New ffdshow build (?)

[foxyshadis]

In that case I'll start the pp tweaking again that I'd set aside before tryouts was established. It's still a goal of mine to get auto-levels working as well, so this "washed out" madness can end for good.

[clsid]

diff file for ffdshow-tryout rev49+ (http://d.turboupload.com/d/879658/faad2.zip.html)Thanks.

Updated builds:
download rev63 (http://d.turboupload.com/d/879982/ffdshow_rev2546-63_20060815_mbaff.exe.html)
download rev66 (http://d.turboupload.com/d/880232/ffdshow_rev2546-66_20060815_mbaff.exe.html)
download rev68 (http://d.turboupload.com/d/880283/ffdshow_rev2546-68_20060815_mbaff.exe.html)

[SeeMoreDigital]

Hi clsid,

Thanks for not hosting these builds on RapidShare ;)

[Egh]

Hi clsid,

Thanks for not hosting these builds on RapidShare ;)

btw why not to put them on ffdshow tryouts sf page?

P.S. @VM9: can you make a newer build with recent changes then?

[SeeMoreDigital]

Hi clsid,

Just installed ffdshow_rev2546-68 and noticed it has "MBAFF" support too. How does SVN68 differ from SVN63?

By-the-way, is there any way of "switching on" MBAFF support otherthan via the installer GUI?


Cheers

[Kostarum Rex Persia]

Changelog for rev2546-68?

[SeeMoreDigital]

Changelog for rev2546-68?My apologies... but where can you refresh my memory as to where the change-logs be found nowadays?

I think it would be kinda handy to include the change-log info together with the main .exe file ;)

[clsid]

The installer contains two versions of libavcodec.dll, one with and one without MBAFF. Support for MBAff was already added in revision 16, but it is turned off by default in the makefile.

Latest few (64 -> 68) changes:
- FAAD2 v2.5
- Some asm fixes for vorbis, snow and fft.

ffdshow changelogs (http://svn.sourceforge.net/viewvc/ffdshow-tryout/?sortby=date&view=log)

[haruhiko_yamagata]

Yes, change log is very important.
I posted changelog.txt to SVN (root directory). It includes only changes of rev 7-13. Lest's edit it and compleate.

I think this is a developer's change log and users would need a documentation from their point of view though.

[_xxl]

I updated theora and x264 codecs:
libtheora-1.0alpha7 20060526
x264 core 49 rev 551
x264 doesn't work, any ideas are welcome and appreciated!
http://rapidshare.de/files/29546625/updates.zip.html

[Liisachan]

http://svn.sourceforge.net/viewvc/ffdshow-tryout/changelog.txt

We can check the Log Message too, for each rev
http://svn.sourceforge.net/viewvc/ffdshow-tryout?view=rev&revision=1

[igor1st]

BTW, I communicate with Menno now about this "speed playback problem". I'll inform the results later.
Bug fixed. Thanks to Menno!

My builds updated to latest code with fix (CVS 2006-08-15, v2.6 beta):
http://forum.doom9.org/showthread.php?p=863967#post863967

2 VM9: when doing update ffdshow-tryout don't forget to update version.c for new license info.

[jffulcrum]

Starting from ffdshow-tryouts-rev25 MPEG-1 video playback via libavcodec.dll became sluggish a little. In later builds near all .MPG files mired and some became totally unplayable (whole movie frames dropped after one or two seconds of playback). It doesn`t depend on used player or output renderer as i tried. I`m use generic builds from vm9. Disabling error concealment has no effect. Via libmpeg2 all ok.

[_xxl]

Bug fixed. Thanks to Menno!
My builds updated to latest code with fix (CVS 2006-08-15, v2.6 beta):
http://forum.doom9.org/showthread.php?p=863967#post863967
Could you make a patch file?

