LAV Filters - DirectShow Media Splitter and Decoders

[NikosD]

It could also be very helpful if someone with an Nvidia or AMD card with hybrid, not pure, HW HEVC decoder could test this clip to compare MS MFT vs LAV in order to check out if it's Intel specific problem or hybrid specifically.

[Manni]

I'm happy to test with my HD7870 and also with the AMD R9 Fury I've ordered which should arrive tomorrow.

I don't know exactly how the HD7870 deals with HEVC, ie if it's CPU only or hybrid (I'm using Crimson 15.1.1), but I too have noticed that MPC-BE_LAV+MadVR (or EVR-CP) struggle more with HEVC files than PowerDVD15. I expect the Fury to use hybrid for 10bits HEVC.

I've noticed that frames tend to drop more at every shot transition, or when a long shot has a slow pan (so a lot of info that changes in each picture).

For example, if you look at the Exodus HEVC HDR 10bits trailer, PDVD 15 plays it almost perfectly, while MPC-BE+LAV struggles at each shot transition and towards the end of the shot with the temple at night, which is a long pan shot. I've set the GPU and CPU to the max (22) in MadVR, it helps but it doesn't get rid of all the stutter.

[nevcairiel]

I don't think the HD7870 supports any kind of HEVC acceleration from the driver side, unless thats a relatively new feature.

[Manni]

I don't think the HD7870 supports any kind of HEVC acceleration from the driver side, unless thats a relatively new feature.

Thanks, that's what I thought. Then is it useful if I test with the upcoming AMD R9 Fury which definitely has hybrid acceleration for HEVC?

[nevcairiel]

I've just pushed VP9 DXVA2 support to LAV, which will be available in the next nightly build tomorrow.
Note that there is some caveats to know about:

- Its limited to 4:2:0 8-bit (Profile 0), like most DXVA2 accelerators
- Its been tested on Intel Braswell only. Skylake will probably also work. Older Intel systems may not.
- While NVIDIA pretends to support VP9 DXVA2 on the GTX960, it does not actually work (black screen). I'm somewhat confident that its a driver problem, since it fails in rather specific ways that it otherwise shouldn't.
- When using DXVA2-Native and playing an unsupported file (4:2:2/4:4.4 or > 8-bits), you may get a black screen, since fallback to software may not catch the unsupported format early enough (Copy-Back does not have this issue, since it can fallback to software at any time)

Due to the limited amount of hardware that can actually make use of this, testing was also limited, so keep that in mind.
Its also disabled by default, due to all this.

[Thunderbolt8]

is it possible to manually update dcadec in LAV filters? I cannot find the .dll file in the installed directories. also, for some reason the DTS-HD bitstreaming field of LAV audio is greyed out for me.

[clsid]

is it possible to manually update dcadec in LAV filters? I cannot find the .dll file in the installed directories. also, for some reason the DTS-HD bitstreaming field of LAV audio is greyed out for me.
DTS-HD gets activated when you enable DTS.

[nevcairiel]

is it possible to manually update dcadec in LAV filters?

LAV is entirely open-source, you can rebuild it with a newer dcadec.
Other kinds of updates are not available.

Note that todays nightly contains the very latest version.

[Thunderbolt8]

DTS-HD gets activated when you enable DTS.right, completely forgot about that -.-

[Nullack]

G'day Hendrik. Very nice to see your commits into FFMPEG in the last few days with VP9 and now on my birthday, a nice little surprise with LAV filters nightly having VP9. I'm on GTX 960 and given your insights into nvidia's drivers I had not expected VP9 dxva2 to work currently. This is indeed the case sir, when playing back VP9 like phfx_4KHD_VP9TestFootage.webm. Using MPC_HC the app exits and will not display. Using MPC-BE the app tries to play, but stops immediately without exiting. Either case, no playback. Im on driver build revision 359.06. Cheers

EDIT: Also the tick in the VP9 checkbox to enable it, in the LAV Video Decoder app, that isnt persisting. I dont know if this is a feature where LAV auto unticks it cos it knows it failed, or if it's a bug where the GUI isnt remembering the user wanting it enabled.

[nevcairiel]

Please try to keep hardware benchmarks separate from this thread (NikosD has a thread dedicated to such things), or at the very least just post numbers and not screen filling screenshots.
Also keep in mind that Braswell is an Atom-CPU, so hybrid decoding of 4K is probably just too much for it.

[sirhaden]

Does the video decoder support decoding faster than real time? It seems that the default settings result in decoded samples being provided based on the sample duration.

[nevcairiel]

Does the video decoder support decoding faster than real time? It seems that the default settings result in decoded samples being provided based on the sample duration.

Content is output as fast as any downstream filter will receive it. Video and Audio renderers of course only receive content in real-time, since they are designed for that.

LAV Audio and LAV Video can deliver content as fast as your CPU can decode it (or your drive read it)

[Manni]

It could also be very helpful if someone with an Nvidia or AMD card with hybrid, not pure, HW HEVC decoder could test this clip to compare MS MFT vs LAV in order to check out if it's Intel specific problem or hybrid specifically.

I have an AMD R9 Fury tonight, it's leaving tomorrow.

Here is what it can do: https://dl.dropboxusercontent.com/u/7097546/LAV/AMD%20R9%20Fury%20DXVA%20Checker.png

When I run the decoding test with LAV set to DXVA copyback, here is what I get:
https://dl.dropboxusercontent.com/u/7097546/LAV/Test%20Fitness%20LAV%20HEVC%20Main.png

Not sure I'm doing it right, first time I'm running these kind of tests.

HEVC Main acceleration seems to work fine, but it looks like it's hardware, so not sure it's what you're after. I have no hybrid acceleration for any Main 10 file, but that might be the way I've set it up.

I can do more tests if you let know what you want me to do tonight, otherwise the GPU is going back tomorrow.

[Manni]

Just to add that I've done some tests with PDVD15 and I seem to get with it some hybrid acceleration for Main10 that I haven't been able to get from LAV.

With PDVD, the CPU is at 50% for my HDR test files of Exodus and even the more demanding Life of Pi one which it plays almost perfectly.
With LAV, CPU is close to 100% and I get dropped frames during transitions with Exodus and unplayable with Life of Pi.

I've set up LAV with DXVA2 copyback (tried native too) and renderer set to EVR-CP.

Maybe it doesn't like 10bits files because it works fine with the fitness test file and many other 8bits files.

[NikosD]

Fury has indeed a pure HW HEVC decoder, not a hybrid one, so it's not the case.

Regarding 10bit HEVC, Fury should have no HW support hybrid or pure HW acceleration.

The cpu usage of 50% of PowerDVD and 100% of LAV is not easy to explain.

DXVA2 copy-back or native with LAV has no difference for 10bit HEVC because they are not actually used with Fury, it falls back to SW.

Try "None" and see of there is any difference.

[Manni]

Fury has indeed a pure HW HEVC decoder, not a hybrid one, so it's not the case.

Regarding 10bit HEVC, Fury should have no HW support hybrid or pure HW acceleration.

The cpu usage of 50% of PowerDVD and 100% of LAV is not easy to explain.

DXVA2 copy-back or native with LAV has no difference for 10bit HEVC because they are not actually used with Fury, it falls back to SW.

Try "None" and see of there is any difference.

Thanks for this, it confirms what I thought. Sorry I couldn't help more.

I knew that the Fury had no hardware acceleration for Main10, but I was hoping for hybrid. As it's not the case - except for some reason with PDVD it seems, which doesn't really matter to me as I'm more interested in MPC-BE+LAV+MadVR performance - I decided last night to keep my HD7870 until Arctic Islands.

As I was only trying to provide the requested data and as the Fury is not relevant to your issue, I won't do more tests, the GPU is packed and going back today :).

[nevcairiel]

PDVD15 implements its own hybrid codec using OpenCL I think.

[Manni]

PDVD15 implements its own hybrid codec using OpenCL I think.

Ah thanks (I couldn't see your post in the other thread).

That explains it then :).

They did a pretty good job, the Main10 videos are anything from stuttering to unplayable without acceleration on my 3770K and play almost perfectly on PDVD with just 50% CPU use (up to about 50mbps).

I wonder how it would have dealt with NikosD's high bandwidth files (crowd run). I couldn't test as the bandwidth is too limited on HDMI 1.4 without the Club3D adapter as the files are 50p.

The best I could do without was 30p.

[huhn]

just set the display to 1080p60 for such a test.

[sirhaden]

Content is output as fast as any downstream filter will receive it.

I have 2 video streams in a proprietary container that get demuxed in a custom demuxer and sent downstream to 2 separate instances of the LAV video decoder. One of the video streams is 30 fps and the second is 5 fps. The decoded streams are sent downstream to a custom picture-in-picture filter where synchronized PIP is performed before outputting a single image for rendering.

The problem is that the 5 fps stream arrives at the PIP filter out of sync, behind the 30 fps stream. A short delay is applied to the faster stream to allow the slower stream to catch up.

What is happening is that the slower stream's packets are not arriving fast enough from the LAV video decoder in order to resync the streams. Using an older version of ffdshow in the same exact architecture does not exhibit this behavior and the slower stream's packets are decoded and delivered fast enough to sync the streams and perform PIP.

Could this be due to the use of 2 instances of the LAV video decoder? Are there any known issues running multiple instances of the LAV video decoder?

[nevcairiel]

LAV Video does not implement any rate limiting at all. Like I outlined in my earlier post, it'll decode content as fast as it gets it from the source and the renderer accepts it.
Note that decoding can have a delay based on number of threads used and some codec properties, so you may allow for a bit more buffering to sync a low fps stream up properly to account for the delay.

There are no known issues running multiple instances at the same time.

[Manni]

just set the display to 1080p60 for such a test.

I'm only interested in UHD HEVC Main10 decoding, so have no interest in testing anything at 1080p.

[nevcairiel]

The resolution or refresh rate is irrelevant if you just want to test *decoding* :)

[Manni]

The resolution or refresh rate is irrelevant if you just want to test *decoding* :)

You're right, I should have said I'm interested in playback performance at native resolution, so downscaling to a lower resolution doesn't really help to assess that.

Plus it's only with PowerDVD, and I care even less about that :)

[LigH]

To (more or less) "benchmark" AviSynth decoders, AVSmeter is quite useful. If DirectShow doesn't have similar tools, one may as well use it with a DirectShowSource script, despite a little overhead.

[NikosD]

DXVA checker can measure both DirectShow and MediaFoundation decoders, in both pure decoding mode (without renderer) and in playback mode using EVR at arbitrary resolutions.

[dansrfe]

How is the appropriate LFE mix level determined for a stereo output to satellite speakers which separates and routes low frequencies to a small sub?

[foxyshadis]

How is the appropriate LFE mix level determined for a stereo output to satellite speakers which separates and routes low frequencies to a small sub?

Your sound card or speakers handles all of that. LAV doesn't do upmixing.

[Nullack]

Something unexpected is happening with VP9 playback on my Nvidia GTX 960

When testing this file phfx_4KHD_VP9TestFootage.webm I had expected to see LAV Filters nightly x64 slam the CPU since I have both VP9 DXVA 2 decoding disabled in the latest nightly of LAV filters, but also because AFAIK the Nvidia VP9 hybrid mode isnt working in current drivers. What I found in MPC-BE is that playing that test 4K VP9 file back, even though the renderer said that it wasnt using DVXA, I was seeing 48% GPU utilisation. I dont know how the GPU got so busy in assisting the playback of the VP9 file. The end result was that the 4K VP9 video played back smooth as silk.

[nevcairiel]

The VP9 software decoder is pretty well optimized, I wouldn't be surprised if it just managed to decode it on its own. Stuff doesn't magically get accelerated even though its all turned off.
The GPU usage was likely just from processing the video.

[Nullack]

Thanks. I wonder what processing it is doing then on the GPU? Its progressive so no deinterlacing needed. Maybe its some sort of colour conversion but I dunno cos I run a HTPC TV display in YCbCr 4:2:0 so I thought it just got all pushed through without fiddling for most content. And the source video is 4K, my display is 4K, so image scaling doesnt seem necessary. I tried to use mediainfo to see the source colour infos and so on but its not listing that information. Maybe the source colour is something strange. Still using a whole 48% of my GPU seems allot, thats like what I see with HEVC L52 DXVA 2 decoding.

[nevcairiel]

PCs don't operate in YCbCr, there will generally always be a RGB step, which is why YCbCr output is generally flawed.
So in your case, there is actually two conversions then, YCbCr from the video -> RGB in the renderer -> YCbCr for output.

[Manni]

PCs don't operate in YCbCr, there will generally always be a RGB step, which is why YCbCr output is generally flawed.
So in your case, there is actually two conversions then, YCbCr from the video -> RGB in the renderer -> YCbCr for output.

True, but if you send RGB to the display most of the time (at least with TVs and Projectors, not necessarily with PC Monitors) there will be an internal conversion to YCB first to apply controls like color and tint which can't be applied in RGB, then another conversion to RGB internally.

The best way to test for this is to see if you can still change color and tint on the display when sending RGB. If you can, then the display is doing an RGB > YCB > RGB internal conversion.

In that case sending YCB to the display means there is one less RGB > YCB conversion.

If color/tint are locked, then the display accepts RGB directly without any further internal conversion, and it can be better to send RGB from the PC as in that case you do get less conversions.

So it's usually a good idea to test what works best for each display.

[wanezhiling]

http://i.imgur.com/S4NkVJq.png
Hi nevcairiel, I suggest adding a 'Restore Default Settings' option in Start Menu, how about?

[nevcairiel]

In that case sending YCB to the display means there is one less RGB > YCB conversion

Not really, you just do the conversion in the GPU instead of the display, still same amount of conversions. I'm not sure which would be of better quality.
In the case of YCbCr output from the GPU you have to hope that the conversion matrix and the inverse matrix in the TV are the same. If its all done inside the TV, you could probably safely assume that its the same.

[Isochroma]

Still no fix for the Doom9 fiefdom's censors, and still no fix or word from the author of buggy LAV Filters which won't install LAVvideo.ax on WinXP SP2.

It errors out instead, complaining about being unable to register the file.

Turns out - after I wasted half my precious afternoon investigating due to complete lack of documentation or software explanation - that the file is dependent on a certain function in Windows implemented in SP3+. That function is called GetLogicalProcessorInformation().

GetLogicalProcessorInformation() is un-necessary for the work performed by LAVvideo.ax. Instead it is another useless dependency designed to force users to change their entire operating system and undocumented on either of the author's websites and not checked for by the installer during installation, thus leaving users with a cryptic error message at the end and a nonfunctional video decoder.

There is no explanation on either of the software's websites and no specification of OS requirements.

The software does no checking during installation and instead trips all over itself finally falling apart on the last mile.

It's time the author took responsibility for his poor software coding and fixed it.

If he doesn't then he will never hear the end of this issue on a certain other site.

[davidsama]

I am pretty sure nevcairiel has said that he no longer supports windows xp or xp sp2 or xp sp3. Just because you refuse to upgrade to windows 7 does not mean that you can try to force other people to still support your 20+ year old os.

[Sparktank]

It is open source, afterall.

Nothing stopping anyone from forking the source to emphasise support for XP (esp. outdated service packs).

Surely, there must be some site out there that specializes in updates for XP ( <SP2 ) compatibility.

[manolito]

The current LAV filters install and work nicely under WinXP SP3. I still am a big fan of XP (but of course I also have a Win7 installation for the ever growing number of software which does not work under XP).

But I really don't see any good reason not to install SP3 under XP. At least for me and everyone I know the SP3 does not break anything.

@ Isochroma
You sure could use a little improvement of your netiquette...


Cheers
manolito

[foxyshadis]

This has to be one of the most insane demands I've ever heard, but at least you aren't cursing out the people trying to tell you it won't happen this time. When any software says it supports an OS without mentioning service packs, that always means it supports at least the latest service pack, but maybe not anything before it. That's obvious and expected; who would bother to keep around all previous versions, if they could even get hold of them, for the two people in the world who would ever try to install it that way? LAV can't possibly be the only thing that doesn't work on SP2.

[nevcairiel]

As others have mentioned XP SP3 still works fine, however any earlier versions are not supported, as they lack key Windows functions that LAV uses.

[Nullack]

Hi. Ive done some research but Im still confused about LAV Video configuration settings. Help would be great please.

Background: Only today did I realise my Samsung UHD 65" display has a special mode that I could have configured ages ago to do more than 2160P60hz with YUV 420. In this special UHD colour mode I can now do lots more settings. Two cases of note:

Case 1:
3840x2160P 60Hz
Output colour format: RGB
Output colour depth: 8bpc
Output dynamic range: Full 0 - 255

Case 2:
YCbCr444 but dynamic range goes to limited without beeing able to adjust it and 8Bpc stays the same

I assume case 1 would be better as a config? Since I get full dynamic range in RGB?

Now onto LAV Video specific settings.

RGB Output levels (for YUV -> RGB Conversion): I have it set currently to PC 0-255? Or should it be untouched?

And output formats. At the moment I have just RGB32 enabled. I'm confused though since 8Bpc on each R G B would be 24. For some reason the TV displays RGB32 output even though the nvidia control panel says it can only do 8Bpc. Should RGB24 be enabled, and RGB32 turned off, or both on as well? And given 8Bpc I think RGB48 should be disabled?

[nevcairiel]

You should generally not have LAV convert video to RGB, especially on UHD content its just going to be rather slow because it uses the CPU for that. Let your GPU do the conversion, its going to do it anyway, much faster at that.
RGB32 and RGB24 are both 8-bit RGB, the reason for 32 is that having data be aligned on powers of 2 is more efficient to work on, so it gets 8-bit of unused data added.

[Nullack]

Ok thanks. Ive set YUV to RGB as leave untouched and stuck with LAV video just having RGB32 as the only ticked output. As I assume that since Im outputting RGB 8Bpc full dynamic range to the display, thats what the display wants to see. Im trying to understand your response, I assume case 1 config is the right config to use for the display and hence I should set lav to just do RGB32 as the only output?

With playing back:

4K HEVC 10 Bit P010: Mixer Output RGB32, Input Type P010 : all seems OK playing this stuff. But then on this:

4K HEVC YUV 420: Input type NV12, Mixer Output NV12 Type NV12

So is that a bug? I told LAV to use RGB32 as the only allowed output yet it says its outputting NV12 for 8 bit HEVC?

Sorry if this is not a bug and its complex than I realise - Im trying to learn this stuff and Im a bit confused. :)

[nevcairiel]

If you use DXVA2-Native, then the output settings have no effect.
And you didn't seem to understand my response, since I said its better to not try to output RGB from LAV. :) The best quality and performance is usually achieved by not disabling any of the output formats and just letting the renderer handle the necessary conversions.
The options are only provided for testing and for renderers which have broken conversions - but many of the conversions in LAV are relatively slow. The CPU is just much slower at processing raw image data than a GPU would be.

[Nullack]

Thanks. So in DXVA 2 native none of the output settings matter. So even if I wanted to do something not recommended like get LAV to do RGB conversion, since Im in native mode it wont ever happen.

Since none of the settings will be obeyed, can I politely suggest it would be less confusing for the user just to disable that whole section of the GUI by greying it out when native is set? :)

I therefore assume when in DXVA native mode, the output mode
is "whatever it is" that gets forced in the whole stack and the user has no choice or decision to make.

[nevcairiel]

Since none of the settings will be obeyed, can I politely suggest it would be less confusing for the user just to disable that whole section of the GUI by greying it out when native is set?

Not every video can be decoded by DXVA2, since DXVA2 only supports a small range of video codecs.
Therefor disabling the settings makes no sense.

[Nullack]

Oh yes, I see ofcourse thats a global setting whereas for example some video codecs will go back to software. Thanks, you really are the master of this stuff :) I appreciate all the insights.

[Nullack]

Hi. Below in this thread is a description of a bug I had assumed to be a MPC-BE / MPC-HC problem. One of the MPC-BE Developers is saying I should raise it for the LAV Filters project. Basically on all three configuration tools for LAV, it works fine in high DPI and 4K situations when the three config tools are run by themselves. WHen however MPC-BE or MPC-HC is playing a video, and I right click into filters then one of the LAV configurators, the formats tab is all garbled up. Screenshots in the below link:

http://forum.doom9.org/showthread.php?p=1749525#post1749525

[Plutotype]

Hi folks,
Do LAVfilters support DCP package playback? xml and vid.mxf and aud.mxf
Thanks
Pluto

[lowenz]

Hello guys!

I enjoy testing LAV internal and other filters for HEVC playback but I can't find a way to disable the GPU acceleration of my 750 Ti (Maxwell 1): it seems the GPU drivers override every other options in MPC-HC / LAV if I use "modern" renderers.

Any suggestions?

[Vasilich]

with LAV you can set it to use CPU only when using decoder type "avcodec".
what do you mean under "modern renderers"? renderer can also use GPU for effects, so refer to wiki/forum for that particular renderers you use how to disable GPU usage by them.

[lowenz]

what do you mean under "modern renderers"? renderer can also use GPU for effects, so refer to wiki/forum for that particular renderers you use how to disable GPU usage by them.
VMR 7,9, EVR (and Sync).

0% GPU usage with a Radeon 7850 (windows 7), high GPU usage with the 750 Ti (windows 10)

[lowenz]

with LAV you can set it to use CPU only when using decoder type "avcodec".
HW acceleration is already disabled in MPC-HC/Lav configuration but the GPU (Nvidia Maxwell) usage is still high (vs AMD GCN @0%, same settings)

[wanezhiling]

Because the player can't detect the gpu usage on ATI/intel platform, so always 0%, that's not correct.
With nvidia, it's normal, you can check GPU-Z -> GPU Load.

[lowenz]

Because the player can't detect the gpu usage on ATI/intel platform, so always 0%, that's not correct.
With nvidia, it's normal, you can check GPU-Z -> GPU Load.
It's 0% with RTSS.

[nevcairiel]

Your stats are just wrong, then. Video Rendering will always take a bit of GPU load.
However, if hwaccel is turned off in LAV, then it will not touch the GPU at all. Anything that happens after LAV has decoded the image is entirely out of its hands, so not relevant to this thread.

[Nullack]

Bug Report: Forced subtitles are not appearing despite being enabled in LAV splitter. During playback LAV splitter identifies the auto subtitle stream for forced subtitles. However no forced subs are displayed such as in star wars episode 1 bluray disc chapter 14. Normal PGS type subtitles display fine.

I see the help text in LAV Splitter has a note saying the tick box for forced subtitles may not work for all bluray discs. Is there a way to fix this? Can I assist with testing? What is the issue causing some blurays to not work with forced subtitles.

Thanks

[sneaker_ger]

PGS subtitles in BluRay can mark certain subtitle pictures (not tracks!) as "forced". So you can have one track that has the complete English + alien dialogue of Star Wars in a single track. LAV can filter the non-alien parts out because they are not marked as forced if the user wishes so. Some authoring houses choose to not mark any subtitle pictures as "forced" at all, instead they just put the full dialogue (including the alien parts) in one track and the alien parts into another track. Since no picture is marked as forced LAV will filter everything out. Your only options:
1.) deactivate the respective LAV option or
2.) choose the respective full subtitle track (not the auto forced-only one) or
3.) re-author the file with some subtitle editor to make use of the flags

You might be forced to do 3.) in case you have not copied the track that only consists of the alien language parts from the BluRay. (or re-rip the BluRay, of course)

The problem is well known - there's nothing you can assist with.

[Nullack]

Thanks for the details. However this doesnt stand up to logical reason to say it cant be done programatically. Since the Xbox One can do it using the AMD x86 CPU thats in it, clearly it can be done in software. Also Cyberlinks PowerDVD does it, again some software compiled for the x86 platform. I assume the problem is the open source community was never given licences from the Bluray folk and hence all the open source stuff was clean room reverse engineered? And theres gaps in whats been discovered from that reverse engineering vs the actual spec?

[nevcairiel]

The information is just not present, the only way to do this is to use the full menu navigation, which is obviously not the goal here.

[foxyshadis]

The only way I can see a player accounting for that is by silently switching to a forced-only subtitle track if forced is selected and no forced subpictures are in the main track. Then you have to guess which track if none of them are marked forced either...

[nevcairiel]

and no forced subpictures are in the main track.

Unfortunately, you don't know that until the end of the movie. ;)

[jkauff]

@nevcairiel
You're working for JRiver now, I'm curious why the Red October version of LAV is always behind the LAV standalone released version. Is there a technical reason? I have LAV set up as external filters, which seems silly given how often MC21 is updated. Just curious.

I didn't post this in the JRiver forums because I didn't want to take the chance that I would embarrass anyone.

[Stereodude]

The information is just not present, the only way to do this is to use the full menu navigation, which is obviously not the goal here.
And yet, nearly everyone still insists there's no compelling use case to implement menu support via libbluray into players. The fact that none of the open source players (potentially excluding VLC) can properly play back many intact Blu-rays (after decryption) without requiring the user to manually futz with subtitles during playback somehow gets overlooked. :scared:

I know I just love having to pause a movie to go figure out if there are forced subtitles (by flagging or a VM command to pick a certain PGS track) in one of the other PGS subtitle tracks that I should be seeing when I hear foreign languages being spoken but aren't seeing any subtitles on the screen. Apparently only the end users, and none of the developers, are actually trying to play back intact decrypted Blu-rays. :confused:

[nevcairiel]

You are free to wish for any feature you want, but its not my responsibility to fulfill your wishes. I do perfectly fine without BD menus, in fact I would never use them even if I could. And yes, I do watch plenty Blu-rays, I just make them MKVs first. My remote control can't go to the PC and put a disc in, after all.

[Stereodude]

You are free to wish for any feature you want, but its not my responsibility to fulfill your wishes. I do perfectly fine without BD menus, in fact I would never use them even if I could.
No one said it was. That doesn't change the validity of the use case. The open source community is happy with half-ass Blu-ray playback. That point became crystal clear a long time ago. Unfortunately, there's no decent commercial software player. Arcsoft was the closest and they've since given up.

And yes, I do watch plenty Blu-rays, I just make them MKVs first. My remote control can't go to the PC and put a disc in, after all.
ROFL... Since when is watching .mkv files with the streams extracted from Blu-rays "watching a Blu-ray"? Am I listening to CDs when I play FLAC files in foobar?

[nevcairiel]

ROFL... Since when is watching .mkv files with the streams extracted from Blu-rays "watching a Blu-ray"? Am I listening to CDs when I play FLAC files in foobar?

Its the same end-result. I buy a disc, and I watch the movie in the evening, without any frustration during watching time.

[Stereodude]

Its the same end-result. I buy a disc, and I watch the movie in the evening, without any frustration during watching time.
Yes, because in the afternoon you spent an hour of your time to rip it, figure out which tracks should be included, and mux a mkv of it.

[nevcairiel]

Yes, because in the afternoon you spent an hour of your time to rip it, figure out which tracks should be included, and mux a mkv of it.

Its more like 5 minutes (10 max on stubborn discs) of unpacking the disc and starting the process and setting up the streams, and 20 for it to finish ripping, but that part is unattended of course.
I rather spent that time while I am working on the PC anyway, rather than when I want to just relax and watch the movie.

Anyway, enough of this discourse. LAV isn't going to offer BD menus now or any foreseeable future, but any other developer is of course free to build this!

[dansrfe]

Your sound card or speakers handles all of that. LAV doesn't do upmixing.

Actually I meant just downmixing to stereo with LFE frequencies. Do I just find the optimal LFE mix level by trial and error in the case where the speaker system "upmixes" 2.0 -> 2.1? Is 1 the max LFE level of the source?

[AlexKane]

Actually I meant just downmixing to stereo with LFE frequencies. Do I just find the optimal LFE mix level by trial and error in the case where the speaker system "upmixes" 2.0 -> 2.1? Is 1 the max LFE level of the source?

1 means no volume scaling.

[videonerd]

And yet, nearly everyone still insists there's no compelling use case to implement menu support via libbluray into players. The fact that none of the open source players (potentially excluding VLC) can properly play back many intact Blu-rays (after decryption) without requiring the user to manually futz with subtitles during playback somehow gets overlooked. :scared:

I know I just love having to pause a movie to go figure out if there are forced subtitles (by flagging or a VM command to pick a certain PGS track) in one of the other PGS subtitle tracks that I should be seeing when I hear foreign languages being spoken but aren't seeing any subtitles on the screen. Apparently only the end users, and none of the developers, are actually trying to play back intact decrypted Blu-rays. :confused:

Hi Stereodude, do not fool yourself to think that nevcairiel can actually add any features to lav filters (which is just a hack using ffmpeg). You are more likely to get that feature developed in lav filters if you chase the ffmpeg devs.

[sneaker_ger]

Nevcairiel is one of the most active ffmpeg developers. Don't mistake lack of interest for lack of ability.

https://github.com/FFmpeg/FFmpeg/commits?author=Nevcairiel

[SeeMoreDigital]

Hi Stereodude, do not fool yourself to think that nevcairiel can actually add any features to lav filters (which is just a hack using ffmpeg). You are more likely to get that feature developed in lav filters if you chase the ffmpeg devs.
How rude :eek:

[e-t172]

Actually I meant just downmixing to stereo with LFE frequencies. Do I just find the optimal LFE mix level by trial and error in the case where the speaker system "upmixes" 2.0 -> 2.1? Is 1 the max LFE level of the source?

The correct value for LFE when downmixing to two stereo channels is +10dB - 6dB = +4dB = 1.58.

The +10dB part is because of ITU-R BS.775 (https://www.itu.int/rec/R-REC-BS.775/en) which states that LFE should be played back 10dB louder than any of the main channels.

The -6dB part is to compensate for the fact that you're mixing LFE into two channels, not one. The reason it's 6dB and not 3dB is because, at these low frequencies, acoustic summation is typically coherent.

I believe the reason why nevcairiel did not make 1.58 the default is to avoid overloading stereo systems that don't have a lot of bass capability.

[ryrynz]

How rude :eek:

Obviously had no idea, was appropriately schooled.

[filler56789]

You are more likely to get that feature developed in lav filters if you chase the ffmpeg devs.

Unlikely. Hendrik Leppkes is good, whereas many /¿most? of the "regular FFmpeg-ers" are just bureaucrats.

[Nullack]

Hi Stereodude, do not fool yourself to think that nevcairiel can actually add any features to lav filters (which is just a hack using ffmpeg). You are more likely to get that feature developed in lav filters if you chase the ffmpeg devs.

Hendrik is an ffmpeg developer, look at his commits in the ffmpeg git log! It aint right to diss him in the way you have! I dont agree with his views on bluray menu playback where he considers it to not be important for the users, but the bloke has done a masterful job not only on lav filters, but also ffmpeg. Were lucky we benefit from his efforts. You should consider apologizing.

[Nullack]

Yes it is, including 4K. Smooth as silk with little CPU utilisation. I did notice a strange value for the DXVA2 variable - I dont usually see that its more descriptive. As seen in the screenshot below:

http://i.imgur.com/nT2vCe0.jpg

[huhn]

Does the latest Nvidia 361.43 WHQL driver released today fix anything with VP9 DXVA2 decoding on GM206 GPUs?

yes and it isn't hybrid.

[Nullack]

For Hendrik:

* I see the Microsoft DXVA2 Spec Paper for VP9 is bundled into VP8 as well. Is there any future intent to add VP8 hardware assist in LAV filters since the two versions seem kinda intertwined? Currently VP8 is done in software.

* I also see on Microsofts DXVA2 Spec Paper website, MPEG 4 part two specification documents. Does LAV Filters intend to support stuff like xvid gpu acceleration?

Thanks

[nevcairiel]

VP8 might happen when hardware starts supporting this, no plans for mpeg4pt2, its too much of a butchered and misused format that would only result in trouble.

[v0lt]

The sample (http://www.mediafire.com/watch/9c7uq38m9b9ixzj/sample_HEVC_1080p_20sec.mkv) has artifacts when HEVC DXVA (Nvidia GTX 950).

[pacuro]

Hi guys,
I do not know what I am doing wrong or maybe what still have to do.
Each time I enable VP9 decoding in LAV Video Decoder Properties tab, "tick" disappears when playing VP9 sample and software encoding is engaged.
Doing this with latest nightly MPC-HC and LAV (both portable) and can not obtain VP9 DXVA playback.:confused: Of course 361.43 drivers are installed.
AVC and HEVC DXVA still works though.

[huhn]

you are most likely using the internal lavfilter version from MPC-HC not the new installed version for playback.

there are multiply ways to change this. one is to disable all internal filter in MPC-HC.

[pacuro]

Thanks huhn. Chose external filter and it works now!
Could you please tell me what is the difference since both (installed in Windows and portable in MPC-HC filters catalog) LAV versions are the same (0.67.0.29). I prefer portable "things" over those with instalators. Portables work fine with AVC and HEVC...

Edit:
Quite interesting as I removed LAV from Windows and choose in external codecs portable one stored in MPC-HC directory. Works as well :)
Is there anything what should be fixed by developers in MPC-HC or LAV ?

[hubblec4]

hi nevcairiel

I have some questions about nested chapters.
I will use ordered chapters too, cause its necessary for chapter segment linking
(which will work perfect, thank you so much).

It seems to me that LAV Splitter use the "main chapter" start and end time and the segment uid only.

1. Which parameter of a nested chapter is used in LAV Splitter?

The duration of an ordered-nested-chapter seems not to be used. Only the start time marks the chapter.
Also the segment uid will be ignored and an existing chapter-segment-link will not work.

I have uploaded a sample folder (http://forum.gleitz.info/attachment.php?attachmentid=98871&d=1450828621) with test files.
There are 3 episodes(separate folder) and a menu folder(all episodes together with two menu-"start".mkv's)
Epsiode 1 has an opening part(which will used for the other two episodes with chapter-segment-linking),
2 chapter parts and a credits part.
The other two episodes has 2 chapter parts and credits part only.
(Note: Only in the menu folder plays the episode 2 and 3 correct with the opening part from episode 1.)

The "without nested chapter menu.mkv" play all chapter-segment-links well and all durations correct.
The nested menu.mkv shows the opening part from episode 1 only one times.
The first nested chapter from episode 2 and 3 has a chapter-segement-link to episode 1 opening part
(00:00:00.000-00:00:02.000).

LAV Splitter menu -> chapters:
You see the episode 2/3 "main chapter", the first and the second nested chapter have the same time stamps.

2. Is this a bug or a not implemented feature?

Nested chapters are not really good documented and I think you has programmed this so good you knew.

3. Or could you find detailed specs?


I hope you have time for this topic.

hubble

[huhn]

Thanks huhn. Chose external filter and it works now!
Could you please tell me what is the difference since both (installed in Windows and portable in MPC-HC filters catalog) LAV versions are the same (0.67.0.29). I prefer portable "things" over those with instalators. Portables work fine with AVC and HEVC...

Edit:
Quite interesting as I removed LAV from Windows and choose in external codecs portable one stored in MPC-HC directory. Works as well :)
Is there anything what should be fixed by developers in MPC-HC or LAV ?

mpc-hc will update the lavfilter version sometime in the feature that will fix it. they are simply "outdated".

overriding may work too.

[Nullack]

The sample (http://www.mediafire.com/watch/9c7uq38m9b9ixzj/sample_HEVC_1080p_20sec.mkv) has artifacts when HEVC DXVA (Nvidia GTX 950).

Yes I too get artifacts on my 960 under copyback DXVA2. In software decoding there is no artifacts.

[foxyshadis]

Is there anything what should be fixed by developers in MPC-HC or LAV ?

MPC-HC only uses official releases, so it'll be fixed shortly after nev releases a new stable version.

[kasper93]

mpc-hc will update the lavfilter version sometime in the feature that will fix it. they are simply "outdated".

overriding may work too.

We are not outdated atm.

[NikosD]

You and nevcairiel should be working closer and update your software synchronised.

[ryrynz]

You and nevcairiel should be working closer and update your software synchronised.

LAV was updated 10 days ago, MPC-HC updated LAV 10 days ago (0.67.30) The MPC-HC devs will follow LAV and get email notifications when new changes
are made so you really can't really get any closer than that unless Nevcairiel updated MPC-HC himself. He could do this if he felt like it (he has rights to do so I believe)
but I doubt this will ever happen given he's not really associated with MPC-HC's development.

If you want bleeding edge then you use the external filters, I can't see anything changing in this regard.

[huhn]

We are not outdated arm.

had no better word for it so i put it in "". doesn't change that his version wasn't the nightly.

[mark0077]

Guys I'm using LAV Splitter / Audio / Video with mpc-hc latest beta. Just reformatted my PC and I'm not sure if my issue was there before or not but its there now. When I play an mkv file I have here which has a default first audio track of PCM s 24 LE, the LAV Audio Decoder isn't showing up in the filters list. When I switch tracks to the 5.1 track, LAV Audio Decoder kicks in. The strange thing is, when I switch back to the 2 channel track, LAV Audio Decoder is still in use.

Whats causing this, is it MPC or LAV Splitter or something else? Just that I use LAV Audio delay functionality as my TV has quite a bad video delay, so its noticeable that LAV Audio isn't in use as the audio is all out of sync.

[clsid]

PCM is uncompressed audio and doesn't require any decoding. The reason why LAV gets used after track switching is because (by default) the decoder isn't removed from the filter graph during switching.

The audio switcher in MPC-HC also has a delay option that you could use instead.

[mark0077]

Ah OK thanks. Its just not ideal as things like upmixing / audio delay all need to be replicated in the likes of mpc-hc then. Is there any other way clsid to force LAV Audio to be used so that I don't have to get those other LAV features working elsewhere for PCM?

[clsid]

Delay works independently of mixing. And while LAV can increase the number of channels, it doesn't actually upmix, it just adds silent channels.

If that PCM track is stereo and you want 5.1, then simply set Windows to 5.1 and your sound driver will do the mixing.

[FreeFall]

nevcairiel,

The latest nightly builds are crashing when playing DVD's, happens with MPC-HC x86 / x64 and Zoom Player. Zoom Player shows a crash report pointing to avutil-lav-55.dll as the problem. The problem seems to affect LAVFilters-0.67.0-14 and later builds.

This is on Windows 7 x64 Pro.

Thanks.

[Nullack]

nevcairiel,

The latest nightly builds are crashing when playing DVD's, happens with MPC-HC x86 / x64 and Zoom Player. Zoom Player shows a crash report pointing to avutil-lav-55.dll as the problem. The problem seems to affect LAVFilters-0.67.0-14 and later builds.

This is on Windows 7 x64 Pro.

Thanks.

Yes, it intermittent but Im replicating it on Win 10 x64 during DVD playback. Heres what the app log spat out from the crash:

Fault bucket 120380086268, type 4
Event Name: APPCRASH
Response: Not available
Cab Id: 0

Problem signature:
P1: mpc-be64.exe
P2: 1.4.6.1053
P3: 567b658a
P4: avutil-lav-55.dll
P5: 55.10.100.0
P6: 01185798
P7: c0000094
P8: 00000000000235cb

[Aleksoid1978]

Hi nevcairiel.
1 - thanks for you code for VP9 DXVA2, work great on GTX 960.
2 - can help deal with a small problem : when use LAVVideo and close playback do first calling CLAVVideo::BreakConnect() with dir == PINDIR_INPUT. When use MPC-BE's Video decoder, close playback - do first calling ::BreakConnect() with dir == PINDIR_OUTPUT.
I can not understand the reason for different behavior. Can you help me ?? :)

[mark0077]

I also tried my Queen Greatest Hits DVD today and it also has issues but with the latest official version. With internal mpcbe mpeg2 decoder there's no issue. This DVD always seemed to cause ffdshow issues in the past but haruhiko had fixed all of those. Is it worth uploading the menu for this somewhere. Basically the menus freeze and windows blue spinning icon appears.

[Aleksoid1978]

https://yadi.sk/i/eFmQ-MWBiHWR3
About 50 second video stream changes (changing the frame size from 720x576 -> 1920x1080). In DXVA mode only Microsoft DTV-DVD Video Decoder can normally handle this situation.

But - if you seeking this place, the resolution is changed and continues to play normally.

[Nullack]

You should generally not have LAV convert video to RGB, especially on UHD content its just going to be rather slow because it uses the CPU for that

Thanks. Should this be default to leave it as untouched as input? Currently LAV is defaulting to PC levels 0-255.

[nevcairiel]

Thanks. Should this be default to leave it as untouched as input? Currently LAV is defaulting to PC levels 0-255.

The option makes no difference if you don't force RGB conversion by disabling everything else. And once you do convert, full range RGB is more standard. There is usually a reason for the defaults to be what they are ;)

[nevcairiel]

1. Which parameter of a nested chapter is used in LAV Splitter?

The duration of an ordered-nested-chapter seems not to be used. Only the start time marks the chapter.
Also the segment uid will be ignored and an existing chapter-segment-link will not work.



Only the primary chapters directly under an edition are used for "ordered chapters", anything else wouldn't really make much sense, IMHO.
You can't "nest" ordered chapters, the interaction with the top level and the levels further downwards would be very bizzare.

For example, in your "00 Menu nested chapters.mkv", you have a top level chapter that says (Episode 2) "Play 0:00.000 to 0:07.487 from Segment #2", which is a clear instruction.
Then nested chapters try to override this - the first nested chapter says "Play 0:00.000 to 0:02.000 from Segment #1". Is this supposed to override the first 2 seconds of the top level chapter? Or make it longer? Its just not clear what its supposed to do, because there is no clear spec.

A more logical way to approach this would be to not have the top level chapters actually be ordered chapters, and only the level below - but alas thats not how it was designed, or how any other ordered chapters MKV works.

So in short, top level chapters inside an edition are used for ordered chapters, any further nested chapters are purely cosmetical for navigation purposes.


2. Is this a bug or a not implemented feature?

Nested chapters are not really good documented and I think you has programmed this so good you knew.

Sounds like you are doing something that its not meant for.

If you really want to shove multiple episodes into one mkv, best would probably to use one edition for each. That way you can even share OP/EDs properly.
Or don't use nested chapters, and just name them differently, ie. name them like LAV would name them if they are nested "+ [Name]" etc, that way it looks like its nested, but its all top-level.

Otherwise, nested ordered chapters are not something that works. The top level chapters under the edition need to be the ordered chapter entries, no others will be used for ordered chapter handling.

[hubblec4]

Only the primary chapters directly under an edition are used for "ordered chapters", anything else wouldn't really make much sense, IMHO.
You can't "nest" ordered chapters, the interaction with the top level and the levels further downwards would be very bizzare.

Ok. I figured out that LAV work like you described but I think thats a limit of your programming and not a limit of the specs.
I have seen other menu chapter.xml files which have nested ordered chapters.

A possible way for ordered chapters is set the ChapterTimeEnd equal to the ChapterTimeStart or let it empty. If you found such top-level ordered chapter and it has nested chapters too, than you could use this start and end times for ordered chapter handling.


For example, in your "00 Menu nested chapters.mkv", you have a top level chapter that says (Episode 2) "Play 0:00.000 to 0:07.487 from Segment #2", which is a clear instruction.
Then nested chapters try to override this - the first nested chapter says "Play 0:00.000 to 0:02.000 from Segment #1". Is this supposed to override the first 2 seconds of the top level chapter? Or make it longer? Its just not clear what its supposed to do, because there is no clear spec.

A more logical way to approach this would be to not have the top level chapters actually be ordered chapters, and only the level below - but alas thats not how it was designed, or how any other ordered chapters MKV works.

So in short, top level chapters inside an edition are used for ordered chapters, any further nested chapters are purely cosmetical for navigation purposes.


Sounds like you are doing something that its not meant for.

If you really want to shove multiple episodes into one mkv, best would probably to use one edition for each. That way you can even share OP/EDs properly.
Or don't use nested chapters, and just name them differently, ie. name them like LAV would name them if they are nested "+ [Name]" etc, that way it looks like its nested, but its all top-level.

Otherwise, nested ordered chapters are not something that works. The top level chapters under the edition need to be the ordered chapter entries, no others will be used for ordered chapter handling.


Normally my Matroska menu editor creates for each episode an edition and an edition with all episode. It works like it should.
Only if in the episodes an existing segment-link to another file then it doesnt work with nested chapters well.
(but I think I can write a workaround)


I know LAV Splitter has other priorities and maybe you have in future a little bit time to enhance the Matroska splitting.
I would help you if you like.

[nevcairiel]

Ok. I figured out that LAV work like you described but I think thats a limit of your programming and not a limit of the specs.
I have seen other menu chapter.xml files which have nested ordered chapters.

You have to ask yourself one thing:
Does it work with Haali Splitter?

If the answer is No, then the specification does not allow it.
There is no written specification for ordered chapters, whatever works with Haali is the specification. And I'm not going to invent any arcane use-cases that are never going to be used by more than 2 people anyway.

[hubblec4]

You have to ask yourself one thing:
Does it work with Haali Splitter?

Im sure it would not work with Haali Splitter, cause the segment-linking works not perfect like you said(many posts before).
The other reason is Haali was not a fan of a menu so i think he don't implement such a feature.



And I'm not going to invent any arcane use-cases that are never going to be used by more than 2 people anyway.

Yes, that can I understand. How many people would be needed so that such a function is incorporated?

[thomaz909]

i have a problem with 0.67.

when i install 0.67 i cant bitstream "Reclock encoded AC3" over "Realtek HDMI Audio" anymore.
with 0.65 all fine.

same settings used for both versions.

i absolutely have no idea what happens.

reclock shows that it gets 6 channels and reclock output say that it is in "ac3 enc mode (wave out)" but my denon avr stays in "analog" mode which indicates that no digital signal is coming.
then i run 0.65 installer and after that it works again immediately.

[nevcairiel]

Thats a ReClock bug. You can fix it by checking the legacy 5.1 layout option in LAV settings.

[thomaz909]

yes,
it works.
thx.
:)

[Nullack]

Thanks very much Hendrik for the commit in the git log fixing DVDs cheers :)

[nevcairiel]

The latest nightly builds are crashing when playing DVD's, happens with MPC-HC x86 / x64 and Zoom Player. Zoom Player shows a crash report pointing to avutil-lav-55.dll as the problem. The problem seems to affect LAVFilters-0.67.0-14 and later builds.


Thanks very much Hendrik for the commit in the git log fixing DVDs cheers :)

I was meaning to comment on that last night but got distracted. But yes, it should be working again in the latest nightly, hopefully.

Do let me know if the issue persists.

[nevcairiel]

https://yadi.sk/i/eFmQ-MWBiHWR3
About 50 second video stream changes (changing the frame size from 720x576 -> 1920x1080). In DXVA mode only Microsoft DTV-DVD Video Decoder can normally handle this situation.

But - if you seeking this place, the resolution is changed and continues to play normally.

The behavior seems to be quite similar in software mode and in DXVA mode for me. It briefly hangs and shows artifacts for a second, then just continues without problems.
Since software mode behaves the same way, it seems like a problem in the FFmpeg decoder. Having said that, resolution changes are a bit on the complicated side as it is. If there is a clean cut in the video stream it should work OK, but not always.

[Aleksoid1978]

It's working on Nvidia, but fail on ATI. But - Microsoft video decoder fine on ATI too.

[mark0077]

I was meaning to comment on that last night but got distracted. But yes, it should be working again in the latest nightly, hopefully.

Do let me know if the issue persists.

Nev do you still have the Queen Greatest Hits I DVD. I remember a while back I posted lots of issues with both it and the Star Wars Episode II DVD with LAV Video Decoder. Just even with the latest nightly using LAV Video for the Queen DVD, it pretty much doesn't work at all anymore. The windows blue spinning icon appears when I navigate to Title 1.

If you have it would you mind trying it with the latest nightly?

[nevcairiel]

The spinning icon would indicate a problem with the player instead, it gets stuck in its main thread, which LAV should have no direct control over.
I don't think I have that DVD.

[mark0077]

Ah right OK no worries. I can't reproduce with other video decoders like ffdshows mpeg2 one or mpc-be one thats the only reason I mentioned it.

[dansrfe]

Is it possible to have a quick shortcut or way to automatically detect current default audio output to turn mixing on/off?

I often switch between bluetooth headphones and the multichannel speaker system and remember to enable/disable mixing however its a bit difficult to get everyone else using the setup to follow suit.

