LAV Filters - DirectShow Media Splitter and Decoders

[roytam1]

It already does output more then one media type. The LAV special one is always the second one in the list.
I can look at them and see if they are "standard", but they'll still not work with the RealAudio DLL based decoder.
quoting my edited message some posts before:
when I revert this commit (http://git.1f0.de/gitweb?p=lavfsplitter.git;a=commitdiff;h=d1db625b604f21eab896d5f9a79686c03005b62a), it works without crashing again

[nevcairiel]

quoting my edited message some posts before:

I already fixed it properly, but i cannot push changes from work :)

[betaking]

NEV last lav git play some h.264 or x.264 give me carsh!
Faulting application mpc-hc.exe, version 1.6.1.4048, time stamp 0x4f321266, faulting module LAVSplitter.ax, version 0.46.0.0, time stamp 0x4f31f2a6, exception code 0xc0000005, fault offset 0x0000860c, process id 0xdb0, application start time 0x01cce63a5cdd1c60.
Git commit: ae4d0e034729731344009c97be61def690ab0864

[Mercury_22]

NEV last lav 0.46 git play some h.264 or x.264 give me carsh!
Faulting application mpc-hc.exe, version 1.6.1.4048, time stamp 0x4f321266, faulting module LAVSplitter.ax, version 0.46.0.0, time stamp 0x4f31f2a6, exception code 0xc0000005, fault offset 0x0000860c, process id 0xdb0, application start time 0x01cce63a5cdd1c60.

Yeah last lavsplitter x64 keeps crashing everything not just h264 e.g. com surrogate too (when exploring my videos folder) (not windows explorer! ):mad: Also the crash message appears after a few sec after closing e.g. MPC not immediately after closing

[ageback]

NEV last lav 0.46 git play some h.264 or x.264 give me carsh!
Faulting application mpc-hc.exe, version 1.6.1.4048, time stamp 0x4f321266, faulting module LAVSplitter.ax, version 0.46.0.0, time stamp 0x4f31f2a6, exception code 0xc0000005, fault offset 0x0000860c, process id 0xdb0, application start time 0x01cce63a5cdd1c60.

Confirmed.Crash caused by LAV Splitter.
Git commit: ae4d0e034729731344009c97be61def690ab0864

[nevcairiel]

Is it too hard to just read the last few posts? It has already been covered. Reporting the same issue 5 times on the same page doesn't help, especially if i said its already fixed. :p

Also, if you use Git builds, its at your own risk. There is no guarantee that it'll work.

PS:
There is no such thing as 0.46

[betaking]

Is it too hard to just read the last few posts? It has already been covered. Reporting the same issue 5 times on the same page doesn't help, especially if i said its already fixed. :p

Also, if you use Git builds, its at your own risk. There is no guarantee that it'll work.

PS:
There is no such thing as 0.46

sorry! can you fix it?

[Kurtnoise]

He said it's already fixed...do you understand ?

[nevcairiel]

sorry! can you fix it?

I cannot. It'll have to keep crashing for ever now. :rolleyes:

[roytam1]

sorry! can you fix it?

It is fixed but the commit doesn't push to git tree yet.
For the time being you can use my build, which reverts a commit to make it working again:
http://roy.orz.hm/lavf-w32-nightlies/lavf-my120208-revert_d1db625b6-ae4d0e0.7z

diff:
http://roy.orz.hm/lavf-w32-nightlies/lavf-my120208-revert_d1db625b6-ae4d0e0.diff

[Aleksoid1978]

It already does output more then one media type. The LAV special one is always the second one in the list.
I can look at them and see if they are "standard", but they'll still not work with the RealAudio DLL based decoder.

No - you set output non standart subtype. Example - COOK codec.
DEFINE_GUID(MEDIASUBTYPE_COOK,
0x4b4f4f43, 0x0000, 0x0010, 0x80, 0x00, 0x00, 0xAA, 0x00, 0x38, 0x9B, 0x71);

but LAV splitter output is different :
subtype: Unknown GUID Name {6B6F6F63-0000-0010-8000-00AA00389B71}

That's what I'm talking about.

[nevcairiel]

That's what I'm talking about.

No its not what you were talking about. You talked about my own special LAV type, which is completely independent from everything else.