[igor1st]

Could you make a patch file?
diff file for ffdshow-tryout rev65-69 (http://d.turboupload.com/d/883767/faad2.zip.html)

[_xxl]

http://rapidshare.de/files/29629541/ffdshow-20060816-rev2546-XXL.exe.html
Updated:
1)FAAD v2.6 beta
2)libtheora-1.0alpha7 20060526

[clsid]

revision 74 (http://www.ffdshow.info)

Does anyone know of some online H.264 sample files that use MBAFF? I want to test if it works ok.

[foxyshadis]

http://www.giusberto.ch/hdtv/
http://forum.doom9.org/showpost.php?p=826082&postcount=5

[_xxl]

Bugs fixed:
1).FAAD2
2).xsharpen + swscaler = green problem
Updates:
1).libtheora-1.0alpha7 20060526
2).Inno Setup Script
Autodetecting best optimizations:
http://rapidshare.de/files/29844729/FFdshow-20060818-rev2546.exe.html
Generic build:
http://rapidshare.de/files/29846447/ffdshow-20060818-rev2546-XXL.exe.html
Inno Setup Script:
http://rapidshare.de/files/29846740/innosetup_script.zip.html

[wyrd]

I seem fixed problem of the following in kurosu's 20060818-18H08 version.
http://forum.doom9.org/showthread.php?p=860593#post860593
btw, why QDM2(mov) is noisy in recent ffdshow-tryouts?
e.g. sample file (http://ffdshow.wyrdic.net/moe/sample/mov/sample_sorenson%5bSVQ1+QDM2%5d.mov)
ffdshow-tryouts-rev24-20060803.exe = work fine.
ffdshow-tryouts-rev38-20060804.exe or later = noisy.
http://forum.doom9.org/showthread.php?p=861459#post861459
http://prdownloads.sourceforge.net/ffdshow-tryout/ffdshow-tryouts-rev60.exe?download
This build crash when playing vorbis audio (in ogm, mkv, ogg audio only too) with libavcodec, tremor is fine.
http://ffdshow-tryout.sourceforge.net/phpBB2/viewtopic.php?t=24

@kurosu and foxyshadis
thanks a lot.:)

[haruhiko_yamagata]

I seem fixed problem of the following in kurosu's 20060818-18H08 version.
http://forum.doom9.org/showthread.php?p=860593#post860593

http://forum.doom9.org/showthread.php?p=861459#post861459

http://ffdshow-tryout.sourceforge.net/phpBB2/viewtopic.php?t=24

@kurosu and foxyshadis
thanks a lot.:)
Vorbis and qdm2 is fixed now. Thank you for bug reports.

//EDIT
New build
ffdshow-20060820-rev80-Q.exe
http://sourceforge.net/project/showfiles.php?group_id=173941

[clsid]

I still have problems with Vorbis :(

When I use tremor in your build MPC will close itself immediately without any errors. It works without problems in my own rev80 build (gcc 3.4.5).

With libavcodec vorbis I get only static noise as output. Both with your build and my own. Same problem as I had before your latest changes. I had no problems in rev47. I suspect something got broken around rev58 when VM9 did some changes to the vorbis code.

Tested on an AMD Athlon (no SSE support). Problem remains after disabling all SIMD sets. So perhaps a bug in the generic code?

[thuan]

haru-han something's wrong with your build, it seems only the ax libavcodec and libmplayer is ok, the other dlls are borked. They don't show up in About->Version details here. BTW libavcodec vorbis works fine now with 2ch audio (I deleted the 5.1ch test file so can't test) don't know what went wrong with clsid though.

[haruhiko_yamagata]

I still have problems with Vorbis :(

When I use tremor in your build MPC will close itself immediately without any errors. It works without problems in my own rev80 build (gcc 3.4.5).

With libavcodec vorbis I get only static noise as output. Both with your build and my own. Same problem as I had before your latest changes. I had no problems in rev47. I suspect something got broken around rev58 when VM9 did some changes to the vorbis code.

Tested on an AMD Athlon (no SSE support). Problem remains after disabling all SIMD sets. So perhaps a bug in the generic code?
I tested again on P4HT and Athlon64x2. Both of the problems are not reproducible. My build use SSE, so it may not work on old Athlon (as for tremor) .