[Nullack]

Hendrik a quick comment on testing. Theres some toolsets in the commercial space that are achieving a very high level of test coverage - and doing so with test frameworks that make it quick and easy. Argon for example have built a model of the various specs such as VP9 and HEVC, and have produced test coverage that is mathematically based on all possible values from the standard:

http://www.argondesign.com/products/argon-streams-vp9/

Then there is intel as well:

https://software.intel.com/en-us/intel-stress-bitstreams-and-encoder/

There is obvious value in this for both FFMPEG and Lav Filters - it goes far beyond what FATE does.

Given your leadership in the field and name recognition, I just wonder if they would be willing to negotiate with you on donating their software "for the cause" to FFMPEG / LAV Filters. It would be good PR for them, and given FFMPEG is the open source leader in the field it would be a strong statement to add to their brand. Theres lots of examples in the past where commercial companies have done this and I would encourage you not to dismiss it out of hand. Cheers

[nevcairiel]

There is no need. There is plenty people running FFmpeg against various automated testing and fault finding tools to identify problems, including people from various "big" organizations like Google and Mozilla who frequently provide results for all sorts of codecs, not just limited to VP9 or HEVC.
FATE is not comparable to such tools (and not wanting to be), since its "just" a deterministic automated regression test, and not out to find "new" bugs, just to prevent old ones from coming back.

In general such commercial tools are usually more marketing than useful. Both VP9 and HEVC provide a wide range of official conformance streams that test the behavior on "correct" bitstreams, and there are plenty of smart tools to create and test "broken" streams available to us today.
So in short - the "obvious value" just isn't there, since such things are already done today, and one more tool that does the same thing just isn't going to make any big difference.

[Nullack]

Good stuff, so its already in play - I get your point about FATE being for a cut down regression test suite and not an attempt to cover functional ranges in a mathematical way - but this type of automated functional testing has a place just as much as regression. What impressed me about these tools was the maths model used in mathematically ensuring test coverage over the whole ranges of possibilities. That goes way beyond a few companies looking at a few validation streams. And its not fuzz testing if that what you mean to imply (I read about what google did in this space it was top stuff). This is different to fuzz testing. This is developing a framework to mathematically analyse functional ranges and cover all such scenarios. Logically, if this is all "marketing" and the companies arent offering anything of value beyond what is already existing for free, then they have no sales and no business; yet they do exist....If your sure this is actually already being done with full range functional test coverage from an IV&V standards point of view, well awesome :) Its very difficult to do - a friend of mine had a whole team doing a formal IV&V process for an aircraft system which took 2 years of intense effort. Most test efforts rarely put more than a fraction of test coverage into their functional tests simply because of the numbers involved unless they get into automated functional test frameworks.

[hdboy]

Hi

I have a TrueHD 7.1 track that ac3filter cannot handle but LAV can. However I like ac3filter's extensive mixer options, including creating a custom mix (why I want to do this is a long story I probably don't need to go into here). Ac3filter has not been updated for a while and I want to use lav for everything (already use it as splitter and video decoder). Is there a way to have lav do the decoding and ac3filter do the mixing? Thanks.

Edit: I figured it out. Duh. In ac3filter's system tab, uncheck truehd in the "use ac3filter for" box. Then put ac3filter above lav audio decoder in external filters list of your player (MPC-HC in my case). Voila.

[Nullack]

Hello Hendrik. Thanks for the commit in your latest nightly on fixing the formats tab with high DPI scenarios when operating under media players like MPC-BE. It was a cosmetic item but nonetheless it looks great now its fixed and I'm thankful for the fix. Regards

[NikosD]

Hello and Happy New Year to all.

It's been confirmed about 4 months ago here in Doom9, that Skylake accelerates in HW 8bit HEVC format up to 8K resolution and I thought that maybe you could add this resolution in LAV video properties.

I don't have a Skylake to test it, but I believe that you allow Skylake to accelerate 8K HEVC, even if you haven't added this resolution to LAV video yet.
Right ?

[nevcairiel]

The 4K checkbox really means "4K and up", so yes, if the driver accepts 8K then it should work.

[nevcairiel]

Ah right OK no worries. I can't reproduce with other video decoders like ffdshows mpeg2 one or mpc-be one thats the only reason I mentioned it.

I had one thought - does this only happen with madVR by any chance?
Maybe I shouldn't be calling this one madVR function the way I do, I'm not sure really.

Unfortunately, none of the test menus I have shows any problems, and the number of full DVD discs I have is rather limited.

[mark0077]

Actually yes when I switch to use EVR-CP the blue spinning icon doesn't appear, but the menus don't really behave properly. There is an "OPTIONS" menu button which when clicked doesn't seem to bring you where it should. Just some animation appears which seems to then get stuck mid animation, and you don't get brought to a different screen. Maybe its just bad luck that the very few DVDs I have always seem to be tricky to deal with.

Heres a screenshot after I click "OPTIONS", the screen seems to be stuck mid animation. I tried getting the DVD menu to work without the bigger .VOB files but it won't so I'm not sure that I'm going to be able to upload any parts of the menu to show you.

http://i66.tinypic.com/358c95u.jpg

EDIT: Actually the menus do work fine, it was SVP that was causing the second issue :) So yeah using evr-cp fixes the first issue so maybe there is a change needed if using madVR as you say.

[clsid]

Are you planning to add a fallback to software decoding for unsupported VP9 profiles?

[nevcairiel]

Are you planning to add a fallback to software decoding for unsupported VP9 profiles?

Fallback works fine today, just don't use DXVA2-Native if you care about that, VP9 does not allow detecing the profile before decoding starts - ie. not early enough to fallback in time.

[nevcairiel]

Actually yes when I switch to use EVR-CP the blue spinning icon doesn't appear, but the menus don't really behave properly. There is an "OPTIONS" menu button which when clicked doesn't seem to bring you where it should. Just some animation appears which seems to then get stuck mid animation, and you don't get brought to a different screen. Maybe its just bad luck that the very few DVDs I have always seem to be tricky to deal with.

Heres a screenshot after I click "OPTIONS", the screen seems to be stuck mid animation. I tried getting the DVD menu to work without the bigger .VOB files but it won't so I'm not sure that I'm going to be able to upload any parts of the menu to show you.

http://i66.tinypic.com/358c95u.jpg

EDIT: Actually the menus do work fine, it was SVP that was causing the second issue :) So yeah using evr-cp fixes the first issue so maybe there is a change needed if using madVR as you say.


SVP breaking DVD menus is not surprising, menus need rather careful handling as every single frame needs to be displayed without delay, while things like SVP tend to add a bit of delay for temporal interpolation.

Anyway, can you try the latest nightly with madVR?
https://files.1f0.de/lavf/nightly/

I hope it should work now.

[mark0077]

Hi nev. I have disabled SVP for the test now, but unfortunately with the latest nightly (LAVFilters-0.67.0-38.exe) when I play the DVD and Navigate -> Jump To -> Title 1 the seekbar seems to show the correct chapter markers in mpc-be and I see the duration of the actual DVD appearing correctly but the blue spinning icon shows up again on what looks like the first frame.

[nevcairiel]

Maybe it is in fact a madVR problem then. LAV uses a special subtitle interface with madVR for DVD subs and menus, maybe it blocks somewhere unfortunate.

[mark0077]

Maybe it is in fact a madVR problem then. LAV uses a special subtitle interface with madVR for DVD subs and menus, maybe it blocks somewhere unfortunate.

OK Thanks for looking into it nev, do you think I should bring this over to the madVR thread to see if madshi can see if its anything from that side thats causing it.

[Gleb Egorych]

Hi, nevcairiel,
Here is a sample that there are artifacts in all hardware-assisted modes (DXVA2/CUDA): https://www.sendspace.com/file/nfhb8p. Software mode is OK.
Win7 x64, LAV 0.67.0-43 x86, GTX660, Forceware 359.06.

[wanezhiling]

Hi, nevcairiel,
Here is a sample that there are artifacts in all hardware-assisted modes (DXVA2/CUDA): https://www.sendspace.com/file/nfhb8p. Software mode is OK.
Win7 x64, LAV 0.67.0-43 x86, GTX660, Forceware 359.06.

Same here.
But works good with Intel GPU.

[oddball]

I have an issue with LAV in MPC that I cannot get my head around. I recently purchased a Denon AVR and connected it using HDMI from an AMD video card. My problem is I cannot get stereo Dolby Surround decoding when I set the AMD to 5.1 or 7.1 channels. I can only get it by setting the AMD speaker configuration to stereo. I then have to set LAV to bitstream. I would prefer to have LAV set to decode audio to multi-channel output (So that I can have multi-channel+DS) but in order to do this I have to set Windows to 5.1 or 7.1 speaker configuration. But this then sends stereo as 2 channels over multi-channel (all channels are active so the amp does not think it's stereo only) and I lose stereo Dolby Surround sound.

It's Kinda annoying really. If I play a MP3 in foobar with 5.1 or 7.1 speaker configuration set it auto switches the AVR to Dolby surround so I know it's possible. I can only think it's the way LAV is splitting the channels to multi-channel.

Any ideas?

[huhn]

try ac3 filter.

[nevcairiel]

I have an issue with LAV in MPC that I cannot get my head around. I recently purchased a Denon AVR and connected it using HDMI from an AMD video card. My problem is I cannot get stereo Dolby Surround decoding when I set the AMD to 5.1 or 7.1 channels. I can only get it by setting the AMD speaker configuration to stereo. I then have to set LAV to bitstream. I would prefer to have LAV set to decode audio to multi-channel output (So that I can have multi-channel+DS) but in order to do this I have to set Windows to 5.1 or 7.1 speaker configuration. But this then sends stereo as 2 channels over multi-channel (all channels are active so the amp does not think it's stereo only) and I lose stereo Dolby Surround sound.

It's Kinda annoying really. If I play a MP3 in foobar with 5.1 or 7.1 speaker configuration set it auto switches the AVR to Dolby surround so I know it's possible. I can only think it's the way LAV is splitting the channels to multi-channel.

Any ideas?

You want an Audio Renderer with WASAPI Exclusive mode that properly switches the output device to the number of channels the decoder outputs. Thats really the only way.

[nevcairiel]

Hi, nevcairiel,
Here is a sample that there are artifacts in all hardware-assisted modes (DXVA2/CUDA): https://www.sendspace.com/file/nfhb8p. Software mode is OK.
Win7 x64, LAV 0.67.0-43 x86, GTX660, Forceware 359.06.

Sounds like a driver or hardware problem if even CUVID shows the same problem.

[Gleb Egorych]

Sounds like a driver or hardware problem if even CUVID shows the same problem.
CoreAVC in CUDA mode shows no artifacts.

[GCRaistlin]

Feature request: show LAV Video settings window (by double click on the system tray icon) on the primary monitor even if the player's window is not on the primary monitor.

[oddball]

You want an Audio Renderer with WASAPI Exclusive mode that properly switches the output device to the number of channels the decoder outputs. Thats really the only way.

None of my audio renderers in MPC are listed as WASAPI capable. Is there any other way to bypass LAV for stereo sources? I want to use LAV for everything BUT stereo. I don't want to use AC3Filter etc thanks.

EDIT: I have found a workaround by using ReClock. It allows me to set my AVR to WASAPI output. I am unsure why LAV/MPC alone is not enough to achieve the same result. I tried setting my AVR to exclusive mode in MPC and it made no difference.

[AlexKane]

@oddball
The 'Internal Audio Renderer' in MPC, is using WASAPI by default.
I tried setting my AVR to exclusive mode in MPC and it made no difference
You are probably just using DirectSound.

[sneaker_ger]

I believe MPC-HC's new renderer always uses Windows' channel count instead of the source count, so it does not work for your purpose. It has already been asked in the MPC-HC thread to change that.

[nevcairiel]

CoreAVC in CUDA mode shows no artifacts.

CUVID in LAV Video also seems fine here. No artifacts.
The stream is damaged though, and I think it just uses Error Concealment to hide the problems, which doesn't work very well in DXVA2 modes tbh.

The avcodec software decoder has the much better error concealment than any Hardware Decoder i've ever seen, but due to the way DXVA2 works, it cannot function there.

[fluffy01]

I believe MPC-HC's new renderer always uses Windows' channel count instead of the source count, so it does not work for your purpose. It has already been asked in the MPC-HC thread to change that.
Yes. I noticed that too. I have a very similar setup to Oddball with a Denon AVR connected through a HDMI cable.
I use the MPC-BE internal renderer in exclusive mode, as this does in fact always try to switch the output layout to match the input layout, and it works great for me.

My exact setup is this: I configured the windows mixer to 2 channels, so I can let the AVR upsample it to surround, if I want to, when playing normal audio in Windows.
Then I configured MPC-BE to use "MPC Audio Renderer", which is the internal audio renderer of MPC-BE. I believe that the audio renderer can be used with other players too, though I haven't tried.
I configured the renderer as such: The sound device is just the primary audiodriver, though you can choose whatever you want to use.
I set WASAPI mode to exclusive, and ticked "Allow bit-exact output" and unticked the "Use system layout channels".
I set the Sync method to "Sync video to audio", though I do not think this has any effect on what channel layout is used.

I hope this helps.

[oddball]

Yes. I noticed that too. I have a very similar setup to Oddball with a Denon AVR connected through a HDMI cable.
I use the MPC-BE internal renderer in exclusive mode, as this does in fact always try to switch the output layout to match the input layout, and it works great for me.

My exact setup is this: I configured the windows mixer to 2 channels, so I can let the AVR upsample it to surround, if I want to, when playing normal audio in Windows.
Then I configured MPC-BE to use "MPC Audio Renderer", which is the internal audio renderer of MPC-BE. I believe that the audio renderer can be used with other players too, though I haven't tried.
I configured the renderer as such: The sound device is just the primary audiodriver, though you can choose whatever you want to use.
I set WASAPI mode to exclusive, and ticked "Allow bit-exact output" and unticked the "Use system layout channels".
I set the Sync method to "Sync video to audio", though I do not think this has any effect on what channel layout is used.

I hope this helps.

Did not work for me. I set MPC-BE the same and I can only have one or the other. MPC-HC actually gives more options in regards to whether you want bitstreaming, just multi-channel decoding etc as far as I can tell. I can't get both multi-channel decoding (not bitstreaming) to HDMI and keep stereo sources sent as just stereo for Dolby Surround. It is always sent as stereo left and right over multi-channel.

[nussman]

You are looking for this (not released yet) https://github.com/alexmarsev/sanear/commit/8e14705e1421851235d7a369c8bc27ea6698414e

This can not be changed by splitter/decoder (LAV). You have to use a wasapi audiorenderer with this feature (reclock should work, not sure about MPC BE).

[fluffy01]

Did not work for me. I set MPC-BE the same and I can only have one or the other. MPC-HC actually gives more options in regards to whether you want bitstreaming, just multi-channel decoding etc as far as I can tell. I can't get both multi-channel decoding (not bitstreaming) to HDMI and keep stereo sources sent as just stereo for Dolby Surround. It is always sent as stereo left and right over multi-channel.

Did you enable the mixer in LavFilters? If so, disable it.

I just bitstream all bitstreamable formats and only decode whats left... Why do you need to decode everything except 2-channel AC3?

[Gleb Egorych]

CUVID in LAV Video also seems fine here. No artifacts.
Which videocard/driver do you use? I do see artifacts in CUVID mode in LAV Video.

[nevcairiel]

Both GTX960 or Titan X, using latest drivers (361.43)

[clsid]

I get artifacts with CUVID using a GTX960 and 361.43 on Win7. LAV 67.0-44.

[Thunderbolt8]

You are looking for this (not released yet) https://github.com/alexmarsev/sanear/commit/8e14705e1421851235d7a369c8bc27ea6698414e

This can not be changed by splitter/decoder (LAV). You have to use a wasapi audiorenderer with this feature (reclock should work, not sure about MPC BE).doesnt wasapi exclusive mode ignore this anyway?

[clsid]

No, wasapi does what you tell it to do. The current MPC-HC renderer simply doesn't do the correct thing for the discussed use case. But it will provide the necessary options in the future.

[Stereodude]

doesnt wasapi exclusive mode ignore this anyway?
Yes, when it's implemented "correctly" (ie: how people want).

[rock]

pardon me to ask something. I have WVC1 files (a file can be downloaded from here (http://media.ch9.ms/ch9/b208/c62d6156-e839-4d1c-8d78-a484a382b208/aspnet26767_Source.wmv)), play with MPC-HC nightly (1.7.10.56). When I play with dxva2 selected, wmv9 mft is used. When I play with intel quicksync selected, quicksync is used. My notebook vga is intel hd 3000. so I wanna ask, dxva2 not support for this kind of file? or actually my vga not support dxva2 for this kind of file?. thanks in advance. :)

[NikosD]

My notebook vga is intel hd 3000. so I wanna ask, dxva2 not support for this kind of file? or actually my vga not support dxva2 for this kind of file?. thanks in advance. :)

For SandyBridge (HD 2000/3000), Intel didn't provide DXVA2 VC-1 HW decoders in drivers.

You need Ivy or better for that.

I have asked them in Intel forums but Intel didn't care to support them in such relative old HW.

QuickSync decoder does support it, as you tested yourself and it's the only way to get HW acceleration for WMV3/VC-1 for Sandy.

[Djfe]

I have a question concerning TV recordings with AC3 sound (Transportstreams)
If the number of channels switches (for example) from 2.0 to 5.1 then the sound is missing for a second or so

if I play the same scene with VLC then there is no sound missing

I have an example with speech in which you can clearly hear the difference

is this issue known already?

@nevcairiel
if not and you want the sample just write me a pm ;)

[nevcairiel]

Changed should be immediate from LAV Audio's side, it is however possible that the audio renderer or even audio device re-inits which causes this.
If this is the case, you could test by enabling the mixer in LAV Audio, which suppresses all channel changes on the output. If the missing sound disappears then, you know that its likely the audio renderer or even audio device.

[Manni]

NevCairiel, I said so already in the MadVR thread, but thanks a lot for implementing 3D support in LAV. This is really great news!

[Djfe]

Changed should be immediate from LAV Audio's side, it is however possible that the audio renderer or even audio device re-inits which causes this.
If this is the case, you could test by enabling the mixer in LAV Audio, which suppresses all channel changes on the output. If the missing sound disappears then, you know that its likely the audio renderer or even audio device.

thx for the explanation :)
you are right, the issue doesn't seem to be caused by your filters
turning mixing on "fixed" it

since it happened with mpc-hc
is the reason for it a misconfiguration in mpc or rather in windows/my audio drivers?

or is there no way to fix it rather than turning mixing on, because the hardware just takes its time to switch between channel modes?

my speakers channel configuration here is 2.1 and they are simply connected to one green audio jack
I'm using the realtek onboard soundcard of my Gigabyte mainboard

[nevcairiel]

my speakers channel configuration here is 2.1 and they are simply connected to one green audio jack
I'm using the realtek onboard soundcard of my Gigabyte mainboard

If you only have Stereo speakers anyway, it would probably be easiest to just set LAV Audio to downmix to stereo at all times, then this problem disappears and you don't have to worry about windows doing proper mixing.

[Nullack]

Gday Hendrik. Certainly pleasing to see the following commits in the log

Support MVC frames in internal frame copy functions
Clear stereo image buffers when de-allocating a frame
Implement basic H264 MVC decoder based on Intel MediaSDK

Is there any plan to also share the love with Nvidia users :thanks:

[nevcairiel]

This works on every system. Please consider testing before commenting :)

[Nullack]

Understood :)

Ok so now that I'm home and not limited to looking at GIT like at work, I've done many hours of testing. When it kicks in, it seems to be working well on the latest nightly 0-54 which is up to commit d27597b as stated in the build log.

Can you please provide input into why it might not kick in under DXVA and resort to software even though the 3D file is in h.264? For example, heres an AVC encoded bit of footage that is not using DXVA:

At http://bbb3d.renderfarming.net/download.html And the bbb_sunflower_2160p_60fps_stereo_abl.mp4 file

ffprobe spits out:

Input #0, mov,mp4,m4a,3gp,3g2,mj2, from 'bbb_sunflower_2160p_60fps_stereo_abl.mp4':
Metadata:
major_brand : isom
minor_version : 1
compatible_brands: isomavc1
creation_time : 2013-12-20 20:17:03
title : Big Buck Bunny, Sunflower version
artist : Blender Foundation 2008, Janus Bager Kristensen 2013
comment : Creative Commons Attribution 3.0 - http://bbb3d.renderfarming.net
genre : Animation
composer : Sacha Goedegebure
Duration: 00:10:34.53, start: 0.000000, bitrate: 12491 kb/s
Stream #0:0(und): Video: h264 (High) (avc1 / 0x31637661), yuv420p, 3840x4320 [SAR 1:1 DAR 8:9], 12006 kb/s, 60 fps, 60 tbr, 60k tbn, 120 tbc (default)
Metadata:
creation_time : 2013-12-20 20:17:03
handler_name : GPAC ISO Video Handler
Stream #0:1(und): Audio: mp3 (mp4a / 0x6134706D), 48000 Hz, stereo, s16p, 160 kb/s (default)
Metadata:
creation_time : 2013-12-20 20:17:06
handler_name : GPAC ISO Audio Handler
Stream #0:2(und): Audio: ac3 (ac-3 / 0x332D6361), 48000 Hz, 5.1(side), fltp, 320 kb/s (default)
Metadata:
creation_time : 2013-12-20 20:17:06
handler_name : GPAC ISO Audio Handler
Side data:
audio service type: main

Here's one which does work nicely with DXVA 3D, the video seems pretty much the same under ffprobe.exe except for the frame rate

Input #0, matroska,webm, from 'LG_3D_Air.mkv':
Metadata:
title : LG_3D_SOUND_DEMO_2_1_1.Title7.Bluray3DRip
encoder : libebml v0.7.8 + libmatroska v0.8.1
creation_time : 2012-09-11 05:16:35
Duration: 00:03:51.68, start: 0.000000, bitrate: 12811 kb/s
Chapter #0:0: start 0.000000, end 231.680004
Metadata:
title : (01)00:00:00:000
Chapter #0:1: start 232.065000, end 231.680004
Metadata:
title : (02)00:03:52:065
Stream #0:0: Video: h264 (High), yuv420p(tv, bt709), 1920x1080, SAR 1:1 DAR 16:9, 23.98 fps, 23.98 tbr, 48003.07 tbn, 47.95 tbc (default)
Stream #0:1(kor): Audio: dts (DTS-HD MA), 48000 Hz, 5.1(side), fltp (default)

The curious thing is that when I use MKVToolNix to make the bbb file into an MKV as "bbb_sunflower_2160p_60fps_stereo_abl.mkv" I still dont get DXVA acceleration. I thought it did support muxing 3D, so if thats true its not a simple container format problem as "LG_3D_Air.mkv" just works fine. Also I tried "bbb_sunflower_2160p_30fps_stereo_abl.mp4" and that uses software too so its not a high frame rate issue. So ffprobe doesnt seem to give me much help as to why some 3D isnt using DXVA. The only visual difference I can see between the 3D AVC encoded files is the working one is side by side, and the non working ones are top by bottom.

[madshi]

The recently added 3D support is for MVC streams, which is how Blu-Ray encodes 3D, so both left and right eye have full resolution. Side-by-side and top-by-bottom are from LAV's point of view simply 2D encodings.

[Nullack]

Thanks for that Madshi, appreciated :) I have a large DVD/Bluray/Bluray 3d collection. All the 3D blurays with MVC streams Ive played in my hours of testing today using the nightly build with MCP-BE and AnyDVD HD have worked super great. What I dont have my head around, is if SBS and TBB AVC encodes are just 2D to LAV, why is seemingly similar video footage falling back to software in some cases?

[madshi]

I can't answer that. But I don't understand what you said about playing MVC Blu-Rays with AnyDVD HD, that shouldn't work yet. Currently only 3D Blu-Rays remuxed into MKV by MakeMKV are supported. Playing 3D Blu-Rays as is in m2ts format shouldn't work yet.

[nevcairiel]

The dimension of the first file you referred to is way too high for DXVA (3840x4320), its 4K OU 3D, which is more than most DXVA devices can decode.
As madshi said, there are no changes to such files (and none planned) since they are "just" 2D files which happen to contain 2 images in the stream, but nothing the video decoder has to care about.

The only new format supported right now is MKV 3D files created by MakeMKV from a 3D Blu-ray.

[Nullack]

Understood. Im happy to fix any error in my test process - test errors occur just like dev errors. Im using the 3D bluray movie "Gravity" from the 3D disc. I know for sure Im entering 3D mode on my UHD display, cos one it has a little window saying so, and also while its a UHD display when in 3D mode it only does 1920x1080. I have 3D clearly visible in the render. I switched temporarily to EVR CP as yours (which is awesome thanks) doesnt have CPU / GPU utilisation in the render stats menu, and I plainly see the GPU being used with little CPU in 3D bluray playback. Please find attached a screenshot of the render stats. I dunno mate.

http://s12.postimg.org/3xeapwqst/mvc.png

[nevcairiel]

MVC 3D decoding also only works with madVR, or you will only get a 2D image.
You could also always decode 3D Blu-ray before, you would simply only get a 2D image. And nothing has changed in that. Like repeated many times now, the only supported format are MakeMKV MKV 3D files created from a 3D Blu-ray, and only if you use the latest madVR release.

[Nullack]

I get what your saying I'm not challenging that. Given I *am* seeing 3D, as I also clearly stated, I'm not sure where it's coming from then if LAV is just doing 2D in such use cases. I'll look into my TV manual as maybe its some sort of 2D->3D conversion it auto does in that mode or something.

[rock]

For SandyBridge (HD 2000/3000), Intel didn't provide DXVA2 VC-1 HW decoders in drivers.


thanks for answering. anyway I do some testing on other player, potp, it show dxva native when using built-in dxva decoder. as you said "For SandyBridge (HD 2000/3000), Intel didn't provide DXVA2 VC-1 HW decoders in drivers.", I wanna ask to answer my curiosity, how to check when player really use h/w acceleration, in this case dxva2? thanks.

sorry for thread owner if its off-topic.

[madshi]

I get what your saying I'm not challenging that. Given I *am* seeing 3D, as I also clearly stated, I'm not sure where it's coming from then if LAV is just doing 2D in such use cases. I'll look into my TV manual as maybe its some sort of 2D->3D conversion it auto does in that mode or something.
As I've explained in the madVR v0.90.0 announcement, if you have 3D activated with NVidia drivers, NVidia itself tries to convert 2D movies to 3D, at least when using D3D11 presentation. Maybe you're seeing that. Or maybe your TV is doing the conversion, I don't know.

[Manni]

@Nevcairiel
Please could you provide the name of the Intel DLL downloaded during setup, and in which folder it's located?
I have to install LAV on a different PC because I use KCP on my HTPC, so LAV isn't installed in the default location, and then copy the installed files into the KCP folder.
I don't think the Intel dll is in the LAV Filters folder, which might explain why the 3D playback with LAV+MadVR doesn't work on my HTPC.

[nevcairiel]

The dll is inside the LAV Filters folder if you used the latest nightly installer and didn't disable installation of the H.264 MVC components (libmfxsw32/64.dll)
I cannot support such a setup however, please just execute the installer on the system its supposed to run on, it does more than just install files.

Anyone running a custom setup for some reason is on their own, I am afraid. I provide the installer to avoid setup issues.

[Manni]

The dll is inside the LAV Filters folder if you used the latest nightly installer and didn't disable installation of the H.264 MVC components (libmfxsw32/64.dll)
I cannot support such a setup however, please just execute the installer on the system its supposed to run on, it does more than just install files.

Anyone running a custom setup for some reason is on their own, I am afraid. I provide the installer to avoid setup issues.

Thanks, and yes to both conditions (latest nightly and both options enabled). I guess I'll have to either wait for KCP to update its components or leave KCP behind and install all the components separately, which I'm tempted to do anyway to move to 64bits. The main drawback for me is that MPC-BE handles most file types internally by default, I'm not sure how to get the same setup as the one I get with KCP, where LAV handles everything.

Last time we discussed this, you said it was fine to simply drop the LAV files in the LAV Filters folder in the KCP folder, hence why I tried this way, but I totally understand you can't support this.

[Nullack]

Manni I run nightlies of MPC-BE x64 and LAV Filters x64. You can easily disable all the internal filters and just run with the LAV ones as external, with priority on all three being splitter / video / audio. It works well, actually very well, stable and robust most of the time for what are nightly builds. Its a real credit to Hendrik and the MPC-BE Team. To do so just go into the various internal filter sections of MPC-BE, right click and choose disable all. Then in the external filters, after installing the LAV filters (I only use x64 builds for both software) manually add the three LAV splitter / audio / video by clicking add filter and choose the prefer option for all three.

[nevcairiel]

Last time we discussed this, you said it was fine to simply drop the LAV files in the LAV Filters folder in the KCP folder, hence why I tried this way, but I totally understand you can't support this.

Most of the time this works fine, however sometimes the setup functions need to be executed again - especially if support for new formats is added like in this case.

[Manni]

Most of the time this works fine, however sometimes the setup functions need to be executed again - especially if support for new formats is added like in this case.

Manni I run nightlies of MPC-BE x64 and LAV Filters x64. You can easily disable all the internal filters and just run with the LAV ones as external, with priority on all three being splitter / video / audio. It works well, actually very well, stable and robust most of the time for what are nightly builds. Its a real credit to Hendrik and the MPC-BE Team.

Thanks, will try to move to x64 and separate installs of MPC-BE, LAV and MadVR tonight. It will certainly make updating each component as needed easier. :)

[clsid]

Has anyone benchmarked the performance of the Intel Media SDK decoder yet?
Does it use hardware acceleration on systems with an Intel GPU?

[akim]

nevcairiel

Please do audio normalization in Potplayer and MPC-BE and then the MPC-HC will must have.
If the answer is not whether it is planned in the future?

[huhn]

mpc-hc/be have an normalize feature.

[filler56789]

If the answer is not whether it is planned in the future?

Nope :)

[Nullack]

Hello Hendrik, with todays nightly I have observed I can no longer seek forward or backward by clicking the mouse in the player's seek bar somwhere else other than current playback and have it work OK in webm / VP9 using this sample for example:

http://www.reduser.net/forum/showthread.php?111230-Google-VP9-4K-HD-Sample

Im not sure when this was last working/if it worked off the top of my head. I tried seeking in other container and video codec formats, all is seeking OK including even similar footage such as VP8 / webm

Thanks

[NikosD]

Has anyone benchmarked the performance of the Intel Media SDK decoder yet?
Does it use hardware acceleration on systems with an Intel GPU?
Is there a sample to test it ?

[huhn]

how to test? it only actives when madVR is the renderer right?

[NikosD]

@nev and anyone else that can help regarding MVC MKV support

1) Your last nightly builds (installers executables) can't be uncompressed properly using latest Inno Setup Unpacker v0.45 with the command innounp -x "installer.exe"
It misses the Intel DLL.

I saw that you are using 7za.exe, possibly to extract the libmfxswXX.dll ?

Is there a proper way to extract the latest nightly installer ?

2) For manually registering properly the latest nightly builds, do we need to register the libmfxsw*.dll besides the .ax files ?

Even on Intel systems ?

I searched my C: disk and didn't find the SW libraries - libmfxsw64.dll

Only the HW dlls, like libmfxhw64.dll

3) Does the use of SW Intel libraries by LAV mean that even on Intel systems which have already installed HW Intel libraries, only SW decoding of MVC can be done ?

Do you plan to add HW acceleration for MVC decoding ?

4) Can we use MVC decoding with EVR/EVR-CP or is it supported by madVR only ?

Thanks!

[madshi]

1) The installer doesn't contain the Intel DLL. The installer downloads it during installation in case you have the MVC decoding option checked. The reason for this is that the DLL is quite big, so it would increase the size of the installer quite a lot, and not all users might want/need it.

4) Currently only madVR has the necessary interfaces in place to properly receive the decoded 3D frames from LAV. Maybe other renderers will add support for that sooner or later, or maybe not. It might be possible for LAV to create a side-by-side view for renderers which don't support 3D, but I don't know if nevcairiel is planning to do that.

[NikosD]

Thanks for the reply.

I don't have a BluRay 3D or a MVC MKV on my HDD (only SBS and top-bottom files) but your reply at 4) looks like that it's impossible to see a MVC MKV file without madVR!

Or is it the implementation of LAV MVC MKV decoder that needs explicitly madVR ?

[madshi]

At the moment madVR is the only way to see an MVC MKV file with LAV, that's correct. LAV isn't written to specifically only support madVR. LAV outputs the frames in a way that any renderer can implement. madVR is simply the first that does. The standard DirectShow APIs don't really support 3D frames, so nevcairiel and I developed a new interface for that. It's not very complicated and easy enough to implement for other renderers, too, they just need to invest the time and effort to do so.

[NikosD]

Very clear answer.

And the question is:

How all the other decoders/players display MVC MKV files without using madVR renderer ?

[madshi]

Which other decoders/players do you mean? PowerDVD? That one has its own 3D decoders and renderers, they're not using LAV or madVR. So they could implement their own private interface between decoder and renderer.

[NikosD]

It's been a long time since my last involvement in MVC.

IIRC, Stereoscopic Player can decode/ display MVC MKV and I think it was using CoreMVC ?

And i think there is another one player that I don't remember its name.

[madshi]

I'm not sure which interface CoreMVC uses to output 3D frames. AFAIK CoreMVC has never been publically released. I think Stereoscopic Player and CoreMVC did their own private thing.

[Manni]

Which other decoders/players do you mean? PowerDVD? That one has its own 3D decoders and renderers, they're not using LAV or madVR. So they could implement their own private interface between decoder and renderer.

It's been a long time since my last involvement in MVC.

IIRC, Stereoscopic Player can decode/ display MVC MKV and I think it was using CoreMVC ?

And i think there is another one player that I don't remember its name.

I'm not sure which interface CoreMVC uses to output 3D frames. AFAIK CoreMVC has never been publically released. I think Stereoscopic Player and CoreMVC did their own private thing.

I don't know about CoreMVC but Stereoscopic Player doesn't work on AMD due to a bug in the drivers (see https://community.amd.com/thread/168983).

It's been a while since I last tried, I hope it won't be the case with the MadVR/LAV implementation, hopefully not as it relies on the Intel DLL.

By the way, PowerDVD doesn't support 3D MVC in mkv files. It simply does a 2D > 3D conversion of the left eye stream which looks awful. Works fine with 3D Bluray though (apart from a levels mismatch issue in 3D vs 2D with AMD).

[madshi]

From the description in that bug report it sounds like they're using the AMD custom 3D API which was made for Windows 7 (but might still work on Windows 8+). madVR is using the official Windows 8+ Direct3D APIs instead. I can't say for sure whether they have the same problem or not. I've only tested if 3D generally works, I've not had a chance to watch a full movie and look out for delays or other issues.

[NikosD]

I remembered the other player I thought it could play MVC MKV files but I'm not sure it does.

I can't test it.

It's called Bino3d
http://bino3d.org

[nevcairiel]

I remembered the other player I thought it could play MVC MKV files but I'm not sure it does.

I can't test it.

It's called Bino3d
http://bino3d.org

That one doesn't seem to support MKV MVC files, only 3D files encoded as "normal" H.264.

[fluffy01]

I was just wondering... Do you have any plans of implementing DXVA2 decoding of MVC?
I once looked at the specs, and it seems like a pretty simple and straightforward interface for it.
That way, we don't have to depend on the Intel API.

[nevcairiel]

There are no plans. Its not quite as simple as you portrait it, I'm afraid.

[fluffy01]

Sorry to hear that. And you are right. I only did a quick readthrough, never actually tried to do it myself.
I am not a C/C++ programmer. I mainly program Delphi and C#.
It looked pretty simple, though, but as you said, there is probably more to it than that.

[fluffy01]

Oh, almost forgot... I reencoded some of my 3D-blurays into a HSBS video, a long time ago, for easy viewing. I tagged the video stream in the mkv-files as being SBS, I was wondering if LAV Splitter/Video is reading and sending this information on to madVR? Because I can't seem to get it to realize, that it is in fact 3D.

[madshi]

Is there an SBS flag in the container? If so, maybe LAV could export it somehow (e.g. IPropertyBag) in some future version, would be helpful for me.

[Nullack]

Hello Hendrik, with todays nightly I have observed I can no longer seek forward or backward by clicking the mouse in the player's seek bar somwhere else other than current playback and have it work OK in webm / VP9 using this sample for example:

http://www.reduser.net/forum/showthread.php?111230-Google-VP9-4K-HD-Sample

Im not sure when this was last working/if it worked off the top of my head. I tried seeking in other container and video codec formats, all is seeking OK including even similar footage such as VP8 / webm

Thanks

Looking a bit deeper at this there seems to be some ffmpeg level discussion about seeking in VP9. Handy that Hendrik is also a ffmpeg contributor :cool:

[Duffy Moon]

Not sure if anyone can help with this, or even if it's an LAV issue (but it doesn't happen with my TiVo) but when I watch some video files via MPC, my amplifier shuts itself off on occasion.
It's not set at a high volume, so I don't think it's a signal overload (although it does seem to happen more often when there's a sudden loud noise on the soundtrack). I have tried checking clipping protection and using DRC, but they made no difference. It only happens on about 1 in 100 files, but when it happens, it's very annoying!

[Nullack]

Hi Hendrik. Say a user has got the tickbox for prefer audio stream with the highest quality enabled in the lav splitter config. Its rare but it can happen that a bluray disc has both a TrueHD and a DTS HD MA track. I've noticed LAV splitter will default to TrueHD. Since TrueHD and DTS HD MA are both lossless, it doesnt make a difference which one is chosen if they can decode that chosen one in their audio system. And I realise you needed to pick one of them in this case. However, it is generally acknowledged that DTS is a better option than AC3. So when a user might have older audio gear that only does the old standards like DTS or AC3, the better outcome in this circumstance would be to default to DTS HD MA since that will fallback to DTS and not TrueHD which would fall back to AC3. Thanks

[nevcairiel]

Looking a bit deeper at this there seems to be some ffmpeg level discussion about seeking in VP9. Handy that Hendrik is also a ffmpeg contributor :cool:

Seeking in that file works fine, its just slow because the file doesn't contain a proper seek index. Nothing that can be done about that, other than remuxing it of course.

[dansrfe]

Bug: AC3 track is prioritized higher and selected instead of ATMOS/TrueHD track of same language.

[nevcairiel]

Bug: AC3 track is prioritized higher and selected instead of ATMOS/TrueHD track of same language.

It is most definitely not, unless the AC3 track has more channels.
Check your player, some players like to override stream selections.

[dude2012]

Is there a consensus as to which is better with Lav 0.67.0? I've read posts including from Nev saying DXVA2 is better than CUVID in Windows 10 but not sure which to pick? BTW I have an Nvidia GTX770.

[fluffy01]

Is there an SBS flag in the container? If so, maybe LAV could export it somehow (e.g. IPropertyBag) in some future version, would be helpful for me.

It is not just a SBS flag. It is actually a Stereoscopy flag, which allows a multitude of different 3D types.

The MKVToolNix GUI offers a dropdown with 15 different types of 3D (Including Mono-2D), so you can look there and see what is supported.

Setting this flag actually causes MKVToolNix to default the extension of the file to .mk3d instead of .mkv, so perhaps add this extension to LAV, so we can double-click this type instead of having to either change the extension back to .mkv or manually drag and drop the file in the player.

[nevcairiel]

I can probably export this metadata in some form, but I haven't seen many files use it. But no harm anyway!
LAV supports the .mk3d extension, file associations are managed by the player however, and not the splitter.

[fluffy01]

I usually add this flag to the 3D movies I recode, and if LAV and madVR starts to support it, then I bet that loads others will start to use it as well :)

[dansrfe]

It is most definitely not, unless the AC3 track has more channels.
Check your player, some players like to override stream selections.

I'm on the latest nightly for MPC-HC, the "Prefer audio stream with highest quality" option is enabled in LAV Splitter.

MediaInfo:

Audio: Dolby AC3 48000Hz 6ch 640kbps [A: AC-3 [eng] (ac3, 48000 Hz, 5.1, 640 kb/s) [default]]
Audio: TrueHD 48000Hz 8ch 9216kbps [A: ATMOS [eng] (truehd, 48000 Hz, 7.1, s24)]

I'm not sure what default audio track means. Did this override priority selection? None of the other files with multiple audio tracks show this.

[nevcairiel]

The "default" flag overrides everything, don't set it if you don't want that behavior.

[dansrfe]

The "default" flag overrides everything, don't set it if you don't want that behavior.

Makes sense, although perhaps an option to override the default flag for audio/subs would be useful. I've yet to find a case where the default flag was used for a purpose other than to ensure decoder compatibility by the author of the file.

[DragonQ]

Makes sense, although perhaps an option to override the default flag for audio/subs would be useful. I've yet to find a case where the default flag was used for a purpose other than to ensure decoder compatibility by the author of the file.

I always set primary audio streams to "default" over secondary streams (e.g. director commentaries).

[Blight]

Probably not out in the wild yet:
Zero-Day FFmpeg Vulnerability Lets Anyone Steal Files from Remote Machines (http://news.softpedia.com/news/zero-day-ffmpeg-vulnerability-lets-anyone-steal-files-from-remote-machines-498880.shtml)

[nevcairiel]

Probably not out in the wild yet:
Zero-Day FFmpeg Vulnerability Lets Anyone Steal Files from Remote Machines (http://news.softpedia.com/news/zero-day-ffmpeg-vulnerability-lets-anyone-steal-files-from-remote-machines-498880.shtml)

IMHO that issue is slightly overrated, but LAV is not affected in any case, unless you use it in an unsafe manner (ie. by directly hooking it up to a web server, potentially exposing its output, which is always a bad idea).

And just to be safe, the "concat" protocol used in those demonstrations is not available in LAV either (and never has been).

[3421]

Installation, disabling rtp. LAV splitter, rtp enabled. The rtp links do not open. Installation, rtp enabled. Lav splitter, rtp disabled. rtp links open. Checked on the player which not plays rtp without lav filter.

Checkbox disable/enable all the components. I disabled everything and I choose what I need and enable.

A difficult captcha from questions I don't know all the answers. And I don't know English.

[ileile]

I can probably export this metadata in some form, but I haven't seen many files use it. But no harm anyway!
LAV supports the .mk3d extension, file associations are managed by the player however, and not the splitter.

do you have any plan to support 3d bluray or ssif files in the near future? it is a important feature. with 3d bluray support, we can throw PowerDVD out of the window.

btw, thank you for your work :thanks:

[3421]

Enable or disable rtp in the menu LAV splitter is ignored. RTP is always as selected during installation LAV. 62

[Warner306]

Is it possible to modify the audio delay in LAV Audio to apply to specific refresh rates (e.g. 24 Hz)?

I have problems with audio sync at 24 Hz when playing 23.976 fps content. Other refresh rates are fine. This problem is reasonably common.

- Plex Media Player has a built-in delay for 23.976 content (see bullet #9 (https://support.plex.tv/hc/en-us/articles/208175907-How-are-Plex-Media-Player-and-Plex-Home-Theater-different-)).

- Kodi offers the ability to set delays for specific refresh rates (described here (http://forum.kodi.tv/showthread.php?tid=223175&pid=2033893#pid2033893)).

- There is an entire thread dedicated to this issue in the Kodi forums (found here (http://forum.kodi.tv/showthread.php?tid=80247)).

Adding a refresh rate-based delay in LAV Audio would open up the ability to apply this type of correction universally to any player that uses LAV Filters.

[Asmodian]

I don't think LAV even knows the current display's refresh rate, does it? That sounds more like a player feature to me, I can see it being useful but I think it is important to keep LAV's feature list as light as possible. :)

If I understand this correctly, these features are compensating for different amounts of input lag in the display due to it using different modes with different refresh rates, not desync between the audio and video as it leaves the computer?

[nevcairiel]

Asmodian is right, LAV Audio of course doesn't know the current refresh rate, it has nothing to do with video afterall, and I'm not exactly eager to add stuff to try to detect the current refresh rate, especially considering things like automatic display mode changers and whatnot.

[ang3l]

Asmodian is right, LAV Audio of course doesn't know the current refresh rate, it has nothing to do with video afterall, and I'm not exactly eager to add stuff to try to detect the current refresh rate, especially considering things like automatic display mode changers and whatnot.

since madVR and more generally any players who uses this feature determines the refresh rate starting by the frame rate, lav could do the same for the audio delay and give two possibilities to set the value, the current method or an advanced text string as madvr does

something like

1080p23@-300, 1080p50@-150 ...

just an idea

[nevcairiel]

LAV Audio also doesn't know the frame rate of the video, it deals with the audio, not the video. And such guesses would be extremely inaccurate anyway, and I don't do broken features. Just get whichever player you are using to do it.

[cengizhan]

hi nevcairiel,
could you make the installer skip downloading decoder if it alreadys exists?

[an3k]

Even if LAV has to be responsible for A/V delay it would be the Splitter but actually the Player has to take care of it and to be honest I never experienced any delay regardless of the displays refresh rate. Maybe because I use digital connection between the playback device and the display.

Do you use an external Amp/Surround system?

[Warner306]

LAV Audio also doesn't know the frame rate of the video, it deals with the audio, not the video. And such guesses would be extremely inaccurate anyway, and I don't do broken features. Just get whichever player you are using to do it.

Understood. I thought adding this feature further up the chain would make it easier to implement for the end user.

[madshi]

At some point I could implement an option to do video delay (instead of audio delay) in madVR, depending on refresh rate. But probably not any time soon.

[Warner306]

I don't think LAV even knows the current display's refresh rate, does it? That sounds more like a player feature to me, I can see it being useful but I think it is important to keep LAV's feature list as light as possible. :)

If I understand this correctly, these features are compensating for different amounts of input lag in the display due to it using different modes with different refresh rates, not desync between the audio and video as it leaves the computer?

Given the cause of this still remains unknown to the developers I've talked to, the player is simply delaying the audio by a preset amount before it outputs to the audio output device.

The Kodi thread has people blaming a number of features and claiming solutions found in the Kodi audio settings. But, in my experience, I think it comes down to the display lagging when it uses its 24 Hz mode. In either case, an easy fix is to add a delay to 23.976 content.

[Warner306]

Even if LAV has to be responsible for A/V delay it would be the Splitter but actually the Player has to take care of it and to be honest I never experienced any delay regardless of the displays refresh rate. Maybe because I use digital connection between the playback device and the display.

Do you use an external Amp/Surround system?

Sure, it could also be the AV receiver struggling to lock-on to 24 Hz content. But I've used a very new and a reasonably old (2-3 old) Onkyo and they both display this flaw.

I think it could be a flaw of some LED TVs. They all struggle with motion in some way. I am using a very common, supposedly high-end Samsung LED.

[huhn]

this issue is called input lag. TV can have easily an input lag of over 100ms there are even TV with an input lag of over 200 ms.

this changes with refresh rates and settings like crappy frame interpolation and TV mode like game(low delay)/cinema(very high delay).

[nevcairiel]

Many TVs have input lag defined in number of frames (due to their processing), which makes it noticeable longer on 24Hz since frames are just so much longer (41.6ms instead of 16.6ms at 60Hz)

[zerowalker]

When it comes to bad frame timing with video and MPC-HC frame time correction helps, can LAV filter also solve this or is it a completely different matter?
Codec is H264.

[Warner306]

Many TVs have input lag defined in number of frames (due to their processing), which makes it noticeable longer on 24Hz since frames are just so much longer (41.6ms instead of 16.6ms at 60Hz)

I've read dozens of posts attempting to diagnose this issue and this is the most logical I've heard. It would mirror my suspicion of unoptimized processing being at fault. It is one of the ongoing factors that have weaned me off the belief that LED TVs are a great investment.

[FDisk80]

I can't answer that. But I don't understand what you said about playing MVC Blu-Rays with AnyDVD HD, that shouldn't work yet. Currently only 3D Blu-Rays remuxed into MKV by MakeMKV are supported. Playing 3D Blu-Rays as is in m2ts format shouldn't work yet.

Shouldn't work yet? :o
So there is hope?

[Nullack]

To clarify, that puzzle existed for a number of days. In the end I finally got an answer from Samsung about how to turn off 3d enhancement that's auto done for 3d content when in 3d mode. To add to the puzzle too, Nvidia drivers also fiddle with it as well. It was a big mission to stop the auto 3d from 2d content when in 3d mode.