Its not my fault that the RM files have "cook" as FourCC inside the file but someone decided to use "COOK" as FourCC in DirectShow. :p

[Aleksoid1978]

Its not my fault that the RM files have "cook" as FourCC inside the file but someone decided to use "COOK" as FourCC in DirectShow. :p

But - you use "COOK" as FourCC in you code :)
also - Atrac3

// RealMedia Audio
{ &MEDIASUBTYPE_COOK, CODEC_ID_COOK },
{ &MEDIASUBTYPE_ATRC, CODEC_ID_ATRAC3 },

[aufkrawall]

I'm sorry to bother you again, nevcairiel, but I'm very sure that there is a bug on decoder site regarding x264 I444 video. ;)
The brightness difference also occurs with 8bit video, so it can't be a dithering issue of the renderer.
I also checked if it could be a problem of Avisynth.
The script is:
LoadCPlugin("C:\Program Files (x86)\AviSynth 2.6\plugins\ffms2.dll")
FFVideoSource("trinergb.avi", colorspace="RGB32")
ConvertToYV24(matrix="Rec709")
AssumeFPS(30)
When I use this script but let x264 output only I420 the results looks absoulutely right (*bright*), it's not a decoder conversion issue.
Here's a I444 8bit sample:
http://www.mediafire.com/?msr5x58laeb0j32

Please take a look at it, probably it's a bug of ffmpeg or x264 encoder? :)

[nevcairiel]

If the RGB output looks correct, its not a decoder issue, but like i said before, a issue in the renderer and its handling of 4:4:4 YUV.
Even if it was a decoder issue, i didn't write the H264 decoder, thats all in the hands of ffmpeg/libav.

[daWsOn_s]

Hello,
I'm trying to get subtitles to work.
The files is m2ts directly from the BD.
The subs are in pgs, I have enabled lav video/audio/splitter and all works, the subs are recognized too. When I select one sub track from the stream with lav splitter they don't show up. I also tried enabling ffdshow sub fillter but no luck.
What is the way to show them?
I'm using MPC HC latest nightly of course.

PS: they work in VLC

[CruNcher]

@Nev

very strange dvcpro 50 decoding issue

http://www.mediafire.com/?ioa60f2np2fbife

avplay = ok
ffplay = ok
lav splitter + lav video = fail

[nevcairiel]

@Nev

very strange dvcpro 50 decoding issue

http://www.mediafire.com/?ioa60f2np2fbife


Fixed

[Pat357]

The script is:
LoadCPlugin("C:\Program Files (x86)\AviSynth 2.6\plugins\ffms2.dll")
FFVideoSource("trinergb.avi", colorspace="RGB32")
ConvertToYV24(matrix="Rec709")
AssumeFPS(30)
When I use this script but let x264 output only I420 the results looks absoulutely right (*bright*), it's not a decoder conversion issue.
I'm afraid it's not that simple to make conclusions like that.

1.What codec is used for your source "trinergb.avi".
If your source is not directly decoded in RGB32, FFMS2 will do a YUV conversion to RGB32 because that's what you ask it to do in the second line of your script.
You will surely get RGB32, but the question is : has it been converted from YUV-> RGB32 first by FFMS2 or not and if it was, how was it converted (matrix ? levels ?)

2.What version of FFMS2 are you using ? The previous version, FFMS v2.16, had major problems with the levels and did always use "full range" (0-255).
The recently released FFMS v2.17 should fix this, but I didn't check it yet. There might be still something not 100% in the first "fixed" release....
I didn't folow the FFMS2 thread, but you can read it here on Doom9 and on Doom10.

3. What version and/or MOD of x264 are you using ?
There have been a lot of discusions about 10bit -> 8bit colorspace/level problems and incorrect levels (not follwing BT709 matrix) with the "official" x264 version.
There are also some x264 MOD's that claim to fix some colorspace conversions/levels. (see for example http://x264.fushizen.eu/ ).
I would recommend the safest way : this is not letting x264 do any 10-> 8 or 8 -> 10 bit conversions and always feeding it "16-235" (lim. level).
I suspect that when feeding it RGB32, it will start with a RGB32 -> YUV conversion, so here could it already go wrong.

4. To check the color visually, I trust MadVR far more than EVR if you talk about 4:4:4 input.
Of course if you use LAV to convert to RGB32, than even EVR or VMR should be able to handle that:p

Note : about x264 vanilla; I just checked with the --version command, the latest build 2146
x264_32b_8b__2146_unpatched --version
x264 0.120.2146 bcd41db
(libswscale 2.1.0)
(libavformat 53.20.0)
(ffmpegsource 2.16.4.0)
built on Jan 19 2012, gcc: 4.6.2
configuration: --bit-depth=8 --chroma-format=all
x264 license: GPL version 2 or later
libswscale/libavformat/ffmpegsource license: GPL version 3 or later
Not sure if this is related to the above mentioned "x264 issues", but it seems like this x264 version was build with the known faulty FFMS v2.16. Maybe the x264 guys patched something to get around this, this should be traceable in the logs.