As for libavcodec, could you upload a sample?

[clsid]

I am testing previous revisions of libavcodec now. Hopefully I can pinpoint when the problem was introduced.

The problem occurs with every ogg file that I try. Here is a small sample that you may already have: http://akapuma.info/test/aka.ogg

[haruhiko_yamagata]

haru-han something's wrong with your build, it seems only the ax libavcodec and libmplayer is ok, the other dlls are borked. They don't show up in About->Version details here. BTW libavcodec vorbis works fine now with 2ch audio (I deleted the 5.1ch test file so can't test) don't know what went wrong with clsid though.
I can't reproduce your problem on both of my PCs. You probably have SSE, I have no idea.

//EDIT
Excuse me, 0820 build included debug build files. I updated my build (http://sourceforge.net/project/showfiles.php?group_id=173941).

[thuan]

My CPU is a T-bread SSE and 3DNow. BTW libsamplerate and libmpeg2's also fine on my computer. Here the two pictures of the version details box.

http://img219.imageshack.us/img219/8698/videogp2.th.png (http://img219.imageshack.us/my.php?image=videogp2.png)
http://img155.imageshack.us/img155/7691/audiojk1.th.png (http://img155.imageshack.us/my.php?image=audiojk1.png)

There's also one more thing I'd like to report. I make a codec pack myself for personal and my friend use and I use ffdshow as the core decoder. With ffdshow revision after audx support inclued it seems that if either one of ff_libmad.dll and audxlib.dll must be present for user to chose a mp1/2/3 decoder. If there's none of them in the ffdshow folder then in the ffdshow audio decoder settings you still can select mp3lib as decoder but after apply and exit the window the mp1/2/3 decoder go back to disabled.

[clsid]

Problem is fixed in rev82 :)

download rev82 build (http://prdownloads.sourceforge.net/ffdshow-tryout/ffdshow_rev2546-82_20060820_mbaff.exe?download)

[jffulcrum]

Problem with sluggish MPEG-1 palayback still present, no matter of build supplier. But they not present in case of MPEG-encoded video in .avi package. May be out of sync between ffdshow audio & video decoders ?

[MacAddict]

Rev82 build has broken the resize again it seems.

[thuan]

Haru-han your new r82 build seems to be perfect. MacAddict what's wrong with the resize filter? Can't find any here.

[MacAddict]

Haru-han your new r82 build seems to be perfect. MacAddict what's wrong with the resize filter? Can't find any here.
Good question, I uninstalled Rev80 to try 82 again. This time both Rev82 builds worked fine with resizing. My first try yesterday was with clsid's build and gave me a green overlay on the movie if 'resize' was selcted within ffdshow. So I'm not sure what happened but now both builds are fine:D

Edit-
clsid's build is indeed giving me the green overlay on 2 or 3 clips here but haruhiko_yamagata's isn't displaying the problem with the same clips.

[_xxl]

Updates:
ffmpeg rev 82
mplayer rev 81
http://rapidshare.de/files/30212873/ffdshow-20060821-rev2546-XXL.exe.html
autodetecting best optimizations:
http://rapidshare.de/files/30259330/FFdshow-20060821-rev2546.exe.html

[Isochroma]

clsid: where did MakeAVIS go? It was in previous builds and is necessary to make AVIS virtual AVI files. VFAPI cannot be used with fft3d.

haruhiko_yamagata: Your builds include MakeAVIS, but it doesn't work. First error is can't find msvcr80.dll, then when the file is copied into System32, the new error is:

"The application failed to initialize properly (0x800000003). Click OK to terminate the application."

drevil_xxl: You win the prize! Your excellent autodetecting builds include MakeAVIS and it works fine. Thanks for allowing me to continue my work with this excellent feature.

But wait... in the vfw encoder window... where did Windows Media 7/8 encoding go? I see the buggy WMV7/8 ffmpeg encoders, but the interface to the original Windows Media encoders is gone! WMV8 happens to be the fastest decode for HD video of all of them (by my tests), and is just fast enough for my system. Hopefully this can be re-added? Thanks!