[LigH]

@ zerowalker:

Usually, decoders decode streams of a media source in sync, according to their multiplexing. It will probably be mainly the player being responsible for either requesting asynchronous decoding, or delaying a stream relative to others.

[XTrojan]

Is there any statistics on how much CPU power is needed for the upcoming HEVC blurays? Will for example 3770k be able to decode in realtime without hardware acceleration?

[huhn]

maybe 100 mbit with 24hz.

[Piyoko]

I've read dozens of posts attempting to diagnose this issue and this is the most logical I've heard. It would mirror my suspicion of unoptimized processing being at fault. It is one of the ongoing factors that have weaned me off the belief that LED TVs are a great investment.
I always assumed this was the cause. The extra processing of the "movie" mode on the TV coupled with the longer frame time at low refresh rates.

I have one of the newer 4K Samsungs and at 60Hz PC/game mode the input lag is extremely low (for a TV) but at 24Hz movie mode I have to apply an almost 200ms audio delay. Thankfully everything I watch is 24fps so I can just leave the value as it is.

[Manni]

Is there any statistics on how much CPU power is needed for the upcoming HEVC blurays? Will for example 3770k be able to decode in realtime without hardware acceleration?

I have a 3770K at 4Ghz with 8Gb RAM and it's not able to decode HEVC 10bits UHD without H/W acceleration. It does it, but it stutters as soon as there is too much info, for example in pans etc.

My HD7870 isn't the bottleneck, I tried with a Fury and it was the same. It's the CPU.

I'm waiting for Arctic Islands now.

[hu1kamania]

I recently "upgraded" to Windows 10. Unsatisfied with MadVR's deinterlacing, and yadif, I wanted to see if I could get CUVID + hardware deinterlacing to work. Years ago this was the best method, but more recently on Windows 8.1 I noticed quality was almost identical to MadVR's, so I just stuck with DVXA + copy back.

Flash forward to CUVID, and right away I was getting very unstable results. After some research, this is a known issue with Windows 10 + CUVID. I know it's not recommended by nevcairiel and the problem was identified by the developers of SVP here:

Windows 10 - video player hangs and/or crashes (http://www.svp-team.com/forum/viewtopic.php?id=2671&p=2)

As you can see, using an driver equal to or before 350.12 fixes the issue. The problem I had was, this driver doesn't support my GTX 980Ti. So I modified and installed the 350.12 driver using the following guide:

Nvidia INF driver modding (http://forums.guru3d.com/showthread.php?t=377158)

This allowed stable use of CUVID with Windows 10, and hardware deinterlacing was significantly improved.

I'd like to see CUVID support improved in newer drivers. Since something was broken after 350.12, it would seem the ball is in Nvidia's court to fix the issue. However, if something can be done on your side to bring compatibility in line with recent drivers, that would be nice!

That was all I had to say. I hope somebody finds this helpful, insightful or interesting!

[clsid]

That instability (only) occurs when something else is also using CUDA/OpenCL at the same time. For example SVP or certain functionality in madVR (like NNEDI and Error Diffusion). NVIDIA is the one that must fix it.

[hu1kamania]

That instability (only) occurs when something else is also using CUDA/OpenCL at the same time. For example SVP or certain functionality in madVR (like NNEDI and Error Diffusion). NVIDIA is the one that must fix it.

Thank you for clearing that up. It's a shame that we need to go back to April of last year to find a driver that works. Even worse, the driver doesn't officially support my hardware or OS!

I will have to keep an eye on OpenCL development from Nvidia, which might be about as interesting as watching paint dry. Nvidia never had a very cozy relationship with OpenCL, to my recollection.

[rock]

sorry, just wanna say thanks for dxva2n fix for old intel gpu (mine is sandy bridge). it fixed playback for my ts file I posted in the past which is actually an interlaced video :)

[khanmein]

guys, when i tick HEVC on LAV Video Decoder to watch x265 .mkv, it will suddenly freeze or hang if i drag fast forward/backward. i'm using GTX 970 + i5-4460 + DXVA2 (copy-back)

[LigH]

No surprise. When your player prefers exact seeking, it will have to decode often from the beginning of the GOP you navigated into to the target frame, and because HEVC is very complex, that will take its time (the longer the GOPs may be, the longer the seeking may take). If you can switch to fast seeking, it will seek always to a GOP start instead, that will be faster but rather coarse, you may then only seek each few seconds apart.

[clsid]

Dragging the seekbar will spawn a lot of seek operations. You will get better response by clicking or using the jump function of the player.

Also, the GTX970 only supports partial acceleration of HEVC. Using software decoding may give better performance.

[khanmein]

thanks so partially support x265 only? damn~

[nevcairiel]

As of 0.67.0-67, LAV Video supports a new software deinterlacer, Weston Three Field Deinterlacing Filter (w3fdif)

w3fdif was developed by the BBC, and as I understand it, is also used by them in real-world usage.

w3fdif comes in two variants: Simple and Complex.
The difference is of course the potential quality, but also the speed.

I will let you guys judge its quality in comparison to YADIF (or even hardware deinterlacing), but there is one thing to be said about it: Its fast.

On 32-bit, Simple is significantly faster than YADIF, but due to lack of 32-bit optimizations, Complex is slower.
On 64-bit, both Simple and Complex are faster than YADIF.

Hopefully the SSE2 optimizations for Complex can be made compatible with 32-bit in the future.

A new build is available here now:
https://files.1f0.de/lavf/nightly/

Note that w3fdif does not have an option to control the number of output frames, it will always produce one frame per input field, ie. its always a double rate deinterlacer (the only proper way to deinterlace, anyway!)

PS:
The Software Deinterlacing options section looks a bit crammed, but never mind that, I have some plans to re-do some areas of that entire config screen.

[madshi]

Interesting! Is there some paper with screenshots or technical description or something about that deinterlacer? Never heard about it yet...

[nevcairiel]

Interesting! Is there some paper with screenshots or technical description or something about that deinterlacer? Never heard about it yet...

I didn't write it, just use what is in FFmpeg, so..
On a quick search I found its patent entry, which at least explains some things:

http://www.google.com/patents/US4789893#v=onepage&q&f=false

[madshi]

Ah, I see, thanks!

[SamuriHL]

That's pretty wild. I think I'll have to try that out on some of my 1080i tivo recordings and see how it looks.

[noee]

That's pretty wild. I think I'll have to try that out on some of my 1080i tivo recordings and see how it looks.

fwiw, I've been using it for a bit now on some old broadcast videos with generally poor interlaced content and to my eye, it seems on par with yadif (for my settings), but as nev says, it is faster.

I do notice on some scenes with vertical panning, there is a shimmering that occurs on edges here and there that are not as pronounced with yadif.

Be interesting to see what you think.

[hu1kamania]

How exciting! I can't wait to play around with this. I hope it beats out yadif, and CUVID hardware deinterlacing.

Thank you for the hard work, and innovation. Nevcairiel, you have my eternal respect. Same goes for madshi. Thank you!

[sneaker_ger]

Not a regression of the new deinterlacer but I think software deinterlacing is broken with 10 bit H.264 files. (picture is black)

[nevcairiel]

Software deinterlacing only works with 8-bit, always has. Don't encode such crap like 10-bit interlaced. :p

[sneaker_ger]

Ok, guessed so. I only vaguely recalled this coming up before but didn't find the older posts.

Let's hope they don't combine it in television now that HEVC 10 bit is gaining.

[SamuriHL]

fwiw, I've been using it for a bit now on some old broadcast videos with generally poor interlaced content and to my eye, it seems on par with yadif (for my settings), but as nev says, it is faster.

I do notice on some scenes with vertical panning, there is a shimmering that occurs on edges here and there that are not as pronounced with yadif.

Be interesting to see what you think.

Thanks for the input. I'll play around with it tonight. I had already updated to the latest nightly for another reason anyway, so, I should be good to go by enabling it.

[SamuriHL]

Software deinterlacing only works with 8-bit, always has. Don't encode such crap like 10-bit interlaced. :p

What the??? Who in their right mind would encode 10 bit interlaced?! LAMO! *NOT SUPPORTED!* LOL

[P.J]

Weston Three Field Deinterlacing Filter http://forums.mydigitallife.info/vb4_style/smilies/3eyes.gif

:thanks:

Edit: Any comparison between it and hardware pixel-adaptive?

[SamuriHL]

Wow, I gotta say, that new deinterlacing is pretty damn sweet. I'm using complex right now. Combined with madvr's reconstruction, my TiVo recordings look incredible. Well done boys! P.S. And DAMN smooth.

[burfadel]

The nightly builds available from https://files.1f0.de/lavf/nightly/ do not work with the latest Windows Insider preview 14251 released today. It fails with regsvr32 error 0x3.

Even when already installed and Windows is updated, the previously registered files no not work.

However, build LAVFilters-0.65.0-24.exe and older installs and works perfectly, so the issue must be in LAVFilters-0.65.0-25.exe and later. I tried the .zip archive version of 0.67 as well, the files did not register with that either. Maybe there is a compiler issue/reason for it not working?

I'm willing to test different compiles of the latest version to see if the issue is resolved. The issue is probably with the new build of Windows, but at least if the issue is isolated it can be reported correctly through the Windows feedback tool. LAV Filters seems to be the only program affected by it, so simply reporting that LAV Filters doesn't work and the circumstance that the issue not affecting any other program, may mean the issue goes unfixed before Windows 10 Redstone release later this year.

That said, the issue may be with a change introduced with LAVFilters-0.65.0-25.exe that has only become apparent with the latest Windows build. Either way, I'm more than willing to test out different builds!

[nevcairiel]

0x3 means its missing some dependency (or some function from some dependency), you can use Dependency Walker (http://www.dependencywalker.com/) to check what it may be missing.
That said, I'm not overly concerned with breakage in insider builds, there have been a lot of issues with the recent builds.

0.65.0-25 was a FFmpeg update, so if something changed it would appear to be in there.

[MrBonk]

I recently started testing using MakeMKV to rip DVDs to watch with better upscaling, but was disappointed in that none of the Deinterlacing, whether yadif or cuvid hardware adaptive (GTX 980 here on W7) seemed to be able to match the PS3's De-interlacing quality.
w3fdif doesn't seem to quite get there yet either.

HW Adaptive DI from Cuvid still seems to work the best.

Images. Same frame

yadif http://u.cubeupload.com/MrBonk/mpchc201601280412047.png
w3fdif http://u.cubeupload.com/MrBonk/mpchc201601280411032.png
HW Adaptive http://u.cubeupload.com/MrBonk/mpchc201601280415355.png
(Ignore glitches/repeated/dropped and render times)
OSD shows DI is off, but I have it set to Force in LAV. Otherwise it doesn't seem to work. It is Interlaced and MediaInfo shows that it is.

[nevcairiel]

Most deinterlacers are designed for real-life video content, not animation, so they will usually not fare very well on that kind of content (too many sharp edges).
Note that DVDs might also be telecined, not interlaced, which a normal deinterlacer would not handle very well at all.

[madshi]

You can try forced film mode in madVR, if the DVDs are telecined content.

[Magik Mark]

I experimented with it by turning off madvr deinterlacer. Image is a lot better. The 32bit really need some optimization though

[MrBonk]

The content has to be interlaced as using forced film mode(+Film mode deinterlacing) produces even poorer quality and constant judder with any camera movement (Where native 23.976FPS and 29.97 Progressive files and Video mode played back do not)

Screenshot of same frame with forced Film mode and Film yadif
http://u.cubeupload.com/MrBonk/mpchc201601280448566.png
Screenshot forced film mode and film Cuvid HW Adatpive
http://u.cubeupload.com/MrBonk/mpchc201601280452128.png

Turning all deinterlacing off, leaving on auto source detection and using only the "Inverse Telecine" option in NVCP while using Cuvid looks like this
http://u.cubeupload.com/MrBonk/mpchc201601280454457.png
But heavy interlacing and judder are still present in motion
http://u.cubeupload.com/MrBonk/mpchc201601280454508.png

Is re-encoding my only option to fix it really?
I wouldn't mind. Is an encoding solution of higher quality?

[madshi]

Screenshot of same frame with forced Film mode and Film yadif
Screenshot forced film mode and film Cuvid HW Adatpive

I was talking about forced film mode in madVR. If you want to use that, you have to completely turn off deinterlacing in LAV Video Decoder.

[MrBonk]

I was talking about forced film mode in madVR. If you want to use that, you have to completely turn off deinterlacing in LAV Video Decoder.

Oh, yes I had that enabled too. But I didn't know you had to disable LAV and it's DI detection and other DI methods off at the same time.

With only that set or video set and DI set to disabled in LAV. Same results. Judder, constant interlacing.

OH:thanks: I have to leave DI set to "Auto" in LAV and leave the other DI in that off and then set force "Film" mode.

And shazam. Works perfectly.
I'm such a dumb dumb.

Thanks a lot!

[clsid]

0x3 means its missing some dependency (or some function from some dependency), you can use Dependency Walker (http://www.dependencywalker.com/) to check what it may be missing.
That said, I'm not overly concerned with breakage in insider builds, there have been a lot of issues with the recent builds.

0.65.0-25 was a FFmpeg update, so if something changed it would appear to be in there.
Looking at the dates of those build, I think you also updated from GCC 4.9.3 to a 5.1 toolchain.

The error appears to be related to failing to load API-MS-WIN-*.DLL files.

[nevcairiel]

Never mind those, only direct dependencies of LAV or the FFmpeg dlls are of relevance, those MS things always show up one or two levels down.

[KoD]

I think I noticed in the commit logs that files which have fps outside some arbitrary magic values will be considered "corrupt". I don't think this is a good idea. If someone has a recording made with a high speed camera, he will encounter issues when trying to play it, and not because the file has a problem, but because the decoder (or was it the splitter?) has a if "fps < 125" then be nice otherwise claim file is bad. Also, the case I keep on making that one can make a mkv of 1 hour with only one or two frames in it would become impossible to play when having a limit on how low the fps is allowed to be. What I am trying to say is that if a container spec says that such files are supported, they should be supported by lav as well.

[nevcairiel]

Files with "uncommon" FPS will not stop it from playing. There is just a heuristic in place to try to select the most logical FPS to offer up as metadata. Any "FPS" values are unreliable anyway, the real determining fact is the timestamps of the video frames.

[r4dius]

The nightly builds available from https://files.1f0.de/lavf/nightly/ do not work with the latest Windows Insider preview 14251 released today. It fails with regsvr32 error 0x3.

Even when already installed and Windows is updated, the previously registered files no not work.

However, build LAVFilters-0.65.0-24.exe and older installs and works perfectly, so the issue must be in LAVFilters-0.65.0-25.exe and later. I tried the .zip archive version of 0.67 as well, the files did not register with that either. Maybe there is a compiler issue/reason for it not working?

I'm willing to test different compiles of the latest version to see if the issue is resolved. The issue is probably with the new build of Windows, but at least if the issue is isolated it can be reported correctly through the Windows feedback tool. LAV Filters seems to be the only program affected by it, so simply reporting that LAV Filters doesn't work and the circumstance that the issue not affecting any other program, may mean the issue goes unfixed before Windows 10 Redstone release later this year.

That said, the issue may be with a change introduced with LAVFilters-0.65.0-25.exe that has only become apparent with the latest Windows build. Either way, I'm more than willing to test out different builds!

Thanks for you analyse, I installed 0.64.0-24 until some other fix :)

[hu1kamania]

I have some questions and comments in regards to CUVID. I don't post often, but I've been reading Doom9 for many years. Over the last couple years I've noticed a trend to wean people off the usage of CUVID.

I understand there are a lot of con's in regards to CUVID. Many are illustrated in this post at the jriver forums: https://yabb.jriver.com/interact/index.php?topic=80258.0 though some concerns seem unfounded.

More recently it seems to be broken in Windows 10, unless you are running CUDA 7.0.0 or older. Thanks to clsid for expanding on an previous post as to why that is.

Are we shying away from CUVID because we won't trust NVIDIA to reliably support it on their end? After all, it takes some less than ideal measures to get it working with Windows 10. The driver that "works" was designed for Windows 8.1 and dates back to April of last year!

Is the move away from CUVID because DXVA support is more universal? As in, if development time is put towards DXVA, it should benefit AMD as well as NVIDIA users?

I realize I probably answered my own questions, but I'm asking because image quality is king for me, and I find CUVID's hardware deinterlacing to be best for real time playback. If I thought CUVID support was at an end, or likely at an end, I might be persuaded to compromise and learn to live with yadif/w3fdif. MadVR's DXVA hardware deinterlacing doesn't cut it right now, but I have faith madshi will get it there eventually. At this point in time, is it safe to say CUVID's place in LAV Filter's is purely a legacy one? An explanation beyond: "forget CUVID plz... (http://forum.doom9.org/showpost.php?p=1755170&postcount=35696)" would be appreciated.

Thanks!

[clsid]

Regarding the Win10 Build 14251 problems:
Faulting application name: regsvr32.exe, version: 10.0.14251.1000, time stamp: 0x56a53e57
Faulting module name: avcodec-lav-57.dll, version: 57.22.101.0, time stamp: 0x00000000
Exception code: 0xc0000005
Fault offset: 0x0076d5d2
0.67.0-67 x86

[foxyshadis]

Are we shying away from CUVID because we won't trust NVIDIA to reliably support it on their end? After all, it takes some less than ideal measures to get it working with Windows 10. The driver that "works" was designed for Windows 8.1 and dates back to April of last year!

The pressure to abandon CUVID is because CUVID appears to be abandoned and no reassurance has come from nvidia that it will ever be actively supported again. They haven't announced that it'll ever be available for win10, or that it'll be supported going forward. It's not terribly surprising that it's still present a recent 8.1 driver, since it's relatively unchanged from the original win8 CUVID support since early 2013; no work on their part to keep it in. Notably, although it's still bundled into drivers, no new formats have been supported, no bugs have been fixed in a year, new bugs have appeared. Instead, nvidia turned on the CUVID processing (deinterlacing and sharpening) on their DXVA2 decoder in the last year or two, so what's the point of not using DVXA now?

I haven't tested on my win10 system, but on my 8.1 the deinterlacing quality seems to be the same in CUVID or DXVA mode. Are there specific differences you're seeing?

[huhn]

I have some questions and comments in regards to CUVID. I don't post often, but I've been reading Doom9 for many years. Over the last couple years I've noticed a trend to wean people off the usage of CUVID.

I understand there are a lot of con's in regards to CUVID. Many are illustrated in this post at the jriver forums: https://yabb.jriver.com/interact/index.php?topic=80258.0 though some concerns seem unfounded.

More recently it seems to be broken in Windows 10, unless you are running CUDA 7.0.0 or older. Thanks to clsid for expanding on an previous post as to why that is.

Are we shying away from CUVID because we won't trust NVIDIA to reliably support it on their end? After all, it takes some less than ideal measures to get it working with Windows 10. The driver that "works" was designed for Windows 8.1 and dates back to April of last year!

Is the move away from CUVID because DXVA support is more universal? As in, if development time is put towards DXVA, it should benefit AMD as well as NVIDIA users?

I realize I probably answered my own questions, but I'm asking because image quality is king for me, and I find CUVID's hardware deinterlacing to be best for real time playback. If I thought CUVID support was at an end, or likely at an end, I might be persuaded to compromise and learn to live with yadif/w3fdif. MadVR's DXVA hardware deinterlacing doesn't cut it right now, but I have faith madshi will get it there eventually. At this point in time, is it safe to say CUVID's place in LAV Filter's is purely a legacy one? An explanation beyond: "forget CUVID plz... (http://forum.doom9.org/showpost.php?p=1755170&postcount=35696)" would be appreciated.

Thanks!


this thread is a way better place to talk about this than the madVR thread.

you are saying madVR DXVA doesn't cut it for you right now.
but it should be the same as CUVID deinterlacing and it all my test it is the same.

so a sample or something that will show the difference between DXVA deinterlancing and CUVID deinterlancing would be really nice!

general CUVID bug/limitations:

windows 10 bug:
CUVID and running any other openCL program will stop CUVID from delivering frames to the renderer.

limitation:
CUVID doesn't support 10 bit output so it can't be used for 10 bit HEVC.

the used decoder/deinterlancer shouldn't be any different from DXVA. why should nvidia write different deintelrancer for DXVA and CUVID that would be a huge waste of money.

[nevcairiel]

Note that deinterlacing through CUVID on Windows 10 may also be slightly lower quality than it is on Windows 7/8.1, because the Direct3D interop stopped functioning, which was previously required to obtain full quality.

[huhn]

getting better and better...

[biship]

The nightly builds available from https://files.1f0.de/lavf/nightly/ do not work with the latest Windows Insider preview 14251 released today. It fails with regsvr32 error 0x3.

However, build LAVFilters-0.65.0-24.exe and older installs and works perfectly, so the issue must be in LAVFilters-0.65.0-25.exe and later

Yup, same problem. I had to go back to LAVFilters-0.65.0-24.exe. Thanks for finding the fix!

[nevcairiel]

Report it to Microsoft instead. Their insider builds have been rather broken as of late.

[burfadel]

For everyone with the 0x3 error installing the LAV Filters (post-version I mentioned earlier of course).

Open this link in Microsoft Edge browser:
Windows-Feedback:?contextid=156&feedbackid=9a9f3867-6e6d-439d-a5ed-d8006d050268&form=1&src=2

If using Chrome, Firefox, or other browser just copy and paste it into Edge.

This will take you directly to a feedback someone (not me) has given regarding this issue. Apparently other programs are also affected.

Upvote way! The more upvotes the better :).

[hu1kamania]

The pressure to abandon CUVID is because CUVID appears to be abandoned and no reassurance has come from nvidia that it will ever be actively supported again.

Thank you for expanding on the situation with Nvidia. It certainly looks like CUVID is in fact abandoned, and their intent is to have DXVA2 take it's place.

I haven't tested on my win10 system, but on my 8.1 the deinterlacing quality seems to be the same in CUVID or DXVA mode. Are there specific differences you're seeing?

you are saying madVR DXVA doesn't cut it for you right now.
but it should be the same as CUVID deinterlacing and it all my test it is the same.

so a sample or something that will show the difference between DXVA deinterlancing and CUVID deinterlancing would be really nice!

I follow what you're saying huhn, but I am definitely seeing a difference!

In Windows 8.1 I did not see a difference between MadVR and CUVID, but I found the quality to be lacking, so I stuck with yadif + DXVA copy-back. After switching to W10 and experiencing the bug related to CUVID I eventually came across the solution I explained a few posts back. With the older 350.12 driver I saw a big improvement in quality with CUVID based deinterlacing.

The interlaced content I watch most is broadcast NTSC, so 480i/30, 720i/30, and 1080i/30 as well as some 480i/30 DVDs. With MadVR's hardware deinterlacing, I see more shimmer/aliasing when compared to CUVID or yadif.

One channel in particular runs at 480i/30 and has a static watermark that flickers constantly with MadVR. Forcing film mode helps with the watermark flicker, but I get a lot of jerky movement. Yadif/w3fdif helps, but CUVID + 350.12 driver completely eliminates the flicker. I do see a bit more aliasing on sharp edges with w3fdif, so yadif is my favorite between the two (I want to test that more this weekend, I'm writing this from work :eek: )

Note that deinterlacing through CUVID on Windows 10 may also be slightly lower quality than it is on Windows 7/8.1, because the Direct3D interop stopped functioning, which was previously required to obtain full quality.

Making my claims appear all the more lofty, I know!

[Mixer73]

One channel in particular runs at 480i/30 and has a static watermark that flickers constantly with MadVR. Forcing film mode helps with the watermark flicker, but I get a lot of jerky movement. Yadif/w3fdif helps, but CUVID + 350.12 driver completely eliminates the flicker. I do see a bit more aliasing on sharp edges with w3fdif, so yadif is my favorite between the two (I want to test that more this weekend, I'm writing this from work :eek: )

I'd be prepared to bet this content is actually shot progressive, but transmitted in interlaced, with the bug added, so it is actually mixed content.

[KoD]

Hi nevcariel. I have an audio-related issue which I suspect is due to lav audio.

With this sample (http://wikisend.com/download/925408/WaveOut_TrueHD.mkv) [25Mb], when using any of the DirectSound options or the sanear audio renderer in a media player, the sound is ok. However, trying to use any of the wave out audio renderers, has as effect a garbled sound, as if it's coming from under water. I get the same behavior using either MPC-HC or ZoomPlayer, and the video renderer does not matter. Only the audio renderer matters. VLC plays this file with any audio renderer (the wave out too) just fine. What's interesting is that VLC says the audio is TrueHD 48Khz 32 bit per sample, while Lav audio says it's 24 bit integer per sample. Lav outputs 24 bit integers too. Btw, I have only stereo audio devices which don't know to handle 32 bit output, only up to 24 bit, but this should not matter, right? I mean, even when lav outputs 6 channel 32 bit float with other files, wave out still works.

Lav audio plays without issue with the direct sound renderers, and its status says the same thing, that the audio samples are 24 bit integer, and that it outputs 24 bit integers.

This behavior is not new, it has been like this with the lav filters since forever, but I decided to finally make it known.

[nevcairiel]

Sounds like your downmixing is just not working properly in all cases, for consistent output you could just let LAV downmix to stereo instead of relying on the OS. WaveOut probably relies on the audio driver to downmix, while DirectSound uses the DS audio mixer.
Why do you want to use WaveOut anyway? DirectSound should generally be superior.

[KoD]

Hi nev, the issue is the same when activating downmixing in lav audio to stereo. It's not a downmixing issue, it's a misdetection of the file format issue, or maybe sending the data out to the renderer with bad headers or something. (I was sure you were going to try and get the easy way out by blaming this... how about testing the file first?)

[nevcairiel]

Hi nev, the issue is the same when activating downmixing in lav audio to stereo. It's not a downmixing issue, it's a misdetection of the file format issue, or maybe sending the data out to the renderer with bad headers or something.

Sounds more like your renderer doesn't like 24-bit audio then, which you can also turn off in LAV for testing. All output modes have been well tested over years of usage, and clearly its not a decoding problem otherwise it would sound bad everywhere, and clearly its not a generic 24-bit output problem or it would also sound bad everywhere, so that leaves me to conclude that your audio renderer doesn't like something it produces, that leaves either downmixing or 24-bit. And since WaveOut is controlled by the audio driver, thats usually the case.

This theory is easy to test, just force other files that generally work to 24-bit output using the LAV options, and see if they now also sound bad.

LAV doesn't behave differently depending on what kind of audio renderer it interfaces with, and if the status page is exactly the same as you say, then it also behaves exactly the same. When one audio renderer fails and the other works on the same audio, why do you search the problem with the decoder in the first place?

(I was sure you were going to try and get the easy way out by blaming this...)

The easiest solution is often the correct one.

[KoD]

24 bit audio works perfectly fine with other files. When lav audio is set to downmix to stereo, it outputs 32 bit floats, and it still has the same issue. When I play any other file which is also decoded to 24 bit ints or 32 bit floats, they play fine too. It's only with this file that the sound is not ok.

Again, please test the file. I would have not posted this issue, without testing first.

[nevcairiel]

I tested your file and it plays fine with WaveOut for me.

The last remaining theory is that its TrueHD's fault. TrueHD is "special" in the sense that it has extremely short audio frames, 0.833ms usually (in contrast, DTS is 10.6ms and AC3 is 24ms). These short audio frames can cause some audio renderers to "screw up".
LAV already has some buffering in place to collect a few and output them as one to avoid issues with such audio renderers. However, maybe the limit I set is not strong enough for your WaveOut device, so I could potentially try to increase that a bit, and see if it helps.

[sneaker_ger]

@KoD
Can you test this remux of your file (https://www.sendspace.com/file/1x2c95)? It has the timecodes redone by mkvmerge.

[KoD]

The issue happens no matter what sample format is forced in lav audio: 16, 24, 32 bit integer, or 32 bit float. It's still there. And any other file (also other TrueHD audio muxes) work fine, no matter the output sample format.

Why is VLC saying that the input sample size is 32 bit while lav says it's 24 bit though?
However, maybe the limit I set is not strong enough for your WaveOut device, so I could potentially try to increase that a bit, and see if it helps.
How do I do that?

@KoD
Can you test this remux of your file (https://www.sendspace.com/file/1x2c95)? It has the timecodes redone by mkvmerge.
Thanks for trying to help. Unfortunately, it does not fix the issue. :(

[sneaker_ger]

How do I do that?
You don't, he has to change it in his code.

Actually, he already did. Try the upcoming nightly version.

[KoD]

Ok, I will try it and report back.

[hu1kamania]

I'd be prepared to bet this content is actually shot progressive, but transmitted in interlaced, with the bug added, so it is actually mixed content.

I'm not sure which bug you're referencing, the "Direct3D interop" bug that nevcairiel mentioned, simultaneous rendering limitations of CUDA with Windows 10 or is it related to MadVR? I get the most inconsistent results with MadVR. I'd be curious to know more.

That particular channel plays a lot of older Film based content. Like old westerns, as well as 80's and 90's action movies. I used the watermark in my example because it's good for a quick test.

What it really comes down to, between CUVID, DXVA2, yadif, and w3fdif deinterlacers, is their ability to handle sharp edges. For example, a lot of text on the screen during a commercial will appear to pulse as the deinterlacer tries to sync up, or a characters glasses might appear more aliased, any hard edges really. Motion further exacerbates the issue. I get the best results from CUVID, followed by yadif. It's a very pronounced difference. This is consistent with most content.

If I can figure out how to consistently capture an exact frame, I will submit screenshots between the four.

This might be a case where DXVA2 deinterlacing is not providing the quality that CUVID used to, or I could be totally insane! :p

[huhn]

could share a sample?

[hu1kamania]

could share a sample?

Sure thing :)

[hu1kamania]

could share a sample?

http://www.filedropper.com/sample1_1

Here is a sample where pretty much anything looks better than hw deinterlacing via MadVR. Comparing yadif to CUVID is more difficult with this sample.

Keep in mind, I'm using Windows 10 and CUVID via Nvidia's 350.12 driver.

[nevcairiel]

http://www.filedropper.com/sample1_1

Here is a sample where pretty much anything looks better than hw deinterlacing via MadVR. Comparing yadif to CUVID is more difficult with this sample.

Keep in mind, I'm using Windows 10 and CUVID via Nvidia's 350.12 driver.

That file seems to be telecined, not interlaced.
You need a IVTC, not a deinterlacer.

Disable all deinterlacers in LAV (both CUVID and Software), and set madVR in processing -> deinterlacing to "disable automatic source type detection" -> "force film mode". It should report a constant 3:2 cadence in the OSD and show the video without artifacts.

The artifacts from telecined and interlaced content often look very similar, but in contrast to interlaced, telecined can be fully restored to its original image without any quality loss, if done properly.
Generic deinterlacers are often not capable of doing this, hence dedicated Inverse-Telecine (IVTC) algorithms.

[hu1kamania]

That file seems to be telecined, not interlaced.
You need a IVTC, not a deinterlacer.

Disable all deinterlacers in LAV (both CUVID and Software), and set madVR in processing -> deinterlacing to "disable automatic source type detection" -> "force film mode". It should report a constant 3:2 cadence in the OSD and show the video without artifacts.

The artifacts from telecined and interlaced content often look very similar, but in contrast to interlaced, telecined can be fully restored to its original image without any quality loss, if done properly.
Generic deinterlacers are often not capable of doing this, hence dedicated Inverse-Telecine (IVTC) algorithms.

Thank you for elaborating on the difference between telecined and interlaced. I'll bet you've had to do it a thousand times for the less knowledgeable folks like myself. No doubt you're hitting at the heart of the matter here.

Force Film Mode works fantastic in this specific clip, but this clip is a commercial for a sports program. Golf to be precise. When the program resumes, it becomes jerky, thus it's mixed content right?

CUVID provides IVTC as well, and it seems to provides more consistent results with most content that I view. It's closer to "set it and forget it", when compared to MadVR's auto setting.

This wasn't the case when I used newer drivers in Windows 8.1. At that time CUVID hw deinterlacing was on par with MadVR set to Auto, so CUVID was pointless. Switching to a much older driver; 350.12 in Windows 10, I got better results. 350.12 should work even better in Windows 8.1, but I'm not going back to find out.

As the only two options for IVTC are DXVA2 and CUVID, and CUVID is practically defunct, this might be an issue to take up with madshi. I don't know how much control you guys have over hardware deinterlacing, so bringing DXVA2 in line with older CUVID might not be feasible. If what I'm reporting is even accurate.

[huhn]

maybe there is just a bug in madVR just check EVR for DXVA deinterlancing.

for all of my test CUVID and DXVA2 where identical they even made the same major mistakes.

[hu1kamania]

maybe there is just a bug in madVR just check EVR for DXVA deinterlancing.

for all of my test CUVID and DXVA2 where identical they even made the same major mistakes.

Could be, I'll check it out. I'd be happy to identify a bug.

What sorts of mistakes?

[hu1kamania]

The more I look at it, the more I realize the problem is tied to MadVR's struggle with mixed content. Here is a second sample, a continuation of the last. Here you can see where MadVR falls short, and CUVID stays smooth throughout.

http://www.filedropper.com/sample2

MadVR set to auto (defaults to Video in this case) struggles with the car commercial, but does well with the golf portion. Force it to Film, and the inverse is true. CUVID does well with both. Could this indicate a bug with MadVR?

[huhn]

if EVR looks better yes.

[KoD]

Regarding the issue (http://forum.doom9.org/showthread.php?p=1755407#post1755407) I reported with waveout on that sample with TrueHD audio, I'm happy to report that the build with increased buffer size has done the trick. Now the sample plays fine through both the optical S/PDIF output and through the nvidia hdmi out, while before it failed with both. Thanks for identifying the cause and fixing the issue!

[Manni]

Feature request:

Would it be possible to select a different bitstreaming option for Atmos and Dolby TrueHD and for DTS-HD and DTS:X?

Some AVRs, including all D&M models, are not able to apply DSU to DTS:HD tracks or Neural:X to Dolby TrueHD tracks after the upgrade to the DTS:X f/w. This is a limitation of the implementation.

The only workaround is to output the audio track as LPCM to be able to apply any upmixer, but then of course you have to change that setting again when you want to decode Atmos or DTS:X, as these need to be bitstreamed for the AVR to get the metadata and decode them.

So the solution would be to enable bitstream only for Atmos and DTS:X tracks, and leave them as LPCM for all the other formats, including HD audio tracks.

I hear jRiver is able to do this, so I was wondering if LAV would be able to make this distinction as well, maybe thanks to progress with Mediainfo or other libraries in reporting the presence of metadata in the HD Audio track.

Thanks for considering this!

[nevcairiel]

I have no plans to make bitstreaming any more conditional than it is right now.
I don't really want to need to parse the entire bitstream just to find out if its supposed to be bitstreamed.

[Manni]

I have no plans to make bitstreaming any more conditional than it is right now.
I don't really want to need to parse the entire bitstream just to find out if its supposed to be bitstreamed.

Thanks for the quick reply.

I understand but why would you need to parse the whole file? Can't you use the reporting from mediainfo? It supports Atmos since December (not sure about DTS:X).

I was only suggesting to split the bitstreaming options (so for example to be able to select different options for Atmos and TrueHD), not for you to add any logic, as different users will want different things depending on their AVRs capabilities/limitations.

[nevcairiel]

I also don't want to add a dependency on some external library just for some obscure feature.

[Manni]

I also don't want to add a dependency on some external library just for some obscure feature.

Thanks, I understand.

[Manni]

One last question, I don't think LAV supports keyboard shortcuts? A shortcut to enable/disable bitstreaming globally would be a fine solution.
I wonder how iRiver and Mediainfo detect Atmos vs TrueHD. I guess they do parse the file, as Mediainfo also reports the bitrate.
By the way it's still obscure now because the DTS:X update only landed on the 28th of January for the first AVR supporting it (the Denon X7200WA).
It will get worse as other models are upgraded and users lose the ability to cross-upmix between formats, especially if other manufacturers like Yamaha have the same limitation.
Of course if you have no interest in immersive audio I can see why there is no motivation to address this.
Again, thanks for the time you took to consider this. I won't ask again, promise :).

[nevcairiel]

Maybe you should address this loss of existing functionality with the manufacturers instead.

[amayra]

Q: Will you write a Subtitle Filter?
A: Its possible, nothing has been determined yet.

can you give me a % for chance you make a VSFilter (or XySubFilter) alternative

[Manni]

Maybe you should address this loss of existing functionality with the manufacturers instead.

I'm not asking you to fix anything for the manufacturers, I'm asking for a feature. It is your right not to be willing to implement it. Again, thanks. I won't ask again. :)

[hu1kamania]

if EVR looks better yes.

Not surprisingly, EVR reacted the same as MadVR set to auto. With the overlay up, there is a clear distinction between CUVID and DXVA2. Here are some screenshots illustrating this point:

CUVID1 (http://i.imgur.com/sITtTsG.jpg)
CUVID2 (http://i.imgur.com/HXJifIf.jpg)
DXVA2_1 (http://i.imgur.com/8Ds92FN.jpg)
DXVA2_2 (http://i.imgur.com/3iinuWY.jpg)

Notice the difference between peaks? There is something going on here. :confused:

[nevcairiel]

Not surprisingly, EVR reacted the same as MadVR set to auto. With the overlay up, there is a clear distinction between CUVID and DXVA2. Here are some screenshots illustrating this point:

CUVID1 (http://i.imgur.com/sITtTsG.jpg)
CUVID2 (http://i.imgur.com/HXJifIf.jpg)
DXVA2_1 (http://i.imgur.com/8Ds92FN.jpg)
DXVA2_2 (http://i.imgur.com/3iinuWY.jpg)

Notice the difference between peaks? There is something going on here. :confused:

That is irrelevant, its only the difference of sending 29.97 or 59.94 to the renderer (before and after deinterlacing), it makes frame timing different and therefor the graph.

[huhn]

it's not the same frame quality can't be judged this way.

[hu1kamania]

That is irrelevant, its only the difference of sending 29.97 or 59.94 to the renderer (before and after deinterlacing), it makes frame timing different and therefor the graph.

This can been viewed in each screenshot, at the top as Frame Rate, correct?

I don't know if this is getting into the realm of a MadVR feature request, and MadVR design philosophy but, I don't to have an option in EVR or MadVR of sending 59.94. Perhaps that is a bad idea?

The best I can do is enable yadif to 50p/60p or disable MadVR's deinterlacing and enable CUVID at 50p/60p.

[huhn]

this part said it get's 29i so it is deinterlanced to 59p. so everything is working as intended.

CUVID fails hard on your first screen BTW.

[hu1kamania]

this part said it get's 29i so it is deinterlanced to 59p. so everything is working as intended.

CUVID fails hard on your first screen BTW.

Are you talking about the logo? I see that too. These screenshots are not taken at the exact same frame. In the CUVID sample the scene is beginning to fade. That said, I wouldn't be surprised if there was some visual differences between the two decoders.

Everything is working as intended, and yet MadVR/EVR struggles to render that sample smoothly from start to finish. CUVID or yadif at 50/60p provides a more consistent experience.

[baii]

Just tried out w3fdif, to me it seems that it definitely have less artifact(horizontal ziggy line thingy) for motion vs HW(AMD direction or vector adaptive), but it kind of have the "soap opera" effect like a real TV. I guess I just got used to HW deinterlace, vs watching real TV.

[huhn]

This can been viewed in each screenshot, at the top as Frame Rate, correct?

I don't know if this is getting into the realm of a MadVR feature request, and MadVR design philosophy but, I don't to have an option in EVR or MadVR of sending 59.94. Perhaps that is a bad idea?

The best I can do is enable yadif to 50p/60p or disable MadVR's deinterlacing and enable CUVID at 50p/60p.

the stats are meaningless they show no delayed, dropped or reapeated frame so everything is fine in the stats.

[hu1kamania]

the stats are meaningless they show no delayed, dropped or reapeated frame so everything is fine in the stats.

Correct, I misunderstood what was happening. With CUVID and yadif deinterlacing it gets converted before it reaches MadVR/EVR. Without CUVID hw deinterlacing or yadif, it happens with MadVR or EVR.

My sample from start to finish plays back more smoothly without MadVR or EVR doing the deinterlacing. Is this by design?

My understanding of what I'm seeing is evolving, thank you for your patience as I work through this.

[huhn]

deinterlancing quality of telecine content is so bad for me right now that there has to be a bug in the way how madVR, EVR or my system is using DXVA2 deinterlancing.

nvidia DXVA2 deinterlancing should have a field matching algorithm but that is clearly not working.

it is currently w3fdif for interlanced and madVR film mode for 3:2 telecine.

[hu1kamania]

deinterlancing quality of telecine content is so bad for me right now that there has to be a bug in the way how madVR, EVR or my system is using DXVA2 deinterlancing.

nvidia DXVA2 deinterlancing should have a field matching algorithm but that is clearly not working.

it is currently w3fdif for interlanced and madVR film mode for 3:2 telecine.

This must be the case. Do you show the same results with AMD?

I think Nvidia was on the right track with CUVID, but it's in the past now. I like the switch to DXVA2 but I believe something was lost somewhere along the way.

Barring an unlikely update, CUVID is dead to me now. I'll be moving back to DXVA copy-back, and I'll suffer along with you.

I'm confident a solution will surface someday. The knowledge displayed here might not show it, but I've been following the development of video play back for close to fifteen years. I'll continue to follow the work of people like nevcairiel and madshi. The work they do is inspiring.

Future concerns with DXVA2 deinterlacing will be directed towards MadVR. Thanks for your patience huhn! I've enjoyed the journey. ;)

[huhn]

deinterlancing is totally broken with crimson.

and because nvidia deinterlancing isn't working for me doesn't mean it isn't working in general and it worked fine on interlanced content.

[unknownsoldierX]

Win 7 x64
AMD HD2400 card

I can't get DXVA to work for VC-1/WMV3 in MPC-HC. Going by this (https://trac.mpc-hc.org/ticket/2518), the internal filters do not work with this card. I was using ffdshow tryouts, and I tried disabling the internal wmv3 filter and adding ffdshow as an external, but that still didn't give me DXVA.

Now I have LAV set as the external video and audio filters. LAV and MPC-HC are both set to DXVA2 (copy-back). When I right click on playing video and check filters, it shows that it is using the LAV video decoder, but I can tell it's definitely not using DXVA.

[nevcairiel]

VC1/WMV3 is not supported on old AMD cards like that one, sorry.

[NikosD]

Win 7 x64
AMD HD2400 card

I can't get DXVA to work for VC-1/WMV3 in MPC-HC.

Interesting card, I copy from Wikipedia:


Several reports from owners of HD 2400 Pro suggest the card do not fully support hardware decoding for all H.264/VC-1 videos.
The device driver, even with the latest stable version, seem to only honor hardware decoding for formats specified in the Blu-ray and HD-DVD specification.
As a result of such restriction, the card is not deemed very useful for hardware video decoding since the majority of the H.264/VC-1 videos on the net are not encoded in those formats (even though the hardware itself is fully capable of doing such decoding work).
This device driver restriction has led to the development of a third party driver patch, "ExDeus ATI HD Registry Tweak", to unlock the potential of HD 2400 Pro for full support of H.264/VC-1 hardware video decoding.


The link for that "ExDeus ATI HD Registry Tweak" are:

http://home.comcast.net/~exdeus/ati-hd2x00/

http://www.avsforum.com/t/872538/ati-radeon-hd-2x00-2400-2600-2900-series-owners-thread/2010#post_11622510

http://www.avsforum.com/t/872538/ati-radeon-hd-2x00-2400-2600-2900-series-owners-thread/2100#post_11659897

Also, you could try MPC-HC v1.6.9 or older.

[omarank]

With the latest LAV nightly, DTS/ DTS-HD audio is broken when played in MPC-HC 32 bit. The problem does not manifest in MPC-HC 64 bit and JRiver MC. By broken, I mean the sound is all garbled.

I am using LAV filters as external filters in MPC-HC. When the internal filters are used in the latest nightly of MPC-HC 32 bit (v1.7.10.101.x86), there is no problem.

[nevcairiel]

Everything works perfectly fine here.
If you are using the Internal Audio Renderer in MPC-HC, I would personally still advise against it, I can easily make it break in various ways.

In any case you need to provide more details. MPC-HC 32-bit is usually the platform I test the most on and generally also develop on, so if that would be broken, I would hope to notice.

[omarank]

Ok, I did more testing and found that the problem is manifested only in WASAPI exclusive output. When I use the Internal Audio Renderer, the sound is garbled with WASAPI excl output. When I use Reclock, I get all noise (static) with WASAPI excl output. DirectSound and WaveOut output work fine.

In case of JRiver Media Center, both the Event Style and Non Event Style WASPI excl output work perfectly.

I was using v0.67.0-66 build before upgrading to v0.67.0-72. I tried reverting to the older build. There is no such problem with that build.

Please let me know what other information I need to provide to help you reproduce/ diagnose the problem.

[nevcairiel]

You could try the other intermediate nightly versions to check when it started specifically, like, does it already happen with 0.67.0-70 from two days ago?
Personally I don't see why or how it should have broken, in particular only with exclusive mode, there is nothing special to that mode for an audio decoder. It just hands out audio.

Also, do all files break with DTS? Just some? What does that LAV Audio status page say when a broken file is playing? Did it say something different in a previous version?

[GCRaistlin]

How do I enable new deinterlacing engine? I see only "YADIF" in the properties.

[sneaker_ger]

What build are you using? You need a recent nightly version to test the new deinterlacer.

[omarank]

You could try the other intermediate nightly versions to check when it started specifically, like, does it already happen with 0.67.0-70 from two days ago?
Personally I don't see why or how it should have broken, in particular only with exclusive mode, there is nothing special to that mode for an audio decoder. It just hands out audio.

Also, do all files break with DTS? Just some? What does that LAV Audio status page say when a broken file is playing? Did it say something different in a previous version?
Just checked v0.67.0-70. The problem does not happen with that build.

Yes all videos with DTS/ DTS-HD track exhibit the problem. LAV audio status page shows the input audio codec correctly – whichever DTS track is there. For example, Input: dts-hd ma; 6 / 0x60f; 48000, 16bit Integer; Output: PCM; 2 / 0x3; 48000; 32bit Float.

Just observed one more thing – the problem happens in seeking only. If I let a movie play from the starting, there is no problem.

When the audio is garbled, video plays in slow motion.

[GCRaistlin]

sneaker_ger
Sorry, my fault. It's available in the latest MPC-HC nightly.

[nevcairiel]

Just checked v0.67.0-70. The problem does not happen with that build.

Yes all videos with DTS/ DTS-HD track exhibit the problem. LAV audio status page shows the input audio codec correctly – whichever DTS track is there. For example, Input: dts-hd ma; 6 / 0x60f; 48000, 16bit Integer; Output: PCM; 2 / 0x3; 48000; 32bit Float.

Just observed one more thing – the problem happens in seeking only. If I let a movie play from the starting, there is no problem.

When the audio is garbled, video plays in slow motion.

I found the problem, it was a rather weird one. Will be fixed in the next nightly.

[omarank]

I can confirm that the problem is fixed in the build v0.67.0-75. Thanks!

[Leader]

Hi, nevcairiel!

I found a bug in "LAV Video Decoder".

Incorrect playback this file: https://torrentreactor.com/torrents/6197645/APC-Princess-Mononoke-DVD-x264-07f4b3d1-ogm in LAV Video Decoder (DXVA2 decoding and software decoding).

[mixus]

can you add hevc multithread?

[nevcairiel]

can you add hevc multithread?

HEVC is already multithreaded.

[mixus]

Not Write HEVC.
I wrote this reason.

http://i.hizliresim.com/jV71jW.png

Sorry for my bad english :(

[LigH]

^ Apparently he only forgot to add this format in the tooltip text.

[nevcairiel]

I don't even remember there is such a list in the tooltip, I should probably just remove it, its quite out of date.