[Reino]

If his source is yuvj420p (normal Fraps recording method), FFVideoSource(colorspace="RGB32") + ConvertToYV24(matrix="Rec709") produce an entirely different image compared to AVISource (Fraps through VFW) and the (almost) equivalent FFVideoSource() + ConvertToRGB(matrix="PC.709").
If his source is bgra (forced lossless RGB in Fraps), and although the recording of FPS1(bgra) files on my computer has issues (http://forum.doom9.org/showthread.php?p=1556898#post1556898), I don't think FFVideoSource would need any matrix or level conversion at all, because, although faulty, opening one of these FPS1(bgra) files with AVISource or FFVideoSource, without any conversion, produces the same result here (png screenshots are identical in filesize).
Or does Fraps record in yet even more colourspaces, with different ranges and matrixes, I don't know of?...

[sexus]

well latest lav workin nicely , thou freezes if loading up a .avi file if i disable mpeg4 in lav videos format listing it works , please fix this nevcairiel id appreciate it as would others too id asume

[Asmodian]

lol @ sexus

A mpeg4 .avi file? I believe Nevcairiel will need a bit more information. ;)

[Sebastiii]

Hi Nevcairiel,

With that sample all player freeze : https://rapidshare.com/files/1846637945/test_low.flv
It didn't happen on version 0.42 :)
Thanks :)

[nevcairiel]

well latest lav workin nicely , thou freezes if loading up a .avi file if i disable mpeg4 in lav videos format listing it works , please fix this nevcairiel id appreciate it as would others too id asume

You posted the exact same message 4 times or so, which already in itself is considered very rude.

In addition, you don't provide a single piece of useful information.
Try again with a sample file, and i might not ignore you.


With that sample all player freeze : https://rapidshare.com/files/1846637945/test_low.flv
It didn't happen on version 0.42 :)


This is probably already fixed in ffmpeg, but i'm not going to update until the new version is released.

[sexus]

my bad shouldve given you the mediainfo xD:

General
Complete name : H:\Movies library III\Yaar Anmulle\Yaar Anmulle - DVDScr - XviD - 1CDRip - [DDR].avi
Format : AVI
Format/Info : Audio Video Interleave
File size : 691 MiB
Duration : 2h 25mn
Overall bit rate : 665 Kbps
Writing application : VirtualDubMod 1.5.10.3 | www.virtualdub-fr.org || (build 2550/release)
Writing library : VirtualDubMod build 2550/release

Video
ID : 0
Format : MPEG-4 Visual
Format profile : Advanced Simple@L5
Format settings, BVOP : 1
Format settings, QPel : No
Format settings, GMC : No warppoints
Format settings, Matrix : Custom
Codec ID : XVID
Codec ID/Hint : XviD
Duration : 2h 25mn
Bit rate : 580 Kbps
Width : 624 pixels
Height : 256 pixels
Display aspect ratio : 2.40:1
Frame rate : 23.976 fps
Color space : YUV
Chroma subsampling : 4:2:0
Bit depth : 8 bits
Scan type : Progressive
Compression mode : Lossy
Bits/(Pixel*Frame) : 0.152
Stream size : 603 MiB (87%)
Writing library : XviD 1.2.1 (UTC 2008-12-04)

Audio
ID : 1
Format : MPEG Audio
Format version : Version 2
Format profile : Layer 3
Mode : Joint stereo
Mode extension : Intensity Stereo
Codec ID : 55
Codec ID/Hint : MP3
Duration : 2h 25mn
Bit rate mode : Constant
Bit rate : 80.0 Kbps
Channel(s) : 2 channels
Sampling rate : 24.0 KHz
Compression mode : Lossy
Delay relative to video : 13ms
Stream size : 83.2 MiB (12%)
Alignment : Split accross interleaves
Interleave, duration : 1000 ms (23.98 video frames)
Interleave, preload duration : 1008 ms


as said mpeg4 enabled in file list in lav video makes mpchc crash with that file type, not all .avi files but the ones with these properties, thanks

[roytam1]

my bad shouldve given you the mediainfo xD:

General
Complete name : H:\Movies library III\Yaar Anmulle\Yaar Anmulle - DVDScr - XviD - 1CDRip - [DDR].avi
Format : AVI
Format/Info : Audio Video Interleave
File size : 691 MiB
Duration : 2h 25mn
Overall bit rate : 665 Kbps
Writing application : VirtualDubMod 1.5.10.3 | www.virtualdub-fr.org || (build 2550/release)
Writing library : VirtualDubMod build 2550/release

Video
ID : 0
Format : MPEG-4 Visual
Format profile : Advanced Simple@L5
Format settings, BVOP : 1
Format settings, QPel : No
Format settings, GMC : No warppoints
Format settings, Matrix : Custom
Codec ID : XVID
Codec ID/Hint : XviD
Duration : 2h 25mn
Bit rate : 580 Kbps
Width : 624 pixels
Height : 256 pixels
Display aspect ratio : 2.40:1
Frame rate : 23.976 fps
Color space : YUV
Chroma subsampling : 4:2:0
Bit depth : 8 bits
Scan type : Progressive
Compression mode : Lossy
Bits/(Pixel*Frame) : 0.152
Stream size : 603 MiB (87%)
Writing library : XviD 1.2.1 (UTC 2008-12-04)

Audio
ID : 1
Format : MPEG Audio
Format version : Version 2
Format profile : Layer 3
Mode : Joint stereo
Mode extension : Intensity Stereo
Codec ID : 55
Codec ID/Hint : MP3
Duration : 2h 25mn
Bit rate mode : Constant
Bit rate : 80.0 Kbps
Channel(s) : 2 channels
Sampling rate : 24.0 KHz
Compression mode : Lossy
Delay relative to video : 13ms
Stream size : 83.2 MiB (12%)
Alignment : Split accross interleaves
Interleave, duration : 1000 ms (23.98 video frames)
Interleave, preload duration : 1008 ms


as said mpeg4 enabled in file list in lav video makes mpchc crash with that file type, not all .avi files but the ones with these properties, thanks

Don't you think you need to upload a sample for testing and debugging?

[sexus]

ok shall do

[fastplayer]

ok shall do
Don't bother. I reported you already.

[sexus]

http://www.mediafire.com/?lnjean4wf2abgd8 < this i used a different format and switching off mpeg4 in format list wont change the freezing only disabling lav video makes it work

your welcome

p.s: im abit a noob sometimes ok xD

update : ill do another one with a smaller size one sec

http://www.mediafire.com/?jeor5lxxr0tl1kw < this is the one where when you turn off mpeg4 in format list it wont freeze mpchc

[aufkrawall]

I'm afraid it's not that simple to make conclusions like that.

1.What codec is used for your source "trinergb.avi".
If your source is not directly decoded in RGB32, FFMS2 will do a YUV conversion to RGB32 because that's what you ask it to do in the second line of your script.

Source's CSP is RGB.


2.What version of FFMS2 are you using ? The previous version, FFMS v2.16, had major problems with the levels and did always use "full range" (0-255).
The recently released FFMS v2.17 should fix this, but I didn't check it yet. There might be still something not 100% in the first "fixed" release....

I'm using 2.17, the range conversion just works fine.
I'm sure about this because if I let x264 output I420 with the same AVS script, there is no brightness issue.
It doesn't matter either if I let Avisynth convert to YV12 or YV24.


I didn't folow the FFMS2 thread, but you can read it here on Doom9 and on Doom10.

I already got there useful advices. :)


3. What version and/or MOD of x264 are you using ?
There have been a lot of discusions about 10bit -> 8bit colorspace/level problems and incorrect levels (not follwing BT709 matrix) with the "official" x264 version.

I always use the latest official version.
And it has really nothing to do with 10bit video, the issue also exists for 8bit.


There are also some x264 MOD's that claim to fix some colorspace conversions/levels. (see for example http://x264.fushizen.eu/ ).
I would recommend the safest way : this is not letting x264 do any 10-> 8 or 8 -> 10 bit conversions and always feeding it "16-235" (lim. level).
I suspect that when feeding it RGB32, it will start with a RGB32 -> YUV conversion, so here could it already go wrong.

I agree with your idea of not feeding it with risky formats, that's why I let Avisynth do all the decoding and conversion stuff.
But this doesn't help in the case of I444. :(


4. To check the color visually, I trust MadVR far more than EVR if you talk about 4:4:4 input.
Of course if you use LAV to convert to RGB32, than even EVR or VMR should be able to handle that:p

Yes, this works as a workaround.
Maybe it works only with madVR because it does a RGB conversion anyway.