Oops, VFW decoder has dual entries for WMV3/9 in the Codecs list.
Also, VFW encoder has 8 'W' items in Encoder dropdown list.

[Rash]

Drevil, what is the difference between your two builds? The Autodeteccting features only?

Thank you very much. :)

[_xxl]

Oops, VFW decoder has dual entries for WMV3/9 in the Codecs list.
Also, VFW encoder has 8 'W' items in Encoder dropdown list.
Ccodecs.cpp
formats.push_back(Tformat(_l("WMV3/9") ,IDFF_wmv3,IDFF_MOVIE_LAVC,_l("incomplete")));
formats.push_back(Tformat(_l("VC-1") ,IDFF_wvc1,IDFF_MOVIE_LAVC,_l("incomplete")));
if (filterMode&IDFF_FILTERMODE_VFW)
formats.push_back(Tformat(_l("WMV3/9") ,IDFF_wmv3,IDFF_MOVIE_WMV9,_l("WMV3, WMVA, MSS2")));
Fix:
static const int movies_wmv3[]={IDFF_MOVIE_LAVC,IDFF_MOVIE_WMV9,0};
formats.push_back(Tformat(_l("WMV3/9") ,IDFF_wmv3,movies_wmv3,_l("incomplete")));
formats.push_back(Tformat(_l("VC-1") ,IDFF_wvc1,IDFF_MOVIE_LAVC,_l("incomplete")));
if (filterMode&IDFF_FILTERMODE_VFW)
//formats.push_back(Tformat(_l("WMV3/9") ,IDFF_wmv3,IDFF_MOVIE_WMV9,_l("WMV3, WMVA, MSS2")));
The same problem can be found in ffdshow_rev2546-82_20060820:
http://i8.tinypic.com/25ezbea.jpg

[_xxl]

Generic:
http://rapidshare.de/files/30442144/ffdshow-20060823-rev2546.exe.html

Oops, VFW decoder has dual entries for WMV3/9 in the Codecs list.
Also, VFW encoder has 8 'W' items in Encoder dropdown list.
Please test the latest build.

[_xxl]

Drevil, what is the difference between your two builds? The Autodetecting features only?
Thank you very much. :)
1).In autodetect are 3 different versions of libavcodec.dll and libmplayer.dll: generic, sse and sse2.
2).Customized Inno Setup script.

[haruhiko_yamagata]

haruhiko_yamagata: Your builds include MakeAVIS, but it doesn't work. First error is can't find msvcr80.dll, then when the file is copied into System32, the new error is:

"The application failed to initialize properly (0x800000003). Click OK to terminate the application."
It was build by VS2005.
Rev 84 : makeAVIS_2005.vcproj fix : link static libray
Now VS2005 can build makeAVIS.exe properly. I hope this change does not have strange side effect. Thank you.

[haruhiko_yamagata]

Ccodecs.cpp

formats.push_back(Tformat(_l("WMV3/9") ,IDFF_wmv3,IDFF_MOVIE_LAVC,_l("incomplete")));
formats.push_back(Tformat(_l("VC-1") ,IDFF_wvc1,IDFF_MOVIE_LAVC,_l("incomplete")));
if (filterMode&IDFF_FILTERMODE_VFW)
formats.push_back(Tformat(_l("WMV3/9") ,IDFF_wmv3,IDFF_MOVIE_WMV9,_l("WMV3, WMVA, MSS2")));
Fix:

static const int movies_wmv3[]={IDFF_MOVIE_LAVC,IDFF_MOVIE_WMV9,0};
formats.push_back(Tformat(_l("WMV3/9") ,IDFF_wmv3,movies_wmv3,_l("incomplete")));
formats.push_back(Tformat(_l("VC-1") ,IDFF_wvc1,IDFF_MOVIE_LAVC,_l("incomplete")));
if (filterMode&IDFF_FILTERMODE_VFW)
//formats.push_back(Tformat(_l("WMV3/9") ,IDFF_wmv3,IDFF_MOVIE_WMV9,_l("WMV3, WMVA, MSS2")));
The same problem can be found in ffdshow_rev2546-82_20060820:
Thank you for the fix. I commented out "if (filterMode&IDFF_FILTERMODE_VFW)" and committed.
btw you have the SVN write access.