[Reign]

I have ran into a problem with the CUVID HW decoding for HEVC, while it has always worked for me a video I downloaded recently causes MPC-HC crash here is a dump I received. https://drdump.com/UploadedReport.aspx?DumpID=6648438 sample video: http://www.nyaa.se/?page=view&tid=779737&showcomments=52aa3f1#c538731

Win 7 x64 sp1
MPC-HC x64 1.7.10
LAV x64 0.67.0
NVidia GTX 970

[nevcairiel]

I have ran into a problem with the CUVID HW decoding for HEVC, while it has always worked for me a video I downloaded recently causes MPC-HC crash here is a dump I received. https://drdump.com/UploadedReport.aspx?DumpID=6648438 sample video: http://www.nyaa.se/?page=view&tid=779737&showcomments=52aa3f1#c538731

Win 7 x64 sp1
MPC-HC x64 1.7.10
LAV x64 0.67.0
NVidia GTX 970

This issue is already fixed in the next version.

[Leader]

Hi, nevcairiel!

I found a bug in "LAV Video Decoder".

Incorrect playback this file: copyright violation removed. in LAV Video Decoder (DXVA2 decoding and software decoding).

nevcairiel, I've updated the link to the torrent file, so you can download it without any problems.

Link: copyright violation removed

With regard to the problem: during playback of the file, there are frames missings and deviations, as well as video frames played unsmooth and intermittently.

[nevcairiel]

OGM is a broken format, I wouldn't expect many improvements, sorry.

[DragonQ]

As of 0.67.0-67, LAV Video supports a new software deinterlacer, Weston Three Field Deinterlacing Filter (w3fdif)

w3fdif was developed by the BBC, and as I understand it, is also used by them in real-world usage.
Well I wonder where they actually use this w3fdif filter because their deinterlacing method when doing slow/stop motion during sports replays is awful - it's basically weave. Even their method of adding real-time graphics to interlaced video doesn't look great and produces a tonne of shimmering and other artefacts.

I assume, and hope, the w3fdif filter is different.

[Alexey1975]

I have tested all of betas in 0.67 series and catch such an issue with some of sources.

http://f19.ifotki.info/thumb/7c92db627144a1abc49403726e4052a325358f237498156.png (http://i-fotki.info/19/7c92db627144a1abc49403726e4052a325358f237498156.png.html)
704x576 AVC (High@L3 CABAC / 4 Ref Frames)

Release version LAV 0.67 just fine, but beta versions are buggy.

[nevcairiel]

As always, please provide a sample, otherwise there is nothing I can do.

[Alexey1975]

Excuse me, here it is http://www.ex.ua/98151805 the stream dump.
(DVBViewer 5.5.2.0 + MadVR 0.90.4 + LAV 0.67.0.76)

[nevcairiel]

Sorry, but your link doesn't work.

[Alexey1975]

Sorry, try this link https://yadi.sk/i/RfCr6RnMoSDhL

[nevcairiel]

Sorry, try this link https://yadi.sk/i/RfCr6RnMoSDhL

The file seems to decode fine, which decode mode do you use?

[Alexey1975]

Strange thing. I tried CUVID, DXVA-CB, DXVA-N modes with the same result...
Third-party decoders produce no any issues (just like release version LAV 0.67) with that stream, but 0.67-betas do!

[nevcairiel]

It breaks only in DXVA2 mode apparently. I get the same issue with 0.67.0 however, and with 0.66 and 0.65.
Same problem with Microsofts DXVA decoder.

Seems like either the stream is not compatible with DXVA for some reason, or its a hardware/driver issue.

[Alexey1975]

Just right now I rollback to the release version and it works just fine. I hope this strange things don't float up into the next release - I just want to help the beta testing. Thank you for attention.

[nevcairiel]

From what I can tell it never worked with DXVA2, same result in all recent release versions.
And like I said in my previous post, same result using the Microsoft decoder as well.

Software Decoding is perfectly fine however.

[Alexey1975]

It breaks only in DXVA2 mode apparently. I get the same issue with 0.67.0 however, and with 0.66 and 0.65.
Same problem with Microsofts DXVA decoder.

Seems like either the stream is not compatible with DXVA for some reason, or its a hardware/driver issue.

Microsofts DXVA2 decoder and etc. works just fine with this stream - i had no any problem before that betas.

[Alexey1975]

Software Decoding is perfectly fine however.

Confirm!

[nevcairiel]

Microsofts DXVA2 decoder and etc. works just fine with this stream - i had no any problem before that betas.

I can only tell you what I am seeing, and that is that its broken with DXVA everywhere. I tested 0.64 to 0.67, and the Microsoft decoder, and all behave *exactly* the same, so from my end, its not a new issue, and considering the Microsoft decoder looks just the same, it smells like a driver problem.

One simple explanation would be that earlier LAV versions for some reason just didn't use DXVA for you, because Software decoding works.

In any case, unless I can find one DXVA decoder that actually works, I have no proof that its actually supposed to work, and no starting point where to start poking.

[DragonQ]

Well I wonder where they actually use this w3fdif filter because their deinterlacing method when doing slow/stop motion during sports replays is awful - it's basically weave. Even their method of adding real-time graphics to interlaced video doesn't look great and produces a tonne of shimmering and other artefacts.

I assume, and hope, the w3fdif filter is different.

Just had a quick check of this with some 576i/25 footage. YADIF definitely handles panning shots with diagonal lines (e.g. painted lines on a pitch) better. I was gonna compare to AMD's Adaptive Sync but it looks very broken with Crimson drivers so I can't...

[wanezhiling]

@Alexey1975, nevcairiel
That is Nvidia's hardware problem, ATI&Intel are just fine.

similar clip (https://forums.geforce.com/default/topic/908792/artifacts-with-dxva-decoding/)

[hoborg]

Hello.
Intel (HD6200) also working just fine.

[NikosD]

@Alexey1975, nevcairiel
That is Nvidia's hardware problem, ATI&Intel are just fine.

similar clip (https://forums.geforce.com/default/topic/908792/artifacts-with-dxva-decoding/)

Both files start with artifacts or black, frozen image but the video stream after a while is decoded just fine in DXVA mode using Win 10 - iGPU 4600 - Drivers 4331.

One thing that differentiates those two interlaced H.264 files, from all the "common" files that work fine using Nvidia HW decoder is this according to MediaInfo:

Color primaries : BT.601 PAL
Transfer characteristics : BT.470 System B, BT.470 System G
Matrix coefficients : BT.601

My other H.264 interlaced samples are BT.709, but I really don't know if that means something.

[nevcairiel]

One thing that differentiates those two interlaced H.264 files, from all the "common" files that work fine using Nvidia HW decoder is this according to MediaInfo:

Color primaries : BT.601 PAL
Transfer characteristics : BT.470 System B, BT.470 System G
Matrix coefficients : BT.601

My other H.264 interlaced samples are BT.709, but I really don't know if that means something.

That info is just metadata, its probably something about that specific way to encode interlaced.

[NikosD]

Yes, maybe that particular way is incompatible with Nvidia's HW decoder.

[huhn]

it there any reason it should work with quicksync and CUIVD but not with DXVA?

[nevcairiel]

it there any reason it should work with quicksync and CUIVD but not with DXVA?

QuickSync is not NVIDIA, so there is that.

And it only works with CUVID on Windows 10 because that doesn't go through DXVA, on earlier Windows versions it goes through DXVA and shows the exact same result.
Its probably a driver issue and not even hardware, but who knows if they'll fix anything.

[huhn]

And it only works with CUVID on Windows 10 because that doesn't go through DXVA, on earlier Windows versions it goes through DXVA and shows the exact same result.

that make a lot more sense to me now.

[Alexey1975]

Here is the my assumption:

The fact that NVIDIA has some specific problems with DXVA is clear, but CUVID engine is not identical to the DXVA.
Here the experiment:
1. When it comes thru the CUVID tract - all just fine.
(pay attention to the log)
http://f19.ifotki.info/thumb/84af893261bff3040a475792875b35e525358f237554083.png (http://i-fotki.info/19/84af893261bff3040a475792875b35e525358f237554083.png.html)

2. When it comes thru the DXVA enviroment it cause a problems. And pay attention to the log now!
http://f19.ifotki.info/thumb/179d5306189d28964ee40e54163dbc6525358f237554105.png (http://i-fotki.info/19/179d5306189d28964ee40e54163dbc6525358f237554105.png.html)
http://f19.ifotki.info/thumb/be00410a8153352f56f99f1afdd7356825358f237554112.png (http://i-fotki.info/19/be00410a8153352f56f99f1afdd7356825358f237554112.png.html)

The fact that the video streams were processed by the graphics card is evidenced by its load levels. So we have not software but hardware decoding in all of the cases.

3. Now check the results of the latest LAV beta:
http://f19.ifotki.info/thumb/00d597fc27b9ad2aab79a4933bf2856525358f237554118.png (http://i-fotki.info/19/00d597fc27b9ad2aab79a4933bf2856525358f237554118.png.html)

At latest betas the CUVID mode calls or passes through the DXVA enviroment instructions and that begets the inconsistency.

[romulous]

Hi nev,

When the next stable version rolls around, will there be an installer available that includes the new MVC dll's that does not require an internet connection? So the dll's will be included in the installer rather than having them downloaded I mean.

Thanks,

romulous

[nevcairiel]

No, having them downloaded is the intended solution since they are quite big and most people probably won't install them.

[Manni]

I found a solution to switch bitstreaming on/off on the fly in LAV in order to resolve the problem of limited cross-upmixing with some AVRs after the DTS:X upgrade, so I thought I'd share it in case others experience the issue. I created a new thread to not clutter this one and make it easier to find the workaround.
Here is the link: http://forum.doom9.org/showthread.php?p=1756727#post1756727

[4h4h270]

When I using latest beta to play MVC 3D with madvr on framepacked 3D mode, after 1hr10min play I will get massive frame drop.
It happened by abnormal cpu use, I'm not sure if it's intel's dll or lavfilter's problem :(

[madshi]

When I using latest beta to play MVC 3D with madvr on framepacked 3D mode, after 1hr10min play I will get massive frame drop.
It happened by abnormal cpu use, I'm not sure if it's intel's dll or lavfilter's problem :(
Can you please check if maybe the process is running out of RAM or something? Are you using an x64 or x86 media player? Try x64, just as a test.

[4h4h270]

Can you please check if maybe the process is running out of RAM or something? Are you using an x64 or x86 media player? Try x64, just as a test.

:O I'll try x64, still using x86 because of reclock.

Thx madshi, I never notice the RAM problem, I tried x64 no more frame drop. :D

[scollaco]

I'm a relative newbie to all this and trying my best to learn.

I have a new Intel Skylake NUC (NUC6i5syk), Using MPC-BE x64, LAV x64 and MadVR (all latest nightlies)

My question is which decoder setting is the best for me? I usually play some 720p, mostly all 1080p and sometimes 4K 8 bit. I get dropped frames with 4K 10 bit (I know my Skylake doesn't support hardware accel with 10 bit).

So in general what would be my best choice to get the most efficiency?

What do I choose in LAV?

None
CUVID
Intel Quicksync
DXVA (Copy Back)
DXVA native

thanks for the help

[DragonQ]

DXVA Native should always be the most efficient. If DXVA Copy Back works that is probably a better choice since it's a bit more flexible.

[scollaco]

DXVA Native should always be the most efficient. If DXVA Copy Back works that is probably a better choice since it's a bit more flexible.

Got it. Both DXVA native and copy back work fine for me. So I guess it's up to me to choose one of them knowing DXVA copy back is more flexible with different setups.

So at least I know to ignore Intel Quicksync, CUVID and none.

[Patrik G]

this is probably a noob question but i have downloaded the latest LAV nightly build for use in MPC-HC
just to test some HDR demos with madvr

but when i start MPC-HC it still uses the build in lav filters
how do i activate the latests lav filters in mpc-hc?

Edit:
problem solved i think
just added the latest lav decoder to the external filters list.
now mpc-hc uses it

[LigH]

The other way would probably be to exchange the files in MPC-HC\LAVFilters, these are the "internal" copy which MPC-HC prefers over system-wide installed DirectShow filters if you don't disable internal filters for a specific format.

[Patrik G]

About HDR playback
is it more to include with the latest nightly builds for HDR metadata that is being sent to madvr or is it complete?
Are all HDR metadata being sent?
HDR peakbrightness and colorgamut (BT2020) are there already but what about gamma?

just watched Exodus and life of Pi HDR demos on a 8 year old KRP-500M plasma monitor with peakbrightness at 220cd/m2 and 93% of DCI P3 gamut
it looked amazing.
im going to compare the regular blu ray version of that movie later.

[nevcairiel]

Are all HDR metadata being sent?
HDR peakbrightness and colorgamut (BT2020) are there already but what about gamma?

The Gamma of HDR content is being defined by SMPTE ST 2084, and the required min/max luminance is supplied as metadata.

So yes, all required metadata is provided to madVR to render HDR content properly.

[Patrik G]

So yes, all required metadata is provided to madVR to render HDR content properly.

thats amazing!
thanks for the hard work it really pays off even on tvs with less peak brightness and no HDR support

so if you put it this way
if you have a tv with 500cd/m2 peakbrightness and 100% DCI P3 coverage but no natvie HDR support.
will this tv output the same HDR performance as a "new" tv with the same specs but with native HDR support?

it clearly is a big difference with the HDR version compared to the Blu Ray
much more details both near black and near white
better and more natural looking colors.

i took two photos here
one with the HDR version of Exodus and one from the regular Blu Ray
its not quite a fair comparsion because the crappy camera still crushes blacks.
http://privat.bahnhof.se/wb192876/500M%20HDR%20COMPARSION.jpg
you can see that the HDR version has more details and better colors.
colors with the wider colorgamut on the 500M with the regular blu ray just looks oversaturated and cartoon.

when watching these two versions its like comparing a Full frame DSLR camera to a crop sensor camera.
the image from the HDR version is much more lifelike.
just when you watcing photos taken with a full frame DSLR.

just waiting for more HDR content to be available :)

[nevcairiel]

thats amazing!
if you have a tv with 500cd/m2 peakbrightness and 100% DCI P3 coverage but no natvie HDR support.
will this tv output the same HDR performance as a "new" tv with the same specs but with native HDR support?


A HDR TV can of course show a much higher dynamic range, since the panels are designed for a much higher range of brightness.
How this will be usable from a HTPC is a question we won't be able to answer until we developers have access to such TVs however.

[Patrik G]

A HDR TV can of course show a much higher dynamic range, since the panels are designed for a much higher range of brightness.


yes but HDR is not all about peak brightness isnt it?
the blacklevel is a more important factor to gain dynamic range.

lower the blacks twice and you have the same dynamic range as a tv with doubled peak brightness but higher blacklevel.

also high peakbrightness is not important if the panel cant show all steps from black to white needed for HDR right.
contrast ratio is one thing and dynamic range is another.

OLED tvs stil lacks "dynamic range" compared to the Kuros.
they just cant show all levels properly near black.

last
do you really need 1000 nits watching in a dim room.?
with sunglasses probably lol

[nevcairiel]

yes but HDR is not all about peak brightness isnt it?
the blacklevel is a more important factor to gain dynamic range.


That depends, any ordinary living room setup won't really benefit much from better black levels because of ambient light, many of todays TVs can easily reach values good enough for such usage, in such a setup moving the dynamic range up with higher peak brightness would be advantageous. If you have a pitch-black batcave, thats another topic, but the majority of consumers do not.

In any case, proper HDR TVs will show you more dynamic range than older TV models. How obvious that is we'll have to wait and see.

This does not really belong into the LAV thread though, if you want to continue to discuss the impact of HDR please consider opening a dedicated thread for that topic.

[kolak]

OLED tvs stil lacks "dynamic range" compared to the Kuros.
they just cant show all levels properly near black.

last
do you really need 1000 nits watching in a dim room.?
with sunglasses probably lol

Consumer OLEDs yes, but not Sony 4K OLED reference monitor. This blows everything on the planet (except prototype laser projectors) and it also does 1000 nits.
Most if not all HDR current titles are mastered on this Sony monitor (+ consumer TV on the side to see how it translate to consumer world).
1000 nits is only for peak (just some scenes or none depending on the content), where average brightness will be way lower, probably <500 nits.
HDR is nothing that special at all- it just brings display technology closer to real world and away from old and very restricted 100 nits mastering. This has been used for bit to long.

[nevcairiel]

There is a new nightly available in the usual place which adds support for demuxing H.264 MVC from Blu-ray SSIF files
https://files.1f0.de/lavf/nightly/

You'll need 0.67-92 or newer.
Bare in mind that its an early version, so it may not function properly on all files/discs yet.

This version does not yet read the eye-order metadata from the Blu-ray playlists, nor does it support opening a Blu-ray playlist directly - only straight SSIF files are supported so far.
It is however a big step towards achieving those missing features, as interleaving the two views into one for the decoder is an important step.

My current plan is to get direct Blu-ray demuxing to work as well before the next release, and hopefully soon as well, so that a new version can be officially released soon.
Lets hope my current plan to integrate this into the libbluray flow works out.

[madshi]

Awesome!! :)

:thanks:

[SamuriHL]

That's pretty sweet. Thanks, Nev!

[FDisk80]

There is a new nightly available in the usual place which adds support for demuxing H.264 MVC from Blu-ray SSIF files
https://files.1f0.de/lavf/nightly/

You'll need 0.67-92 or newer.
Bare in mind that its an early version, so it may not function properly on all files/discs yet.

This version does not yet read the eye-order metadata from the Blu-ray playlists, nor does it support opening a Blu-ray playlist directly - only straight SSIF files are supported so far.
It is however a big step towards achieving those missing features, as interleaving the two views into one for the decoder is an important step.

My current plan is to get direct Blu-ray demuxing to work as well before the next release, and hopefully soon as well, so that a new version can be officially released soon.
Lets hope my current plan to integrate this into the libbluray flow works out.

Awesome! :D Can't wait.

[SamuriHL]

Just tried it out and it doesn't seem to be working for me. Since MC21 doesn't open ssif files, I used MPC-HC to open the ssif off a mounted BD ISO. It's playing back in 2D and LAV video is showing avcodec as the active decoder. Opening an MKV with the same setup works just fine.

[nevcairiel]

MPC-HC is doing something evil and if you try to open any file from a Blu-ray (mounted or actual disc) it opens the main playlist instead.
Copying it off the disc first works, even as full structure, as long as its not at the root of a drive.

[SamuriHL]

Oh joy. Ok, I'll give that a shot when I get a moment.

[SamuriHL]

Yea, that worked. SSIF plays fine in 3D just like the MKV. Very nice!

[nevcairiel]

I also just pushed support for demuxing the H.264 MVC content directly from the Blu-ray using LAVs Blu-ray demuxer, so you can open it for title playback like any 2D Blu-ray.
It'll be in the next nightly (in about ~5 hours).

As the SSIF change before, its in a very early state and I do not know how well it works across a wide variety of discs. I only tested 3 or so which I had sitting around, although one of those uses seamless branching, so thats at least tested.
It will also export the eye-order metadata from the Blu-ray now if its available, so madVR can switch the eyes if needed.

Seeking is a bit complicated because it has to sync the main Blu-ray stream and the MVC extension stream which comes from a different file, but it seems to work decently fast in my tests (off of a HDD, not an optical drive, though).

It might be important to note that every 3D Blu-ray automatically gets this treatment, although due to how H264 works any decoder unaware of MVC support *should* just ignore the extra data and decode the 2D image like before.
At least that would be the hope.

One of these days I should sit down and actually watch an entire movie in 3D to test this more thoroughly.

[clsid]

You forgot to push the latest libbluray changes.

[nevcairiel]

Its all there now.

[madshi]

Wonderful, that actually sounds better than I had expected! So you had no trouble extending libbluray for 3D?

Is there anything still missing for 3D Blu-Ray support? Only thing coming to my mind right now would be subtitle specific 3D stuff (I think there is an SEI for that).

[clsid]

Just tested with a mounted 3D Bluray ISO and it is working great. Awesome work!

[nevcairiel]

Wonderful, that actually sounds better than I had expected! So you had no trouble extending libbluray for 3D?
I didn't extend it really, just use it to parse the metadata (only extended it slightly to expose more metadata).
3D Blu-rays are a bit funny. The .ssif files aren't actually the ones you are supposed to use for 3D playback, instead there is two m2ts files, one with the 2D Base, Audio and subtitles, and another one with just the MVC extension.

The SSIF is created through some magic on the file system level by combining the sectors of those two files into a virtual file (which makes a "naive" rip of such a disc twice as large)

So I just let libbluray read the 2D file like it usually does, and open the 3D file manually on the side and sync up the MVC frames to the base view.
Luckily libbluray sends me events when it switches clips, so I can also switch to the next 3D clip on seamless branching titles. Works surprisingly well!


Is there anything still missing for 3D Blu-Ray support? Only thing coming to my mind right now would be subtitle specific 3D stuff (I think there is an SEI for that).

I haven't looked into subtitles much yet, but from what I briefly skimmed over its a combination of BD metadata (ie. defining the available planes) with a combination of SEI (assigning planes to events)

[madshi]

3D Blu-rays are a bit funny. The .ssif files aren't actually the ones you are supposed to use for 3D playback, instead there is two m2ts files, one with the 2D Base, Audio and subtitles, and another one with just the MVC extension.
If you're not supposed to use the SSIF files, whey do they exist? Not important, just wondering...

So I just let libbluray read the 2D file like it usually does, and open the 3D file manually on the side and sync up the MVC frames to the base view.
Luckily libbluray sends me events when it switches clips, so I can also switch to the next 3D clip on seamless branching titles. Works surprisingly well!
Sounds good!

I haven't looked into subtitles much yet, but from what I briefly skimmed over its a combination of BD metadata (ie. defining the available planes) with a combination of SEI (assigning planes to events)
Subtitles aside, is there anything else missing for 3D Blu-Ray? Nothing coming to my mind atm.

[nevcairiel]

If you're not supposed to use the SSIF files, whey do they exist? Not important, just wondering...

Who knows. Maybe they are supposed to be used under some circumstances which I didn't find yet. But the playlist has an extension link to the MVC-only clips which lets me identify them directly and just open them in parallel. Doing it this way is so much easier - but it does require two MPEG-TS demuxers to run, maybe hardware doesn't like that, who knows.

[Casshern]

Who knows. Maybe they are supposed to be used under some circumstances which I didn't find yet. But the playlist has an extension link to the MVC-only clips which lets me identify them directly and just open them in parallel. Doing it this way is so much easier - but it does require two MPEG-TS demuxers to run, maybe hardware doesn't like that, who knows.

Just speculatiing here but maybe it's because optical drives (remember that blurays are supposed to be on optical disks) have absymal seek times. Reading two files on different locations on the disks is extremly slow. So they interleave the two m2ts files sector wise. The SSIF file might be a view to those (combined) interleaved files and makes seeking and sequential reading much easier. So i guess that by parsing the SSIF and reading from the ssif file may provide a means to read 3d blurays from optical drives efficiently..... both views are probably required - as opposed to ssif only - to provide backward compatibility for old players for 2d playback

[IanD]

Its all there now.

Can someone briefly document the requirements and limitations for playing Bluray 3D?

I gather it requires the latest MadVR and LAV filters, but can't seem to find whether it will work with XP and if using an Nvidia GT730 for example, whether Cuvid, DXVA or DXVA copyback should be selected or whether it works from a mounted ISO or only files copied to HDD.

[nevcairiel]

XP is not supported. You can safely assume that any new advanced features will generally not work on XP.
It does not use any kind of hardware acceleration, so it makes no difference what you select.

More specifically, the HDMI 1.4 3D Output through madVR requires Windows 8 or newer, although it could output SBS/TB or some interleaved format on earlier versions of Windows as well, at a loss of resolution.
Decoding will likely not work on XP however.

[IanD]

XP is not supported. You can safely assume that any new advanced features will generally not work on XP.
It does not use any kind of hardware acceleration, so it makes no difference what you select.

More specifically, the HDMI 1.4 3D Output through madVR requires Windows 8 or newer, although it could output SBS/TB or some interleaved format on earlier versions of Windows as well, at a loss of resolution.
Decoding will likely not work on XP however.
Thanks for the clarification.

Interestingly, Stereoscopic Player (Trial) works for me under XP to decode 3D SSIF and output as SBS/OU/interlaced/anaglyph. I can't test frame packing output as I don't have a suitable 3D TV yet, but even if that doesn't work, interlaced might still be useful for LG passive 3D TVs. You seem to suggest that MadVR could output interlaced, so I guess I should ask there, but I couldn't find any options when I tried the latest version recently (albeit hurriedly) with the latest nightly LAV.

Would be great to have this working under XP, even if limited in output, from open source.

[clsid]

The limitations are not directly in LAV itself, but in the Intel Media SDK that it uses for decoding MVC video. That most likely requires Vista+ and SSE2 capable CPU.

[nevcairiel]

Indeed, the decoder is not actually opensource, and it works where it works, but historically Intels MediaSDK libraries have not worked on XP.

[IanD]

The limitations are not directly in LAV itself, but in the Intel Media SDK that it uses for decoding MVC video. That most likely requires Vista+ and SSE2 capable CPU.

Indeed, the decoder is not actually opensource, and it works where it works, but historically Intels MediaSDK libraries have not worked on XP.

I have been quite partial to Cuvid hardware acceleration with XP because it seems to get around the XP limitation with Microsoft and Intel, so it is disappointing that this is no longer being developed into the MVC arena, leaving only CPU decoding (and that no longer supporting XP).

If the decoder isn't open source, I might as well just use Stereoscopic Player as that does work with XP, however I kind of like the MadVR/LAV PQ.

[Stereodude]

If the decoder isn't open source, I might as well just use Stereoscopic Player as that does work with XP, however I kind of like the MadVR/LAV PQ.
Well, you could get a modern OS instead of expecting programs to keep supporting it. XP is almost 15 years old now. I still have a few infrequently used computers running XP floating around my house, but I don't expect new software and features to work on them. What compelling reason do you have for still using XP as a media playback OS instead of Windows 7?

[ryrynz]

What compelling reason do you have for still using XP as a media playback OS instead of Windows 7?

Not to derail, but I don't know why people keep recommending or mentioning 7.. it's basically a dead OS. 8.1 minimum IMO, I've heard it does a better job of media presentation it's also considerably more up to date with some nice new features, a lot faster and from my experience 100% stable. Anyway I'm sure he has a good reason, IanD knows what he's doing, bit slow on the OS uptake though.. Windows 98SE in 2006 :3

[Konrad Klar]

Not to derail, but I don't know why people keep recommending or mentioning 7..
For some the GUI of Windows 8 and newer looks atrocious.

[Manni]

SSIF playback works great, well done.

The main upside of playing .mpls is to keep chapter info.

When playing the SSIF, it's like when playing the m2ts, you lose the ability to skip chapters.

Apart from that, it seemed to work great, at least with the movie I tested.

[Stereodude]

Not to derail, but I don't know why people keep recommending or mentioning 7.. it's basically a dead OS. 8.1 minimum IMO, I've heard it does a better job of media presentation it's also considerably more up to date with some nice new features, a lot faster and from my experience 100% stable. Anyway I'm sure he has a good reason, IanD knows what he's doing, bit slow on the OS uptake though.. Windows 98SE in 2006 :3
Because 7 is the next step after XP if you want to lag behind (We'll ignore Vista). Sure, 8.1 with Classic Shell would be the logical choice of a Windows OS for media playback. I put 8.1 x64 on my HTPC rebuild that I just put together in the past week.

[FDisk80]

Finally got home from work and tested the new Full 3D Bluray thing. IT'S MAGIC! RIP PowerDVD.
Everything just works. Supports ISO like a champ. No need to mount. It even enables and then properly disables the 3D feature in the nvidia control panel after the player is closed.
Picture quality is superb with madVR.

Had to drop from NNEDI 128 to Jinc for Full 3D playback though because there is no DXVA being used in 3D so the GPU went all the way to 100% but there is probably no point using it for Bluray's anyway. But I'm actually amazed that all the madVR picture enhancements like Sharpen Edges and SuperRes can be used in 3D mode.
That's sick! :)

A small issue that I notices. With exclusive mode everything works perfect but if I disable exclusive mode and switch between windowed and full screen the 3D picture freezes. And the only way to unfreeze it is to pause or rewind the video while in full screen. Then it continues playing perfectly.

Amazing work!

[ileile]

Finally got home from work and tested the new Full 3D Bluray thing.

does it support external subtitle for 3D Bluray?

I'm not at home, so haven't tested it myself.

[nevcairiel]

External subtitles would be handled by your player, not LAV. Just like with any 2D content.

[ileile]

External subtitles would be handled by your player, not LAV. Just like with any 2D content.

I see.

so if madvr can properly display external subtitles (from player) in 3D mode, then everything is fine.

[Sunset1982]

nev, is dolby atmos supported by lav?

I#m using latest lav nightly and mpc-hc to passthrough the dolby atmos soundtrack to my atmos capable receiver, but it only shows truehd...

[nevcairiel]

Passthrough should work, but I have no way to test it. Decoding is not supported, and will likely not be for a long time to come.

[scollaco]

I can confirm that Pass through works correctly for me and I get Atmos on my receiver. I'm also using latest LAV nightlies.

Also...Thanks so much for 3D support!!

[Sunset1982]

Passthrough should work, but I have no way to test it. Decoding is not supported, and will likely not be for a long time to come.

ok, thanks for the info. What is the reason for not supporting decoding it?

[FDisk80]

does it support external subtitle for 3D Bluray?

I'm not at home, so haven't tested it myself.

Didn't test yet. I think nevcairiel said he is working on it.

Edit: ah, nevermind. I missed a whole page of comments :D

[madshi]

ok, thanks for the info. What is the reason for not supporting decoding it?
The reason is that you didn't write a decoder yet, Sunset1982... ;) If you want to try, be my guest. It will be hard, though, because there's no detailed tech spec available for Atmos.

[Sunset1982]

The reason is that you didn't write a decoder yet, Sunset1982... If you want to try, be my guest. It will be hard, though, because there's no detailed tech spec available for Atmos.


ah, that's a good reason. I would code one if I where a talented coder like you or nev, but sadly I'm not a software developer. :(

Is DTS:X also supported for passthrough by lav?

[ileile]

this statement on the first page:

full DTS-HD support depends on the ArcSoft decoder DLL being installed

Is it still true?

I remember that the ArcSoft decoder is not needed anymore.

[nevcairiel]

No, external components are no longer required for full DTS-HD support.

[Aleksoid1978]

Hi nevcairiel.
About MVC output from LAV Source - as i see (debug LAV Source) you output combined H264+MVC Extension data even if mediatype is different from MEDIASUBTYPE_AMVC. I think that this is unnecessary, since video decoder anyway will not handle MVC Extension data.
And it only gives extra load on the disk system and RAM.

P.S. In MPC-BE i check output mediatype and output combined data only if equal MEDIASUBTYPE_AMVC. :)

And - thanks for you idea about muxing method 3D H264 data from Blu-ray.

[nevcairiel]

About MVC output from LAV Source - as i see (debug LAV Source) you output combined H264+MVC Extension data even if mediatype is different from MEDIASUBTYPE_AMVC. I think that this is unnecessary, since video decoder anyway will not handle MVC Extension data.
And it only gives extra load on the disk system and RAM.

I thought about doing that before, but I didn't implement it yet. Its done now though.

[LigH]

Is there any way that LAV Filters could help me reducing the annoying loudness difference in DVB, especially during commercials in private broadcaster programmes? I tried using DRC and preferring AC3 audio streams, but that doesn't help; I even wonder if that may be counterproductive?
__

Found a solution: Normalizing in mixing should be disabled. That was a mistake in my setup.

[FDisk80]

Is there any way that LAV Filters could help me reducing the annoying loudness difference in DVB, especially during commercials in private broadcaster programmes? I tried using DRC and preferring AC3 audio streams, but that doesn't help; I even wonder if that may be counterproductive?
__

Found a solution: Normalizing in mixing should be disabled. That was a mistake in my setup.

Isn't the whole point of Normalizing is to reduce loudness difference?

I don't get how disabling it solved your problem.

[LigH]

I am not sure what exactly happens when you activate this feature; but the tooltip warns that when enabled, downmixed multi-channel sources will get a lower volume than stereo sources. So it is probably not a kind of AGC, just a "pessimistic" scaling of the matrix values (just a coarse description, inexact: pessimistic ~ 1/n volume for n channels to ensure no clipping; optimistic ~ 1/sqrt(n) volume for n channels because negative mixdown factors will usually avoid excessive clipping).

[Boardlord]

Is there any way that LAV Filters could help me reducing the annoying loudness difference in DVB, especially during commercials in private broadcaster programmes? I tried using DRC and preferring AC3 audio streams, but that doesn't help; I even wonder if that may be counterproductive?
__

Found a solution: Normalizing in mixing should be disabled. That was a mistake in my setup.

For me this method is still the best:

http://forum.doom9.org/showthread.php?t=162433

I use LAV Filters to decode the audio and use ffdshow to process the raw audio, utilizing the RockSteady Winamp plugin. Unless of course you bitstream the AC3 streams to an external amplifier...

[LigH]

"I sue"? ... I guess that means "I use". ;)

[Boardlord]

"I sue"? ... I guess that means "I use". ;)

Yup, thanks, corrected :D

[Gravitator]

Салам!
I discovered hardware VP9 decoding in the new version of the LAV. Please tell me, GTX 750 supports hardware decoding vp9?

[clsid]

Only the 750 v2, which is a rebranded 950.

[ryrynz]

Only the 750 v2, which is a rebranded 950.

Only 512 cores enabled with that apparently rather than the 950's 768 cores, it would seem to be an OEM part? as I've seen no physical cards and the naming is odd.

Nvidia is however is readying the 950 SE/LE which will have 640 cores and be out in March.

[Sunset1982]

Are these the right settings to use the GTX960's built in HEVC hardware decoder for 4K HEVC files? I want the other codecs to be decoded by software...

http://fs5.directupload.net/images/160224/temp/olgp22co.jpg (http://www.directupload.net/file/d/4274/olgp22co_jpg.htm)

[nevcairiel]

Personally I would leave HD on aswell, even a 1080p file can bring many CPUs to its knees.

[FDisk80]

Are these the right settings to use the GTX960's built in HEVC hardware decoder for 4K HEVC files? I want the other codecs to be decoded by software...

http://fs5.directupload.net/images/160224/temp/olgp22co.jpg (http://www.directupload.net/file/d/4274/olgp22co_jpg.htm)

Why not enable everything? Except maybe VP9. That one could be problematic.

I don't see any downsides to this.

[Sunset1982]

ok, I thought I read that softwaredecoding gives better image quality.

[nevcairiel]

ok, I thought I read that softwaredecoding gives better image quality.

No, the quality is exactly the same.

[huhn]

isn't software better at error handling?

[ryrynz]

isn't software better at error handling?

I guess it would make some sense that it was, but there should be adequate error handling built into the spec. And how often is it that errors would be encountered that software would make sense to use over hardware?
Maybe almost never?

[huhn]

broadcast is full with errors.

[DragonQ]

broadcast is full with errors.
Shouldn't be, unless you're talking about different "errors" to what I'm thinking. I frequently run recordings to TS-Doctor to trim them and it almost never finds stream errors.

[rock]

could enable/disable system tray icon applied immediately even in the middle of playback?

[nevcairiel]

could enable/disable system tray icon applied immediately even in the middle of playback?

You can enable it on the fly, and it'll just show up then, but disabling is intentionally delayed because it can cause problems since the tray icon interacts with the playback to some degrees.
Not likely to change.

[omarank]

While testing 3D with the latest LAV nightly and madVR 0.90.12, I found issues with two samples:

Sample 1 (https://www.mediafire.com/?6k0e88m2rj6xa0u) shows image corruption in the right eye view.

Sample 2 (http://memepuffs.com/indoor.MTS) shows a black screen. It’s an MTS file though. I am not sure if it is supported, but LAV splitter is able to identify the video stream as H.264 stereo high L4.1.

I tried reverting to older nightlies with 3D support. I see the same issue with all of them.

[jkauff]

Now that the latest major version of ffmpeg has been released, can we expect a new official LAV release soon?

[nevcairiel]

While testing 3D with the latest LAV nightly and madVR 0.90.12, I found issues with two samples

I fixed the first one, for the second one, only Blu-ray compatible muxes are supported, which this one is not. It lacks timestamps on some frames and the views are interleaved in the wrong order (the extension has to come first). Don't have immediate plans to try to support any more kind of muxes of such content.

Now that the latest major version of ffmpeg has been released, can we expect a new official LAV release soon?

ffmpeg versioning or releases are of no consequence to LAV.

[omarank]

I fixed the first one
Confirmed. Thanks!

for the second one, only Blu-ray compatible muxes are supported, which this one is not. It lacks timestamps on some frames and the views are interleaved in the wrong order (the extension has to come first). Don't have immediate plans to try to support any more kind of muxes of such content.
Ok.

[jkauff]

ffmpeg versioning or releases are of no consequence to LAV.
I was thinking more of your workload, you being an ffmpeg contributor.

[P.J]

Can't choose HD4600:

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

[nevcairiel]

Can't choose HD4600

Only devices with a connected display are available.

[P.J]

No way to create a fake connected display?

[nevcairiel]

Windows needs to see it as an active display, that's all. If its fake or real doesn't matter, as long as windows thinks its active.
But the GTX960 is a much better choice for decoding anyway, so why bother.

[wanezhiling]

http://i.imgur.com/ZnxlEPb.png
PotPlayer's dxva cb mode supports dx11 feature and can select both gpus.
Maybe lav can add too?

[nevcairiel]

I have no plans to add d3d11 soon. Just enable the GPU you want to use.

[andybkma]

Hi nev, I am finally getting around to reporting this long standing bug I am having with LAV and Speex audio. In the download link I am including two files:

1) the orig AVC/Speex file that pixelates when using LAV to audio/video decode it (unwatchable due to video pixelation)
2) the same exact file "ran through" a lossless video editor (in this case Machete Video Splitter) that after I run in through selecting the start/end points now plays fine (nothing transoded). I can't explain why if plays fine now but that's the only way to play the file short of transcoding the audio to AAC but then it goes out of sync. Definitely something to with Speex

http://depositfiles.com/files/ao7x6y5an

I am hoping that by including both of these files (especially the original) you can figure out why the video is pixelating with Speex. I have many files like this, please let me know if you need more

Cheers

[3421]

Opencl is turned off or not supported, no code?

[nevcairiel]

Hi nev, I am finally getting around to reporting this long standing bug I am having with LAV and Speex audio. In the download link I am including two files:

1) the orig AVC/Speex file that pixelates when using LAV to audio/video decode it (unwatchable due to video pixelation)
2) the same exact file "ran through" a lossless video editor (in this case Machete Video Splitter) that after I run in through selecting the start/end points now plays fine (nothing transoded). I can't explain why if plays fine now but that's the only way to play the file short of transcoding the audio to AAC but then it goes out of sync. Definitely something to with Speex


Both files play identically for me. No pixelation. Maybe something specific to your settings?

[LigH]

^ Confirming, MPC-HC 1.7.10 with internal LAV Filters of shipping date plays both well.

In addition: NSFW! If you mentioned that earlier, I already forgot, should have been repeated.

[3421]

I say: %HYGF%&t^**(())PQWEGHJKL:#$%^^&&UJDFTTHY&dv

https://www.cubbyusercontent.com/pl/text.7z/_a1f29a1629b5400da7a0804c3e49f9e7

[andybkma]

Both files play identically for me. No pixelation. Maybe something specific to your settings?

Wow, that is really strange...

Well, I can only say that I am using two different media players to duplicate the problem:

1) Zoom Player with mVR and LAV set up as the file's splitter & audio/video decoder with Reclock

2) PotPlayer with EVR and LAV set up as the file's splitter & video decoder only (no Reclock). If I remove either LAV splitter or LAV video decoder from the Potplayer chain then I have no pixelation/video corruption. But as soon as both are added back in (as the set up is in Zoom Player) then I have the problem again.

I have troubleshot this for a long time... can you give me any pointers on what to look for next? The only common usage between the two media players is LAV as the splitter/decoder. They both have different video and audio renderers so that is what led me to believe it was a LAV problem.

Really appreciate it....

[LigH]

@ 3421:

Instead of publicly quoting a halfway anonymized chat snippet, posting a password protected archive of it ... that's one of the most complicated ways to discuss a matter. :rolleyes:

Using OpenCL to decode UHD video would depend a lot on the available hardware. If there is no GPU available with a really powerful implementation, and the OpenCL algorithms are executed on the CPU instead, that will probably be slower than CPU optimized decoding algorithms with support for SSE2 or even better.
__

@ andybkma:

"Pixelation" ... hard to imagine if I don't see any, but its appearance would be important to understand. Is it like decoding errors of video streams cut not at GOP borders, so that the first GOP is decoded against a blank background as reference?

[andybkma]

@ andybkma:

"Pixelation" ... hard to imagine if I don't see any, but its appearance would be important to understand. Is it like decoding errors of video streams cut not at GOP borders, so that the first GOP is decoded against a blank background as reference?

I guess I should say extremely corrupted looking video instead of pixelation... you would know what I mean if you saw it. Blurred, totally washed out, unwatchable instead of a usual looking compressed cam video.

[LigH]

You could possibly give us an impression if AviSynth with DirectShowSource loads the same corrupt video, to convert a brief Trim() to a sample.

[nevcairiel]

With settings i meant things like disabling all hardware acceleration, trying different hardware accelerators if available for your hardware, etc. Everything behaves differently.
Thats really the first thing anyone with decode corruption should do, turn off hardware decoding, that way it can point to the issue more clearly.

[3421]

Very good captcha. I wrote only for one person. Who needs that will unpack the archive. GPU OK support decode format H264 SD /HD /FHD /4K, other formats SD /HD /FHD. For me, there is nothing to discuss.

[andybkma]

With settings i meant things like disabling all hardware acceleration, trying different hardware accelerators if available for your hardware, etc. Everything behaves differently.
Thats really the first thing anyone with decode corruption should do, turn off hardware decoding, that way it can point to the issue more clearly.

Ah yes, that was the push I needed in the right direction... after changing LAV video decoder from CUVID to software decoding the problem went away. So ummm, it's Nvidia's fault? Odd though that the files play fine with Cuvid after running it once through Machete Video Splitter. And that the problem only occurs with Speex audio. So CUVID, Speex and AVC don't play well together? I guess that's the bottom line...

[rseiler]

Does LAV Audio Decoder have an audio gain function hidden away somewhere like AC3Filter's? It's handy when using players that don't have the capability when playing videos that have abnormally low volume.
http://www.ac3filter.net/wiki/AC3Filter:Main_page

[Anima123]

There are some videos which LAV (mpc-hc invoked) reports fps double of the fps reported by mediainfo, while mpc-be is correct.

[wanezhiling]

There are some videos which LAV (mpc-hc invoked) reports fps double of the fps reported by mediainfo, while mpc-be is correct.
Such report means nothing, posting a sample file means everything.

[Anima123]

Such report means nothing, posting a sample file means everything.

Please check the following url for HJSplitted files (the first and the last)

https://www.dropbox.com/s/qo797legsyw012l/%5B1pondo%5D%202015-02-10%20%28021015_025%29.avi.001?dl=0

https://www.dropbox.com/s/ma1xqkr7hljg8w1/%5B1pondo%5D%202015-02-10%20%28021015_025%29.avi.1326?dl=0

LAV reported as 119.880 fps
MPC-BE reported as 59.940 fps

and I tend to believe that both are wrong.

[Sparktank]

Please check the following url for HJSplitted files (the first and the last)

LAV reported as 119.880 fps
MPC-BE reported as 59.940 fps

with "Deinterlacing Mode Auto", it doubles.
With "Deinterlacing Mode Disbaled (Progressive)", it singles.

Deinterlacing Mode Auto pin info in MPC-HC with LAV nightly (0.67.0-136)
Pin: [OUT] Output
Filter : LAV Video Decoder - CLSID : {EE30215D-164F-4A92-A4EB-9D4C13390F9F}

- Connected to:

CLSID: {04FE9017-F873-410E-871E-AB91661A4EF7}
Filter: ffdshow Video Decoder
Pin: In

- Connection media type:

Video: YV12 1280x720 119.88fps

Deinterlacing Mode Disabled (progressive) pin info in MPC-HC with LAV nightly (0.67.0-136)
Pin: [OUT] Output
Filter : LAV Video Decoder - CLSID : {EE30215D-164F-4A92-A4EB-9D4C13390F9F}

- Connected to:

CLSID: {04FE9017-F873-410E-871E-AB91661A4EF7}
Filter: ffdshow Video Decoder
Pin: In

- Connection media type:

Video: YV12 1280x720 59.94fps

[Anima123]

Sparktank, thank you for the info.

Would someone please check why IVTC of madVR won't work on this file? As I remembered, I have inverse telecined one video that reported as 59.94 fps and resulted 23.976 fps, so a more GPU intensive image doubler can be applied without dropped frames.

[huhn]

Sparktank, thank you for the info.

Would someone please check why IVTC of madVR won't work on this file? As I remembered, I have inverse telecined one video that reported as 59.94 fps and resulted 23.976 fps, so a more GPU intensive image doubler can be applied without dropped frames.

inverse telecine can only be used on telecine content this is not telecine because it is 59p.

and ivtc can be used to decimate for example 24p in 60p.

but the content must be 24p in 60p it doesn't work on real 60p.
i can't play your second sample so i can't check it but what so ever i don't see on issue in lavfilter.

[Anima123]

I always dreams of a decimator which can output half the framerate from decoder side, so I can enjoy the video with higher scaling settings.

@huhn, the second one is just the tail part of the video, in case it was needed.

[ipanema]

Some MP4 files from smartphones have a rotation indicator that states which way up the camera was held when the video was recorded. So if the phone was held vertically the rotation indicator would be 90 degrees or maybe 270.

Playing in VLC seems to correctly rotate the video so that the recorded scenery is always the right way up.

Is there any way to get the LAV decoder to rotate the video frame in the same way?

Incidentally, I noticed that the Intel Media SDK 2016 can do rotation by 90, 180 and 270 degrees but I guess that would only apply to the Intel Quicksync component of the LAV decoder.

https://software.intel.com/en-us/forums/intel-media-sdk/topic/601236

[sneaker_ger]

I think LAV may support this already, but only passes on the rotation info as metadata and it's up to the video renderer to do the actual work. Try madVR renderer.

[nevcairiel]

LAV will export the rotation metadata, its up to the video renderer and/or player to actually rotate it. Some already do.

[ashlar42]

Hi! Does LAV source/splitter support authentication headers in https streams?

[nevcairiel]

Hi! Does LAV source/splitter support authentication headers in https streams?

Yes, as long as you pass the username/password in the URL in the common format (ie. http://user:pass@host.tld/)

[ashlar42]

Yes, as long as you pass the username/password in the URL in the common format (ie. http://user:pass@host.tld/)Thanks, do you see anything in the Kodi log I posted here: http://forum.kodi.tv/showthread.php?tid=177557&pid=2264762#pid2264762 that could justify the failure?

[ashlar42]

On the same subject, that plugin's developer is asking me to post this, as he's got problems registering here:

"KODI allows passing along additional header information in a HTTP/HTTPS request by passing it along in the URL with a |. Such as a URL "https://www.google.ca|Authentication: Bearer 1234" which would attach Authentication: Bearer 1234 to the header of the request. Is there an equivalent way to pass long authentication information that can be appended to the header of the http request? See http://forum.kodi.tv/showthread.php?tid=77500 -- note this has been implemented in KODI since v12"

An extra bit of important information: all this concerns Kodi DSPlayer use, which is a DirectShow implementation of it I think you might be familiar with.

[nevcairiel]

I already told you how authentication data can be passed to HTTP/HTTPS requests, which is a pretty standard syntax any browser would also understand, no plans to add any other weird ways, sorry.
LAV supports Basic and Digest auth, but you need to tell it the credentials.

[P.J]

Anyway to force LAV to load a specific settings for a certain video?
Something like .lav file with the same name of the video, just like .srt subtitle files
I need to force de-interlacing + TTF/BFF settings for some videos :(

[DragonQ]

Anyway to force LAV to load a specific settings for a certain video?
Something like .lav file with the same name of the video, just like .srt subtitle files
I need to force de-interlacing + TTF/BFF settings for some videos :(
If you're using MadVR as your renderer, it supports settings parameters in file names.

[clsid]

Export the LAV settings from the Registry. Then create .reg files with default and custom settings and import the desired one before playback.

[TechnIx]

I've been experimenting with a bunch of demuxers and LAV Splitter plays back my files perfectly, but has one crucial flaw for my use - it doesn't support multiple output pins for multiple video streams (i.e. 1 pin for each stream)

My files tend to be between 1 to 4 video streams, with 1 audio stream, muxed as MPEG-TS.