[nevcairiel]

Yes, this works as a workaround.
Maybe it works only with madVR because it does a RGB conversion anyway.

Every renderer does a RGB conversion, its just how it works in PCs. The question is how good or bad the conversion is. ;)

[Aleksoid1978]

Hi Nev.
You have bug in HRESULT CLAVAudio::ParseRealAudioHeader(const BYTE *extra, const int extralen)
and some COOK fail decode.
Need
int ra_extralen = min((extra + extralen) - (fmt+4), *(DWORD*)fmt);
replace with
int ra_extralen = AV_RB32(fmt);

also here
if (codec == CODEC_ID_COOK) {
uint8_t *extra = (uint8_t *)av_mallocz(extralen + FF_INPUT_BUFFER_PADDING_SIZE);
add Atrac codec:
if (codec == CODEC_ID_COOK || codec == CODEC_ID_ATRAC3) {

[Sebastiii]

This is probably already fixed in ffmpeg, but i'm not going to update until the new version is released.

Thanks :)

[avih]

sexus, striked. Please read the forum rules. You clearly don't own this movie.

[sexus]

ummm actually i do own the movie i just tested it out before i purchased it on dvd thats for what torrents are for and its easier to make a sample file from a already pre-ripped movie than having to make one from a dvd from ground up -.-', anyhow if you think i own it or not isnt the issue at hand that ive mentioned , ill be waiting to see what you have to say nevcairiel , im countin on ya bro, peace

p.s: that strike was uncalled for only cause you "clearly" "asume" that i dont own a movie -.-', pro moderation here i can tell

[egur]

Version 0.26 beta is out with the following changes:
* Added option to disable SW decoding when HW can’t decode. Default is not to decode in SW.
* Even faster memory copy function
* FFDShow rev4313

Download from SourceForge home page (http://sourceforge.net/projects/qsdecoder/)

[Cyber-Mav]

ummm actually i do own the movie i just tested it out before i purchased it on dvd thats for what torrents are for and its easier to make a sample file from a already pre-ripped movie than having to make one from a dvd from ground up -.-', anyhow if you think i own it or not isnt the issue at hand that ive mentioned , ill be waiting to see what you have to say nevcairiel , im countin on ya bro, peace

p.s: that strike was uncalled for only cause you "clearly" "asume" that i dont own a movie -.-', pro moderation here i can tell

post the proof that you own it then

[andyvt]

Is there a way to tell which decoding method is actually being used by LAV Video? I'm trying to figure out if a clip is falling back to the software method.

[Andy o]

If you have HW decoding available, it should say if it's being used right next to the option, in the LAV video control panel, while you're playing the video.

[andyvt]

Is there a way to tell which decoding method is actually being used by LAV Video? I'm trying to figure out if a clip is falling back to the software method.

Turns out there is, I just wasn't looking close enough.

When playing a clip that can be HWA the "available" text next to the select box changes to "active". If the clip can't be HWA it stays "available".

[sexus]

post the proof that you own it then

post proof? lols :rolleyes:

heres your proof, the cover scan

http://i1033.photobucket.com/albums/a413/breakdown23/KGrHqRmIE8FFTiZTfBPLueRMDw60_12.jpg

[XadoX]

Is there a way to open the filter settings directly from windows, without going through a special programm like MPC-HC?

[roytam1]

Is there a way to open the filter settings directly from windows, without going through a special programm like MPC-HC?

rundll32 LAVAudio.ax,OpenConfiguration
rundll32 LAVSplitter.ax,OpenConfiguration
rundll32 LAVVideo.ax,OpenConfiguration

[Andy o]

post proof? lols :rolleyes:

heres your proof, the cover scan

http://i1033.photobucket.com/albums/a413/breakdown23/KGrHqRmIE8FFTiZTfBPLueRMDw60_12.jpg

A picture taken from the ebay listing? You know the first rule of digging holes, right?

[sexus]

atleast i tried ,yarrr!! lols xD, anyhow wont change the problems currently with lav as said , hope this can be fixed

[XadoX]

rundll32 LAVAudio.ax,OpenConfiguration
rundll32 LAVSplitter.ax,OpenConfiguration
rundll32 LAVVideo.ax,OpenConfiguration

:thanks:

[nevcairiel]

:thanks:

If you use the installer, it also creates shortcuts like that in the start menu.

[pankov]

Slightly off-topic, but does anyone know if it is possible to run the Intel 2000 IGP at the same time as a Nvidia 450 GTS?