//EDIT
Excuse me, I forgot to write "patch by drevil_xxl" on SVN log. I'm sorry.

[clsid]

rev85 build (http://www.ffdshow.info)

* Updated installer. Now also contains all app plugins.
* MBAFF support is now always enabled.
* ICL 9.1 build. Please test all ffdshow filters (pp, sharpen, etc.).

[_xxl]

rev85 build (http://www.ffdshow.info)
ICL 9.1 build. Please test all ffdshow filters (pp, sharpen, etc.).
ffdshow.ax: doesn't like any optimizations of ICL for certain files (/O2 /O3 /Ox all breaking certain x264 files).
test.mkv AVC1
http://rapidshare.de/files/30475548/test.mkv.html
windows media player 2 crashed!
http://i8.tinypic.com/25f5a1w.jpg
ffdshow.ax is compiled by icl9.

[Isochroma]

Wow, never expected the builds to come so soon! I will test yours, drevil_xxl!

[clsid]

I know about the crash issue. It only happens with files created by some old x264 revision. I don't think many encodes created with that revision are floating around. Just a few test samples.

There exist a very strange workaround to play the problematic samples without crashes. Just play some other file first that also uses ffdshow, then reuse the player instance to play the test sample.

[_xxl]

Autodetecting best optimizations:
http://rapidshare.de/files/30491317/FFdshow-20060823-rev2546.exe.html
Video samples:
ftp://mplayerhq.hu/MPlayer/samples/V-codecs/

[bob0r]

Question to all compilers.
Since the lightning struck here i am kinda out of the loop.
Since you are all so active, i need to know:

What version should i use to update ffdshow on x264.nl?
I need target the most possible number of users.
I am testing FFdshow-20060823-rev2546.exe

Does FFdshow-20060823-rev2546.exe have multicore support?
Because unlike coreavc, with the program timecode.exe, ffdshow is not using more then 50% of my cpu.

[igor1st]

What version should i use to update ffdshow on x264.nl?
I need target the most possible number of users.
Latest revisions of ffdshow-tryout are good (and it already contains all important patches). Single problem I know - after updating libavcodec (rev30) it become up to 30% slower when decoding ASP.

[clsid]

There is no multithreaded decoding in ffdshow afaik. The only thing that is made multithreaded is the resize filter in ffdshow. But you need to enable 'Queue output samples' in the options first.


Revision 85 - Generic build (http://prdownloads.sourceforge.net/ffdshow-tryout/ffdshow_rev2546-85_20060823.exe?download)
* Custom InnoSetup installer
* Contains both non-unicode and unicode versions
* H.264 MBAFF support is enabled

[haruhiko_yamagata]

Hello, bob0r. Nice to see you again.

There is no multithreaded decoding in ffdshow afaik. The only thing that is made multithreaded is the resize filter in ffdshow. But you need to enable 'Queue output samples' in the options first.
Resize always uses multithread if you have dual core or multipule CPU regardless of "Queue output samples".
"Queue output samples" is an accelerator rather than multithreading, but I belive it is effective on P4HT or dual core.

Another known bug : mpeg2 is sluggish after rev 30. I'm trying to fix this.

bob0r, if you are distributing for end users, I recommend rev 13. Patch rev 37 should be added to rev 13.
Updating libavcodec is important work, it is being debuged and needs more testing for release.

[Egh]

LOL, it seems I've just discovered quite a FAIL in ffdshow GUI :P

In the "Output" page, DV checkbox can be checked but can't be unchecked. Thus if you checked
you screw everything up, and can't remove it :)

Luckily I found out "reset" button on same page to revert settings to default. But I guess better solution would be possibility to uncheck that box :)

P.S. Last build from clsid has been used for testing.