I know this is a bit of a specialized feature, but is it on the horizon? If not, how hard would it be to implement in the code myself? I have looked at the source code and not quite sure where to start. I have written a few programs that use DirectShow filters in a graph, and even written a very basic filter, but actually meddling with an advanced demultiplexer filter like LAV Splitter just seems a bit out of my depth.

I have used AV Splitter (which I believe was a fork of LAV Splitter from way back) which has multiple output pins, but doesn't quite play my files back correctly (long delays when seeking)

[P.J]

If you're using MadVR as your renderer, it supports settings parameters in file names.

No, I'm using EVR :(


Export the LAV settings from the Registry. Then create .reg files with default and custom settings and import the desired one before playback.

Not a good solution.

[nevcairiel]

I know this is a bit of a specialized feature, but is it on the horizon? If not, how hard would it be to implement in the code myself?

LAVs entire architecture is designed around having one output pin per type, and I don't see that ever changing. Similarly, its not going to be easy to change that, sorry.

PS:
AV Splitter is not related at all.

[rock]

hi, I wonder if this is limitation of my spec or already known issue. My notebook is using Windows 8.1 with Sandy Bridge GPU. With combination of LAV Splitter + WMAudio Decoder DMO, I easily get audio out of sync whenever I seek this video (http://media.ch9.ms/ch9/b208/c62d6156-e839-4d1c-8d78-a484a382b208/aspnet26767_Source.wmv) around 20s or before that. With LAV Splitter + LAV Audio Decoder or MS Splitter + WMAudio Decoder DMO, it's fine.

[AngelGraves13]

nevcairiel, can you please increase the buffering of Dolby Atmos tracks a little? If I skip forward in an MKV file, there's a hiccup in the audio, like it's trying to catch up for a second and then it's fine. Haven't experienced the issue with TrueHD or DTS-HD MA.

[nevcairiel]

nevcairiel, can you please increase the buffering of Dolby Atmos tracks a little? If I skip forward in an MKV file, there's a hiccup in the audio, like it's trying to catch up for a second and then it's fine. Haven't experienced the issue with TrueHD or DTS-HD MA.

Thats not possible. Atmos is just some fancy metadata in TrueHD streams.
Sounds like your Atmos receiver is just being slow.

I don't know what you think "buffering" does.

[AngelGraves13]

Thats not possible. Atmos is just some fancy metadata in TrueHD streams.
Sounds like your Atmos receiver is just being slow.

I don't know what you think "buffering" does.

I'm just playing it on PC with Downmix.

Seems to be fine after running the file through mkvtoolnix.

[Leader]

Hi, nevcairiel!

In Windows XP incorrectly displayed and positioned options (checkboxes) shown in the screenshot:
http://fs5.directupload.net/images/160305/gnb4mozw.png (http://www.directupload.net)

[LigH]

May depend on the default UI font.

[Vasilich]

@Leader, what version is shown on screenshot?

[sneaker_ger]

It's one of the recent nightlies. Nev changed the GUI a bit to make room for a drop-down list of hardware decoding devices. (Same cosmetic problem in Win7 without Aero, btw.)

[ocyl]

This is what I see in v0.67.0.142-git.

http://i.imgur.com/5j73NSl.png

[nevcairiel]

The latest nightly should give those checkboxes a bit more room, hopefully it looks better now. On my system I can fit another full character in there now without clipping, but oh well. :)

[Leader]

The latest nightly should give those checkboxes a bit more room, hopefully it looks better now. On my system I can fit another full character in there now without clipping, but oh well. :)

I checked the latest nightly build 0.67.146-git - now everything appears normal and looks neat.

Many thanks! :thanks:

[Leader]

And yet, nevcairiel, I discovered that Windows XP, using "NVIDIA CUVID"-Decoder, incorrectly playback video files at a frame rate = 60 FPS.
Jerks observed deviations and missing frames. Perhaps the case in timestamps. Do you have the ability to fix this?
Just because of this bug is inconvenient to use "CUVID-Decoder".
P.S. Link to sample: https://dropmefiles.com/c9WKm

[nevcairiel]

Looks fine here, and I'm not supporting XP anymore, especially when it comes to HWAccel, sorry.

[Mikounou]

Hello,

Sorry for my english but i have a big problem with the latest lavfilters:
I can't bitstream Dolby Digital + (eac3) and DTS-HD HRA.

With Dolby Digital +, DD logo appear on the Onkyo but no sound.
With DTS-HD HRA, i have only the DTS Core.

No problem with Dolby Digital, TrueHD, DTS or DTS-HD MA.

Windows 10
NVIDIA GTX 960
Onkyo TX-RZ900
Latest MPC HC nightly build

Can you help me ?
I have tested many things, change all options in Lav Filters, but no luck :confused:

Thanks

[PPPPark]

Hello,

Sorry for my english but i have a big problem with the latest lavfilters:
I can't bitstream Dolby Digital + (eac3) and DTS-HD HRA.

With Dolby Digital +, DD logo appear on the Onkyo but no sound.
With DTS-HD HRA, i have only the DTS Core.

No problem with Dolby Digital, TrueHD, DTS or DTS-HD MA.

Windows 10
NVIDIA GTX 960
Onkyo TX-RZ900
Latest MPC HC nightly build

Can you help me ?
I have tested many things, change all options in Lav Filters, but no luck :confused:

Thanks
I tested EAC3 No problem, DTS-HD HRA just DTS core. POT也只是输出DTS核 。@万恶之灵
Windows 10
NVIDIA GTX 760
Onkyo TX-NR646
LAVFilters-0.67.0-146.exe

[nijiko]

Can you supply a portable version of nightly lav? In most cases, we use lav with mpc-hc. And if you directly pack the files, I think it's better and more convenient than to compile to a installer.

[nevcairiel]

Can you supply a portable version of nightly lav? In most cases, we use lav with mpc-hc. And if you directly pack the files, I think it's better and more convenient than to compile to a installer.

There are no plans to offer LAV nightlies in any other form, sorry.

[filler56789]

@nijiko: you can use the Inno Setup Unpacker for extracting the contents from the installer.
If you prefer a GUI, download the Universal Extractor package, and then update its innounp.exe to the newest version.

[mzso]

Is there a possibility of Atmos decoding with LAV in the near future?
I imagine it as DTS-HD MA used to be for a long time, with an external dll loaded by LAV.
(Somehow I don't think that we'll see Atmos decoding in ffmpeg anytime soon.)

[AngelGraves13]

Is there a possibility of Atmos decoding with LAV in the near future?
I imagine it as DTS-HD MA used to be for a long time, with an external dll loaded by LAV.
(Somehow I don't think that we'll see Atmos decoding in ffmpeg anytime soon.)

I'm sure it's planned, but just like with DTS:X, we don't know when it will be possible. It's basically nothing more than metadata on top of the TrueHD stream and DTS-HD Ma stream (for DTS:X).

[mzso]

I'm sure it's planned, but just like with DTS:X, we don't know when it will be possible. It's basically nothing more than metadata on top of the TrueHD stream and DTS-HD Ma stream (for DTS:X).

The stream might be. But I'm sure decoding is not as easy. It needs HTRF and who knows what sort of other processing to sound right.

[nevcairiel]

LAV Filters 0.68

LAV Splitter
- NEW: H.264 MVC 3D demuxing from MKV 3D, Blu-ray discs and Blu-ray SSIF files
- Fixed: The bitdepth of DTS-HD MA streams is reported properly in the stream information

LAV Video
- NEW: H.264 MVC 3D decoding (requires madVR 0.90 or newer)
- NEW: HEVC HDR streams export the HDR metadata to the video renderer (requires madVR 0.89.18 or newer)
- NEW: VP9 DXVA2 Hardware Acceleration
- NEW: Weston Three Field Deinterlacing filter (w3fdif)
- NEW: Ability to choose which GPU is used for DXVA2-CopyBack decoding
- Changed: Removed the CUVID HQ processing option, it is now automatically used when appropriate
- Fixed: CUVID would crash when playing certain HEVC files
- Fixed: Some pixel format conversions were much slower than expected
- Fixed: Pixel Format changes mid-stream were not re-negotiated with the renderer in all cases
- Fixed: Changes in the framerate due to software deinterlacing are properly reflected in the media type

LAV Audio
- Fixed: Increased PCM buffering to avoid stuttering with TrueHD on some audio devices


Download: Installer (both x86/x64) (https://files.1f0.de/lavf/LAVFilters-0.68.exe) -- Zips: 32-bit (https://files.1f0.de/lavf/LAVFilters-0.68.zip) & 64-bit (https://files.1f0.de/lavf/LAVFilters-0.68-x64.zip)
Download 3D Plugin: 32-bit (https://files.1f0.de/lavf/plugins/libmfxsw32-v1.7z) & 64-bit (https://files.1f0.de/lavf/plugins/libmfxsw64-v1.7z)

H.264 MVC 3D Support

This version introduces support for decoding H.264 MVC 3D streams, as found on 3D Blu-rays (or in MKVs created from those Blu-rays with MakeMKV).
LAV Splitter supports reading H.264 MVC from Blu-rays directly, when opening a Blu-ray SSIF file, or from MKVs. Note that MVC in MPEG-TS which is not created Blu-ray compliant is not supported at this time.

To enable decoding of the H.264 MVC extensions, an external decoder plugin is required, which the Installer will automatically download if the appropriate option is set, or which you can download manually from the links above (place in the same folder as LAVVideo.ax)
LAV Video currently does not allow "processing" the 3D video in any way, it can only output it to a 3D aware video renderer - madVR 0.90 or later. In the future it might be possible to have LAV Video render the video as Side-by-Side, but not in this release.

The interface to receive 3D video and accompanying metadata is documented in the developer info package, if other renderer developers want to support this.

HEVC HDR support

Also new in this release is the ability to export the HDR metadata from HEVC HDR streams. This metadata is automatically send to the video renderer, so it can render the video properly. This is available with madVR 0.89.18 or newer.
LAV Video does not process the HDR information in any way, and any video renderer unaware of HDR will still render the typical "washed out" image.

The interface for the HDR metadata is also documented in the developer information.

DXVA2 VP9 Hardware Acceleration

Late last year Microsoft published a specification on VP9 DXVA2 hardware acceleration. Shortly after, both Intel and NVIDIA started to support this in their hardware.
This version of LAV Video enables using the hardware acceleration in the latest NVIDIA GPUs (GTX950/960), and recent Intel GPUs (Braswell and Skylake)

Please note that only VP9 Profile 0 is supported, 4:2:0 8-bit.

There is more!

On top of all these features, LAV Video also added a new deinterlacer, Weston Three Field Deinterlacing Filter (w3fdif), which can achieve quality similar to YADIF at much higher speeds (simple on 32-bit, complex on 64-bit)
For multi-GPU systems, you can now choose which GPU DXVA2-CopyBack should use for decoding. Please note that only GPUs with a connected display are available (Direct3D9 limitation)
And of course there is also a list of stability and bug fixes that may not even have made the changelog.

Enjoy one of the bigger LAV releases of the recent past, and have fun!

PS:
The download links have switched to HTTPS, if you're having trouble with access due to an old browser and/or OS (I'm looking at you, XP users), you can manually switch the URL back to HTTP if you must.

[madshi]

Awesome release - many thanks! :)

[LigH]

:thanks:

This will be the first "stable" release with w3fdif to recommend for DVBViewer users, to test different deinterlacers regarding speed and quality.

The only annoying part of "TV on a PC" are the flickering top and bottom lines. They are not always encoded cleanly by the broadcaster. I wonder how to treat them in a special way; but I know you have never been a fan of implementing post-processing filters, pure decoders only...

[NikosD]

LAV Filters 0.68


DXVA2 VP9 Hardware Acceleration

Late last year Microsoft published a specification on VP9 DXVA2 hardware acceleration. Shortly after both Intel and NVIDIA started to support this in their hardware.
This version of LAV Video enables using the hardware acceleration in the latest NVIDIA GPUs (GTX950/960), and recent Intel GPUs (Braswell and Skylake)

Please note that only VP9 Profile 0 is supported, 4:2:0 8-bit.


Thanks for the release.

I remember talking about "VP9_VLD_10bit_Profile2" support and adding this info to a nightly build.

[nevcairiel]

I remember talking about "VP9_VLD_10bit_Profile2" support and adding this info to a nightly build.

Still no hardware that exposes this, and Microsoft actually updated the spec again and removed it again.

[LigH]

Typo:

For multi-GPU systems, you can now choose which GPU DXVA2-CobyBack should use for decoding.

[manolito]

Quick question:
Is the ArcSoft dtsdecoderdll.dll still supported?

[nevcairiel]

Is the ArcSoft dtsdecoderdll.dll still supported?

Yes, but I don't really consider it useful anymore. The built-in DTS decoder has proven to work quite well, and if there is a sample where it doesn't work, we would love to get a copy.

The only thing it doesn't support yet is DTS Express (which is its own little codec, not much to do with the remainder of DTS), but those are commentary streams only, and rare at that.
But someone was actually working on DTS Express decoding, so it may show up in the future as well.

Anyway, as long as it doesn't start causing problems, no reason to remove it.

[Mikounou]

Yes, but I don't really consider it useful anymore. The built-in DTS decoder has proven to work quite well, and if there is a sample where it doesn't work, we would love to get a copy.

The only thing it doesn't support yet is DTS Express (which is its own little codec, not much to do with the remainder of DTS), but those are commentary streams only, and rare at that.
But someone was actually working on DTS Express decoding, so it may show up in the future as well.

Anyway, as long as it doesn't start causing problems, no reason to remove it.

For me the built-in DTS decoder not bitstream DTS-HRA, only core :(

My messsage :

I can't bitstream Dolby Digital + (eac3) and DTS-HD HRA.

With Dolby Digital +, DD logo appear on the Onkyo but no sound.
With DTS-HD HRA, i have only the DTS Core.

No problem with Dolby Digital, TrueHD, DTS or DTS-HD MA.

Windows 10
NVIDIA GTX 960
Onkyo TX-RZ900
Latest MPC HC nightly build

Any ideas ?

I will try with the ArcSoft dtsdecoderdll.dll

[huhn]

what type of connector are you using.

bitstreaming has nothing to do with a decoder decoders aren't used for that.

[manolito]

Yes, but I don't really consider it useful anymore. The built-in DTS decoder has proven to work quite well, and if there is a sample where it doesn't work, we would love to get a copy.

For me it still is extremely useful... :p

The built-in DTS decoder fails miserably with a very common DTS 6-ch audio track when piping the decoded 6-ch wav to Aften.

Source info:
Audio
ID : 2
Format : DTS
Format/Info : Digital Theater Systems
Format_Settings_Mode : 16
Format-Einstellungen für Endianess : Big
Codec-ID : A_DTS
Dauer : 1h 28min
Bitraten-Modus : konstant
Bitrate : 755 Kbps
Kanäle : 6 Kanäle
Kanal-Positionen : Front: L C R, Side: L R, LFE
Samplingrate : 48,0 KHz
Bildwiederholungsrate : 93,750 FPS (512 spf)
BitDepth/String : 16 bits
Stream-Größe : 479 MiB (11%)
Titel : Surround 5.1
Sprache : Deutsch
Default : Ja
Forced : Nein

Using Wavi or AVS2PipeMod and piping the result to Aften causes Aften to detect a wrong channel layout. The 6-ch source is either detected as mono or 2-ch depending on the Aften build (Wisodev or KurtNoise).

Using the ArcSoft DTS DLL fixes it, the source is correctly detected as 6-ch. I cannot believe that Aften is to blame since it always detects raw wav files correctly when using them as STDIN. Somehow the built-in DTS decoder seems to create non-standard PCM audio streams.

So please do not disable support for the ArcSoft DTS decoder...


Cheers
manolito

[nevcairiel]

Sounds like Aften (or your AviSynth step) is just faulty and doesn't like the channel layout or sample format, maybe the float output that the built-in decoder produces for lossy DTS, while ArcSoft produces 24-bit integer.
If there would be something fundamentally wrong, it would happen in all sorts of circumstances, during playback, or who knows where. Especially if behavior changes depending on which build you use, it points to the software just being faulty.

Easy enough to test by forcing integer output in LAV for example, or switching the 5.1 channel layout using the option.

I never used anything like that to pipe audio, nor would I have a reason to, but there is various reasons such a tool could break in one circumstance and not another.
Maybe they just don't handle format changes well, which are always a possibility with DTS decoding as you don't know beforehand if its going to be lossy core only, or 16-bit HD, or 24-bit HD. ArcSoft "bypasses" that problem to some degree by using 24-bit output for lossy, instead of (the more accurate) float output.

[manolito]

Did some more tests, but no difference...

The built-in decoder does not produce float output. The LAV Audio status window reports 24bit integer 6-ch for input and output, no matter if the built-in DTS decoder is used or the ArcSoft decoder. To make absolutely sure I retested it disabling float output first, but the results were identical.

Of course it is easy to blame Aften. After all there is no real standard for headerless raw PCM. But since Aften handles this for all other decoders (AC3, MP3, AAC, WMA, you name it), and the ArcSoft DTS decoder also handles it correctly, I do have a point when I say that the built-in DTS decoder is responsible for the issue.

http://i.imgur.com/Q3UnVE4.png

Aften is no longer maintained, you can consider it static. But a lot of applications use it, I do see it as kind of a standard.

And I am a result oriented person, I do not care too much for principles. It just has to work. Period. So again please continue supporting the ArcSoft DTS decoder... (because it just WORKS)


Cheers
manolito

[nevcairiel]

Like I said, as long as it works, there is no reason to remove it. If it breaks, its leaving.
And no, your reasoning is not very useful. Just because something is broken with a "static" component doesn't automatically put blame on everything else. I don't care if Aften breaks, I care that my software functions according to specifications. If Aften can't deal with that, thats not something I concern myself with.

There is a long list of things between LAV and Aften as well, some AviSynth DirectShow source plugin, AviSynth itself, and your CLI tool that reads from AviSynth and pipes it out, those could equally be faulty and just not handle DirectShow properly.
There is a reason DirectShow is not very commonly used for processing or transcoding, its architecture isn't all that well suited to that task.

In any case, its not going anywhere soon. But its dead either way, since ArcSoft vanished into obscurity. So it would seem to be in your long term interests as well to figure out why its broken, and not just accept defeat and go back to some other solution.

PS:
I just tried AVStoDVD and it works perfectly fine and creates a 5.1 AC3 from my 5.1 DTS input.

[clsid]

The LAV Audio status window reports 24bit integer 6-ch for input and output, no matter if the built-in DTS decoder is used or the ArcSoft decoder.Is the channel mask also the same?

You should try the "Use Legacy 5.1 channel layout" option in LAV.

[ddjmagic]

LAV Filters 0.68


Great release :cool: Thanks!

[Anima123]

LAV Filters 0.68
On top of all these features, LAV Video also added a new deinterlacer, Weston Three Field Deinterlacing Filter (w3fdif), which can achieve quality similar to YADIF at much higher speeds (simple on 32-bit, complex on 64-bit)

What exactly do you mean here by "complex on 64-bit"?

[huhn]

http://forum.doom9.org/showthread.php?p=1755017#post1755017


On 32-bit, Simple is significantly faster than YADIF, but due to lack of 32-bit optimizations, Complex is slower.
On 64-bit, both Simple and Complex are faster than YADIF.

it means that w3fdif complex is only faster with 64 bit than YADIF.

[rock]

if I remember correctly I read it somewhere, output formats on LAV Video Decoder only work on s/w decoding. Beside output formats, any other option that don't work on h/w decoding?

[ileile]

does this release support hardware acceleration for msdk mvc? when I play a 3D Bluray, the "active hardware accelerator" shows none.

if hw accel for stereo 3D is not supported at current, do you have a plan to support it in the near future?

[nevcairiel]

if I remember correctly I read it somewhere, output formats on LAV Video Decoder only work on s/w decoding. Beside output formats, any other option that don't work on h/w decoding?

The only hardware accelerator with any kind of limitations is DXVA2-Native, and the answer is simple there: Nothing works.

All other hardware accelerators have no limitations at all.

does this release support hardware acceleration for msdk mvc? when I play a 3D Bluray, the "active hardware accelerator" shows none.

if hw accel for stereo 3D is not supported at current, do you have a plan to support it in the near future?

3D is not hardware accelerated and there are no immediate plans to change that. Its a 1080p H.264 decoder only, any half-recent PC can handle those with ease.

[Sharc]

LAV Filters 0.68
[CODE]
LAV Splitter
- NEW: H.264 MVC 3D demuxing from MKV 3D, Blu-ray discs and Blu-ray SSIF files
.....
H.264 MVC 3D Support

.... In the future it might be possible to have LAV Video render the video as Side-by-Side, but not in this release.


Thanks for this great release!
... and looking forward to the future Side-by-Side rendering .... :)

[P.J]

For multi-GPU systems, you can now choose which GPU DXVA2-CopyBack should use for decoding. Please note that only GPUs with a connected display are available (Direct3D9 limitation)
Thanks a lot :thanks:

Now waiting for Direct3D11 :)

[ocyl]

The latest nightly should give those checkboxes a bit more room, hopefully it looks better now. On my system I can fit another full character in there now without clipping, but oh well. :)

Confirmed. This is what I see now.

http://i.imgur.com/nR68ifq.png

Thank you, nevcairiel =)

[macycat]

LAV Filters 0.68

LAV Audio
- Fixed: Increased PCM buffering to avoid stuttering with TrueHD on some audio devices



nevcairiel,

I have been having an audio glitch every minute or so when bitstreaming TrueHD audio streams. I tried the new version, and I still get the occasional audio glitch, so it doesn't look like the above fix addresses this problem. I don't observe the problem when decoding to PCM.

I am using madVR, and I don't see and dropped, repeated, or delayed frames, nor do I see any presentation glitches ... just the audio problem.

Let me know if you need me to test anything for you ...

[nevcairiel]

Bitstreaming needs extremely careful setup to work properly, either it works or it doesn't, there isn't anything I can do.

[ryrynz]

nevcairiel,
I have been having an audio glitch every minute or so when bitstreaming TrueHD audio streams.

Get someone else with your amplifier to test.. also try another HDMI cable.

[mixus]

I waiting HEVC QuickSync Decoding Support.

When will it arrive?

My System : I5 6600K, Asus Z170 Pro Gaming

[nevcairiel]

I waiting HEVC QuickSync Decoding Support.

When will it arrive?

Use DXVA2, there are no plans to further develop QuickSync decoding.

[mixus]

Use DXVA2, there are no plans to further develop QuickSync decoding.

OK, Thanks for very fast reply.

[manolito]

PS:
I just tried AVStoDVD and it works perfectly fine and creates a 5.1 AC3 from my 5.1 DTS input.

Tried the same conversion on my Win7 laptop (Core i5 CPU, 4 GB RAM), and everything worked fine.

So the issue must have something to do with my ancient desktop PC. WinXP, CPU without SSE2, only 576 GB RAM.


Cheers
manolito

[JarrettH]

Is x265 decoding still getting faster? I only update LAV when it's integrated in the latest MPC release.

[ryrynz]

I only update LAV when it's integrated in the latest MPC release.

Why? Just copy it over its the same thing..

[NikosD]

Does LAV video require SSE2 for DXVA (native or cb) ?

I have an old Win XP SP3+ system with Radeon 3650 AGP and AthlonXP (SSE only NOT SSE2) and LAV Video works only in SW decode mode using latest LAV v0.68

Also DXVA Checker reports LAV Video as DXVA2 only and not DXVA1/2 decoder.

[Aleksoid1978]

Does LAV video require SSE2 for DXVA (native or cb) ?

I have an old Win XP SP3+ system with Radeon 3650 AGP and AthlonXP (SSE only NOT SSE2) and LAV Video works only in SW decode mode using latest LAV v0.68

Also DXVA Checker reports LAV Video as DXVA2 only and not DXVA1/2 decoder.

LAV don't support DXVA 1.0 decoding.

[LigH]

x265 decoding

x265 is one specific of many different encoders. The video format being decoded is called HEVC.

http://cosgan.de/images/smilie/figuren/a045.gif

[NikosD]

If LAV Video doesn't support HW acceleration on Windows XP, maybe it should be noted in the first page of this thread which says that it supports Win XP SP3.

[huhn]

than you have to do that with all features that work on some OS and on some not and that would create a real mess.

and in the end DXVA2 can be used under XP but that's not reliable to say it friendly.

[wanezhiling]

https://www.sendspace.com/file/pmnymp
LAV says the duration as 9min49sec but actually it's 10min.

[nevcairiel]

https://www.sendspace.com/file/pmnymp
LAV says the duration as 9min49sec but actually it's 10min.

I can't download the file, it gets stuck at 5% or so, but alas, not really need more:

The math is quite simple, since its nice and CBR.
The stream is DTS at 1536kbit/s (or 192kB/s), for 10 minutes that would require 112.5 MB of data - but its only 107.9, so clearly the file isn't big enough for 10 minutes of audio at that bitrate.

[wanezhiling]

I can't download the file, it gets stuck at 5% or so
Try this host: https://drive.google.com/file/d/0B2LewMkGuMYtU203anF6U1QzZlE

The math is quite simple, since its nice and CBR.
The stream is DTS at 1536kbit/s (or 192kB/s), for 10 minutes that would require 112.5 MB of data - but its only 107.9, so clearly the file isn't big enough for 10 minutes of audio at that bitrate.
Ok, so on the contrary displaying 10min is incorrect?

[nevcairiel]

Ok, so on the contrary displaying 10min is incorrect?

MediaInfo and ffmpeg seem to disagree about the DTS bitrate (1536 vs 1509), that probably explains the difference. There is a slightly variation of "actual bitrate" and "nominal bitrate".
The thing is that to stay fully CBR, a dts encoder is *supposed* to pad the DTS stream to the nominal bitrate, but apparently most don't seem to do that, so you end up with slight differences.

[wanezhiling]

I see.
But there is a problem, when the dts is as an external track of a movie, then it can't match..hmmm...

[nevcairiel]

If you wanted this to be accurate and function properly, you would put it into a container (ie. a mka file). raw audio files are never going to be very fun to handle.

[NikosD]

than you have to do that with all features that work on some OS and on some not and that would create a real mess.

and in the end DXVA2 can be used under XP but that's not reliable to say it friendly.

We are not talking about limitations of the OS or other factor here (for example the HW acceleration of VP9) but about limitations of LAV Video.

So, if LAV supports SW decoding only, on Windows XP and not HW decoding at all on that OS, it should be mentioned as a feature or limitation of LAV and not OS (Windows XP)

[nussman]

That's not true at all.
LAV supports DXVA2 and if it's not working on WinXP this is a limitation of the OS.

Imho every minute spend for WinXP is wasted time ...

[huhn]

We are not talking about limitations of the OS or other factor here (for example the HW acceleration of VP9) but about limitations of LAV Video.

So, if LAV supports SW decoding only, on Windows XP and not HW decoding at all on that OS, it should be mentioned as a feature or limitation of LAV and not OS (Windows XP)

you can use lavfilter with DXVA2 on windows XP. at least i'm pretty sure you can do that because i'm pretty sure i did this with an 6770 or ion1 or both many many years ago.

and it depends on the point of view if XP is missing a feature or lavfilter.

on XP DXVA2 isn't really supported.
lavfilter doesn't support the very old DXVA1.

windows XP is 15 years old so if someone wants to use this they have to life with broken or missing features.

[NikosD]

Both of you seem to misunderstand my point.

We don't judge here Windows XP and its capabilities, but the announcement of LAV Video supporting Win XP SP3.

If someone can use HW acceleration of LAV Video in Win XP SP3, please tell me how.

[nevcairiel]

LAV has no OS specific checks for HW acceleration, and CUVID should also work on XP.
I once disabled DXVA2 on XP and some people complained that it worked for them on some codecs, so I enabled it again.

Quite simply, as others have said, LAV supports DXVA2 and not DXVA1, which OS that works on is not something LAV concerns itself with. It just tries to use it, and falls back to SW if its not available.
There is many things with a OS requirement, not only hwaccels, for example HD audio bitstreaming has never worked on XP because the OS doesn't support it, and there is probably quite a list of other things, which I do not really care to document, or remember.

Everything should work starting with Vista, and XP is practically dead to me, everyone that still runs XP is just lucky I didn't break it by accident yet. I haven't tested against XP in years. ;)

[NikosD]

According to Microsoft, DXVA2 requires Vista or later.

" The DXVA 2 API requires Windows Vista or later. The DXVA 1 API is still supported in Windows Vista for backward compatibility. An emulation layer is provided that converts between either version of the API and the opposite version of the DDI:

If the graphics driver conforms to the Windows Display Driver Model (WDDM), DXVA 1 API calls are converted to DXVA 2 DDI calls.

If the graphics drivers uses the older Windows XP Display Driver Model (XPDM), DXVA 2 API calls are converted to DXVA 1 DDI calls.

The following table shows the operating system requirements and the supported video renderers for each version of the DXVA API.

DXVA 1
Windows 2000 or later
Overlay Mixer, VMR-7, VMR-9 (DirectShow only)

DXVA 2
Windows Vista
EVR (DirectShow and Media Foundation)"

[nevcairiel]

According to Microsoft, DXVA2 requires Vista or later.
If you install some version of the .net framework on XP you get EVR and the most basic dxva2 support (i forgot which .net exactly) - you just need a graphics driver which also supports it, which seems to be rare, but some people claim to have used it.

[NikosD]

I have the unofficial "SP4" installed with all .net installations possible and the latest 14.4 Catalyst drivers, which are pretty old of course.

Maybe Nvidia provided suitable drivers for that.

Thanks for the replies.

[huhn]

try way older driver. maybe some driver before windows 8 was released.

isn't 14.4 one of the lastest CCC driver?

[Volfield]

Some commit between LAVFilters-0.67.0-67.exe and LAVFilters-0.67.0-69.exe broke playing any avi files with external txt subs for me. When using .67 is ok subs show ok, but on .69 sync is broken (they show one after the other, without any interval). Tested on MPC-HC, MPC-BE.

[nevcairiel]

LAV isn't really involved when playing external subtitles, thats the players job.
I don't see what might have changed that would disrupts a players abilitity to do so. So I would suggest to contact the developers of said subtitle renderers and ask them to determine why their subtitle renderers don't obey subtitle timing information anymore.

[Volfield]

Ok. I check more. Problem exist when i select Weston Three Field (complex and simple). Testes on internal subtitles renderer and xy-subfilter on evr-cp.

Media info:
Format : AVI
Format/Info : Audio Video Interleave
File size : 175 MiB
Duration : 20mn 55s
Overall bit rate mode : Variable
Overall bit rate : 1 169 Kbps
Writing application : transcode-1.0.6

Video
ID : 0
Format : MPEG-4 Visual
Format profile : Advanced Simple@L5
Format settings, BVOP : 2
Format settings, QPel : No
Format settings, GMC : No warppoints
Format settings, Matrix : Default (H.263)
Codec ID : XVID
Codec ID/Hint : XviD
Duration : 20mn 55s
Bit rate : 1 023 Kbps
Width : 624 pixels
Height : 352 pixels
Display aspect ratio : 16:9
Frame rate : 23.976 (23976/1000) fps
Color space : YUV
Chroma subsampling : 4:2:0
Bit depth : 8 bits
Scan type : Progressive
Compression mode : Lossy
Bits/(Pixel*Frame) : 0.194
Stream size : 153 MiB (87%)
Writing library : XviD 1.2.0SMP (UTC 2006-01-08)

Audio
ID : 1
Format : MPEG Audio
Format version : Version 1
Format profile : Layer 3
Mode : Joint stereo
Mode extension : MS Stereo
Codec ID : 55
Codec ID/Hint : MP3
Duration : 20mn 55s
Bit rate mode : Variable
Bit rate : 128 Kbps
Channel(s) : 2 channels
Sampling rate : 48.0 KHz
Compression mode : Lossy
Stream size : 20.0 MiB (11%)
Alignment : Aligned on interleaves
Interleave, duration : 24 ms (0.58 video frame)
Writing library : LAME3.90.
Encoding settings : -m j -V 4 -q 2 -lowpass 17.6 --abr 128

[nevcairiel]

Ok. I check more. Problem exist when i select Weston Three Field (complex and simple). Testes on internal subtitles renderer and xy-subfilter on evr-cp.


With YADIF everything is fine? Specifically with YADIF in 50/60p mode?

[Volfield]

You're right. YADIF in 50/60p same problem.

[nevcairiel]

Your file isn't even interlaced, do you have LAV set to force deinterlacing?

[Volfield]

No it's set "AUTO"
My current settings: http://imgur.com/YR6qHUY

[nevcairiel]

Well even if its doing something that it shouldn't be doing, it should not affect the subtitle renderer, so I would still suggest you contact them about that.
What if the file would actually be interlaced, then this behavior would be correct, and subtitles wouldn't work? How odd.

[Alexey1975]

- Changed: Removed the CUVID HQ processing option, it is now automatically used when appropriate

Please, restore the ability to turn off the High-Quality Processing of CUVID HW Deinterlacing function. Because of it the problem with a certain video streams still remains.
http://f20.ifotki.info/thumb/2b59327709a9e4fd747bdddf2fc666f02ec925240438943.jpg (http://i-fotki.info/20/2b59327709a9e4fd747bdddf2fc666f02ec925240438943.jpg.html)

Here the sample https://yadi.sk/i/JfjhbVSmq73ks

[wanezhiling]

- Changed: Removed the CUVID HQ processing option, it is now automatically used when appropriate

Please, restore the ability to turn off the High-Quality Processing of CUVID HW Deinterlacing function. Because of it the problem with a certain video streams still remains.
http://f20.ifotki.info/thumb/2b59327709a9e4fd747bdddf2fc666f02ec925240438943.jpg (http://i-fotki.info/20/2b59327709a9e4fd747bdddf2fc666f02ec925240438943.jpg.html)
Offer a sample file instead.

[nevcairiel]

I have no plans to re-add that option again, sorry. Contact NVIDIA and have them fix their driver.
The effect you are seeing was an unintended side-effect of the option, and not what it was supposed to do in any case. It had more far reaching impact on decoding than just deinterlacing, which is why I removed it. It was extremely misleading.

[Alexey1975]

Can you at least leave the ability to disable it through the registry tweak?
For me parley with NVIDIA too complicated.

[kiwijunglist]

Hi

My question is regarding playback of mkv remux on my htpc with atmos 7.1. (Receiver does Dolby HD MA but not atmos) Can LAV decode the atmos object based channel?

A search of this thread suggests, No it can't decode the object based channel.

I have passthrough enabled, with an HDMI connection to my receiver (Pioneer Elite LX-SC87) which supports Dolby HD-MA 7.1 but not atmos support. I assume I get the 7.1 audio, but the object based channel is completely lost and I miss out on some of the sound effects?

Can't find the answer with search, but I assume it would be discarded.

What happens if I let LAV decode the audio, what happens to the object based channel, is it discarded, or is it mixed down into the 7.1 LPCM output?

Can't find the specific answer with search, but I assume it's discarded as it can't be decoded.

Are there any directshow audio decoders for PC (Open source or commercial) that support Atmos decoding?

?????????

Thanks
Kiwi

EDITED MY POST

[khanmein]

@nevcairiel, i tick codes for HW decoding HEVC (DXVA-CB) & watch h.265 .mkv will freeze & cause the mpc-hc unresponsive? i5-4460 + GTX 970 here.

0.68.0.5 lav
1.7.10.107 mpc
0.9.15 madvr

[huhn]

@kiwijunglist

with an receiver that isn't supporting atmos it is totally normal that all atmos data is ignored so don't think to much about it.

the 7.1 audio is already like it should be.

@khanmein

have you tried EVR with that file and settings?

in the end hybrid decoding of HEVC isn't worth it and it will highly load the GPU maybe even to a point where madVR can't be used and the performance win is very usually little.

[khanmein]

@huhn thanks so i currently un-tick it.

[P.J]

I have noticed that LAV can't play a DVsd video file correctly unless I change the Field Order to Top-Field first.
No problem with VLC/Quicktime.

[LigH]

That's strange, because DVSD usually should contain Bottom Field First interlaced.

[P.J]

That's strange, because DVSD usually should contain Bottom Field First interlaced.

Exactly :confused:
I have another video which is DV dvc BFF that plays fine.

[LigH]

Well, I guess it is not impossible to store TFF in DV, ignoring that BFF is the default. The AVI container does not store a field order, IIRC. Video streams may store a field order, depending on the format. Therefore, decoders usually have to assume a default if there is no explicit flag.

[P.J]

MediaInfo: http://pastebin.com/eRxF4Nmp
Something is wrong in LAV while de-interlacing :confused:

[huhn]

what deinterlancer are you using?

[nevcairiel]

The chroma in that file is 4:1:1, which is not a format LAV natively supports, so it converts the chroma before deinterlacing would see it, which might result in interlacing artifacts. It gets converted to 4:2:2, which should be mostly safe as it only expands it horizontally, but you never know. On the other hand, 4:2:2 deinterlacing may not be well supported by whatever is deinterlacing for you.

4:1:1 is extremely uncommon, so I wouldn't expect improvements, really.

[P.J]

what deinterlancer are you using?
Using the hardware de-interlancer makes problem but software de-interlancers work fine, lower quality though without pull-down detection.

The chroma in that file is 4:1:1, which is not a format LAV natively supports, so it converts the chroma before deinterlacing would see it, which might result in interlacing artifacts. It gets converted to 4:2:2, which should be mostly safe as it only expands it horizontally, but you never know. On the other hand, 4:2:2 deinterlacing may not be well supported by whatever is deinterlacing for you.

4:1:1 is extremely uncommon, so I wouldn't expect improvements, really.
I think all LAV needs is to detect the Field Order flag well.
I have many sd musicvideos in 4:1:1 format :(

[nevcairiel]

I think all LAV needs is to detect the Field Order flag well.

MediaInfo also says BFF, so maybe your file is just bad.

[P.J]

MediaInfo also says BFF, so maybe your file is just bad.

Just tried PotPlayer only to detect the issue, same problem happens as soon as I select GTX960's hardware deinterlacer :confused:
The only working way is to set TFF. I think HW has better quality (it has pulldown detection too)

[nevcairiel]

Just tried PotPlayer only to detect the issue, same problem happens as soon as I select GTX960's hardware deinterlacer :confused:
The only working way is to set TFF. I think HW has better quality (it has pulldown detection too)

Sounds like your hardware deinterlacing gets confused when it has to process 4:2:2, you could always use software deint.

[P.J]

Sounds like your hardware deinterlacing gets confused when it has to process 4:2:2, you could always use software deint.

But it deinterlaces other 4:2:2 videos correctly:

H.264 10-bit 4:2:2 TFF
MPEG2 8-bit 4:2:2 TFF
DV dvc 8-bit 4:1:1 BFF: http://pastebin.com/p709gXVN

Maybe something is wrong with Sony DVsd :confused:

[kiwijunglist]

@huhn - i thought i might be missing out on some important sounds from the object based channel.

[LigH]

@ P.J.:

Again ... decoders may assume BFF just because it's "DV in AVI". There may be no flag at all, neither in the AVI container, nor in the DV video stream, reporting whether TFF or BFF is present. Even MediaInfo may assume only due to the format.

Unfortunately, I am not sure if there is a flag. And if there is any, it may even be wrong. All I know is: The video should not have been saved as NTSC DV in AVI. In addition to the uncertain field order, it also lost chrominance precision.

[P.J]

Unfortunately, I am not sure if there is a flag. And if there is any, it may even be wrong. All I know is: The video should not have been saved as NTSC DV in AVI. In addition to the uncertain field order, it also lost chrominance precision.

How can I ensure if the video is really BFF or TFF?
If not in avi then what? :confused:

[LigH]

Never rely on flags. Always check the content: Compare the results of

AviSource(...)
AssumeTFF()
Bob()

and

AviSource(...)
AssumeBFF()
Bob()

The field order which does not judder is the correct one. If neither is worse, it may not be interlaced at all... For more complex issues (Telecine, blending), watch the result slowly and try to discover the temporal progress attentively.

[P.J]

I think this helps to find the correct information, detects the right field order and pulldown after I press "Guess 3:2 Pulldown":

http://phota.me/3etN.png

[LigH]

Automatic detections may help when they work reliably. But I bet there is footage where this automatic detection fails. The human brain is able to recognize image content and motion better than most current software solutions with reasonable complexity. You should not miss at least a brief glance after an automatic detection, just to confirm.

[P.J]

Never rely on flags. Always check the content: Compare the results of

AviSource(...)
AssumeTFF()
Bob()

and

AviSource(...)
AssumeBFF()
Bob()

The field order which does not judder is the correct one. If neither is worse, it may not be interlaced at all... For more complex issues (Telecine, blending), watch the result slowly and try to discover the temporal progress attentively.

Well, I saved that script in an avs file and played it with MPC-HC x86.
AssumeTFF() fixes the problem, a bit stretched/blurry and lower quality though maybe due to Bob.

And with AssumeBFF(), the problem happens again, skippy frames.

[LigH]

Of course it is a bit blurry, because fields are interpolated to frames; but these scripts are for testing only, not for processing!

So you discovered that some moron stored TFF video as DV, ignoring that DV is usually expected to contain BFF by default. Blame him (also for losing chrominance precision due to 4:1:1 subsampling). Or better: Teach him which other video formats would have been appropriate, instead of NTSC DV.

[NikosD]

I return for a little to my XP-DXVA2-EVR issue to report an interesting test I did.

According to DXVA Checker, ATI installs DXVA1/2 device decoder for VC1_VLD, but only DXVA1 for H.264.

So, there is no luck to HW accelerate H.264 using DXVA2.

Using GraphStudioNext 0.7.0, I tried a VC-1 clip with LAV Video and MPC Video decoder (MPC-BE standalone) using EVR renderer.

Well, both decoders connected to EVR and managed to decode the file, but LAV Video felt back to SW decoding while MPC Video used HW acceleration for VC-1 clip!

How could you explain that besides thinking that LAV Video lacks DXVA2 acceleration while MPC Video seems to have ?

[nevcairiel]

As I have stated before, I have no interest in HWAccel on XP (or in functionality on XP in general). It works as it works. I don't remember when I last tested on XP, its been years, and I certainly do not have any plans to resume doing so (especially for hwaccel, which I would even need to run on real hardware, not just a VM)
Note that MPC video decoder supports DXVA1, so maybe thats in use.

[NikosD]

OK...Maybe you have disabled DXVA2 for Win XP, but years ago.

WinXP is a very old system, I was just doing some tests.

But are the DXVA1 and EVR compatible ?

[foxyshadis]

OK...Maybe you have disabled DXVA2 for Win XP, but years ago.

WinXP is a very old system, I was just doing some tests.

But are the DXVA1 and EVR compatible ?

Microsoft has never supported DXVA2 under XP in any way, there was nothing to disable. The only way it ever worked in some software was with ugly hacks.

Along the same lines, EVR has never been supported on XP. It's DXVA2 (and WDDM) only.

[NikosD]

If you read previous replies only a few posts earlier, a few people here don't seem to agree with you, including nevcairiel I'm afraid.

[Sharc]

Well, I saved that script in an avs file and played it with MPC-HC x86.
AssumeTFF() fixes the problem, a bit stretched/blurry and lower quality though maybe due to Bob.

And with AssumeBFF(), the problem happens again, skippy frames.
You may find this script convenient for testing. It displays the fields (half height) stacked vertically as TFF and BFF.

AviSource("........")
bottom = AssumeBFF().SeparateFields()
top = AssumeTFF().SeparateFields()
bottom = subtitle(bottom, "BFF - bottom field first")
top = subtitle(top, "TFF - top field first")
StackVertical(top, bottom)

[madshi]

Officially XP doesn't support DXVA2, but .Net 3.5 (IIRC) retrofitted XP with DXVA2 API functionality and EVR. Microsoft just lay the groundwork, though. How much of that is actually supported by the GPU drivers is up to NVidia/AMD/Intel. From what I remember, DXVA2 deinterlacing usually works in XP, while hardware decoding is a hit-and-miss, with only some codecs being supported by some GPU drivers. There are no ugly hacks needed, IIRC, deinterlacing just worked out of the box for me, once I had .Net 3.5 installed.

In any case, as nevcairiel has already stated, all of this is rather unofficial, and what works that works, and what doesn't work doesn't. So there's no use even complaining about anything not working. Just consider everything that works as an unexpected bonus. Anyone still using XP shouldn't expect a lot of support these days. It's just too much hassle for us devs to still test on XP, especially since usually using a VM won't work because we need hardware acceleration to work for doing useful tests.

My recommendation for the best media playback OS atm would be Windows 8.1. Windows XP is *REALLY* outdated. Windows 8.1 has several important improvements over Windows 7 (native 3D support, much improved desktop composition). Windows 10 is still too buggy and unstable, which may change in the future, though.

[NikosD]

The whole discussion is more academic and technical for me, than anything else.

It's not that I actually use WinXP in an every day process.

My first post was questioning WinXP HW acceleration usability of LAV Video since its support is explicitly to DXVA2.

But there were posts supporting WinXP's use of DXVA2 and usability of LAV Video HW acceleration on XP.

That's it more or less.

[khanmein]

720p .mkv h.265 is working on CUVID + enable smooth motion

i tried to drag the seek bar, it won't freeze or cause any shuttering or even unresponsive issue on mpc-hc.

DVXA-CB & native still got issue while playing .mkv h.265

[madshi]

The whole discussion is more academic and technical for me, than anything else.

It's not that I actually use WinXP in an every day process.

My first post was questioning WinXP HW acceleration usability of LAV Video since its support is explicitly to DXVA2.

But there were posts supporting WinXP's use of DXVA2 and usability of LAV Video HW acceleration on XP.

That's it more or less.
From what I remember, when I still ran XP, my AMD GPU supported h264 DXVA2 decoding, but neither MPEG2 nor VC-1. Or was it MPEG2, but not h264 and VC-1? I'm not sure. In any case, one codec was supported, the others were not. It all depends on the GPU drivers. The OS support for DXVA2 decoding is there, once .NET 3.5 is installed. It's up to the GPU drivers to support hardware decoding for specific codecs.

So if you're interested in knowing which GPU and which GPU drivers support which codecs for DXVA2 hardware decoding in XP then I suggest you open a new thread with a poll. But I don't think you'll get a lot of response because I think most users have moved on from XP in the meanwhile.

[NikosD]

If you read a previous post, I describe in detail that VC1 is only supported by drivers using DXVA2, but unfortunately only MPC-BE decoder worked with EVR and DXVA2.

LAV Video fell back to SW.

MPEG2 was always hardware assisted by ATI, till HD 6000 series that added VLD and LAV (like most decoders) doesn't support hardware assisted acceleration.

It is probably VC-1 what you remember.

[P.J]

You may find this script convenient for testing. It displays the fields (half height) stacked vertically as TFF and BFF.

AviSource("........")
bottom = AssumeBFF().SeparateFields()
top = AssumeTFF().SeparateFields()
bottom = subtitle(bottom, "BFF - bottom field first")
top = subtitle(top, "TFF - top field first")
StackVertical(top, bottom)

Yes, BFF has problem but TFF is ok.
Don't know why the problem happens with hardware deinterlacer but software deinterlacers :confused:
Maybe because hardware deinterlacer uses inverse telecine since only 2 of 5 frames are interlaced.

[LigH]

What? Did you just say it is telecined? And why did you conceal this fact from us until now?

[P.J]

What? Did you just say it is telecined? And why did you conceal this fact from us until now?

Ain't most of ntsc music videos telecined? :confused:

I was shown it here too: http://forum.doom9.org/showpost.php?p=1760831&postcount=20665

[LigH]

But, well, to deal with telecine, you never use a "deinterlacer". This is a job for IVTC.

[huhn]

try madVR IVTC and life a happy life. (control+shift+alt+t until you see "film" mode.)

[Vincent Vega]

how can i control conversion coefficient with this filter for RGB output?

i have an old dvd that looks better with 709, but any h/w decoding applies 601 apparently.

because of lack of 601/709 toggle in LAV settings i even had to install ffdshow to watch this movie.

PS: i actually wish there was a separate settings sections in LAV for old/problematic SD content where user can define s/w deinterlacer, desired output colorspace and coefficients for full software decode/processing, unchecking h/w SD decoding in main config.