I mean, connect the display to the 450 and just take the audio from the Intel 2000 IGP.

I've search everywhere but I can't really find an answer.

I think I've already posted it here but I'll do it again.
Yes, it works just fine. I'm using a Z68 mobo but I don't think that will make a difference.

[wanezhiling]

http://www.gokuai.com/f/xFm30Q4AHAV5dUks

General
Complete name : D:\Others\[CASO][A-Channal][02][BDRIP][1920x1080][X264xAAC][A4D07E42].mp4
Format : MPEG-4
Format profile : Base Media
Codec ID : isom
File size : 539 MiB
Duration : 24mn 36s
Overall bit rate mode : Variable
Overall bit rate : 3 062 Kbps
Encoded date : UTC 2011-08-12 11:22:12
Tagged date : UTC 2011-08-12 11:22:12

Video
ID : 1
Format : AVC
Format/Info : Advanced Video Codec
Format profile : High@L4.1
Format settings, CABAC : Yes
Format settings, ReFrames : 4 frames
Codec ID : avc1
Codec ID/Info : Advanced Video Coding
Duration : 24mn 36s
Bit rate : 2 670 Kbps
Maximum bit rate : 15.0 Mbps
Width : 1 920 pixels
Height : 1 080 pixels
Display aspect ratio : 16:9
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) : 0.054
Stream size : 470 MiB (87%)
Writing library : x264 core 112 r1867 22bfd31
Encoding settings : cabac=1 / ref=4 / deblock=1:1:1 / analyse=0x3:0x133 / me=umh / subme=7 / psy=1 / psy_rd=0.40:0.00 / mixed_ref=1 / me_range=16 / chroma_me=0 / trellis=1 / 8x8dct=1 / cqm=0 / deadzone=21,11 / fast_pskip=1 / chroma_qp_offset=-2 / threads=6 / sliced_threads=0 / nr=0 / decimate=1 / interlaced=0 / constrained_intra=0 / bframes=5 / b_pyramid=0 / b_adapt=2 / b_bias=0 / direct=3 / weightb=1 / open_gop=0 / weightp=2 / keyint=300 / keyint_min=20 / scenecut=40 / intra_refresh=0 / rc_lookahead=40 / rc=crf / mbtree=1 / crf=18.0 / qcomp=0.60 / qpmin=10 / qpmax=51 / qpstep=4 / vbv_maxrate=38000 / vbv_bufsize=30000 / crf_max=0.0 / nal_hrd=none / ip_ratio=1.40 / aq=1:1.00
Encoded date : UTC 2011-06-05 18:04:11
Tagged date : UTC 2011-08-12 11:22:31

Audio #1
ID : 2
Format : AAC
Format/Info : Advanced Audio Codec
Format profile : LC
Codec ID : 40
Duration : 24mn 36s
Bit rate mode : Variable
Bit rate : 192 Kbps
Maximum bit rate : 238 Kbps
Channel(s) : 2 channels
Channel positions : Front: L R
Sampling rate : 48.0 KHz
Compression mode : Lossy
Stream size : 34.0 MiB (6%)
Encoded date : UTC 2011-08-12 11:22:25
Tagged date : UTC 2011-08-12 11:22:31

Audio #2
ID : 3
Format : AAC
Format/Info : Advanced Audio Codec
Format profile : LC
Codec ID : 40
Duration : 24mn 36s
Bit rate mode : Variable
Bit rate : 192 Kbps
Maximum bit rate : 227 Kbps
Channel(s) : 2 channels
Channel positions : Front: L R
Sampling rate : 48.0 KHz
Compression mode : Lossy
Stream size : 34.1 MiB (6%)
Encoded date : UTC 2011-08-12 11:22:29
Tagged date : UTC 2011-08-12 11:22:31




LAV Splitter select audio 2 not 1 by default.. audio 2 is just some commentary.
http://i.imgur.com/G3Hg3.png

[egur]

@Nev,
I quick question.
this clip (http://www.digital-digest.com/movies/The_Avengers_1080p_Theatrical_Trailer.html) works great using LAV splitter buy Haali reports a frame rate of 90,000fps.
The SPS header indeed reports this fps:
sps.time_scale = 180000
sps.num_units_in_tick = 1

LAV splitter reports correctly a value of AvgTimePerFrame = 417225 within the video info header. Haali reports zero.
Since the SPS data is obviously wrong, how do you find the real frame rate? From looking at several samples?