[thuan]

Nah it's in there long ago, and it's default ffdshow behavior seems weird though. Just try to check one of the colorspace above and DV is unchecked but better use the reset button.

[pandy]

Hello, bob0r. Nice to see you again.

Another known bug : mpeg2 is sluggish after rev 30. I'm trying to fix this.


haruhiko-san probably not only MPEG-2 is "sluggish", MPEG-1 to...

[igor1st]

Another known bug : mpeg2 is sluggish after rev 30. I'm trying to fix this.
libmpeg2 is always the better decoder for mpeg1/2 in ffdshow (and it's faster). So it's not big problem.

bob0r, if you are distributing for end users, I recommend rev 13. Patch rev 37 should be added to rev 13.
Updating libavcodec is important work, it is being debuged and needs more testing for release.
I think adding patches from revs 38, 64, 70 also has sense in that case.

[_xxl]

libmpeg2 is always the better decoder for mpeg1/2 in ffdshow (and it's faster).
Autodetecting best optimizations:
http://rapidshare.de/files/30567750/FFdshow-20060823-rev2546.exe.html
Libmpeg2 is default MPEG1/2 Decoder.

[clsid]

I compared a few files with the latest ffmpeg revision and there are plenty of differences. So updating to latest ffmpeg code might solve a few problems. I only don't know if there are intentional differences in the ffdshow code. If there are any, then I think it is a good idea to document them in a readme file. That would make future updating of the code a whole lot easier.

[haruhiko_yamagata]

I compared a few files with the latest ffmpeg revision and there are plenty of differences. So updating to latest ffmpeg code might solve a few problems. I only don't know if there are intentional differences in the ffdshow code. If there are any, then I think it is a good idea to document them in a readme file. That would make future updating of the code a whole lot easier.
Yes, it's a good idea. There are intentional differences. Though it is sometimes hard to understand why such changes are needed, ffdshow doesn't work without them.

[Isochroma]

drevil_xxl: just tested 20060823, looks great! Excellent work I must say. WMV7/8 encoding works fine, extra WMV3 entry in the DS decoder is gone, and makeAVIS is present & working (note: previous versions of ffdshow put a link to it in the Start menu, this one doesn't).

[bob0r]

I'll just patiently await a more stable build.

[clsid]

@drevil_xxl, I don't know if it still applies to your current script, but here are some pointers based on the last one that you posted:
- Don't use uninsclearvalue. Use uninsdeletevalue instead. Empty values in the driver32 regkey are not good.
- You don't need to delete/clear each subvalue in the ffdshow registry keys separately. Just use uninsdeletekey on keys such as HKCU\Software\GNU\ffdshow.
- You don't need "MinVersion: 0,4.0;" on all items. You already enforce that globally in the [Setup] section.


rev87 build (http://prdownloads.sourceforge.net/ffdshow-tryout/ffdshow_rev2546-87_20060825.exe?download)

AviSynth plugin now automatically gets placed in AviSynth plugin. Added some more a/v formats to the tasks page on the installer. More can be added on request.

[akupenguin]

libmpeg2 is always the better decoder for mpeg1/2 in ffdshow (and it's faster). So it's not big problem.
Anyone want to put together a list of cpus where that's true? (assuming it is cpu-dependent)
Because I have heard several times on mplayer-dev-eng that libmpeg2 is faster, but ffmpeg12 always beats it by a fair margin on my computer (athlon64 3400 in 64bit mode, linux, mplayer).

[jffulcrum]

haruhiko_yamagata
mpeg2 is sluggish after rev 30. I'm trying to fix this
Seems that problem with MPEG1&2 is in FFDShow filter/parser, not in libavcodec themselves, because with video in AVI or MKV container all work normal

[_xxl]

@drevil_xxl, I don't know if it still applies to your current script, but here are some pointers based on the last one that you posted:
- Don't use uninsclearvalue. Use uninsdeletevalue instead. Empty values in the driver32 regkey are not good.
- You don't need to delete/clear each subvalue in the ffdshow registry keys separately. Just use uninsdeletekey on keys such as HKCU\Software\GNU\ffdshow.
- You don't need "MinVersion: 0,4.0;" on all items. You already enforce that globally in the [Setup] section.
Done. Thanks for help!