[P.J]

But, well, to deal with telecine, you never use a "deinterlacer". This is a job for IVTC.

But seems the field order flag affects on it :confused:
Any software IVTC?

try madVR IVTC and life a happy life. (control+shift+alt+t until you see "film" mode.)

Thanks, will try it.

[LigH]

"Offline", in AviSynth: Try tritical's TIVTC (http://avisynth.nl/index.php/TIVTC) package = TFM (http://avisynth.nl/index.php/TIVTC/TFM)(...).TDecimate (http://avisynth.nl/index.php/TIVTC/TDecimate)(...) to revert video material from telecined to the original progressive content.

[nevcairiel]

how can i control conversion coefficient with this filter for RGB output?

You can't.

Or more specifically, by not letting LAV do this conversion. LAV intentionally does not provide such options, and will likely never get them either.
You could use post-processing filters, or use a renderer that lets you choose (madVR would be one, or possibly any renderer that accepts custom pixel shader to convert the video)

In general, LAV Video aims to be a decoder first and foremost, some conversions are only provided for convenience and/or interoperability with other components, but anything thats not native output is not a priority for LAV, and as such options for conversions are intentionally limited.
For RGB conversion this means that it either uses the information from the file, if it contains any, or guesses based on the resolution - but it does not allow you to override.

[balkerman]

Get someone else with your amplifier to test.. also try another HDMI cable.
Alter the refresh rate to as close as possible to the Media frame rate. If close enough it may work. Worked for me! [emoji14]

[macycat]

I had already done what you suggest, and when I was doing my testing, the madVR OSD indicated that I got only 1 frame repeat every 1.5 hr. So I don't think that my problem is due to frame drops/repeats/glitches (see my original post below).

It is interesting to note that I don't get any audio glitches when decoding with LAV and sending the same audio to my receiver via PCM.

The audio track plays as Dolby TrueHD. I should do some more testing to see if the problem occurs with other audio formats. It might have something to do with the fact that the audio track is (or perhaps was) Dolby Atmos (not sure what MakeMKV does for those tracks).

Thanks for the reply. :)

Alter the refresh rate to as close as possible to the Media frame rate. If close enough it may work. Worked for me! [emoji14]

nevcairiel,

I have been having an audio glitch every minute or so when bitstreaming TrueHD audio streams. I tried the new version, and I still get the occasional audio glitch, so it doesn't look like the above fix addresses this problem. I don't observe the problem when decoding to PCM.

I am using madVR, and I don't see and dropped, repeated, or delayed frames, nor do I see any presentation glitches ... just the audio problem.

Let me know if you need me to test anything for you ...

[wanezhiling]

http://forum.doom9.org/showpost.php?p=1761323&postcount=3687
LAV doesn't work too.

[balkerman]

I had already done what you suggest, and when I was doing my testing, the madVR OSD indicated that I got only 1 frame repeat every 1.5 hr. So I don't think that my problem is due to frame drops/repeats/glitches (see my original post below).

It is interesting to note that I don't get any audio glitches when decoding with LAV and sending the same audio to my receiver via PCM.

The audio track plays as Dolby TrueHD. I should do some more testing to see if the problem occurs with other audio formats. It might have something to do with the fact that the audio track is (or perhaps was) Dolby Atmos (not sure what MakeMKV does for those tracks).

Thanks for the reply. :)
Every 1.5 hours is 1-2 glitches per movie. That was never good enough for Dolby hd for me, after adjusting the refresh rate I now have about 1 per day in mad vr. Dts hd never glitched, only Dolby hd. I used a program called CRU I think... Can't remeber, am away from my pc.

Also depends on what your gpu and TV can do!!

[macycat]

How often did you get Dolby TrueHD glitches before you fine-tuned your timing? I get them every minute or so ...

Every 1.5 hours is 1-2 glitches per movie. That was never good enough for Dolby hd for me, after adjusting the refresh rate I now have about 1 per day in mad vr. Dts hd never glitched, only Dolby hd. I used a program called CRU I think... Can't remeber, am away from my pc.

Also depends on what your gpu and TV can do!!

[mzso]

Hi!

What's the reason for HW deinterlacing being restricted to HWA decoding?

[huhn]

Hi!

What's the reason for HW deinterlacing being restricted to HWA decoding?

adding hardware deinterlancing for software decoding is possible too.

and if i'm not mistaken it is a planned feature with no ETA at all.

[chros]

How often did you get Dolby TrueHD glitches before you fine-tuned your timing? I get them every minute or so ...
That's interesting. I've never had any glitches with any kind of audio (TrueHD, Atmos, DTS, DTS-HDMA, PCM) with my system (see my signature), and the timings are way off: 24.000xx , 30.000xxx , 50.000xx , 60.000xx Hz (since I can't create custom refresh rates with intel driver).

[mzso]

adding hardware deinterlancing for software decoding is possible too.

and if i'm not mistaken it is a planned feature with no ETA at all.

I see. So it's just not done yet.

[arrgh]

Hi
after having posted these requests where they don't belong (madVR thread), I'm finally coming over here...:)

I have a rather weak HTPC but the machine is able to do everything I need, including recording simultaneously 2 payTV-HD-Sat-channels while playing back a HD-recording...

But it fails on playing back 3D-MVC (lots of dropped frames). I know that the Hardware is good enough because PowerDVD and StereoscopicPlayer don't have these Problems...

therefore my feature request (1) : Hardware Decoding for MVC-mkv files

I’m of the opinion : let the Hardware do what it is payed for; more maneuvering room for software/CPU to do other things

I was using up to now MPC-BE with its internal codecs; all my movies are mkv with aac-audio; MPC-BE audio codec allows for on-the-fly aac->ac3 transcoding, which my receiver understands (aac not);

therefore my feature request (2) : on the fly transcoding of aac to ac3 and streaming through SPDIF or HDMI
Thanks

PS: why am I using aac?
- E.g in the case of a BR and DTS-audio (2 languages: original and my language), I safe easily 2GB disk-space
- Very often the audio dynamic is so extreme (“speech” vs. “plane crash”) that it becomes impractical (my neighbors are not interested to know which movie I’m watching)
- In the case of DVD (PAL) I can do during the conversion also a slowdown to 23,976 (original framerate) without any later trouble with third party tools like Reclock…everything is 23,976

[nevcairiel]

H.264 MVC Hardware decoding is not likely to appear anytime soon, and even then likely only for Intel GPUs, anything else will probably take a looong time.
For audio transcoding, LAV Audio will never offer that, sorry.

[gpower2]

Hi nev, I am using latest stable version 0.68 with MPC-HC and madVR 0.90.13 with DXVA2 native on NVidia GTX 560 Ti and it works like a charm!

However I am having serious desync issues with ProgDVB (http://www.progdvb.com/) when watching live DVB-T 1080p H264 video stream with MPEG Audio as stated here (http://forum2.progdvb.com/viewtopic.php?p=67264&sid=6136dfce9351322c4af3439b3714c7dd#p67264).
Since ProgDVB's author didn't continue with this issue, I decided to try my luck here. :p

I wouldn't bother you with this issue here, unless Microsoft's DTV codec performed greatly with no desync issues.

Is there anything I could try to test if this is indeed a problem with LAV Filter? Is there anything you could suggest me to do in order to help me pinpoint the issue and perhaps any bugs (if there are) in LAV filter's code?

Thanks for your great work all these years! :)

[balkerman]

How often did you get Dolby TrueHD glitches before you fine-tuned your timing? I get them every minute or so ...
It's been a few years but it started at every few minutes at or so at 24hz then 1-2 per movie at 23.?hz, now doesn't happen anymore. My current refresh rate is a close as can get to the third decimal. [emoji41]

When I adjusted the tv refresh rate my ps4 couldn't recognise the display resolution and reset itself to default screen resolution. I had to manually set it back...

[GCRaistlin]

I'm trying to play 2160p video downscaled to 1080p: LAV Video with software decoding + madVR + MPC-HC, WinXP SP3. The decoding speed is too slow - madVR reports: "decoder queue 1-3/16". Though MPC-HC takes only 30% of CPU so more than half of my Intel Xeon E5430 resources are unused. I tried to manually set number of threads to 4 - the same result. What can be done about it?

[nevcairiel]

If your system is too slow, there is nothing I can do about that. Your CPU is extremely old, I wouldn't expect much from a 9 year old model when handling 4K content.
Multi-threading can only work when the individual threads work fast enough on their own, because most frames depend on the previous frames, and if those don't get done, no number of cores is going to make it run faster.

Even more so, you are running 32-bit, wasting half of the speed already. A top of the line CPU can barely manage to decode HEVC 4K content in 32-bit, because its just so much slower than 64-bit for the HEVC decoder.

[GCRaistlin]

What is the minimal hardware requirements to play 2160p downscaled to 1080p? Will it help if I just change HD5770 to HD7770? Will migrating to LAV x64 help?

[nevcairiel]

What is the minimal hardware requirements to play 2160p downscaled to 1080p? Will it help if I just change HD5770 to HD7770? Will migrating to LAV x64 help?

I don't provide hardware advice, it usually greatly depends on the video in question more than anything. Resolution is not everything.

That GPU upgrade would not help anything, though.

[sneaker_ger]

Will it help if I just change HD5770 to HD7770?
No. You need something like a GTX960 that has a pure hardware decoder capable of crunching 4K HEVC. You must be able to decode faster than real-time before you worry about downscaling. And you will probably have to update to a newer Windows version as well to make use of LAV DXVA2 decoding.

Will migrating to LAV x64 help?
It will be much faster but maybe even that is not fast enough. Like nevcairiel said 4K can be extremely demanding. If you are bound to 32 bit try Lentoid HEVC decoder (http://xhevc.com/en/hevc/decoder/download.jsp), it's much faster than LAV for 32 bit.

We have a separate thread for HEVC decoder performance comparison:
https://forum.doom9.org/showthread.php?t=171219

[GCRaistlin]

If you are bound to 32 bit try Lentoid HEVC decoder, it's much faster than LAV for 32 bit
The video I try to play is H.264, Lentoid HEVC Decoder doesn't support it.

[huhn]

for 8 bit UHD h264 you can try a geforce card.

UHD was a long term problem of AMD.

but you are not getting far with this if you care about decoding UHD you will need a HEVC 10 bit decoder sooner or later.

UHD h264 decoding is usually doable with a beefy CPU.

and 32 bit could be a real problem. decoded UHD frames are huge!

[mzso]

The video I try to play is H.264, Lentoid HEVC Decoder doesn't support it.

I could decode a 2016p30 AVC video (http://bbb3d.renderfarming.net/download.html) real time with a not much faster E5450.

Of course I'm not on obsolete XP, and I'm using the x64 version.

[GCRaistlin]

with a not much faster E5450
My CPU is overclocked to 3.71 GHz so I believe it's even faster than yours (if you haven't done the same trick of course). Thanks for the info, will try Win8.1 x64.

[jebediah]

I was trying to watch a 4K movie for the first time and noticed severe lags.
Opened up GPU-Z and the GPU usage stayed at 12-22% the whole time, so i gave Hardware Acceleration a try for the first time but still the GPU usage says at max 22%.
Any idea what i'm doing wrong?

MPC-HC
LAV filters as external filter
madVR
@
i3 4330
ASUS GTX 960 4GB

[P.J]

I was trying to watch a 4K movie for the first time and noticed severe lags.
Opened up GPU-Z and the GPU usage stayed at 12-22% the whole time, so i gave Hardware Acceleration a try for the first time but still the GPU usage says at max 22%.
Any idea what i'm doing wrong?

MPC-HC
LAV filters as external filter
madVR
@
i3 4330
ASUS GTX 960 4GB

HEVC or H.264? Try EVR and watch for Video Engine Load in GPU-Z
Make sure you have enabled HEVC and H.264 in LAV

[jebediah]

HEVC or H.264? Try EVR and watch for Video Engine Load in GPU-Z
Make sure you have enabled HEVC and H.264 in LAV

What do you mean by EVR?
It's a 4K demo in H.264 and here's some screens while playing the movie.

http://s9.postimg.org/59uw956hr/Namnl_s1.jpg

http://s9.postimg.org/8s6w5j7dr/Namnl_s2.png

[huhn]

most likely 10 bit H264. and that's not supported.
can you post media info information?
if this file is h265 10 bit (HEVC) than CUVID can't deal with it.
but DXVA can decode 10 bit HEVC.

[jebediah]

most likely 10 bit H264. and that's not supported.
can you post media info information?
if this file is h265 10 bit (HEVC) than CUVID can't deal with it.
but DXVA can decode 10 bit HEVC.

http://s12.postimg.org/4mmiol5gd/Namnl_s.jpg

Should i choose Copy-back or Native?

[P.J]

Should i choose Copy-back or Native?

Yes..

[huhn]

http://s12.postimg.org/4mmiol5gd/Namnl_s.jpg

Should i choose Copy-back or Native?
this isn't media info at all.

but look at the picture i guess it is HEVC not h264. madVR usually detects h264 and shows this in the OSD.

with madVR i would use copyback for better picture quality.

[jebediah]

this isn't media info at all.

but look at the picture i guess it is HEVC not h264. madVR usually detects h264 and shows this in the OSD.

with madVR i would use copyback for better picture quality.

Sorry, my bad.
The 4K demo is in .ts but copy back seems to work with it.
I still have some lags time to time and GPU usage stays at 55-60%

Think it's time to OC the CPU soon :)

[P.J]

Sorry, my bad.
The 4K demo is in .ts but copy back seems to work with it.
I still have some lags time to time and GPU usage stays at 55-60%

Think it's time to OC the CPU soon :)

Why? :scared:

GPU usage should be very low, maybe due to MadVR.
Also check your Video Engine Load.
And put the report of video file information with MediaInfo here.

[NikosD]

Do we know something about this .encm video format ?

http://demo-uhd3d.com/fiche.php?cat=uhd&id=141

It looks like it's working only on Panasonic TVs.

LAV (ffmpeg) seems to not recognize it.

[LigH]

I would read "ENCM" as "Encrypted movie" and conclude: "You don't want to, you don't have to."

[chros]

Cosmetic bug regarding to the new hardware accelerator device selection in LAV video decoder?
- MPC-HC latest nightly (1.7.10.130, lav filters 0.68.0.5)
- optimus sytem (see my signature)
- I'm using nvidia for hardware acceleration (I can confirm it using gpu-z or nvidiainspector)
Properties page shows:
- Active Hardware Accelerator: Intel HD Graphics 4000
- Hardware device to use: (available selections)
-- Automatic
-- NVIDIA Geforce GT 650m
-- NVIDIA Geforce GT 650m

[nevcairiel]

It just lists all devices that D3D9 enumerates, if Optimus is being funky, then thats just how it is.

[jebediah]

Why? :scared:

GPU usage should be very low, maybe due to MadVR.
Also check your Video Engine Load.
And put the report of video file information with MediaInfo here.

http://s12.postimg.org/9lqizc6ot/image.jpg

http://s12.postimg.org/meemz9ial/image.jpg

[mzso]

What do you mean by EVR?
It's a 4K demo in H.264 and here's some screens while playing the movie.

Sorry, my bad.
The 4K demo is in .ts but copy back seems to work with it.
I still have some lags time to time and GPU usage stays at 55-60%

Think it's time to OC the CPU soon :)

You don't know what EVR is? That's the renderer you replaced with madVR...
Anyway I'd guess that you chose too brutal algorithms for scaling in madVR and your system can't handle the downscaling from 4k. It's also evidenced by the fact that on your screenshot the decoder queue is full and the render queue is low.
CPU is usually maxed out when your system can't handle the decoding. So OC-ing the CPU won't help you at all.

Disable all image enhancements (and other fancy things) and set scaling algorithms to nearest neighbor to test.

[P.J]

What do you mean by EVR?

Try EVR/EVR-CP:

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

Are you using madVR?

[huhn]

try to lower the GPU queue to 5 this should be all you need.

at least for that file in question.

[hubblec4]

Hi nevcairiel

Before I start with tests, a short question.
Is LAV able to handle the ChapterSegementEditionUID entry? If so, will then played all chapters from the linked editon?

[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.

[nevcairiel]

Those, but only the second version of the 750Ti, too, the original one didn't, which shops may not mark clearly (NVIDIA ran out of the chip the original was build from, so they silently replaced it)

[DragonQ]

I see, thanks. Anyone know a reliable way of distinguishing the second editions of 750 Ti from the originals? I see "750 SE" listed on Wikipedia but pretty much nowhere else on the internet...

[ryrynz]

I see, thanks. Anyone know a reliable way of distinguishing the second editions of 750 Ti from the originals? I see "750 SE" listed on Wikipedia but pretty much nowhere else on the internet...

OEM only card? Don't bother, get something that's actually available. Get the Asus GTX 950 Mini (http://wccftech.com/asuss-latest-geforce-gtx-950-graphics-card-requires-6pin-connector-features-1228-mhz-boost-clocks/) if you must get something that fits into the 75W envelope.

[3421]

The decoder uses SSE4.2 or AVX1 command by decoding the video on CPU?

[hubblec4]

Hi nevcairiel

I think there is an issue in LAV Video Filter.

I have StarTrek TNG (Bluray) series with ordered chapters.
A part of the series should not play, the new timeline is correct but when play the next ordered chapter, the video freeze for 5 seconds(audio play fine).
Then I disabled the LAV Video Decoder in the MPC-HC settings and all plays fine.
As video decoder is Microsoft DTV-DVD Video Decoder in use.
Then I try CoreAVC-Decoder, the video don't freeze but stuttering for 5 seconds.

[nevcairiel]

Sounds to me like your video has a gap in the middle. Don't do that if you want fluent playback everywhere.

[hubblec4]

Sounds to me like your video has a gap in the middle. Don't do that if you want fluent playback everywhere.

a gap??

You mean ordered chapters like this:
00:00.00.000 - 00:02:00.000
00:04:00.000 - 00:10:00.000

The gap are the 2 minutes (after 2 minutes), that is the reason for playing is broken?

[oddball]

I cannot find an answer by searching this thread.

I am playing back a Prores trailer with 8 channel audio.

I am unable to get all 8 channels sent to the correct speakers at the same time to my Denon AVR via HDMI. When I right click on LAV Splitter it shows 8 audio channels but only 1 is selected. This equates with me being able to hear only the center channel being played. If I switch to one of the other channels it plays that to the center channel instead. LAV Splitter and audio decoder only seems to send one channel at a time and only to the center channel from what I can tell.

I have tested with some other files and they send multi channel audio to all speakers over HDMI fine (bitstreamed or just decoded depending how I have LAV audio set).

Screencap provided.

Any ideas?

[nevcairiel]

Some "pro" files have every channel as an individual audio track. LAV does not offer anything to combine them into one, sorry. Such files are honestly not mean for consumer consumption.

[Qaq]

Yeah, I have similar prores too. In case someone would want to provide a patch, what would be correct order to join tracks?

[AngelGraves13]

Yeah, I have similar prores too. In case someone would want to provide a patch, what would be correct order to join tracks?

You can use eac3to to extract to individual wavs and then re-encode to FLAC.

[Aegwyn11]

I have a very high bitrate (~1.4 Gbps) UHDp59.94 Prores422 HQ file that I'm having trouble getting to play smooth. I think it has something to do with the decoder...if I pull up the MadVR stats, I can watch the decoder queue drop and drop until it gets really low (around 2), at which point frames start dropping like crazy. If I increase the CPU queue size, it takes longer for frames to start dropping (since there's more decoder queue to deplete), but it doesn't prevent the issue. It seems to happen with high complexity scenes. I never see my CPU go past 70-80% utilization (i7-5930K), usually in the mid 70's at the complex scenes.

If I take the file and decode/re-encode it to UTvideo in FFmpeg, it plays back smooth (but i lose the 10 bit-ness of the source since UTvideo in FFmpeg only supports 8 bit).

Also, if I decode/re-encode the file as UHDp29.97 Prores 422 HQ (half frame rate, 30p), it plays back smooth....

EVR-CP acts a bit differently...if I display the stats, I can see it plays back at a fairly constant 49-50 frames per second (dropping ~10 frames a second), and the green line in the lower right corner graph climbs at about a 30-40 degree angle out of the screen then drops back to the red line for a bit before climbing again. Increasing the EVR buffers delays the climb but doesn't stop it from happening.

Any ideas? My main goal is to be able to play back pristine quality UHDp60 4:2:2 10 bit content. I don't have to stay Prores if there are other good options, but it doesn't seem there are. I tried just going to v210, but my SSD isn't quite fast enough to handle that much throughput without some frame drops here and there (although not nearly as many as with Prores...).

[sneaker_ger]

Are you using 32 bit or 64 bit LAV? The 64 bit prores decoder seems to be much faster.

[Aegwyn11]

I believe 64 bit....I installed using the installer that installs both. I'm using MPC-HC 64 bit, so I'm assuming that forces all the filters to also be 64 bit.

Could I be running into a max number of threads issue? In other words, I've got spare juice left in my CPU, but the decoder isn't able to take advantage of it?

[AngelGraves13]

I believe 64 bit....I installed using the installer that installs both. I'm using MPC-HC 64 bit, so I'm assuming that forces all the filters to also be 64 bit.

Could I be running into a max number of threads issue? In other words, I've got spare juice left in my CPU, but the decoder isn't able to take advantage of it?

There is currently no hardware capable of playing back 60p at that resolution smoothly. We'll need a new generation of video cards for that.

[Vasilich]

There is currently no hardware capable of playing back 60p at that resolution smoothly
according to intel specs http://ark.intel.com/products/88185/Intel-Core-i5-6400-Processor-6M-Cache-up-to-3_30-GHz
Max Resolution (HDMI 1.4)‡ 4096x2304@24Hz
Max Resolution (DP)‡ 4096x2304@60Hz
so if the monitor is connected via displayPort, then it supports 4k@60Hz. SOme MoBo manufacturers can install external HDMI 2.0 transmitter using DP lines, but that is still rare now.
according to HW decoding - all testing web sites (anandtech, tomshardware etc.) writing that it supports smooth decoding for 4k@60Hz (while using 2% of CPU only), but ONLY 8Bit.

[Aegwyn11]

Guys - thanks for the responses. Please note that in my original ask for help, I said that I'm able to do UHDp60 just dandy if I reencode using UTvideo. It also seems that raw playback (v210) would work as well if my SSD had just a bit more throughput. RAID nVME drives would probably do it, but that isn't a tenable solution for me. I'm 100% confident my hardware has the capability to do this (after all, I can decode two Prores UHDp30 files simultaneously with no problems and CPU to spare), just not sure if the software decoder is optimized for this yet (hence my questions).

[Aegwyn11]

Also, FWIW, I'm using an i7-5930K (Haswell-E, no onboard video) and GTX 950. I can use either the HDMI 2.0 output or a Club3d DP 1.2 to HDMI 2.0 adapter to get UHDp60 RGB to the TV. Either works fine.

[nevcairiel]

> 1Gbps bitrates are just extremely high, don't expect that to get much faster without faster hardware. ProRes uses slice-based threading, which means performance scaling depends on the video encoding. If it has enough slices it should theoretically be able to get to 100% usage, if nothing else bottle-necks the process.

[Aegwyn11]

Thanks nevcairiel, you gave me an idea. I'll try forcing fewer macroblocks per slice in ffmpeg (should result in more slices) and try again.

Edit - I reduced the macro blocks per slice to 4 (default is max, 8) and it seemed better. File size was very slightly larger and CPU usage was slightly higher. I just tested 30s, I'll do a bigger test next week.

[P.J]

I have a very high bitrate (~1.4 Gbps) UHDp59.94 Prores422 HQ file that I'm having trouble getting to play smooth.

It's huge, usually need fast 8-core o.O
Try Quicktime x86, it usually plays a bit faster than ffmpeg x64.

[Aegwyn11]

It's huge, usually need fast 8-core o.O
Try Quicktime x86, it usually plays a bit faster than ffmpeg x64.

I did try QuickTime, it couldn't manage it either. I don't think more cores would help, as my 6 core processor isn't fully utilized. Maybe faster cores would help, but I'm already at 3.5 Ghz. Maybe a Skylake i7 could do it, who knows. Or maybe Broadwell-E when it comes out.

[baii]

I did try QuickTime, it couldn't manage it either. I don't think more cores would help, as my 6 core processor isn't fully utilized. Maybe faster cores would help, but I'm already at 3.5 Ghz. Maybe a Skylake i7 could do it, who knows. Or maybe Broadwell-E when it comes out.
Get some better cooling and pump it to 4ghz, that should be more than 10% performance increase.

Quad i7 pretty much struggle for most high bit rate 4k. Let alone 60fps ones.

Sent from my SM-T700 using Tapatalk

[Reino]

It appears some hls-streams require a user-agent nowadays. Do you think it's possible to add an option to enter a useragent, or have it send a user-agent by default?

[filler56789]

Regarding the recent LBR DTS support in dcadec...
¿isn't it good-enough for being added to LAV Audio?

[nevcairiel]

LAV doesn't use dcadec anymore, and its not going to again just for this.
Once its finished it'll be integrated into FFmpegs dca decoder, at which point LAV will get it.

[nevcairiel]

It appears some hls-streams require a user-agent nowadays. Do you think it's possible to add an option to enter a useragent, or have it send a user-agent by default?

LAV already sends a user-agent. Probably just not one your source likes.
I thought about making a configurable user-agent before, but not sure yet. Those stream providers should realise that user-agents are such a weak protection, everyone can just fake them. Why make us go through all this trouble.

[wanezhiling]

http://112.17.0.10:8085/ysten-business/live/fhchinese/yst.m3u8
Hi, I tried latest nightly mpc-hc and it seemed not work.

[P.J]

Just tried to play an HEVC 8-bit 4K video with Atom x7-Z8700 but got square artifacts: :confused:

http://1.t.imgbox.com/3b2aIiqh.jpg (http://imgbox.com/3b2aIiqh)

[LigH]

Not only squares. You can also see jagged lines in the left half, at the top of the blue tint, like the devil's horns.

But I believe it's a playback error in a way that chroma planes are shifted and scaled. Maybe not a decoder but a renderer issue, misinterpreting the decoder's output or not handling the dimensions well?

[nevcairiel]

This is a bug in the Intel driver.

[P.J]

This is a bug in the Intel driver.

It's a Surface 3, where can I find the latest driver for Atom x7-Z8700? :confused:

[Reino]

LAV already sends a user-agent. Probably just not one your source likes.
I've just installed the latest nightly and I see you've fixed it. Thanks!

[nevcairiel]

I've just installed the latest nightly and I see you've fixed it. Thanks!

I didn't really change anything though, but if it works I won't complain. :)

[Reino]

I just checked with FFMpeg and it seems they removed the 'user-agent-protection'. *sigh*
Guess they read your post. ;)

[mbordas]

I have a question about what seems to be an old/known problem, but I figured it wouldn't hurt to ask in case anyone has suggestions about possible workarounds.

I have two blu rays with mono LPCM soundtracks (Badlands, French Lieutenant's Woman). The soundtrack in FLW has been remuxed to mono AAC.

Lav splitter correctly identifies the BL track as mono, and plays it through the left speaker. This seems to be a bug in the driver which has been around a long time:

https://forum.redfox.bz/threads/reclock-1-8-5-7.37473/#post-239907

But with FLW, Lav Splitter says:

A: Mono [eng] (aac he-aac, 48000 Hz, stereo) [default]

and LAV audio is trying to play it as stereo. I can work around this by enabling mixing and setting the output speaker configuration to mono. Then it plays through the left speaker.

I have "expand mono to stereo" unchecked in both Lav audio and reclock.

Any thoughts? The desired behavior is to have these soundtracks play from the center speaker. But short of rewiring my receiver, I can't find any way of doing this.

[nevcairiel]

If the stream info says its stereo, then your AAC track is in fact stereo, or at least the decoder says it is. Maybe an encoding stage screwed up? I could only comment on the real stream with a sample.
If the audio driver has a bug that assigns the channel wrong, what do you expect to change however? The driver doesn't care if its PCM source or AAC source.

As a workaround for mono sources, you could tell LAV to mix to 5.1, if the mono channel is flagged as "Center", then it'll remain in the center and silent channels will be added, since LAV doesn't actually support upmixing audio.

[3421]

This video file video decoder only plays one second. The audio decoder processes the sound with distortion. Another player plays it as it should be, but not all players norm play. 0-26

https://www.cubbyusercontent.com/pl/ex%d0%b0mple.7z/_23416bebc04c431fbc74af2d3e6d4de3

[chros]

I'm using MPC-HC.1.7.10.180.x64 (latest nightly with LAV 0.68.0-16-git) and :
1. "Prefer audio stream with Highest Quality" doesn't seem to work: it always picks up the 2ch AAC stream (default stream)
- media info from mkv
Video: MPEG4 Video (H264) 1916x1076 23.976fps [V: English [eng] (h264 high L4.0, yuv420p, 1916x1076) [default]]
Audio: AAC 48000Hz stereo [A: English [eng] (aac lc, 48000 Hz, stereo) [default]]
Audio: Dolby AC3 48000Hz 6ch 384kbps [A: English [eng] (ac3, 48000 Hz, 5.1, 384 kb/s)]
Subtitle: UTF-8 [S: CC [eng] (subrip)]
Subtitle: UTF-8 [S: No subtitles]

2. Is it possible to remove all additional filters from DS chain or reinit the chain when "Remove audio decoder on Audio Stream Switch" is selected and switch happens?
Reason behind is that I'm using AC3filter (x64) for upconvert 2ch sources only for decoded stream (it gets PCM data). But when I switch stream (with the example in the 1. point) from AAC to AC3 manually then AC3filter stays in the chain and breaks playback. Only solution is block AC3filter in MPC-HC then restart playback (or modify the default track info in header in the file itself).

Thanks

[nevcairiel]

1. "Prefer audio stream with Highest Quality" doesn't seem to work: it always picks up the 2ch AAC stream (default stream)

Default tags take precedence over everything else, except language choices.


2. Is it possible to remove all additional filters from DS chain or reinit the chain when "Remove audio decoder on Audio Stream Switch" is selected and switch happens?

Not with LAV only, a player could theoretically do that for you, since its really the players job primarily to build the graph.

[mbordas]

If the stream info says its stereo, then your AAC track is in fact stereo, or at least the decoder says it is. Maybe an encoding stage screwed up?

sure, it could be bad encoding, but mediainfo shows the following:

Audio
ID : 2
Format : AAC
Format/Info : Advanced Audio Codec
Format profile : HE-AAC / LC
Codec ID : A_AAC
Duration : 2h 3mn
Channel(s) : 1 channel
Channel positions : Front: C
Sampling rate : 48.0 KHz / 24.0 KHz
Frame rate : 23.438 fps (1024 spf)
Compression mode : Lossy
Delay relative to video : -105ms
Title : Mono
Language : English
Default : Yes
Forced : No

here's a sample:

http://speedy.sh/BqYh5/mono-audio-sample.mkv

[mbordas]

As a workaround for mono sources, you could tell LAV to mix to 5.1, if the mono channel is flagged as "Center", then it'll remain in the center and silent channels will be added, since LAV doesn't actually support upmixing audio.

Thanks, that seems to work (for the pcm mono track anyway). It loses the bit-exactness of the PCM track, but oh well - you can't have everything :)

But, maybe I'm confused about something. If I check "Convert Ouput to Standard Channel Layouts", shouldn't it always set the channel mask to 0x63F in a 7.1 system? Then if the track is stereo or mono or whatever, it will fill in the missing channels with silence? That's what your workaround seems to be doing. What am I missing?

[nevcairiel]

But, maybe I'm confused about something. If I check "Convert Ouput to Standard Channel Layouts", shouldn't it always set the channel mask to 0x63F in a 7.1 system? Then if the track is stereo or mono or whatever, it will fill in the missing channels with silence? That's what your workaround seems to be doing. What am I missing?

That option doesn't look at your speaker settings, it just converts it to one of a few "standard" layouts that are generally known to work well over HDMI, thats generally Mono, Stereo, 5.1, 6.1 and 7.1, always choosing the one that has enough channel for the source.

So that option in itself doesn't help a mono case, since its still going to pick Mono as a valid choice. I could of course remove Mono from the list of standard layouts, which would automatically promote it to 5.1, since Stereo has no center speaker.
But people with a fully working Mono setup probably wouldn't appreciate that.

If you want to force one partiular layout for all combinations, thats what the mixer setting is for.

[LigH]

There may be AAC encoders which encode mono automatically as doubled stereo due to some compatibility reasons (e.g. with poor decoders failing for center-only mono tracks), unless you disable this feature with a parameter, I guess. But I don't remember in detail.

[mbordas]

OK that makes sense, and it gives me a working option which is fine. But I think you could also reverse the logic - set the default to the highest number of channels? It wouldn't break anything and it would make the mono case work better.

[nevcairiel]

OK that makes sense, and it gives me a working option which is fine. But I think you could also reverse the logic - set the default to the highest number of channels? It wouldn't break anything and it would make the mono case work better.

Its meant to be a safe option to use, which wouldn't affect 99% of all files, since most are already in one of those layouts, but then you encounter one of those pesky 5.0 or 3.0 files, and one of the channels comes out as the LFE and you miss the dialog .... thats really the reason I build it a while ago, something I can turn on always and make sure all channels go where they are supposed to, without having any side-effects on proper files.

sure, it could be bad encoding, but mediainfo shows the following:

Audio
ID : 2
Format : AAC
Format/Info : Advanced Audio Codec
Format profile : HE-AAC / LC

The reason is that it uses HE-AAC for a Mono track, which makes the decoder think it should be applying Parametric Stereo decoding, resulting in stereo again.

It is technically a bug in the FFmpeg aac decoder, but I wouldn't hold my breath on a quick fix, since its a rather odd situation.
If you must use AAC, should probably encode it in plain AAC LC, not using HE-AAC.

[mbordas]

bugs everywhere :(

Thanks for the info.

[nevcairiel]

bugs everywhere :(

Thanks for the info.

I checked the AAC specification some more, and apparently its not a bug but just defined behavior in this case:
Since a received mono stream will result in a stereo output if Parametric Stereo data is present in the stream, the
HE AAC v2 Profile decoder shall assume that PS data is available and decide the number of output channels to be
two for a single channel element containing SBR data, and thus also possibly PS data. If no PS data is found the
mono output shall be mapped to the two opened channels for every single channel element.

Its not what you wanted to hear, but apparently its not wrong what we're doing.
As said, you can avoid this by using AAC LC, ie. not using SBR.

[3421]

LAVSplitter: witn mpc mpeg source splitter file play normally.

[Sp00kyFox]

I have a problem here, maybe someone knows whats going on here. with avisynth I scaled a fullhd source down to 1280x532 and encoded it with x265 (16bit in, 12bit out, i444) but when playing it the resolution is 1280x528, the lower 4 pixel lines are cropped away. even though mediainfo says the resolution is how it should be. I'm using the latest stable version of lavfilters v0.68.0 with madvr v0.90.17 and zoom player v11.1.

edit: ok it seems the problem is fixed in the lavfilters version provided by the latest k-lite pack which is v0.68-26-g5be9d.

[chros]

Default tags take precedence over everything else, except language choices.

Not with LAV only, a player could theoretically do that for you, since its really the players job primarily to build the graph.
Thanks for the answers!

[Sp00kyFox]

to the problem I mentioned before, this must be the one which was fixed quite recently:
https://trac.ffmpeg.org/ticket/4980
Nevermind, ffmpeg always multiplies by 2 the default window offset (assuming 4:2:0), while it should use a coordinate system that depends on the chroma format.


I tried it again with the latest nightly build (v0.68.0-28, 2016-04-21) but the issue is still there. I don't understand why it is fixed in v0.68-26-g5be9d (which is only available in the k-lite pack to my knowledge) but not in the nighly builds. is this an unofficial patch or an oversight? again, here is a test clip:
https://mega.nz/#!WJxBAJBR!P1xhjdkSPC-eIlY998PEELj2Rflc0USw4HcxwXEQC4M

maybe the integrated ffmpeg libraries are not quite up to date?

[clsid]

FFmpeg was last updated on February 1st in LAV. K-Lite uses that fix on top.

That sample plays with green colors here with EVR-CP. So colorspace conversion seem b0rked. Plays ok with madVR,

[VictorLS]

Sorry for offtop: can anyone explain why lastest Lentoid HEVC Decoder http://www.strongene.com/en/downloads/downloadCenter.jsp (like Elecard HEVC Video Decoder) shows color abracadabra in any SAT program with any Renderer while receiving 10bit 3840x2160p50 h265 http://mir.cr/E4UZKWXX, but when playing recorded video with LAV Splitter both decoders shows good picture? LAV Video Decoder shows good while receiving and playing though.
PS. I need "Skip non-key frames" in decoder but Elecard can not correct color abracadabra with 10bit (8bit shows good always). I plan to write to Stongene to add "Skip non-key frames" in their Lentoid if they correct 10bit while receiving.

[nevcairiel]

If you start your post with "Sorry off-topic", maybe just don't post at all.

[VictorLS]

nevcairiel
If you add "Skip non-key frames" feature earlier (or nVIDIA produced single slot passive-cooled videochip with HWAccel like GTX950/960) I'd never ask about other decoder here ;)

[nevcairiel]

I said this before, this will never happen. I don't implement features with the whole purpose of destroying image quality. So please take your off-topic discussions elsewhere.

[Sp00kyFox]

FFmpeg was last updated on February 1st in LAV. K-Lite uses that fix on top.
ah, got it. so I guess we can expect it sometime in the future in the official build.

[VictorLS]

I said this before, this will never happen. I don't implement features with the whole purpose of destroying image quality.
Thanks for reply but It's a pity once more. If that feature not too difficult to implement in LAV Video Decoder may be you'll spend some your free time to write special version of decoder with "Skip non-key frames" feature on/off, upload it somewhere and send me link in PM? I'll never say about it anyone )
So please take your off-topic discussions elsewhere.
I promise I'll never ask you about this feature in the future.

[P.J]

How to properly play HDR videos on regular monitors?
They look washed-out/dull :(

[Manni]

How to properly play HDR videos on regular monitors?
They look washed-out/dull :(

Use MadVR with LAV. MadVR does a very good HDR to SDR conversion using the metadata provided by LAV, with many options to get the best possible result on your non HDR display.

[aufkrawall]

Would it be possible that 1080p 60fps get blacklisted for DXVA2 CB/native with < GCN 1.2 GPUs? I get unstable playback with it with madVR (with R9 390) and it's not possible to just uncheck 60fps (videos with lower requirements run fine).

And maybe RGB output levels and dithering method options should be grayed out when all formats are checked? Or am I mistaken and these options still have an effect then?

[nevcairiel]

Would it be possible that 1080p 60fps get blacklisted for DXVA2 CB/native with < GCN 1.2 GPUs? I get unstable playback with it with madVR (with R9 390) and it's not possible to just uncheck 60fps (videos with lower requirements run fine).


I do not plan to add any performance-related hwaccel trigger options, sorry.


And maybe RGB output levels and dithering method options should be grayed out when all formats are checked? Or am I mistaken and these options still have an effect then?

Just because a format is enabled doesn't mean its supported by the renderer, it could still use something with a lower bitdepth or fallback to RGB.

[aufkrawall]

Alright, thanks for your feedback. :)

[Magik Mark]

Use MadVR with LAV. MadVR does a very good HDR to SDR conversion using the metadata provided by LAV, with many options to get the best possible result on your non HDR display.

May I ask which options work best? I could only see "hdr luminance option" for now.

HDR video on regular displays look brighter. I guess there is little we could about this without affecting the video quality of non-HDR videos

[mbordas]

A quick follow up on my earlier question about mono audio processing.

I just realized mpc-hc now supports WASAPI with the latest 1.7.10:

Completely new Internal Audio Renderer:
- Outputs sound through WASAPI (shared or exclusive)
- Employs automatic channel downmixing

It helps to read the changelog sometimes :)

(FYI, you have to set Output->Audio Renderer = Internal and the checkbox Internal Renderers->Audio Filter = Exclusive mode)

But it seems to set the channel mask to match your speaker setup, so everything works perfectly. And I can easily correct for that HE-AAC encode by just downmixing to mono in LAV audio. All is good, and no more reclock!

[NikosD]

On an old laptop with Intel 965 Express chipset and Vista x86 SP2, latest LAV filters do not provide HW acceleration for MPEG2_VLD.

I tried MPC-HC which falls back to SW and DXVA Checker which can't connect to EVR renderer.

On the other hand, MS DS MPEG2 decoder works on the same MPEG2 clips and provides HW acceleration.

[wanezhiling]

MS DS MPEG-2 decoder may provide DXVA_IDCT not VLD?
Try PotPlayer to check by its OSD

[NikosD]

It's VLD because DXVA Checker says so and the CPU usage is almost 0% in Decode mode.
Also, the decoding speed is impressive even for 1080p MPEG2 clips.

[Dodgexander]

I am using MPC-HC and LAV filters to watch live TV. When using the Intel quicksync decoder in LAV Video the aspect ratio of mpeg2 channels are always read incorrectly as 4:3 rather than 16:9.

This can be resolved by setting LAV to ignore aspect ratio, to set LAV to not use hardware decoding for mpeg2 or by using any other decoder apart from quicksync.

I have tested this with EVR to make sure it wasn't MadVR causing the problem. The problem doesn't occur with h264 channels.

[NikosD]

The only weird thing I noticed about MS DS MPEG2 decoder is that it has IMC3 output.

I don't know if that helps anywhere.

[nevcairiel]

You could always check the DXVA device with DXVA Checker if it supports NV12 output, thats a requirement for LAV (and generally also the only format any other drivers/GPUs seem to output, short of P010 for 10-bit of course). There are no other rules that would block the Intel GPU from doing its job.

[NikosD]

Maybe it's one of those proprietary decoding devices that Intel supported in older HW

[Aleksoid1978]

As i remember - old Intel do DXVA 2.0 'IMC3' output. MPC-BE support it.

[NikosD]

You could always check the DXVA device with DXVA Checker if it supports NV12 output, thats a requirement for LAV (and generally also the only format any other drivers/GPUs seem to output, short of P010 for 10-bit of course). There are no other rules that would block the Intel GPU from doing its job.

It supports only IMC3, maybe that's the reason LAV doesn't support it.

As i remember - old Intel do DXVA 2.0 'IMC3' output. MPC-BE support it.

Right!

It works perfectly with both MPC-BE and standalone filters.

Update:

MPC-Be standalone filters use output NV12 and the acceleration works perfectly.

So, I don't know what's going on with LAV!

[Aleksoid1978]

LAV support only NV12/P010 for DXVA.

[P.J]

Use MadVR with LAV. MadVR does a very good HDR to SDR conversion using the metadata provided by LAV, with many options to get the best possible result on your non HDR display.

May I ask which options work best? I could only see "hdr luminance option" for now.

HDR video on regular displays look brighter. I guess there is little we could about this without affecting the video quality of non-HDR videos

I tried madVR but it doesn't still look good compared to the non-HDR version =/

[rseiler]

Any idea what might be happning here? In order to get HEVC support in all players (not just VLC, for example), I switched from CoreAVC to LAV (this is in Win10, which doesn't actually have built-in support unless you have select few video cards).

All was well, until I noticed new behavior: mkv's (x264 or HEVC) all have a glitch, for lack of a better word, exactly two minutes before the end. It amounts to only a few seconds of the video essentially being skipped (video and audio), but it is annoying. It doesn't happen with mp4 files.

It's not a problem with the videos since the problem doesn't show up in VLC or MPC-HC or WMP. In fact, so far I've only been able to reproduce it in Bsplayer, but it most definitely did not have that problem when using CoreAVC, so it's something to do with LAV in conjunction with at least Bsplayer.

[mel2000]

Windows 10 Home/x64
Avisynth+ (x86/x64)
VirtualDub 1.10.4 (x86/x64)

Just installed Avisynth+ (x86/x64) and VirtualDub 32 or 64 bit.

Afterwards VirtualDub threw an error with my DirectShowSource scripts until I installed LAV filters. I didn't touch any LAV configurations. However although I hear audio, I only see a partially blackened video screen.

Before PLAY button
http://i.imgur.com/1XdDhbt.jpg

After PLAY button
http://i.imgur.com/GEAkImj.jpg

Can someone recommend any configuration settings that might help? Thanks.

[LigH]

Afterwards VirtualDub threw an error with my DirectShowSource scripts...

Which error exactly?

Which scripts exactly?

MediaInfo analysis of the source files to be loaded?

Why DirectShowSource, despite repetitive recommendations during the last years to avoid it in favour of native AviSynth source filters (like L-SMASH Works, FFMS2, DGDec*) and use it only in cases where all native source filters fail?

[Groucho2004]

Which error exactly?

Which scripts exactly?

MediaInfo analysis of the source files to be loaded?

Why DirectShowSource, despite repetitive recommendations during the last years to avoid it in favour of native AviSynth source filters (like L-SMASH Works, FFMS2, DGDec*) and use it only in cases where all native source filters fail?
Some context here (http://forum.doom9.org/showthread.php?t=173462).

[mel2000]

Thanks for your response LigH. I recommend that you refer to the link Groucho2004 supplied for more backstory details. Most of my Avisynth scripts are written using DirectShowSource and I would rather find out why it's no longer working than change all my scripts. I know DirectShowSource has its critics but I'm not one of them.

[mel2000]

This issue has been resolved with the guidance of posters in the previously mentioned thread.

I used CodecTweakTool to set the Preferred Splitters and Preferred Codecs to their default and recommended options. DirectShowSource is now working as expected in Avisynth+ and VirtualDub x86/x64. Thanks to all to replied with assistance on this issue.

[nevcairiel]

LAV Filters 0.68.1

LAV Splitter
- NEW: Support for reading metadata tags from MKV files (including the "3d-plane" MVC 3D tag from MakeMKV 1.9.10 and newer)
- Fixed: Improved handling of QuickTime codecs in MKV files
- Fixed: S302M audio streams with non-pcm content did not work anymore (requires an external decoder)

LAV Video
- Changed: Flags for BT.2020 are exported in DXVA2_ExtendedFormat (previously only reported to madVR, since the Windows SDK did not specify them)
- Changed: Prevent connections to the "AVI Decompressor" filter which could result in a broken image or crashes otherwise
- Fixed: Some progressive files could report the wrong FPS when software deinterlacing was turned on
- Fixed: MPEG-2 streams using "intra refresh" did not decode properly

LAV Audio
- Fixed: Decoding MPEG Audio streams from the Microsoft MPEG-1 Splitter could result in playback starting with an audible glitch


Download: Installer (both x86/x64) (https://files.1f0.de/lavf/LAVFilters-0.68.1.exe) -- Zips: 32-bit (https://files.1f0.de/lavf/LAVFilters-0.68.1.zip) & 64-bit (https://files.1f0.de/lavf/LAVFilters-0.68.1-x64.zip)

This is mostly a minor bugfix release on top of 0.68, check its release notes (http://forum.doom9.org/showthread.php?p=1759920#post1759920) if you have not done so already!

[andyvt]

LAV Filters 0.68.1

LAV Splitter
- NEW: Support for reading metadata tags from MKV files (including the "3d-plane" MVC 3D tag from MakeMKV 1.9.10 and newer)
- Fixed: Improved handling of QuickTime codecs in MKV files
- Fixed: S302M audio streams with non-pcm content did not work anymore (requires an external decoder)

LAV Video
- Changed: Flags for BT.2020 are exported in DXVA2_ExtendedFormat (previously only reported to madVR, since the Windows SDK did not specify them)
- Changed: Prevent connections to the "AVI Decompressor" filter which could result in a broken image or crashes otherwise
- Fixed: Some progressive files could report the wrong FPS when software deinterlacing was turned on
- Fixed: MPEG-2 streams using "intra refresh" did not decode properly

LAV Audio
- Fixed: Decoding MPEG Audio streams from the Microsoft MPEG-1 Splitter could result in playback starting with an audible glitch


Download: Installer (both x86/x64) (https://files.1f0.de/lavf/LAVFilters-0.68.1.exe) -- Zips: 32-bit (https://files.1f0.de/lavf/LAVFilters-0.68.1.zip) & 64-bit (https://files.1f0.de/lavf/LAVFilters-0.68.1-x64.zip)

This is mostly a minor bugfix release on top of 0.68, check its release notes (http://forum.doom9.org/showthread.php?p=1759920#post1759920) if you have not done so already!

