to half and feature AV1 support. The NVENC AV1 encode is being adopted by OBS, Blackmagic Design DaVinci Resolve, Discord and more.
https://www.nvidia.com/en-us/geforce...tent-creation/

Quote:
GeForce RTX 40 Series graphics cards feature our eighth generation NVIDIA video encoder, NVENC for short, now with support for AV1.

For livestreamers, the new AV1 encoder delivers 40% better efficiency. This means livestreams will appear as if bandwidth was increased by 40% — a big boost in image quality. AV1 also adds support for advanced features like high dynamic range.
https://developer.nvidia.com/nvidia-video-codec-sdk

Quote:
Introducing AV1 encoding with Video Codec SDK 12.0 on NVIDIA’s Ada architecture. AV1 is the state of the art video coding format that supports higher quality with better performance compared to H.264 and HEVC. On Ada, multiple NVENC coupled with AV1 enables encoding 8k video at 60fps alongside a higher number of concurrent sessions.
For livestreamers, the new AV1 encoder delivers 40% better efficiency. This means livestreams will appear as if bandwidth was increased by 40% — a big boost in image quality. AV1 also adds support for advanced features like high dynamic range.

Last edited by GTPVHD; 20th September 2022 at 18:07.
GTPVHD is offline   Reply With Quote
Old 20th September 2022, 18:13   #2457  |  Link
benwaggoner
Moderator
 
Join Date: Jan 2006
Location: Portland, OR
Posts: 4,736
Big news! Sounds like it has a lot of great stuff for gamers and content creators.

I wish claims like "For livestreamers, the new AV1 encoder delivers 40% better efficiency." would specify "compared to." ! If it is only 40% over AVC, that's a bit underwhelming. There's lots of good tools in AV1 for encoding game content that probably aren't being used by the hardware.
__________________
Ben Waggoner
Principal Video Specialist, Amazon Prime Video

My Compression Book
benwaggoner is offline   Reply With Quote
Old 20th September 2022, 20:50   #2458  |  Link
Yups
Registered User
 
Join Date: Sep 2011
Posts: 362
Quote:
For livestreamers, the new AV1 encoder delivers 40% better efficiency.

It sounds like it's compared to h264 because livestreamers are using/have to use h264.
Yups is offline   Reply With Quote
Old 21st September 2022, 06:49   #2459  |  Link
nevcairiel
Registered Developer
 
Join Date: Mar 2010
Location: Hamburg/Germany
Posts: 10,336
The 40% is a comparison against H264 at 1080p, since that what majority of streaming is using right now, and what they hope to push AV1 into, since HEVC never took off in that space.
Higher resolutions would of course benefit more, but due to H264 still being prevalent there, it was not used often. Maybe that'll change when modern codecs take over?

Of course streaming is a special use-case for encoding wanting extra low latency, which precludes some features, and hardware encoding also tends to avoid some features that are not well suited to hardware processing. But 40% improvements for streaming should still be quite nice.
__________________
LAV Filters - open source ffmpeg based media splitter and decoders
nevcairiel is offline   Reply With Quote
Old 22nd September 2022, 01:20   #2460  |  Link
benwaggoner
Moderator
 
Join Date: Jan 2006
Location: Portland, OR
Posts: 4,736
Quote:
Originally Posted by nevcairiel View Post
The 40% is a comparison against H264 at 1080p, since that what majority of streaming is using right now, and what they hope to push AV1 into, since HEVC never took off in that space.

Higher resolutions would of course benefit more, but due to H264 still being prevalent there, it was not used often. Maybe that'll change when modern codecs take over?
I think H.264 remained the standard as it was the one universally compatible codec. Using anything else for upload means the ingest stream needs to be reencoded for some customers. Which can be a good thing if the reencode still looks better due to a higher quality source. More advanced codecs have tools not in H.264 that can really help for gaming style content. Variable partition sizes is big. Transform Skip can be a big help for really sharp lines and text. Tiles to break HUD from 3D viewport for faster encoding. SAO in HEVC can be very effective in reducing ringing from sharp details at high QP.

Quote:
Of course streaming is a special use-case for encoding wanting extra low latency, which precludes some features, and hardware encoding also tends to avoid some features that are not well suited to hardware processing. But 40% improvements for streaming should still be quite nice.
And 40% would be just the start, I think.
__________________
Ben Waggoner
Principal Video Specialist, Amazon Prime Video

My Compression Book
benwaggoner is offline   Reply With Quote
Reply

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT +1. The time now is 05:33.


Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2024, vBulletin Solutions Inc.