[haruhiko_yamagata]

I only don't know if there are intentional differences in the ffdshow code. If there are any, then I think it is a good idea to document them in a readme file. That would make future updating of the code a whole lot easier.
The slugish mpeg2/libavcodec is fixed by rev 89. And that was exactly "intentional differences in the ffdshow code".

@pandy : Perhaps mpeg1 is fixed too, though I don't have any sample of mpeg1 that shows the problem.

[igor1st]

Anyone want to put together a list of cpus where that's true? (assuming it is cpu-dependent)
Because I have heard several times on mplayer-dev-eng that libmpeg2 is faster, but ffmpeg12 always beats it by a fair margin on my computer (athlon64 3400 in 64bit mode, linux, mplayer).
I tested only inside ffdshow on my old AXP 2.2GHz with WinXP.
my 2543 build: 208fps vs 183fps
my tryout build: 208fps vs 120fps (so lavc update influenced not only for ASP decoding speed)

[jffulcrum]

haruhiko_yamagata
Thank you, MPEG-1 playback via libavcodec now good.

[Isochroma]

1920x1080i .ts decode using libmpeg2 is causing mpc to instant-crash (disappear) when seeking, using haali's renderer (2006-07-07). With haali's renderer, libavcodec works fine, but decodes at about 1/2 the speed (unplayable). Using MPC's internal .ts splitter, seeks work using libmpeg2, but first 1/2 second has weird green blocks.

Playback using Elecard's MPEG-2 decoder (reference) works perfectly, but it only outputs YUY2, so is unusable with the haali renderer (see my last post in the haali render thread for why).

[clsid]

I can confirm the green blocks/artifacts when using Gabest's splitter. Happens only on one of my MPEG-2 files. Using Overlay Mixer as renderer.

Here is the file that has the block issue:
http://www.nextcomwireless.com/r5000/download/hbo_trailer_6ksd.zip

Other that that I haven't had any problems playing MPEG files with the latest revision. Performance is the same as rev2543.

[haruhiko_yamagata]

I can confirm the green blocks/artifacts when using Gabest's splitter. Happens only on one of my MPEG-2 files. Using Overlay Mixer as renderer.

Here is the file that has the block issue:
http://www.nextcomwireless.com/r5000/download/hbo_trailer_6ksd.zip

Other that that I haven't had any problems playing MPEG files with the latest revision. Performance is the same as rev2543.
Your sample shows a problem with hard ware deinterlacing. I don't see green blocks on my hard ware, but it's slugish. Software deinterlacing(ffdshow) gives the best result, followed by HD deinterlacing by old ffdshow.

//EDIT
Now, I can see the green block when libmpeg2 is selected. So both of the decoders are broken.
libmpeg2 seems to have built in deinterlace.

[toby77jo]

dvd navigation in mpc when using dscaler5 video decoder in all of your builds is broken, it's working fine in ffdshow rev. 2543. is there a fix for this?

[KoD]

Am I correct in assuming that libavcodec cannot still be used for decoding mpeg2 when playing a DVD ? At least, it doesn't seem to work for me (plays a batch of frames, then it stops then plays another batch and so on) I've tseted this only with clsid's rev81 build on my old P4 2.4Ghz, no HT, SSE2.

[thuan]

Same with clsid's r.91.

[KoD]

I wanted to say r91 myself, but I misstyped it as r81 ^^;

[igor1st]

"Queue output samples" is an accelerator rather than multithreading, but I belive it is effective on P4HT or dual core.

I tested "queue output samples" on my single-core CPU today (Athlon-XP). Tested with almost full CPU loading (my special AVC file with resize). Using queue is definitely bringing benefit over non-queue output in such situation.

[haruhiko_yamagata]

Thank you, igor1st.