Does LAV splitter support self-signed HTTPS urls?

Thanks

[nevcairiel]

Does LAV splitter support self-signed HTTPS urls?

Due to lack of a clear way to interact with the user to confirm verification of certificates etc, LAV does currently not verify certificates, so any HTTPS cert will work, instead of failing on untrusted certs with no way for the user to override on a case-by-case basis.

So in short, yes, it works.

[andyvt]

Due to lack of a clear way to interact with the user to confirm verification of certificates etc, LAV does currently not verify certificates, so any HTTPS cert will work, instead of failing on untrusted certs with no way for the user to override on a case-by-case basis.

So in short, yes, it works.

Thanks!

[sdancer75]

Hi

I use lav filters successfully in my direct show chain using a mfc/c++ application. I would like to give me some directions of what COM Interfaces should I use to change the audio track inside my player.

Each audio track represent a specific language.

Best regards

[aufkrawall]

Just for the record, as I talked about the video processor in AMD's Hawaii GPU before:
A few days ago I had a random blue screen of death while watching a Youtube 360° H.264 video in Firefox. Some years ago, AMD also had blue screen trouble with just plain 4k H.264 video (they simply turned it off in the drivers). I'm having the strong feeling that they forgot to turn off DXVA2 accel for 360° video or they didn't check it/aren't aware that this kind of video is affected as well. :devil:
As a consequence, I turned off DXVA2 video decoding in Firefox via about:config.

Btw: If Polaris 10 supports VP9 decoding in the drivers, will it automatically work without changes in LAV?

[nevcairiel]

Btw: If Polaris 10 supports VP9 decoding in the drivers, will it automatically work without changes in LAV?

If they support the standard VP9 DXVA2 specification from Microsoft, then it should. Assuming there are no bugs in the implementation that didn't have an impact on Intel or NVIDIA.

[3421]

http://www.streamingmedia.com/Articles/ReadArticle.aspx?ArticleID=110383
https://aomedia.googlesource.com/aom/

[strumf666]

Is this codec supported by LAV?

Format: CineForm
Codec ID: V_MS/VFW/FOURCC / CFHD
Codec ID/Info: CineForm 10-bit Visually Perfect HD (Wavelet)
Duration: 48mn 22s
Width: 4 096 pixels
Height: 2 304 pixels
Display aspect ratio: 16:9
Frame rate mode: Constant
Frame rate: 23.976 fps
Scan type: Progressive
Default: Yes
Forced: No

[LigH]

CineForm is a proprietary codec (used by GoPro cameras) but "closed-source", which means that the developers did not release source code. Reverse engineering would be required to support it by libavcodec based software. In the current full help of a fresh "non-free" ffmpeg build created using media-autobuild_suite, there is no appearance of "CineForm"; but the Wikipedia article about libavcodec (https://en.wikipedia.org/wiki/Libavcodec) lists CineForm as supported for decoding (if that is true, then LAV Filters could support it as well, if enabled). I wonder if support for this codec requires an additional explicit option while compiling ffmpeg.
__

The matching ticket in their issue tracker (https://trac.ffmpeg.org/ticket/1087) is still open, so I believe the reverse engineering is not yet complete...
__

P.P.S.: "ffmpeg -codecs" lists support for CineForm HD. This list was not part of the "full help". That means that LAV Filters may support CFHD if it uses a current libavcodec base.

[strumf666]

Ok, thanks for answering.

[mandarynek82]

Why hardware accelerator is not active? Win 10 x64, geforce gtx 580.

http://s6d7.turboimg.net/t/27537384_Snap1.jpg (http://www.turboimagehost.com/p/27537384/Snap1.jpg.html) http://s6d7.turboimg.net/t/27537385_Snap2.jpg (http://www.turboimagehost.com/p/27537385/Snap2.jpg.html)

[huhn]

not all files can be hardware decoded.

what file are you trying to play?

[kolak]

CineForm is a proprietary codec (used by GoPro cameras) but "closed-source", which means that the developers did not release source code. Reverse engineering would be required to support it by libavcodec based software. In the current full help of a fresh "non-free" ffmpeg build created using media-autobuild_suite, there is no appearance of "CineForm"; but the Wikipedia article about libavcodec (https://en.wikipedia.org/wiki/Libavcodec) lists CineForm as supported for decoding (if that is true, then LAV Filters could support it as well, if enabled). I wonder if support for this codec requires an additional explicit option while compiling ffmpeg.
__

The matching ticket in their issue tracker (https://trac.ffmpeg.org/ticket/1087) is still open, so I believe the reverse engineering is not yet complete...
__

P.P.S.: "ffmpeg -codecs" lists support for CineForm HD. This list was not part of the "full help". That means that LAV Filters may support CFHD if it uses a current libavcodec base.

Decoding has been "cracked" and there is support in lastest ffmpeg (since 3 months or so): 10bit YUV as well as 12bit RGBA.

[sdancer75]

The following error happens VFW_E_CANNOT_CONNECT (0x080040217) on a machine with XP Service Pack 3 when I try to manually connect LavSplitter output with LavVideo input. I tested this directly using graphstudionext_0_7_0_430 and with no any media players software.

I DECLARE from the beggining that I have done a brand new DirectX 9 installation and I restarted the machine. So advices like missing directx files are not correct. Do I NEED to install an extra codecs like for mp4, mkv ie to make it work ?

Can LavFilters create log files to debug such kind of errors ? I tried with older LAV filters version ie .60 as well as with the newest one 0.68.1.

http://forum.doom9.org/attachment.php?attachmentid=15408&stc=1&d=1464023368

15408

[filler56789]

The following error happens VFW_E_CANNOT_CONNECT (0x080040217) on a machine with XP Service Pack 3 when I try to manually connect LavSplitter output with LavVideo input. I tested this directly using graphstudionext_0_7_0_430 and with no any media players software.

Possibly the actual problem is in GraphStudioNext itself, and not in LAV Filters... I have just seen that GsNEXT has 2 problems, one with the Time Measure filter + CLCVD.AX from PDVD-14, and another with LAV Splitter + MainConcept's VC-1 Decoder. This latter issue does not happen when I use the original Graphstudio.

FWIW — OS = Windows 7 x64.

[sdancer75]

Possibly the actual problem is in GraphStudioNext itself, and not in LAV Filters... I have just seen that GsNEXT has 2 problems, one with the Time Measure filter + CLCVD.AX from PDVD-14, and another with LAV Splitter + MainConcept's VC-1 Decoder. This latter issue does not happen when I use the original Graphstudio.

FWIW — OS = Windows 7 x64.

Hi,

Thanks for your answer. This is not true. I am using this DShow filter chain in my C++ app and it has the same problem. I just use graphedit to simplify and isolate the problem. I wanted to be sure that the problem is not caused from my app.

Regards,

PS: I am pretty sure that if I install a banch of codecs like CCCP the problem will be solved but I dont want to do this since I can not force the user the install codecs that maybe will not use. I just need to find the root of the problem.

[mandarynek82]

not all files can be hardware decoded.

what file are you trying to play?

Any file, vob, mkv, mp4, mpeg etc.

Only in DXVA copy-back mode I'am able to select hardware device and acceleration is active.

[sneaker_ger]

DXVA native requires a direct connection to a DXVA compatible renderer like EVR or madVR. "Direct" means no filters like vsfilter/DirectVobSub between LAV Video and renderer. CUVID on Windows 10 is known to be problematic. (https://github.com/Nevcairiel/LAVFilters/issues/69)

Just use DXVA copy-back or software decoding unless you have a good reason not to.

[mandarynek82]

DXVA native requires a direct connection to a DXVA compatible renderer like EVR or madVR. "Direct" means no filters like vsfilter/DirectVobSub between LAV Video and renderer. CUVID on Windows 10 is known to be problematic. (https://github.com/Nevcairiel/LAVFilters/issues/69)

Just use DXVA copy-back or software decoding unless you have a good reason not to.

I don't have vsfilter/DirectVobSub between LAV Video and renderer. I use MPC HC and madvr, nothing more.

[clsid]

Only in DXVA copy-back mode I'am able to select hardware device and acceleration is active.That is normal. DXVA2 native always uses the primary device. Do you have multiple devices?

DXVA2 copy-back is the better choice anyway if you use madVR. The copy-back implementation of LAV is faster than the internal one of madVR.

[sdancer75]

The following error happens VFW_E_CANNOT_CONNECT (0x080040217) on a machine with XP Service Pack 3 when I try to manually connect LavSplitter output with LavVideo input. I tested this directly using graphstudionext_0_7_0_430 and with no any media players software.

I DECLARE from the beggining that I have done a brand new DirectX 9 installation and I restarted the machine. So advices like missing directx files are not correct. Do I NEED to install an extra codecs like for mp4, mkv ie to make it work ?

Can LavFilters create log files to debug such kind of errors ? I tried with older LAV filters version ie .60 as well as with the newest one 0.68.1.

http://forum.doom9.org/attachment.php?attachmentid=15408&stc=1&d=1464023368

15408

Any recommendation ? I can not use LAVfilters in XP machines.

Regards,

[Qaq]

Any recommendation ? I can not use LAVfilters in XP machines.
Try to disable HWA for video decoding in LAV Video (set it to 'none'). LAV doesn't support DXVA1 and XP doesn't support DXVA2.

[sdancer75]

Try to disable HWA for video decoding in LAV Video (set it to 'none'). LAV doesn't support DXVA1 and XP doesn't support DXVA2.

Thanks, I will try it tommorow in the office, but I think default is none right ?

PS 1:I will try to play with the LavVideo Options
PS 2:I see a tab in Lav Filters with the name DbgLog. I can not make it to work. Can someone help on this ?

Regards,

[mandarynek82]

That is normal. DXVA2 native always uses the primary device. Do you have multiple devices?

DXVA2 copy-back is the better choice anyway if you use madVR. The copy-back implementation of LAV is faster than the internal one of madVR.

Multiple devices? That's mean GPU in CPU and PCI-E GPU? Yes. But I can't select hardware device in LAV. Only in DXV2 mode (copy-back) I am able to select GeForce.

[sdancer75]

The following error happens VFW_E_CANNOT_CONNECT (0x080040217) on a machine with XP Service Pack 3 when I try to manually connect LavSplitter output with LavVideo input. I tested this directly using graphstudionext_0_7_0_430 and with no any media players software.

I DECLARE from the beggining that I have done a brand new DirectX 9 installation and I restarted the machine. So advices like missing directx files are not correct. Do I NEED to install an extra codecs like for mp4, mkv ie to make it work ?

Can LavFilters create log files to debug such kind of errors ? I tried with older LAV filters version ie .60 as well as with the newest one 0.68.1.

http://forum.doom9.org/attachment.php?attachmentid=15408&stc=1&d=1464023368

15408

Hi,

Problem solved. I had to check, all the video formats in the tab "Formats" of the Lav Video Decoder Properties. So, now I have to do this programmatically under my C++ application and not manually using the properties tab.

Another issue it was the logging of ALL Lav filters that someone can find inside the tab "DbgLog". As I read in the documentation, this is enabled only in debug compilation of the filters, so it will not work in the release compilation that everyone is using in their system.

Regards,

[filler56789]

Problem solved. I had to check all the video formats in the tab "Formats" of the Lav Video Decoder Properties.

http://forum.videohelp.com/attachments/2671-1279232225/uglylol.gif

Another issue it was the logging of ALL Lav filters that someone can find inside the tab "DbgLog". As I read in the documentation, this is enabled only in debug compilation of the filters

Obviously :)

[sdancer75]

http://forum.videohelp.com/attachments/2671-1279232225/uglylol.gif



Obviously :)

Thanks to Qaq which gave me the idea to search for adjustments inside the LavFilter. I have been searched for general system issues like missing libraries .....:)

[mzso]

How do the software deinterlacing algorithms compare to the CUVID one quality wise? (I'm guessing Weston complex is the best?)
I'm forced to use SW deinterlacing because the HW deinterlacer wouldn't work with software decoding. (And these days my gpu decoding is slower than my cpu, and it's too much of a nuisance changing to and back for interlaced video)

[huhn]

your video renderer should be able to run hardware deinterlancing.

[DragonQ]

How do the software deinterlacing algorithms compare to the CUVID one quality wise? (I'm guessing Weston complex is the best?)
I'm forced to use SW deinterlacing because the HW deinterlacer wouldn't work with software decoding. (And these days my gpu decoding is slower than my cpu, and it's too much of a nuisance changing to and back for interlaced video)
Based on my tests, hardware > YADIF > Weston.

[SamuelMaki]

Based on my tests, hardware > YADIF > Weston.

Which manufacturer (AMD, Intel or nVidia)? I myself find YADIF x2 to be the best, but Weston Complex gets so close I just use it due to being more lightweight.

[DragonQ]

Which manufacturer (AMD, Intel or nVidia)? I myself find YADIF x2 to be the best, but Weston Complex gets so close I just use it due to being more lightweight.

Depends which machine I'm using but my HTPC uses nVidia and that is slightly better than YADIF for my test material (sports). The difference between YADIF and Weston is larger IMO.

[Kotik]

Hi Nev,

Any chance on adding support for demuxing MVC muxed into a .m2ts

Since all my collection is muxed with tsmuxer into .m2ts and that includes 3D movies aswell.
Previously Lav splitter was demuxing only the H.264 stream and i was able to watch the movie in 2D but i would also like to test the newly added 3D capabilities of MADVR and remuxing my whole movie collection into .mkv will take ages.

[mandarynek82]

Multiple devices? That's mean GPU in CPU and PCI-E GPU? Yes. But I can't select hardware device in LAV. Only in DXV2 mode (copy-back) I am able to select GeForce.

Can somebody tell me why I'm able to use my GeForce for hardware acceleration only in DXVA2 copy-back mode?

[Kotik]

Can somebody tell me why I'm able to use my GeForce for hardware acceleration only in DXVA2 copy-back mode?

Because as far as i remember DXVA Native can work only with ur primary and active vga.

So i guess ur monitor/tv is currently connected to your CPU iGPU.

[VictorLS]

nevcairiel
I have found software only Mainconcept HEVC Video Decoder 5.0.14.053 http://www.mediafire.com/download/kts8845luxkk1xq/mc0416.7z that pretty well for my purposes in DVBViewer - I see online slideshow of all UltraHD channels with good soundtrack when I set SkipPB and sometimes Synchronization Off instead off PTS in Mainconcept decoder's settings.
Why Mainconcept HEVC Video Decoder could not connect to LAV Splitter? Can you correct it because your LAV Splitter much more universal?

[3421]

I don't use hevc but I decided to watch. VictorLS if you are using the hevc decoding on CPU use http://xhevc.com/en/hevc/decoder/download.jsp I got minus 40% from 80-96% on the CPU (amd 3391x2 hz avx1). With Strongene Lentoid decoder I have ~40% on CPU on Win 32 bit, decoder without opencl with this file. Exemple file: encoder N-80129-ga1953d4 https://www.cubbyusercontent.com/pl/exemple.7z/_1beeecf5357647c5b44a3030b14ae001

UltraHD - hmm, look at, they have decoder: cpu + gpu on opencl.

[filler56789]

.......
Why Mainconcept HEVC Video Decoder could not connect to LAV Splitter?

Can you correct it because your LAV Splitter much more universal?

1) Because it was designed so by the Mainconcept developers? :)

2) If you pester the Mainconcept people, then maybe you'll get what you want.
The Lentoid HEVC decoder became "connectable" to LAV Splitter only after I sent an e-mail to Strongene themselves.

[3421]

I don't have hevc files, I have hevc files, just to see how they look. Me don't need 2-4K. 4K on the CPU is not really use not with vp9, not with h264/avc, not with h265. Maybe can only with cpu frequencies above 3000 Hz Phenom 2 x6 or Bulldozer x8, or multi-core Xeon processors with avx512 4k will work acceptable or not bad. I never thought about it seriously use 4k. About Lentoid HEVC I know about ~8 months.

Video hevc 1080p 25 fps 1000 kbit/sec with CPU load 3391x2 hz avx1 80-96% is too much. With Lentoid HEVC decoder minus 40-56%, decode on cpu.

I have so: Lentoid hevc decoder and lav hevc decoder use the CPU is 30-50% more intensive compared with lav h264/avc decoder (s/w). Checked files with the same resolution and with the same frame rate, and with the same bitrate.

[VictorLS]

filler56789
Thank you for your e-mail to Strongene.
3421
1. Lentoid (last three version) never decrease CPU load on my computer neither in XP nor in 7x64. I don't know why. You can try my files http://forum.doom9.org/showthread.php?p=1726845#post1726845
2. Only last v2.1.0.2_2015_12_02_r4644 of Lentoid can decode 10bit;
3. Even last Lentoid decodes bad in sat programs like DVBViewer both 8bit and 10bit (like Elecard HEVC Video decoder decodes 10bit in sat programs):

[3421]

VictorLS I сommented personal message.

Lentoid hevc decoder can't create miracles. Video 3840 x 2160 (4k) 50 fps 30 MB/s this much for normal decode on the CPU. Such video files need checked for 6 or 8 (not logical cores, real cores) core processor with a frequency above 3000 Hz. On their website they have information the test, with a maximum file size of 2560 x 1600 30 fps 32937 Kbit and tested on Win. 7 Cpu i7-2600 8GB RAM from a year 2013.11

With Lentoid HEVC decoder minus 40-56%, decode on cpu, for this file and this cpu: Video hevc 1080p 25 fps 1000 kbit/sec with CPU load AMD 3391x2 hz avx1 80-96% is too much. = ~40% with Lentoid HEVC decoder load cpu for this file hevc 1080p 25 fps 1000 kbit/sec.

[VictorLS]

VictorLS I сommented personal message.
I has answered.

Lentoid hevc decoder can't create miracles. Video 3840 x 2160 (4k) 50 fps 30 MB/s this much for normal decode on the CPU. Such video files need checked for 6 or 8 (not logical cores, real cores) core processor with a frequency above 3000 Hz.
I agree, middle Core i7 plays that files well.

[kirakami]

I would like these features implemented in LAV Filters: Normalize, Max Amplification 400% and Regain Volume.
Reason is sound is too low for my normal hearing on my sony headphone even with all volume settings on max and loudness equalization enabled in sound manager, equalizer did nothing too.
On ffdshow audio decoder with these settings enabled, my problem is solved but ffdshow is outdated.

http://i.imgur.com/A9RK3Mq.png

[nevcairiel]

I would like these features implemented in LAV Filters: Normalize, Max Amplification 400% and Regain Volume.
Reason is sound is too low for my normal hearing on my sony headphone even with all volume settings on max and loudness equalization enabled in sound manager, equalizer did nothing too.
On ffdshow audio decoder with these settings enabled, my problem is solved but ffdshow is outdated

Sorry, but there are no plans to offer volume processing like that.

[amayra]

there something is annoying me for awhile what is an advantage for using LAVFilters tMod (http://tmod.nmm-hd.org/LAVFilters/ ) over vanilla builds ?

[3421]

kirakami Use the "ffdshow raw audio processor" priority prefer, with the lav audio decoder or with another audio decoder.

[Sparktank]

there something is annoying me for awhile what is an advantage for using LAVFilters tMod (http://tmod.nmm-hd.org/LAVFilters/ ) over vanilla builds ?

Taro mentions the differences for his builds in an old blog post.
https://astrataro.wordpress.com/2014/10/22/lavfilters-tmod/
But that was back in 2014, now sure how the two builds compare today.

[ashlar42]

Some questions regarding deinterlacing.

1) Is there a quality difference between CUVID deinterlacing and what madVR uses for deinterlacing (which I guess it's DXVA2, judging from OSD info)? In favour of which approach?

2) Is Weston Complex supposed to be lighter on CPU usage than YADIF?

3) Is it supposed to lead to better results or simply different results and one has to judge subjectively?

Thanks nevcariel for all the good work you put in LAV Filters. If you are still in touch with tiben20, tell him Kodi users miss him enormously now that Aracnoz has thrown the towel on Kodi DSPlayer (we were missing him before too but now... it's a desperate situation).

[DragonQ]

Some questions regarding deinterlacing.

1) Is there a quality difference between CUVID deinterlacing and what madVR uses for deinterlacing (which I guess it's DXVA2, judging from OSD info)? In favour of which approach?
Both use the GPU's deinterlacing algorithms, so should be identical.

2) Is Weston Complex supposed to be lighter on CPU usage than YADIF?
Yes.

3) Is it supposed to lead to better results or simply different results and one has to judge subjectively?
Different for sure. Worse, IMO.

[ashlar42]

Both use the GPU's deinterlacing algorithms, so should be identical.Isn't CUVID using CUDA, thus being programmable and potentially different from standard built in DXVA deinterlacing? Otherwise I wouldn't understand why is deinterlacing through LAV only selectable when using CUVID and not when using DXVA2 CP with Nvidia cards.

Looking at your previous posts, do I understand correctly that you place hardware deinterlacing as the best solution?

[huhn]

DXVA deintelancing is possible too with DXVA decoding but lavfilter doesn't support that.

nvidia would be pretty dumb to write two different deinterlancer.

and who said that DXVA decoding can't use CUDA too. the hybrid decoder are using CUDA nearly for sure.

[ashlar42]

DXVA deintelancing is possible too with DXVA decoding but lavfilter doesn't support that.

nvidia would be pretty dumb to write two different deinterlancer.

and who said that DXVA decoding can't use CUDA too. the hybrid decoder are using CUDA nearly for sure.
Nobody said anything. I think I was pretty clear about my ignorance, by asking information here. :)

[VictorLS]

Nobody said anything. I think I was pretty clear about my ignorance, by asking information here. :)
nevcairiel did not answer my question (http://forum.doom9.org/showthread.php?p=1769481#post1769481) about Mainconcept too
I am sure he saw it
May be it is impossible to realize in LAV Splitter (but I am not believe in this version).

[Kotik]

Nev,

Thank you for implementing MVC support in .m2ts container i am happy to report that in your last nightly my 3D .m2ts files are indeed reported as 3D in madvr, BUT (there is always a "BUT" :D ) playback is jerky and picture is often getting corrupt with pixels and recovers after a while.

FYI same movie inside .ssif container is playing fine, the moment i remux it into .m2ts the issue appears.

[nevcairiel]

LAV only supports official Blu-ray m2ts files, if you remux them they probably do not match this strict specification anymore. So, don't. LAV does not support anything else.

[Kotik]

LAV only supports official Blu-ray m2ts files, if you remux them they probably do not match this strict specification anymore. So, don't. LAV does not support anything else.

Nev,

This is the FILE (http://www.filedropper.com/1_145) and somehow it is working fine in stereoscopic player which as far as i know is using a modified version of your splitter (not sure tho).

If you cannot implement it then i guess i will have to remux all my 3D movies once again with MakeMKV this time.....

[PPPPark]

nevcairiel,write a Subtitle Filter,LAV has stalled.

[Aleksoid1978]

Nev,

This is the FILE (http://www.filedropper.com/1_145) and somehow it is working fine in stereoscopic player which as far as i know is using a modified version of your splitter (not sure tho).

If you cannot implement it then i guess i will have to remux all my 3D movies once again with MakeMKV this time.....

Confirm. This file play fine with MPC-BE's Mpeg Source(with LAV Video + madVR = 3D playback), but not with LAV Source.

[Kotik]

Well it seems i got issues with 3D,

In order to test it i force MPC-HC to use EVR-CP and enable MVC decoding in Lav Video config.

Indeed Lav Video Decoder reports msdk mvc and since i am using EVR-CP i get a 2D video which is fine cause i am just testing for issues and issues i get.... lots of them actually, picture is getting corrupt and it is never smooth, mind u i am using a .sffi file so no remuxing.

Same video with Madvr and same issues. The difference here is that i have to pay attention to the video since i cannot activate graphs and observe them. But i do notice that max rendering times jump to 100+.

All this with latest stable Lav or the latest nightly, both give me same results.

The moment i disable MVC decoding in Lav Video everything is playing fine and smooth. Go figure.

I will go ahead and download Life Of Pi 3D full bluray and test it aswell.

[LigH]

download Life Of Pi 3D full bluray

Nope.

:readrule:

[Kotik]

Nope.

:readrule:

I already own a legal version of the movie which i remuxed, and since i want to test a non remuxed version i have to download it.

Original disk not at my office currently. So don't you worry.

[Kotik]

After futher investigation it seems that tsmuxer is corrupting mvc streams or lav splitter/mvc decoder dont like what tsmuxer is doing to the mvc stream.

I need to perform more tests but by the looks of, it if tsmuxer comes in contact with mvc then lav wont play it smooth. If on the other hand i remux a disc with makemkv or play directly the .ssif file then everything is fine.

Somehow stereoscopic player is able to play my 3d remuxes that were done with tsmuxer.

In any case i will perform more tests, but what i would like to know if Nev is willing to impliment .mvc support in .m2ts files in the future, if not then i will remux all my 3d discs with makemkv.

EDIT: I just now remuxed a movie a movie from .m2ts with mvc included (the one that lav doesnt like) to .iso and after that i remuxed the .iso to .mkv with makemkv and after that lav is working fine. My point is that lav doesnt like it if both h.264 and mvc are muxed into one .m2ts.

So question remains: are u planning in supporting h.264 and mvc muxed into a single .m2ts?

EDIT2: Mpeg Splitter from MPC-BE is able to properly handle H.264 and MVC inside a single .m2ts, so if i force it to be used instead of LAV Splitter then all my remuxes are playing smoothly.

[nevcairiel]

There are no plans right now to support anything but official Blu-ray streams or MakeMKV MVC-MKV files. Remuxing to m2ts makes no sense as you lose a bunch of extra metadata from the Blu-ray structure, not to mention that its really not a good format to store video in, so I have zero interest in such a format.

[Kotik]

There are no plans right now to support anything but official Blu-ray streams or MakeMKV MVC-MKV files. Remuxing to m2ts makes no sense as you lose a bunch of extra metadata from the Blu-ray structure, not to mention that its really not a good format to store video in, so I have zero interest in such a format.

Thanx for the reply.

[baii]

nevcairiel,write a Subtitle Filter,LAV has stalled.
What feature is lacking in existing subtitle filters?

Sent from my XT1575 using Tapatalk

[kalehrl]

Just a n00bie question.
How do I force MPC-HC to use separately installed LAV filters instead of the ones that came with it?

[huhn]

go to option -> external filters add all four lavfilter filter (lav* audio decoder, splitter, splitter source, video decoder) set them to prefer.

alternative or additionally go to options-> internal filters and uncheck everything.

for reason unknown i have to deselect all source filter or mpc-hc is always using the internal source splitter.

[LigH]

Disabling all internal decoders and source filters should be the first step to use system-wide installed DirectShow filters instead. Forcing external filters to be preferred is not necessary.

[huhn]

Disabling all internal decoders and source filters should be the first step to use system-wide installed DirectShow filters instead. Forcing external filters to be preferred is not necessary.

the number of installed directshow filters on a PC is unknown and there merit too. or with other word other directshow filter could be used for various reasons.

so preferring lavfilter is a reliable way to make sure it is used and not something else.

[LigH]

Reminds me on the duality of necessary and sufficient preconditions in maths. Disabling internal filters is certainly a necessary precondition to have MPC-HC use any external DirectShow filter. Preferring LAV Filters may be sufficient to ensure their use among all external filters, but only if the use of internal filters was disabled first.

[robl45]

I posted on the MadVR thread, thought maybe I should post here as well. I'm trying to get 3D MVC MKV working. I got a small test file from Kodi website and at first I was getting side by side coming up and it was using the mvc decoder in Lav. If I put my projector in side by side mode, it was giving me 3d. I updated my video drivers and now have the 3d option in windows and now when I try the file, windows switches into the 3d resolution. However the image on the screen is now one image but it appears to be just 2d. My projector is locking on and going into 3d mode, but the picture is just flat. I'm not sure what could be wrong at this point. Its also not exiting out of 3d resolution when the file is done.

[nevcairiel]

LAV just performs decoding, it has nothing to do with rendering the video in 3D on your screen. Decoding clearly works as you got 3D in SBS mode. Posting the same question in multiple areas of the forum is usually considered against the forum rules, just FYI.

[robl45]

LAV just performs decoding, it has nothing to do with rendering the video in 3D on your screen. Decoding clearly works as you got 3D in SBS mode. Posting the same question in multiple areas of the forum is usually considered against the forum rules, just FYI.

I wasn't sure which one could be causing the issue. So the lav is using the mvc decoder so its working correctly and I can rule that out correct?

[nevcairiel]

I wasn't sure which one could be causing the issue. So the lav is using the mvc decoder so its working correctly and I can rule that out correct?

If you got working SBS, then its working fine indeed.
You can always set madVR back to SBS mode to see that.

[dbukhan]

Hello,
Are HTTP Cookies supported by the LAV filters? It appears they are not supported in my first test.
Akamai sends Set-Cookie with a token in response to a request for a m3u8. If the next request does not have a cookie it is rejected by Akamai as a security measure.

Thanks!

[NikosD]

Microsoft added once again VP9 10bit (profile 2) support in DXVA specification.

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

Does Pascal or any other GPU support VP9 10bit HW acceleration ?

[huhn]

Reminds me on the duality of necessary and sufficient preconditions in maths. Disabling internal filters is certainly a necessary precondition to have MPC-HC use any external DirectShow filter. Preferring LAV Filters may be sufficient to ensure their use among all external filters, but only if the use of internal filters was disabled first.

the question was how to use an external installation of lavfilter not how to stop internal filter.

while preferring transform filters is enough for MPC-HC to use them it is not enough for source filters.

while MPC-BE is using the preferred filters even the source filter "correctly".

i guess it is time to ask in the MPC-HC thread.

[baii]

Mpc-hc using prefer option pick up external filter perfectly without the need to disable internal, at least on my end. So does most if not all direct show video player, like pot player etc.

Sent from my XT1575 using Tapatalk

[clsid]

TrueHD (with Atmos) is broken in current nightly.

[Mercury_22]

@NEV Are there any "telemetry codes" (https://www.reddit.com/r/cpp/comments/4ibauu/visual_studio_adding_telemetry_function_calls_to/) in the nightly / release ?

P.S. More info at "Visual Studio 2015 Secretly Inserts Telemetry Code into C++ Binaries" (http://news.softpedia.com/news/visual-studio-2015-secretly-inserts-telemetry-code-into-c-plus-plus-binaries-505113.shtml)

[nevcairiel]

@NEV Are there any "telemetry codes" (https://www.reddit.com/r/cpp/comments/4ibauu/visual_studio_adding_telemetry_function_calls_to/) in the nightly / release ?

P.S. More info at "Visual Studio 2015 Secretly Inserts Telemetry Code into C++ Binaries" (http://news.softpedia.com/news/visual-studio-2015-secretly-inserts-telemetry-code-into-c-plus-plus-binaries-505113.shtml)

Please stop with the FUD, the use of the word "telemetry" is entirely misleading. Those are debugging hooks, nothing else, and on top of all things it only applies to executables, not dynamic libraries.

[DigitalLF]

Hello!

I really want to be able to use bitstream for DTS and DD

But if there is a DTS-MA or TrueHD track those should be preferred and decoded!


Is this possible?

[ddjmagic]

I really want to be able to use bitstream for DTS and DD
But if there is a DTS-MA or TrueHD track those should be preferred and decoded!
Is this possible?

Yes, open LAV Audio settings and under Bitstreaming Formats uncheck DTS-HD and TrueHD and check DD and DTS.
Then open LAV Splitter settings and make sure "Prefer Audio Stream with the Highest Quality" is checked.

[nevcairiel]

Does Pascal or any other GPU support VP9 10bit HW acceleration ?

Pascal doesn't support it with the current drivers.
If the hardware can do it ... only NVIDIA knows.

[Prinz]

Since LAVFilters-0.68.1-9 I'm getting the wrong fps reported to MadVR (= displaymode switching doesn't work correctly)

In Versions LAVFilters-0.68.1-4 and before I get the correct fps:
http://i64.tinypic.com/2i0ao42.jpg

LAVFilters-0.68.1-9 and higher I get:
http://i68.tinypic.com/xf6vsp.jpg

Is this a known issue or a new setting that I'm overlooking?

[nevcairiel]

This is a bug, however it should be noted that FPS information is generally not very reliable and absolutely relying on it will fail eventually.

[videonerd]

Hi nevcairiel, it is generally accepted that ASF container seeking doesn't work too well with lavfilter's splitter. I have recently noticed that another ffmpeg-based player, mpv seems to handle ASF container seeking without glitch. Is the issue therefore at lavfilters level i/o ffmpeg? If so, any plans to have a go at fixing this? Cheers.

[LigH]

I'd like to point at the options of MPC-HC: There used to be an option (not sure if it is currently still present, don't have it right here) to choose between two different ASF/WMV splitters. One seeks faster, but can't seek at all when the file is incomplete (e.g. doesn't contain a keyframe index). So I believe this container format is a bit "brittle", either trust it completely or not at all.

[hubblec4]

Hi nevcairiel

Feature request:
When I open a BDMV folder via right click -> play in MPC-HC, then play the longest movie automatically (normally is this the main movie).
But I have some Blurays where the longest movies are the "Extras" or "Specials". The "Extras or Specials" are not in full HD.
Is it possible to check first the resolution and then the longest playlist?

[clsid]

Some Matroska files trigger a crash due to this change:
http://git.1f0.de/gitweb?p=ffmpeg.git;a=commitdiff;h=ead46dfd7eac4b154b94f3642091f9fcc8deb89b
To be more specific the TARGET_TRACK change. Crash happens in avutil.

Don't have a small sample file available right now, but maybe you don't need it.

[nevcairiel]

I always need a sample. :)

[clsid]

Check your PM.

[nevcairiel]

Check your PM.

Should be fixed.

[rock]

hi, is there any plan to implement more codecs that supported by Media Foundation? Also I would like to know when you will implement HW fallback from DXVA2 Native to Copy-Back as I remember I read the discussion somewhere in this thread.

[nevcairiel]

hi, is there any plan to implement more codecs that supported by Media Foundation? Also I would like to know when you will implement HW fallback from DXVA2 Native to Copy-Back as I remember I read the discussion somewhere in this thread.

I don't intend to do either of those things. Sorry. There are no MF codecs of value and its easy enough to configure DXVA2 to the needs of your setup without some complicated fallback.

[3421]

Add enable DTS-HD.

[clsid]

If you are talking about bitstreaming. DTS-HD gets enabled if you first enable DTS.

[3421]

I only once enable DTS, DTS-HD was gray. Yes now I can enable DTS-HD when enable DTS.

[3421]

I only once enable DTS, DTS-HD was gray. Yes now I can enable DTS-HD when enable DTS. I maybe I didn't notice. I don't remember. This difficult not to notice.

[P.J]

LAV/ffmpeg can't play ProRes well :(

Quicktime (smooth):
http://7.t.imgbox.com/UmhiKkFO.jpg (http://imgbox.com/UmhiKkFO)

LAV (not smooth):
http://2.t.imgbox.com/3BhyjP2Z.jpg (http://imgbox.com/3BhyjP2Z)

[rock]

Video lag (DXVA2 Native, Copy-Back, Software Decoding) with EVR and EVR-CP, play smooth unless I try one of these three way below
- Frame Time Correction enabled in MPC-HC
- Intel Quick-Sync enabled
- madVR as Video Renderer

Tried other DS Filters, latest ffdshow-tryouts and MPC Video Decoder (from MPC-BE), both play smooth.

Video
https://www.youtube.com/watch?v=QbhmTM1cRro (download as mkv 360p using Internet Download Manager)

MediaInfo
http://pastebin.com/UfWTa7um

OSD Screenshot
http://i.imgur.com/nyfY2Xn.jpg

[clsid]

It is the fault of IDM. The original 360p WebM file plays fine and has a framerate of 23.976 instead of the 24.000 shown in your MediaInfo log.

In your 'working' cases the bad framerate value is getting 'corrected' or is ignored.

[rock]

I still don't know much about video things. Thanks for the explanation. :)

[amayra]

Video lag (DXVA2 Native, Copy-Back, Software Decoding) with EVR and EVR-CP, play smooth unless I try one of these three way below
- Frame Time Correction enabled in MPC-HC
- Intel Quick-Sync enabled
- madVR as Video Renderer

Tried other DS Filters, latest ffdshow-tryouts and MPC Video Decoder (from MPC-BE), both play smooth.

Video
https://www.youtube.com/watch?v=QbhmTM1cRro (download as mkv 360p using Internet Download Manager)

MediaInfo
http://pastebin.com/UfWTa7um

OSD Screenshot
http://i.imgur.com/nyfY2Xn.jpg
this doesn't look like madVR

[Nerva]

Is there a reason to use "Expand 6.1 to 7.1" vs. "Enable Mixing" with the Output Speaker Configuration being either 5.1 or 7.1? I have a PC with 5.1 and a HTPC with 7.1. I'm just thinking, if I'm already using the mixer and I'm playing a 6.1 file, wouldn't the mixer serve the same function as the "Expand 6.1 to 7.1" option.

[jerryleungwh]

I am trying to use LAV with madVR but whenever I play a MVC 3D movie the audio bitstream stops working. It also happens when I enter DSD11 FSE mode too. How can I fix this?

[JungleBoy]

LAV 0.68.1 is still having trouble decoding 12bit HEVC 4:4:4

Here's some short 10 second test files:
http://bit.ly/1NxTIUy

This file shows green vertical lines.
video-x265-12bit-444.mp4

All the other files including 12it HEVC 4:2:0 play ok.

[MrGlasspoole]

Hi,

i have some Dolby Pro Logic audio CDs.
In the Status tab i only see pcm_s16le and the channel output only shows left and right.
Does LAV not decode Dolby Pro Logic?

[nevcairiel]

i have some Dolby Pro Logic audio CDs.
In the Status tab i only see pcm_s16le and the channel output only shows left and right.
Does LAV not decode Dolby Pro Logic?

No, it does not. Dolby Pro Logic is not an audio format, CDs are just uncompressed PCM. You would need an external Dolby audio processor to make use of DPL processing.

[MrGlasspoole]

To bad. I thought i can skip adding a receiver to my setup.
But it seems like there are also no 9.1 sound cards.
So no more Asus Essence and it looks like there is no way around HDMI and bitstream today if somebody also wants Atmos, DTS:X and Auro-3D.

[CarlosCaco]

I m on a optimus system intel hd 5500 / nvidia 930m, using latest mpc-be, madvr and lav nigthly on windows 8.1 i also use reclock so is the x86 versions
Both laptop display and hdmi outputs by intel, so nvidia only process the image

I tried to use hardware accelaration on lav, all options are avaliable to me... if i set dxva cb lav shows nvidia 930m as avaliable but no matter what option i choose: cuvid, dxva native oe dxva-cb when i play any file lav shows the option selected but on decoder shows the normal codec and on hardware accelaration shows none... only quicksync works...
Is this because optimus? Is this normal?
In madvr thread people said that cuvid and dxva native is not the ideal and can degrate image quality... same for quicksync? I ready that the latest drivers for broadwell allows hevc hybrid decoding but in lav quicksync hevc is greyd out... im doing something wrong? I want to use dxva cb with nvidia, it is possible?

[Plutotype]

Some users in the madvr forum reported, that with AMD RX480 DXVA CB does not work with HEVC and 10bit files ( using latest LAV filters ). Is this a driver problem with AMD?

[nevcairiel]

Is this because optimus? Is this normal?
In madvr thread people said that cuvid and dxva native is not the ideal and can degrate image quality

With Optimus you are on your own, those systems are terrible. Neither the NVIDIA nor the Intel GPU can be used properly on those things.
Note that all HWAccels result in the exact same decoding quality, with one tiny exception when using madVR - DXVA2 Native, which is due to the way madVR accesses the image which can result in chroma blurring. But all others are absolutely identical.

Some users in the madvr forum reported, that with AMD RX480 DXVA CB does not work with HEVC and 10bit files ( using latest LAV filters ). Is this a driver problem with AMD?

Only AMD can answer you that.

[MrGlasspoole]

Ok I'm lost.
If i understand it right then there is no way to decode Dolby Pro Logic on the PC. The same thing than also applies to "Dolby Digital EX" and "DTS-ES Matrix" where channel six is encoded into other channels?

What i can't find out if the receiver should auto detect Dolby Pro Logic and switch to it (and if does it work with RCA or just SPDIF)? Also how can i tell if a file has Dolby Pro Logic?

[vood007]

Ok I'm lost.
If i understand it right then there is no way to decode Dolby Pro Logic on the PC.

FFDShow Audio can do this.

[LigH]

It is certainly possible to decode matrix surround formats on a PC. The decoder logic is usually known, at least in general, just as the encoder logic is (as used in DPL downmixes when converting multichannel audio to a format like MP3 which supports 2 channels at most); exceptions may exist, and license restrictions may exist as well. But there may be no way to detect it certainly in every digital audio stream. Some audio formats may support bitflags in their headers to indicate that the content may be matrix surround encoded, but on one hand this flag may not have been set by the encoding operator, and on the other hand there are also formats without such flags. So eventually it may be in the responsibility of the listener to enable surround decoding and check whether it sounds pleasant or not.

[MrGlasspoole]

Ok Wikipedia says DTS-ES Matrix has a flag and it seems right because MediInfo says about my file: Matrix and 7 channels.
But LAV Audio in MPC only shows DTS 6 channels in the status/output.

The DTS-ES Discrete file shows a BC...
But if you have 7.1 speakers shouldn't the BC not be copied to BL/BR?

I wonder how others deal with that kind of stuff if you want to use the 7.1 output on the sound card.
The thing is that the speakers i build are active and it makes no sense to buy a expensive AVR if you don't need or can use the amplifiers.
Then a lot of them also have no 7.1 input or mine is 7.1 but has just 5.1 in. So if i use the 5.1 in then 2 amplifiers are not used.

Then switching always by hand the modes on the AVR is also unpractical. Lets say i have a playlist mixed with Dolby Pro Logic music, stereo, mono...

If the PC could to Dolby Pro Logic/DTS-ES Matrix then i guess it would be possible to make a tag in JRiver MC and based on that tag let the PC switch?

[LigH]

Dolby ProLogic and Dolby ProLogic II are made not to hurt material too much which does not contain ProLogic encoded surround information. There is a chance that it may be misinterpreted if it is just stereo, but the risk should be rather low. And if it is pure mono (left and right channels are identical), the result will be exactly in the center as desired, because the sum of both is all, and the difference of both is zero.

[e-t172]

The DTS-ES Discrete file shows a BC...
But if you have 7.1 speakers shouldn't the BC not be copied to BL/BR?

In theory, what's supposed to happen in the case of DTS-ES Discrete is that LAV Audio will output 6.1 channels to the audio renderer, which (in the case of the default, standard renderer) will output 6.1 channels to the Windows audio engine, which (if the active speakers are configured as 7.1) will take care of upmixing BC to the appropriate rear speakers. I have no idea if that actually works in practice though, since I never play 6.1 files.

[MrGlasspoole]

Dolby ProLogic and Dolby ProLogic II are made not to hurt material too much which does not contain ProLogic encoded surround information. There is a chance that it may be misinterpreted if it is just stereo.I thought stereo gets upmixed if you set the decoder to Dolby ProLogic?

From Wikipedia:
DPL II processes any high quality stereo signal source into five separate full frequency channels (right front, center, left front, right rear and left rear).

[LigH]

Of course, using a ProLogic decoder always upmixes from 2 to more channels; whether or not their content makes sense, is a different issue. Ideally, the surround channels should contain audible content only in case the source was created from an original multi-channel audio with a ProLogic encoder. But due to the nature of the ProLogic systems, even stereo which has never been multi-channel audio before might get "decoded" to contain surround output, which would be a misinterpretation of general stereo with some unfortunate phase differences, and such scenes may sound weird. Or not. Depends on the specific case. And with "dual mono" sources, the output will still be multi-channel, but everything except the center channel should be rather silent.

[chros]

In theory, what's supposed to happen in the case of DTS-ES Discrete is that LAV Audio will output 6.1 channels to the audio renderer, which (in the case of the default, standard renderer) will output 6.1 channels to the Windows audio engine, which (if the active speakers are configured as 7.1) will take care of upmixing BC to the appropriate rear speakers. I have no idea if that actually works in practice though, since I never play 6.1 files.
Well, I'm not sure either that windows will upconvert it. It definitely doesn't for pure stereo sources, only if the soundcard supports it by hardware and driver (which most integrated ones don't).

Then switching always by hand the modes on the AVR is also unpractical. Lets say i have a playlist mixed with Dolby Pro Logic music, stereo, mono...
Not really, if you have a network capable AVR with full web setup interface, like my Yamaha RX-A830: you can set everything (! , every small bit of it) on the web inteface. After that you only need just a short time to come up with 3 small batch scripts for e.g. music, DPL, and pure movie profiles, like this:
REM set to Movie DPL mode: Vol -45 DB, Drama mode, Enhancer On, Straight Off , Adaptive_DRC Auto , Dialogue_Lvl 0
C:\Windows\wget.exe "http://192.168.1.71/YamahaRemoteControl/ctrl" --quiet -O NUL --header="Content-Type:text/xml" --post-data="<YAMAHA_AV cmd=\"PUT\"><Main_Zone><Volume><Lvl><Val>-450</Val><Exp>1</Exp><Unit>dB</Unit></Lvl></Volume><Sound_Video><Adaptive_DRC>Auto</Adaptive_DRC><Dialogue_Adjust><Dialogue_Lvl>0</Dialogue_Lvl></Dialogue_Adjust></Sound_Video><Surround><Program_Sel><HDMI_1><Enhancer>On</Enhancer><Sound_Program>Standard</Sound_Program><Straight>Off</Straight></HDMI_1></Program_Sel></Surround></Main_Zone></YAMAHA_AV>"

After that you can just double click on them, make them as scheduled task, or put it on your openwrt router (with a slight mod) and run it from cron. :) That's how I can wake up every morning listening to music as an alarm (with winamp+clockamp). :)

(I even control my milight/easybulb light bulbs in a similar way. :) )

[e-t172]

Well, I'm not sure either that windows will upconvert it. It definitely doesn't for pure stereo sources, only if the soundcard supports it by hardware and driver (which most integrated ones don't).

That's not the same thing. If you're playing stereo (LR content) over 5.1, no processing is necessary - Windows just maps input channel L to speaker L and input channel R to speaker R, as it should. The situation is different when playing 6.1 over 5.1/7.1 - there the input contains a BC channel, but there is no BC speaker. In that case Windows has to reroute BC to other speakers (such as BL or BR), otherwise you would lose the channel.

The same situation occurs when you play 5.1 content on a stereo setup - the Windows audio engine will automatically mix C, SL, SR and LFE into the L and R speakers as appropriate.

When playing stereo over a 5.1 setup, Windows will play the content as-is. It will not apply any kind of "clever" upmixing algorithm. Which is perfectly reasonable: stereo is stereo, it was mixed in stereo, mastered in stereo, and is meant to played in stereo. You might disagree, but that would be a matter of preference/taste, not accuracy. Windows is just being conservative and following the principle of least surprise.

[nevcairiel]

Windows likes to choke on 6.1 content, the back channel might vanish without mixing even on a 7.1 output, which is why LAV has the option to expand the back channel to two back surround channel for 7.1.

[Sunset1982]

Some users in the madvr forum reported, that with AMD RX480 DXVA CB does not work with HEVC and 10bit files ( using latest LAV filters ). Is this a driver problem with AMD?

I think it was me.

Figured this out when comparing RX480 with GTX1070.

Is anyone here already using a new AMD rx480? Are you experiencing the same issues? (green screen when selecting copyback decoding in LAV when running HEVC files)

Is HEVC decoding working on older AMD cards?

If this problem can be corrected, I would buy an RX480...

[blaubart]

Lately I stumbeled over a rare 10-bit UHD demo containing some banding in the backgrounds.

To try it with madVR I had to switch LAV from DXVA2 native to copy-back (GTX960). First i thought madVR did the imressive dithering. But later I switched back to MPC-HC's EVR Custom and the dithering was still the same. Then LAV back to DXVA2 native and the banding came back.

So I wondered 'is copy-back so good or native so bad?' and switched LAV to 'None' > good - nearly the same as copy-back.
Since native at the moment is best for UHD-TV with DVBViewer (Nvidia GTX950/960 and AMD RX480) - would it be possible to give native the dithering of copy-back?

Has anybody a working 10-bit setup? How about the banding in a 10-bit TV?

In the video I collected the banding scenes - sorry for some bad cuts - my cutting soft is still not so good with HEVC without rerendering.
https://dc2.safesync.com/FdSVjnr/offen/4K/Snakes_4K-10-bit_banding_cut.zip?a=zAINxh7PTtc

native-banding:
https://dl.dropboxusercontent.com/u/8424544/pics/4K/native-banding.png

copy-back
https://dl.dropboxusercontent.com/u/8424544/pics/4K/copy-back.png

[nevcairiel]

LAV just outputs the image as close to the original as possible. With DXVA2 Native its output entirely untouched, no conversion and no dithering, and any further processing is up to the video renderer (this also cannot be changed, its just an inherent property of native decoding)
With DXVA2-CopyBack, the video renderer doesn't even know that LAV is performing hardware decoding and just gets an ordinary image (just like software decoding), in whichever format it accepts. For madVR this will be pure 10-bit, for EVR this will be dithered 8-bit.

Its certainly possible that 10-bit DXVA2 Native results in a poor conversion in the video renderer, for madVR depending on the settings chosen (specifically the DXVA settings in the trade quality for performance section), and for EVR in general.
Note that madVR on NVIDIA cards will always produce a bit of chroma bluring when used with DXVA2-Native, so its not recommended for the highest quality results.

[blaubart]

Thanks for the info.
Well, scenes as seen in the video are really rare in 'normal' UHD-life at the moment. So I'm happy with 'native'.

Note that madVR on NVIDIA cards will always produce a bit of chroma bluring when used with DXVA2-Native, so its not recommended for the highest quality results.

On my Nvidia and reported on a AMD RX480 madVR delivers a black screen with 32-bit DVBViewer at 3840*2160 fullscreen + LAV native. Also madVR's D3D fullsreen exclusive modes sometimes cause heavy crashes with copy-back. So I take EVR Custom.

[blaubart]

p.s. just tested
..black screen not only using 32-bit DVBViewer but also MPC-HC x64 - nothing changed since (remember?):
http://www.dvbviewer.tv/forum/topic/55897-quo-vadis-htpc-richtung-hevc/?p=432470

[nevcairiel]

p.s. just tested
..black screen not only using 32-bit DVBViewer but also MPC-HC x64 - nothing changed since (remember?):
http://www.dvbviewer.tv/forum/topic/55897-quo-vadis-htpc-richtung-hevc/?p=432470

That's not a LAV issue, it just delivers the image , it doesn't display it.

In my personal experience madVR with DXVA2 Native just isn't the best combination. Luckily copy back is darn efficient these days , at least on NVIDIA .

[madshi]

Works fine here, are you using the latest madVR build, blaubart? Anyway, if you still get the black screen with the latest madVR build, this is likely a madVR issue, so OT here.

[blaubart]

@ nevcairiel
..no need to defend LAV at all ! As you may have noticed from earlier (mostly German) posts I'm the greatest admirer of your work! Ok I must confess I sometimes see threads not only as topic-knight but as voice to the world hoping to attain sth. unexpected...

@ madshi
sorry, just updated from 0.90.21 to 0.90.22 with no success - LAV native 3840*2160 fullscreen still black (Win10, GTX960, newest NVidia driver, madVR set to default). Black even if everything in 'general settings' is unchecked (D3D9, no FSE).
You're right that's OT here but in my defense I've never written anything in your thread here before and since our war years ago. Your madVR thread is also too fast for me. New posts come within seconds. I hate that..

[MrGlasspoole]

Windows likes to choke on 6.1 content, the back channel might vanish without mixing even on a 7.1 output, which is why LAV has the option to expand the back channel to two back surround channel for 7.1.So for a 7.1 system it's always right to set "Expand 6.1 to 7.1"?

But there is still one thing i wonder about.
If i play a DTS-ES Discrete file then i have output on BL/BR/BC..
Isn't 5.1 not SL/SR:
http://ask.creative.com/wwimages/general/glossary_audio_im13..gif

Sure with the setting above most stuff comes from SL/SR and BL/BR has the same level (mono).

Then i have DD+ 7.1 files where MediaInfo says E-AC-3 but LAV shows only ac3 and SL/SR in the status tab...

Sorry for all the questions. But if you search the web for this stuff most answers are what everybody thinks is right.
And on the DTS site i can't find a setup (speaker placement) guide.

[blaubart]

Ok, some progress, I'm not so often working on the 4k-PC so i've forgotten some important - the restricted output (YCbCr420) of the GTX960 on 3840*2160.
I remembered when I saw the 2 screenshots in post #20939 (http://forum.doom9.org/showthread.php?p=1773808#post1773808) on a RGB outputting Intel GPU - no differences! Strange enough what code a png image, shot with 'Native' contains to show banding only through YCbCr420 and the other 'Copy-back' png not!

So next test - switching the GTX960 from 2160p to 1080p + RGB output - playing the demo and actually no more differences between DXVA2 Native and Copy-back!
What causes 'Native' to produce stronger banding through YCbCr420 using any renderer and 'Copy-back' not - no idea!

Now I hope the always remaining slight banding might vanish through a real 10-bit output into a real 10-bit TV or Monitor, that would make me and my Snake-demo happy ;)


@ madshi
..and for a real 10-bit output we need a madVR performing a reliable 2160p! :
Naturally I always test madVR as well.
When the GTX960 was at 1080p madVR had no probs switching fullscreen with LAV 'Native'. So its just the screen resolution 2160p causing a black fullscreen. And extremely slow FSE switching at Copy-back (MPC-HC).

[madshi]

Hmmmm... So you get it working with madVR with a 1080p resolution? The problem with the black screen only occurs with 2160p? Which refresh rate are you using? At 60Hz some GPUs and/or displays are limited to YCbCr 4:2:0, but at 24Hz RGB 4:4:4 should be possible.

[nussman]

2160p RGB (4:4:4) @60hz (50hz?) @10bit is limited by hmdi bandwidth isn't it?

[blaubart]

2160p RGB (4:4:4) @60hz (50hz?) @10bit is limited by hmdi bandwidth isn't it?

yes I heared sth. like HDMI 2.0a and b, only new a is 4:4:4?
My 'old' GTX960 has just HDMI 2.0 nothing known abot a/b?


Hmmmm... So you get it working with madVR with a 1080p resolution? The problem with the black screen only occurs with 2160p? Which refresh rate are you using? At 60Hz some GPUs and/or displays are limited to YCbCr 4:2:0, but at 24Hz RGB 4:4:4 should be possible.

Yes, I even forgot this - just switched from 50Hz to 30Hz and now Nvidia sais 2160p 'RGB full' and 8 bpc. But with LAV at Native madVR fullscreen still black. The sound is playing and normal OSD on a black screen.

But for 10-bit FSE were needed and Copy-back needs the half demo to switch FSE. I'm not dreaming to get 10-bit 2160p out of my GTX960. I not even have a 10-bit TV (not sure) this here is all mentioned for the future..

[DragonQ]

I'm using WebMediaPortal to watch TV outside of my home and it has an option to generate an m3u file to play transcoded streams in external players. I have tried playing it in MPC-HC using LAV Filters but almost every time I end up with no video, although the audio works fine. LAV Filters doesn't even recognise there is a video track. The weird thing is that sometimes it works fine, and playing the RTSP stream directly (no transcoding) works fine every time.

Debugging this is proving difficult. Is there a way to look at debug logs or something from LAV Filters to see if it's encountering an issue with the stream, or whether the issue is purely server-side?

[madshi]

Yes, I even forgot this - just switched from 50Hz to 30Hz and now Nvidia sais 2160p 'RGB full' and 8 bpc. But with LAV at Native madVR fullscreen still black. The sound is playing and normal OSD on a black screen.

But for 10-bit FSE were needed and Copy-back needs the half demo to switch FSE. I'm not dreaming to get 10-bit 2160p out of my GTX960. I not even have a 10-bit TV (not sure) this here is all mentioned for the future..
Strange thing. If only I could reproduce any of that on my PC. I'm getting a 4K display soon, maybe that will help. But nobody else is reporting this issue atm, so I have my doubts I'll be able to reproduce it with the 4K display, either. Just to be safe: So you get a black screen even if use 1080p display resolution, correct?

[blaubart]

Just to be safe: So you get a black screen even if use 1080p display resolution, correct?

No. Remember #20947 ?
>>> When the GTX960 was at 1080p madVR had no probs switching fullscreen with LAV 'Native'. So its just the screen resolution 2160p causing a black fullscreen<<<
Later I tried 2160p @ 30Hz 'RGB full' - still black fullscreen.

Just found out - naturally a GTX960 supports 2160p 50/60Hz RGB full 4:4:4 over HDMI 2.0 ! It was my LG Display just accepting YCbCr420 at 2160p 50/60Hz.


Strange thing. If only I could reproduce any of that on my PC. I'm getting a 4K display soon, maybe that will help.

Good to hear! 4k will knock you down, that's for sure! Windows complete super tiny, will need 150%..


But nobody else is reporting this issue atm, so I have my doubts I'll be able to reproduce it with the 4K display, either.

hmmm, at least one else came from the LAV chief himself:
http://www.dvbviewer.tv/forum/topic/55897-quo-vadis-htpc-richtung-hevc/page-4#entry432470

At this time madVR performed also windowed black. Today only 4k fullscreen remaining. Of course it may depend on numerous factors. But I'm quite confident '4k in madshis house' will soon change everything.
At <30Hz 2160p madVR performs ok (except the LAV Native fullscreen). But at 50/60hz Copy-back FSE becomes critical. And please regard also DVBViewer, he is very unstable on 4k-channels.

We could furthermore talk about this e.g in madVR topic on German DVBViewer forum (there I'm craig_s, remember? ;)) or whatever Forum you like. But please not in horrible doom9-madVR thread..:scared:

Thank you nevcairiel for your patience! :)

[madshi]

Last post on this topic in this thread: That nevcairiel post in the dvbviewer forum was from August 2015. At that time madVR did not support receiving 10bit native DXVA surfaces from the decoder yet, which is why there was a black screen. Support for 10bit native DXVA was added in madVR v0.90.0 in January 2016. So the black screen problem you have now must have a different cause. I'll follow up in the dvbviewer thread once I have that 4K screen.

[blaubart]

Just switched from Win10 to Win7, same issue. *here (http://www.dvbviewer.tv/forum/topic/37505-madvr-renderer-in-dvbviewer-nutzen/?p=449338)

[chros]

Just found out - naturally a GTX960 supports 2160p 50/60Hz RGB full 4:4:4 over HDMI 2.0 ! It was my LG Display just accepting YCbCr420 at 2160p 50/60Hz.

I'm getting a 4K display soon
Can I ask which one? Monitor or TV?
Choosing the right hardware chain (!) will be more crucial then ever was. Maybe we should create a whole new thread (e.g. called "Which hardware chain for lav+madvr playback?") for this purpose with a checklist in the first post.

[mandarynek82]

So which settings in LAV and MadVR will be the best for GTX 580? Btw still I can enable hardware acceleration only in dxva copy-back mode.

http://s6d2.turboimg.net/t/28548436_Snap1.jpg (http://www.turboimagehost.com/p/28548436/Snap1.jpg.html) http://s6d2.turboimg.net/t/28548437_Snap2.jpg (http://www.turboimagehost.com/p/28548437/Snap2.jpg.html)

[ryrynz]

So which settings in LAV and MadVR will be the best for GTX 580? Btw still I can enable hardware acceleration only in dxva copy-back mode.


Your settings are fine. MadVR is a whole different kettle of fish and should not be discussed here. There are plenty of guides online to assist you with making decisions with that.

[mandarynek82]

Your settings are fine. MadVR is a whole different kettle of fish and should not be discussed here. There are plenty of guides online to assist you with making decisions with that.

OK but how I can enable hardware acceleration? I'ts working only in copy-back mode.

[strumf666]

Copy-back is the preferred mode for most, unless some special circumstances apply.

[alexacolor]

There is an old problem with the non-occurrence of sound in the rtmp stream.
Also in such streams synchronization problem.
Here is an example of the flow:
rtmp://tv-penza.ru/live/1

Flash and VLC playing this stream with sound.

[LigH]

It's pretty hard to find reliable reports about the minimum hardware requirements to decode HEVC 1080p (as used in DVB-T2 HD in Germany) without GPU acceleration. Most discussions I found so far immediately switch the topic to "better get some CPU with GPU incl. hardware decoding support". That may mean exchanging the complete PC for some... and the question was rather if existing hardware is still good enough.

So - which CPU class would you consider a minimum for decoding 1080p HEVC with LAV Filters?

In my experience, an AMD Phenom-II X4 is just not fast enough, most probably due to the lack of instruction sets much better than SSE2. I guess an FX era QuadCore may be able... And regarding intel, a Core i# Quad should be a rather safe bet; I read a Core2 Duo is not.

P.S.: Another fresh report about an E7500 DualCore being close to the threshold.

[sneaker_ger]

In my experience, an AMD Phenom-II X4 is just not fast enough, most probably due to the lack of instruction sets much better than SSE2. I guess an FX era QuadCore may be able... And regarding intel, a Core i# Quad should be a rather safe bet; I read a Core2 Duo is not.
Read some reports about mobile Intel Core i5 dual core being more than sufficient. My Core i7-860 (1st gen i7, 2009) does some 500 fps on German DVB-T2 with LAV 64. Unfortunately the popular DVB Viewer is 32 bit only. Lentoid decoder is much faster on 32 bit and older hardware in general, it seems. But it may have some bugs. It's worth a try, at least.

[Nevilne]

1) on your current pc, decode a vid to -null in ffmpeg and note the speed
2) check your cpu score on cpubenchmark.net
3) do some simple math
4) see which minimal cpu score you require

[huhn]

than you have an idea for one file...

it is heavily based on the used bit rate.

and than you can't ignore CPU instractions

[sneaker_ger]

Is ffmpeg now as fast as LAV in hevc decoding? Nevcairiel cherry-picked optimizations from OpenHEVC, originally.

[Nintendo Maniac 64]

Hey, is there any way to specify LAVfilters to not decode 1080p 60fps? I have an Nvidia NVS 3100m that can decode 720p 60fps and 1080p 30fps, but 1080p 60fps makes it choke. The CPU is plenty fast (i5 540m) to decode 1080p 60fps in software though, but I only want to use software decoding when necessary and not for everything.

tl;dr version:

720p 60fps = OK
1080p 30fps = OK
1080p 60fps = fail

Currently, AFAIK, LAVfilters only has options for specifying resolution (like 720p and 1080p), but not the framerate, which is no good when 1080p 30fps is perfectly fine but 1080p 60fps is not.

For reference the iGPU on my AMD E-350 also has this exact same performance issue (though the CPU portion is much slower and definitely can't do 1080p 60fps in software).


Oh and to clarify, this is all with 8bit AVC @ high profile.

----------------------------------------------------------------

It's pretty hard to find reliable reports about the minimum hardware requirements to decode HEVC 1080p

In my experience, an AMD Phenom-II X4 is just not fast enough

I'm assuming you're talking about codecs newer than AVC, right? Because I could decode 1080p 30fps AVC in software on a lowly Athlon 64 x2 4800+ (2.5GHz)...

As for newer codecs, I've been able to do YouTube-encoded VP9 at 1440p 30fps and 1080p 60fps on a 2.4 GHz Penryn Core 2 Duo, but a key point is to use "D3D Fullscreen" in MPC-HC because otherwise the el-lame-o Intel 965GMA just frame-skips like no tomorrow. No idea about HEVC though...

[LigH]

Yes, I explicitly named HEVC = H.265, because it is used in DVB-T2 HD.

[Nintendo Maniac 64]

Yes, I explicitly named HEVC = H.265, because it is used in DVB-T2 HD.

Well if you could provide a specific video to use for testing, I could test it on a few older CPUs including a...

1.2GHz Pentium M
1.5GHz Celeron (Pentium M-based)
1.6GHz Intel Atom (1st generation single core + SMT)
2GHz Core 2 Duo (Merom)
2.4GHz Core 2 Duo (Penryn)
2.4GHz i5 520m (Westmere)

(turns out the previously-mentioned i5 540m was actually faulty believe it or not - swapped it out and now everything works including 1080p 60fps without any random BSoD-ing)

If I include CPUs and PCs that don't currently have an OS, I also have a...

1.73GHz Athlon XP (Thoroughbred)
2GHz Pentium 4 (Northwood)
2.7GHz Celeron (Northwood)
1.6GHz Core Duo (Yonah, not Core 2 Duo)
1.73GHz Celeron (Merom)
1.9GHz Athlon 64 x2 (Brisbane G1)
2.5 GHz Athlon 64 x2 (Brisbane G2)
3GHz Athlon 64 x2 (Windsor)

[LigH]

I believe sneaker_ger may be able. There may be material up to 1080p@50/60fps, probably with limited complexity.

I asked on behalf of people in the DVBViewer and German doom9/Gleitz boards, but I don't have reception without roof antenna, too far away from supported towns.

[Nerva]

Is there a reason to use "Expand 6.1 to 7.1" vs. "Enable Mixing" with the Output Speaker Configuration being either 5.1 or 7.1? I have a PC with 5.1 and a HTPC with 7.1. I'm just thinking, if I'm already using the mixer and I'm playing a 6.1 file, wouldn't the mixer serve the same function as the "Expand 6.1 to 7.1" option.

Could someone answer this? Thanks.

[DragonQ]

Yes, I explicitly named HEVC = H.265, because it is used in DVB-T2 HD.
Where? In the UK DVB-T2 HD is AVC.

[nevcairiel]

There is dedicated threads on this forum about performance talks, please direct such discussions there! Thanks!

[robl45]

quick question. I've used MPC-HC for a long time now with lav filters, I set up external filters in MPC-HC as that is what i prefer and it works fine. I wanted to try out MPC-BE and I did the same thing adding the external filters and turning off all internal filters yet it is using Lav splitter and Lav video but won't use lav audio. Is there a special way to setup mpc-be?

[VictorLS]

nevcairiel
Prog add MainConcept HEVC Video Decoder support in its Demultiplexor
http://i.cubeupload.com/HDmGWT.png
Is it difficult to add it in LAV Splitter?

[VictorLS]

robl45
In MPC-BE 1.4.4 (build 286) under XP I unchecked all in Audio Tab Internal Filters and add LAV Audio set point to Prefer:
http://i.cubeupload.com/c16Bzn.png
What do you like MPC-BE for?

[LigH]

@ VictorLS:

You want LAV Splitter to "support" MainConcept HEVC Decoder ...

a) LAV Video Decoder can decode HEVC too, what else can MainConcept what LAV Video Decoder cannot?
b) Does MainConcept need a specific support at all, e.g. a special authentication, to be used in DirectShow filter chains? If not, I guess you could just disable HEVC support in LAV Video Decoder, and DirectShow shall connect to the next available filter?

[owlsroost]

Where? In the UK DVB-T2 HD is AVC.

HEVC is being used for the new DVB-T2 service in Germany - only test transmissions at present, but the full service is due to start in May 2017 (AFAIK).

From the test recordings I've played with, it's mostly 1920 x 1080p @ 50fps, around 2 - 2.5 Mbps e.g. -

Video
ID : 48 (0x30)
Menu ID : 137 (0x89)
Format : HEVC
Format/Info : High Efficiency Video Coding
Format profile : Main@L4.1
Codec ID : 36
Duration : 1mn 2s
Bit rate : 2 182 Kbps
Width : 1 920 pixels
Height : 1 080 pixels
Display aspect ratio : 16:9
Frame rate : 50.000 fps
Color space : YUV
Chroma subsampling : 4:2:0
Bit depth : 8 bits
Bits/(Pixel*Frame) : 0.021

[robl45]

robl45
In MPC-BE 1.4.4 (build 286) under XP I unchecked all in Audio Tab Internal Filters and add LAV Audio set point to Prefer:
http://i.cubeupload.com/c16Bzn.png
What do you like MPC-BE for?

BE does stereo scopic subtitles and they work using madvr for rendering MVC 3d.

[VictorLS]

LigH
I have not strong enough computer to decode h265 3840x2160p60 10bit smoothly (h265 1920x1080p50 I can watch absolutely smoothly using CyberLink Video Decoder (PDVD Generic) with 20% load of each three cores of my CPU and using LAV Video Decoder with 50% CPU load). So I wanted to see slideshow h265 3840x2160p60 10bit directly from SATs but LAV Video Decoder can not decode I-frames only - I strongly asked nevcairiel add this feature in LAV Video Decoder two times earlier but (((
Also I asked Cyberlink - they are not answered me at all. Elecard HEVC Video Decoder decodes recorded on HDD h265 3840x2160p60 10bit perfectly but directly from SAT - decodes well 8bit only.
MainConcept HEVC Video Decoder can decode I-frames only with SkipPB settings in DVBViewer many times ago and in ProgDVB from yesterday, because I think MainConcept HEVC Video Decoder has non-standart pin which can connect to MainConcept's MPEG and MP4 Demultiplexer, but can not to LAV Splitter, so I can not use MainConcept HEVC Video Decoder with i.e. mkv files before nevcairiel add support of MainConcept HEVC Video Decoder in LAV Splitter.
PS. I don't want to keep many demultiplexers and decoders in my WinXP and Win7x64 )

[3421]

I removed the message.

[GTPVHD]

https://developer.nvidia.com/nvidia-video-codec-sdk

http://on-demand.gputechconf.com/gtc/2016/presentation/s6226-abhijit-patait-high-performance-video.pdf

http://i.imgur.com/KXdG6YH.png

Maybe nev can retire CUVID decoding since Nvidia doesn't support it much anymore and it doesn't work on Win10 anyway, prefering a unified SDK approach with Nvidia Video Codec SDK 7.0.

[wawa79]

Hello,
Until today, I have been prefering FFDShow audio to LAV audio in my Mediaportal CODEC settings.
The reason is that I never managed to get LAV audio to output Dolby TrueHD 5.1/7.1 as multichannel AC3 to my SPDIF output (my 5.1 AVP has no HDMI / HD audio): I tried the last 0.68.1 version, I tried to enable mixing and in best case, I only get 2.0 (despite GraphStudio status shows 8 channels IN and 6 channel OUT when I use LAVF :devil:)

FFDShow properly handles direct 5.1 TrueHD output and downmixes 7.1 TrueHD (using its Mixer panel) and reencodes it as AC3 to fit my SPDIF output.

My life with FFDShow was peaceful until Dolby's marketing decision to release Dolby Atmos to public. FFDShow just get stuck on these streams so I cannot use it anymore. And because Atmos is the default stream selected when starting playback, there is no way to play these streams from MP...

I found a workaround which is to setup LAVF audio as the decoder and add FFDShow as postprocessor but I am pretty sure that I am missing something to properly setup LAVF to handle properly TrueHD on my setup... or that there is definitely a bug.

Could anyone help me?

Thank you in advance!


But

[nevcairiel]

Maybe nev can retire CUVID decoding since Nvidia doesn't support it much anymore and it doesn't work on Win10 anyway, prefering a unified SDK approach with Nvidia Video Codec SDK 7.0.

They didn't change anything, its still the same thing in a new zip file.

[PeQuE]

I've searched all thread for the word "EXIF" and I only found two posts dated at 2012, which have nothing to do with what I want to ask, so here we go...

I know EXIF metadata has been used to "declare" picture orientation for many years... What I didn't know is that videos (at least .mov and .mp4 ones) had also this EXIF info. The fact is that both smartphones currently at home (IPhone 6s and Nexus 5X) uses "Rotation" EXIF tag to declare actual orientation for videos taken with them...

The question is: who in directshow filter chain is responsible to process this info and finally play correctly oriented video? splitter? video filter? player itself (if it has any sense)?

My currently set up is LAV Splitter + LAV Video inside MediaPortal 1, and it's failing at playing this videos correctly: vertical ones taken with IPhone 6s, all of them taken with right hand using Nexus 5X, because the're all tagged with rotation=180 as sensor is upside-down mounted for space matters for this smartphone.

Thanks a lot. :)

[sneaker_ger]

LAV can read and pass on such rotation info but leaves the actual work to the renderer. madVR should support rotating.

http://forum.team-mediaportal.com/threads/madvr-support-in-mp1.133972/

[PeQuE]

Wow! So it's renderer which does the job?!

I'll tesd madVR then... :)

Thanks a lot for the info.

[suanm]

Mr.nevcairiel
It's very popular with HD fans that one splitter can output and display in-built two and more language subtitles (for example:english and chinese or french) at the same time when playback.I don't find LAV splitter source or LAV splitter can do the same task as I imagine.The in-built splitter in potplayer can splits two different language subtitles at the same time when playback,though.
Could you improve LAV splitter source or LAV splitter?
Thank you in advance.

[aufkrawall]

Received a GTX 1070, 4k HEVC 10 bit 50fps works fine with DXVA2 copyback on the Sandy Bridge PCIe 2.0 platform.
(Sandy till death... or at least Zen)
Can't test anything else anymore, card is already out of the PC and will go back soon (not satisfied with the fans and performance in some games).

[paranoya7]

i cant play some avi files when i choose default splitter of lav splitter. i can play all avi files when i choose avi splitter. Here im adding 2 avi files nfo. First one cant play second one can play. What should i do and what is wrong? I hope someone help me.

I'm using mpc be and codec pack last versions.

[LigH]

Attachments take a while to be approved; what kind of "nfo" is this, a screenshot from player internal dialogs? A MediaInfo analysis in text form could have get pasted in a CODE block much easier.

How does it "not play"? Is there any error message, or does it just not start playing?

[paranoya7]

This one cant play. I click video and mpc hc window open then i have black screen just. Nothing happen. (If i choose avi splitter, everything fine)

Format : AVI
Format/Info : Audio Video Interleave
File size : 490 MiB
Duration : 1 h 36 min
Overall bit rate : 708 kb/s
Writing application : Lavf56.25.101

Video
ID : 0
Format : AVC
Format/Info : Advanced Video Codec
Format profile : Main@L3
Format settings, CABAC : Yes
Format settings, ReFrames : 3 frames
Codec ID : H264
Duration : 1 h 36 min
Bit rate : 564 kb/s
Width : 854 pixels
Height : 480 pixels
Display aspect ratio : 16:9
Frame rate mode : Variable
Frame rate : 50.000 FPS
Color space : YUV
Chroma subsampling : 4:2:0
Bit depth : 8 bits
Scan type : Progressive
Bits/(Pixel*Frame) : 0.028
Stream size : 390 MiB (80%)

This one is playing without no problem.

Format : AVI
Format/Info : Audio Video Interleave
File size : 174 MiB
Duration : 20 min
Overall bit rate mode : Variable
Overall bit rate : 1 200 kb/s
Writing application : Nandub v1.0rc2
Writing library : Nandub build 1853/release

Video
ID : 0
Format : MPEG-4 Visual
Format profile : Advanced Simple@L5
Format settings, BVOP : 2
Format settings, QPel : No
Format settings, GMC : No warppoints
Format settings, Matrix : Default (H.263)
Codec ID : XVID
Codec ID/Hint : XviD
Duration : 20 min
Bit rate : 1 054 kb/s
Width : 512 pixels
Height : 384 pixels
Display aspect ratio : 4:3
Frame rate : 25.000 FPS
Color space : YUV
Chroma subsampling : 4:2:0
Bit depth : 8 bits
Scan type : Progressive
Compression mode : Lossy
Bits/(Pixel*Frame) : 0.214
Stream size : 153 MiB (88%)
Writing library : XviD 1.0.3 (UTC 2004-12-20)

[egandt]

Received a GTX 1070, 4k HEVC 10 bit 50fps works fine with DXVA2 copyback on the Sandy Bridge PCIe 2.0 platform.
(Sandy till death... or at least Zen)
Can't test anything else anymore, card is already out of the PC and will go back soon (not satisfied with the fans and performance in some games).

I have a GTX 1070, using DXVA2 copyback I've tested 4k30p HEVC10, up to 400mbps, without issues.

http://jell.yfish.us/


ERIC

[3421]

GTX 1070 150 watt. Hot. GTX 980 165 W, GTX 580 244 W. 150 W is still a lot.

[Sunset1982]

Any news of the RX480 HEVC decoding issue? Anyone here with a RX card?

[chros]

@nevcairiel: thanks for the Weston Three Fields (simple) method, it does wonders with the caps of Olympics on my underclocked laptop!

[mixus]

I5 6600K + GTX 1060

DXVA2 Copy-Back

Success! ( HEVC Main10 4K 120 FPS )

Video is Converted with StaxRip.
Used AssumFPS and NVENC (rigaya)

http://i.hizliresim.com/MkE162.png


Video
ID : 1
Format : HEVC
Format/Info : High Efficiency Video Coding
Format profile : Main 10@L6.2@Main
Codec ID : V_MPEGH/ISO/HEVC
Duration : 55 s 383 ms
Bit rate : 202 Mb/s
Width : 3 840 pixels
Height : 2 160 pixels
Display aspect ratio : 16:9
Frame rate mode : Variable
Frame rate : 120.001 FPS
Original frame rate : 120.000 FPS
Color space : YUV
Chroma subsampling : 4:2:0
Bit depth : 10 bits
Bits/(Pixel*Frame) : 0.203
Stream size : 1.30 GiB (100%)
Default : Yes
Forced : No

(Sorry for bad english, informational purposes only.)

[xerces8]

Hi!

I installed 0.68.1 on Windows 8.1 64 bit and it works (had to associate files with WMP myself), but the "Play to" feature of WMP ("sending" the files to a DLNA/UPnP device like a TV) does not work. The files not natively supported by Windows do not send. (so MP3 and M4A files work fine)
I tried *.flac , *.ogg and *.opus files.

Can this be made to work? With LAV Filters? Or should try with some other codecs? (I only interested in streaming music file, the xiph dshow codec pack does not work either, the author confirmed it is a limitation of their woftware)

Regards,
David

[nevcairiel]

One guess would that "Play To" uses Media Foundation, so nothing a DShow decoder would handle, otherwise no idea what that does.

[xerces8]

Yes, that might be the case, as "Media Foundation FLAC Codec" works for PlayTo.
So LAV is not the solution for me?
Any ideas for MF for ogg/vorbis and opus?

[chros]

Yes, that might be the case, as "Media Foundation FLAC Codec" works for PlayTo.
So LAV is not the solution for me?
Any ideas for MF for ogg/vorbis and opus?
I think a normal DNLA service would be the right solution, on Windows e.g. Universal Media Server (http://www.universalmediaserver.com) , on Linux (e.g. on OpenWRT) minidlna (https://sourceforge.net/projects/minidlna/).

[Sunset1982]

nev, you have an idea what causes the rx480's HEVC decoder to get greenscreen? Is it LAV incompatibility or a madvr problem?

[nevcairiel]

nev, you have an idea what causes the rx480's HEVC decoder to get greenscreen? Is it LAV incompatibility or a madvr problem?

If history with AMD tells us anything, its their drivers. But I do not plan to buy such a card.

[Schwartz]

Not sure if this would be useful for anyone else, but for my laptop I would love the option to set the resolution limit for LAV hardware decoding by hand instead of those three checkboxes, since it's an older chip. What it can handle falls squarely between SD and HD.

[Gravitator]

Привет.
CUDA when decoding AVC consumes 15% CPU : 12% GPU;
DXVA2 (copy-baсk) when decoding AVC uses 15% CPU : 52% GPU.
Of course, I will use CUDA! Only every time I have to switch manually to DXVA2 (copy-baсk) for HEVC.
Please make ability to automatically switch.

[sneaker_ger]

CUDA when decoding AVC consumes 15% CPU : 12% GPU;
DXVA2 (copy-baсk) when decoding AVC uses 15% CPU : 52% GPU.
Did you compare GPU clocks using GPU-Z?

[huhn]

CUDA is using a high powerstate while DXVA can be used with a low power state.

so GPU usage is pointless.
on top of it only hybrid decoding is using the CPU and the GPU.

real hardware acceleration in nvdia isn't using the GPU shaders.

[LigH]

There is instead the "PureVideo" decoder.

[Gravitator]

Did you compare GPU clocks using GPU-Z?
I use desktop gadget. The photo shows a big difference.
Like any ordinary user begins to worry about the rape of equipment :)

General
Unique ID : 242932764535472548733391206744208490955 (0xB6C32608A0F49ED889F89E75961CC1CB)
Complete name : U:\Кингсглейв Последняя фантазия XV.mkv
Format : Matroska
Format version : Version 4 / Version 2
File size : 3.96 GiB
Duration : 1h 55mn
Overall bit rate : 4 903 Kbps
Encoded date : UTC 2016-08-29 18:21:29
Writing application : mkvmerge v8.3.0 ('Over the Horizon') 64bit
Writing library : libebml v1.3.1 + libmatroska v1.4.2
DURATION : 01:55:32.896000000
NUMBER_OF_FRAMES : 216653
NUMBER_OF_BYTES : 332779008
_STATISTICS_WRITING_APP : mkvmerge v8.3.0 ('Over the Horizon') 64bit
_STATISTICS_WRITING_DATE_UTC : 2016-08-29 18:21:29
_STATISTICS_TAGS : BPS DURATION NUMBER_OF_FRAMES NUMBER_OF_BYTES

Video
ID : 1
Format : AVC
Format/Info : Advanced Video Codec
Format profile : High@L4
Format settings, CABAC : Yes
Format settings, ReFrames : 5 frames
Codec ID : V_MPEG4/ISO/AVC
Duration : 1h 55mn
Bit rate : 4 422 Kbps
Width : 1 920 pixels
Height : 796 pixels
Display aspect ratio : 2.40:1
Frame rate mode : Constant
Frame rate : 23.976 fps
Color space : YUV
Chroma subsampling : 4:2:0
Bit depth : 8 bits
Scan type : Progressive
Bits/(Pixel*Frame) : 0.121
Stream size : 3.57 GiB (90%)
Language : English
Default : Yes
Forced : No
Color range : Limited
Color primaries : BT.709
Transfer characteristics : BT.709
Matrix coefficients : BT.709

Audio
ID : 2
Format : AC-3
Format/Info : Audio Coding 3
Mode extension : CM (complete main)
Format settings, Endianness : Big
Codec ID : A_AC3
Duration : 1h 55mn
Bit rate mode : Constant
Bit rate : 384 Kbps
Channel(s) : 6 channels
Channel positions : Front: L C R, Side: L R, LFE
Sampling rate : 48.0 KHz
Compression mode : Lossy
Stream size : 317 MiB (8%)
Language : Russian
Default : Yes
Forced : No

Menu
00:00:00.000 : en:1
00:12:05.975 : en:2
00:15:47.529 : en:3
00:20:22.845 : en:4
00:25:40.996 : en:5
00:32:13.722 : en:6
00:39:01.212 : en:7
00:43:43.786 : en:8
00:50:05.292 : en:9
00:54:58.418 : en:10
01:02:12.518 : en:11
01:12:54.284 : en:12
01:20:59.227 : en:13
01:26:23.759 : en:14
01:37:17.370 : en:15
01:41:56.148 : en:16


Prompt method of placing text in a spoiler?

[huhn]

like said before the GPU usage is not important.
it simply looks higher with DXVA because the GPU isn't running with it's max clock. but in the end it is saving power this way.

both are using the video engine not the shaders.

[Gravitator]

Then pay attention to the video memory consumption on the photo? CUDA 143MB and DXVA2 (copy-back) 880MB. :scared:

[Asmodian]

Then pay attention to the video memory consumption on the photo? CUDA 143MB and DXVA2 (copy-back) 880MB. :scared:

Is that bad? You aren't using that video memory for anything else and your GPU is using less power when using DXVA2 (copy-back).

[captaiŋadamo]

Then pay attention to the video memory consumption on the photo? CUDA 143MB and DXVA2 (copy-back) 880MB. :scared:

Yeah, and? What exactly is the problem? The GPU has that memory specifically for the purpose of it being used.

[3421]

I see MediaInfo writes video file YUV 8 bit. Graph writes this video file for lav splitter: video biBitcount 12 bit. MediaInfo writes YUV 10 bit, graph writes this video file for lav splitter: video biBitcount 15 bit. LAV Splitter convert color or?

[huhn]

no this is just bits per pixel and bits per component.

[Gravitator]

Did you compare GPU clocks using GPU-Z?
Indeed DXVA2 trying to reduce/save power consumption. Meaning marketing (nVidia-CUDA or DXVA2) also affects the choice/decision-making human :)

In such cases it would be good to enlighten users LAV/LOVE - display text explaining the effectiveness of decoders.

[chros]

Indeed DXVA2 trying to reduce/save power consumption. Meaning marketing (nVidia-CUDA or DXVA2) also affects the choice/decision-making human :)
In certain edge cases CUVID is still faster (= usable) than DXVA2 (probably thanks to its memory usage), e.g. with old hardware and h264 4K content. And you can also restrict powerstate of the GPU e.g. with nvidiainspector.

[cnckane]

Hi!

I just started to use MPC HC and LAV decoders a few weeks ago - everything works fine, but I noticed that some HEVC encodes are not playing well when hardware accelaration (DXVA2-both or CUVID) is enabled.

The TV Show rips in 720p and 1080p are okay with DXVA but there are occaisonal small stutters when Alternate Vsync is enabled - this can be an MPC HC/EVR issue though.

However a movie encode (1 minute sample can be downloaded here: https://mega.nz/#!ONhEkDxb!n3Dks1EeYa5of-LMbO3V-Xopzhqy2LOYr47zFuKeltI ) and "jellyfish" test videos are unwatchable - constant lags with black screen and video desync issues with DXVA2 enabled.

With CUVID it plays the movie but Hardware Accelaration doesn't work.


I have a GTX 970 which as far as I know does not fully support HEVC decoding - could this the cause of the issues ? I thought Cuvid supposed to offer a "hybrid" decoding mode using CUDA.

I also tried the latest nightlies but the same happens.

^^ You can use nVidia inspector to limit power states, even cuvid can work at the lowest 135-405 MHz that way.

[P.J]

Stay away from Hybrid.

/done

[CruNcher]

it seems the error here is somewhere in the muxing/demuxing/parsing but EVR + Lav Splitter + Lav CUVID is very robustly ignoring the issue it seems while EVR + Lav Splitter + Lav (any)DXVA falls still completely apart with it in a sync trauma with Nvidias Hybrid Decoder on GM204.

Lav DXVA
http://i1.sendpic.org/t/At/At4eSr6J11rSXnR5AmVFEivwCMX.jpg (http://sendpic.org/view/1/i/x5ERPCphwTWrBqi7KeJplN6Drmm.png)

Lav CUVID
http://i1.sendpic.org/t/3O/3OBwZ4AOuq35DPsP7GEGDERunWE.jpg (http://sendpic.org/view/1/i/ckxAfn9rhhqAfbcjWqIczYimz10.png)


Complete name : F:\hevc\8bit\LG_4K_View-the-Feeling.mp4
Format : MPEG-TS
File size : 347 MiB
Duration : 1mn 56s
Overall bit rate mode : Variable
Overall bit rate : 24.9 Mbps
Maximum Overall bit rate : 35.5 Mbps

Video
ID : 4113 (0x1011)
Menu ID : 1 (0x1)
Format : HEVC
Format/Info : High Efficiency Video Coding
Codec ID : 6

Audio
ID : 4352 (0x1100)
Menu ID : 1 (0x1)
Format : AAC
Format/Info : Advanced Audio Codec
Format version : Version 2
Format profile : LC
Muxing mode : ADTS
Codec ID : 15
Duration : 1mn 56s
Bit rate mode : Variable
Channel(s) : 2 channels
Channel positions : Front: L R
Sampling rate : 48.0 KHz
Compression mode : Lossy

http://demo-uhd3d.com/fiche.php?cat=uhd&id=101

Stay away from Hybrid.

/done

Sorry but i find this a totally wrong viewpoint you should rather use it when beneficial and leave it away when not if you don't plan to update your hardware with a full fixed function decoder it still can be beneficial depending on your systems overall configuration and Bitstream Complexity.
So why wasting CPU time and Energy for everything if you doesn't need to ?.


Also from a business point it would be not very wise ignoring GTX 970 user configurations GM204 VP6 Feature Set E as your Online Streaming target Platform.

So you have to provide either a High Quality H.264 Bitstream or a Complexity Nvidia Hybrid Decoder Adapted H.265 Bitstream

Also his Encoded Bitstream is not very complex @ all on the complexity scala the GM204 Hybrid Decoder can actually handle it's very very low and works without issues either with DXVA or CUVID, though it seems to be violating the rules.


Lav DXVA
http://i1.sendpic.org/t/vH/vH4spqE1uwUxR7X9up4d93e6J5a.jpg (http://sendpic.org/view/1/i/5YbZRZszPmpEGupv3zLb1aTJ76u.png)

[VictorLS]

I have bought GM206-150-A1 based videocard Palit nVIDIA GTX750 1GB (for MadVR 2GB better) NE5X750THD01-2065F about 100$ and it accelerates perfect under Win7x64 until h265 3840x2160 10bit (excluding h264 10bit) with DXVA2 (native or copy-back). But under XP it accelerates until h265 3840x2160 8bit only with NVIDIA CUVID. I knew it before buying videocard, but
if this file is h265 10 bit (HEVC) than CUVID can't deal with it.
Question to all: why CUVID can't deal with h265 3840x2160 10bit? NVIDIA wrote bad SDK? Or something else?

[huhn]

it doesn't support p010

[VictorLS]

huhn
Question is why. Who is wrong? NVIDIA? Many people sure that CUVID is almost same as DXVA (but I am not believe in it).

[huhn]

NVIDIA.

why should nvidia even change that DXVA is way better supported.

they are using the same decoders. the Nvidia PureVideo it would be very dumb to add 2 on the same chip to say it friendly.

CUVID is a different way to use these decoder.

[VictorLS]

huhn
Thanks for explanation. CUVID is much more XP friendly way )

[CruNcher]

huhn
Question is why. Who is wrong? NVIDIA? Many people sure that CUVID is almost same as DXVA (but I am not believe in it).

CUVID could be rather tricky for high quality content because it has no real protective path that DXVA has ;)

XP as a target is not taking into account 1 second anymore because it cant 1 second guarantee content protection.

Only Windows 7/8/8.1/10 can guarantee a fully content protected environment with the underlying UEFI structure over WDDM 1.1+.

So it is overall questionable if CUVID is going to continue anyway under Windows 10 i doubt it or Nvidia makes it PMP WDDM 2.0 ready but i really have my doubts they will invest any resources into this.

CUVID is very useful for Windows 7 users though seeing how much problems DXVA has their in Reality with Nvidia Hardware and Drivers especially for the Hybrid use cases, though Nvidia also continusly improved the Boost System Arbitrator and 3.0 seems to further reduce overall latency with it's new Optimal Performance Setup :)

http://www.alex-ionescu.com/?p=31

proved to be wrong and so still PMP is being taken as secure especialy with Secureboot and Intel AES (MME though lot of controversy the last years around it from russian attacks) and AMD Trustedzone.

And currently Nvidias Falcon Core has gotten as first Windows GPU Hardware a PlayReady certificate from Microsoft/Netflix for 4K 10bit HDR content Deployment at least under Windows 10 AMD is next with Polaris and Vega Hardware.

[Sunset1982]

If history with AMD tells us anything, its their drivers. But I do not plan to buy such a card.

seems you're right. HEVC decoding is now working on RX480 and LAV with latest crimson 16.9.1 drivers. :thanks:

[CruNcher]

Nice contribute results please to the Decoder Thread, we know some stuff from Reviews already but it's always good to have more input data, especialy since the driver changed especialy in the power efficiency domain and those changes never have been taken into account yet by reviewers :)

http://forum.doom9.org/showthread.php?t=171219

[hoborg]

Hello Nevcairiel.
I just want to say Thank you for continuing on this great piece of software!

[NikosD]

seems you're right. HEVC decoding is now working on RX480 and LAV with latest crimson 16.9.1 drivers. :thanks:
Can you decode this at 60fps ?

Netflix_TunnelFlag_4096x2160_60fps_x265_8bit_700Mbit.mkv (https://yadi.sk/i/swMdlJHMuB6yR)

[CruNcher]

According to the Reviewer Data the current UVD Core used shouldn't have enough Performance reserves todo it ;)

based on that Data i would say it would be nice allready to reach 30 fps ;)

[oddball]

I have a problem using LAV splitter in MPC-BE. I have a line to only show forced English subtitles.

eng:eng|f;eng:off;*:eng;*:*

But it is being ignored in MPC-BE. Subtitles continue showing for every video with subtitles even if it's English language audio and English language subs. It works correctly in MPC-HC but I have other issues with that player so wanted to use MPC-BE instead.

LAV splitter is set to high priority and is definitely being loaded as I can see/use it in the system tray.

[nevcairiel]

Some players simply override the stream choices, check its settings for anything related to that.

[oddball]

Some players simply override the stream choices, check its settings for anything related to that.

Wow! Fast reply!

I have but can't find anything in there regarding it. In MPC-HC there is an option 'Prefer forced and/or default subtitle tracks' which does not exist in BE. I am not sure if that is anything to do with why it works in one and not the other?

[nevcairiel]

LAV can only do an initial selection of streams, if the player then decides to select different streams, its not something LAV has any control over. So it absolutely depends on the player, if it honors the splitters choices or just does its own selection. Different players, different behavior.

[oddball]

LAV can only do an initial selection of streams, if the player then decides to select different streams, its not something LAV has any control over. So it absolutely depends on the player, if it honors the splitters choices or just does its own selection. Different players, different behavior.

OK thanks. I will have to ask in the BE player thread then.

[Aleksoid1978]

OK thanks. I will have to ask in the BE player thread then.

Uncheck this.
http://imgur.com/a/Jwtu7

[oddball]

Uncheck this.
http://imgur.com/a/Jwtu7

That did the trick thanks.

[CruNcher]

Soory nevcairiel for being of topic but some RX 4xx users are here.

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 ?

[filler56789]

Originally Posted by filler56789 View Post
apparently LAV Splitter doesn't support multichannel DTS Express

samples: http://www.mediafire.com/download/8i...SX-samples.zip

And that won't change until libdcadec supports DTS Express.

Now it does support :D — thanks again, Nev :thanks:

Hopefully we all can do without the Arcsoft DLL, from now on :)

[Clemens]

Hello,

I have heavy artifacts if I use dxva2 (native) with my GTX 960 4gb, with Software decoding or QSV it's no problem.

The content is of a TV channel, some of the times its just fine on that channel sometimes I get the artifacts.

Here you can download the sample (is SD and I think H.264): https://www.dropbox.com/s/2zxhgqx2dht68qg/DXVA_NVidia_bug.ts?dl=0
Is this a known bug? Is there a workaround? Have others also problems with Nvidia graphics playing this short sample?

If needed I want to make screen record of the artifact...

Thanks and kind regards,
Clemens

[huhn]

the same happens with my 960.

disable SD in hardware acceleration and you should be fine.

[NikosD]

I thought only very old ATI cards had that bug with SD H.264 clips.

And ATI has fixed that a long time ago.

And now Nvidia has the same bug ?

Is it only for Maxwell or Pascal too ?

[huhn]

this looks like broadcast. so this could be a simple error in the stream it can't handle.
there could be a bug in lav filter too because CUVID is handling this properly.

there is more than one clip that is know to make trouble with DXVA.

so just lay back and wait. we will see where the real issue comes from.

[clsid]

That file has invalid data at the start. Happens often with recorded .ts files. The software decoder can usually discard that and recover. You will notice black screen first two seconds. Hardware decoding doesn't have good error resilience.

CUVID also fails here (with a pretty old driver version).

[Clemens]

Thats not the problem I think, when looking at it live (and much longer) then the same artifacts are there no matter how much I restart the stream. (on that particular channel it isn't a problem all of the time, depends on what they broadcast haven't found a pattern yet).

But its no big deal off course, I can switch hardware decoder to another or turn hardware decoding off for SD. If the bug gets fixed nice if not then no problem, just wanted to report the bug also not quite sure if this is the right place for reporting bugs, if I can do something else please tell.

And thanks for looking at it...seems to be a very nice forum!

Kind regards

[clsid]

Well, FFplay shows errors. You can test for yourself if you want.

https://ffmpeg.zeranoe.com/builds/