Rev 95:
Bug fix : hardware deinterlace : mpeg12.c based on 5578/ffmpeg

Known bug : slow decoding of MPEG4-ASP.
Are there any other bugs related to libavcodec?

[jffulcrum]

igor1st
Tested with almost full CPU loading (my special AVC file with resize). Using queue is definitely bringing benefit over non-queue output in such situation.
But on latest P3`s , and old P4 CPU`s with Willamete core enabling this option in case of 720 AVC/HQ ASP playback (almost 95+% CPU Usage) bail-out any player, even simplest mplayer2.exe . So is not good idea to always enable this trick.

[igor1st]

Known bug : slow decoding of MPEG4-ASP.
Are there any other bugs related to libavcodec?
Not only ASP but mpeg2 and maybe mpeg1 too:
http://forum.doom9.org/showthread.php?p=867826#post867826

[LoRd_MuldeR]

Today I wanted to play an old H.264 video and I noticed that with the current ffdshow builds it doesn't play smooth. I get full CPU load and a lot of dropped frames. The video is only 640x480 at 25fps, so it should play fine on my machine. I got hundreds of similar files, that all play 100%. When I play the same file in MPlayer or VLC there's no problem. So it's an ffdshow problem. Furthermore I remember that I played the file a while ago and there was no problem; no idea what ffdshow version I used at that time...

Here is the sample file:
http://uploaded.to/?id=bd326d (PW is "Doom9")

Note: It's from an old video tape, so don't care about the visual quality ^^

[foxyshadis]

Mulder, are all filters off?

[LoRd_MuldeR]

Mulder, are all filters off?

I use LanczosResize to 1920 × 1080 plus High Qulaity Denoise :p
Nope, they are all off ;)

Could you please test my video and check for unexpected high CPU usage?

[foxyshadis]

You set a password on it. >.>

I can say that with my personal build I have the exact same decode times on several very-high-bitrate test files I made as in versions over the past 4-5 months (I use them to track any improvements to AVC), as well and normal cpu usage in normal movies, so maybe yours just have optimizations turned off somehow. (Or were built in msvc only?)

(I should have hit post before I went to sleep.)

[LoRd_MuldeR]

You set a password on it. >.>

Sorry, I forgot :o PW is "Doom9"

I can say that with my personal build I have the exact same decode times on several very-high-bitrate test files I made as in versions over the past 4-5 months (I use them to track any improvements to AVC), as well and normal cpu usage in normal movies, so maybe yours just have optimizations turned off somehow. (Or were built in msvc only?)

(I should have hit post before I went to sleep.)

It's only that one specific file that makes trouble...

[pandy]

i dont know that anybody use a motion vector visualization but this functionality seems to be lost from few last ffdshow builds...
(quantizers and graph working ok)

[haruhiko_yamagata]

i dont know that anybody use a motion vector visualization but this functionality seems to be lost from few last ffdshow builds...
(quantizers and graph working ok)
It's working for me. What kind of decoder is in trouble?

[_xxl]

Applied patches:
accuracy.diff,aspect.diff,dts.patch,faad2.patch
ffdshow_mbaff.patch,ogg-5.1ch.diff,tsampleformat.patch
indicate we are using high accuracy tremor in gui,
fixed volume normalization by Kurosu,
faad 2.6 beta,xsharpen + swscaler = green problem.It was a bug caused xsharpen that did not call EMMS
by h_yamagata,libsamplerate 0.1.2.
http://prdownloads.sourceforge.net/ffdshow-tryout/FFdshow-20060828-rev2543.exe?download

[pandy]

It's working for me. What kind of decoder is in trouble?

Ver. from Aug 23 2006 13:32:09 (icl 9, x86, unicode)...

[_xxl]

Applied patches:
accuracy.diff,aspect.diff,dts.patch,faad2.patch
ffdshow_mbaff.patch,ogg-5.1ch.diff,tsampleformat.patch
indicate we are using high accuracy tremor in gui,
fixed volume normalization by Kurosu,better management of buffers of the queue,faad 2.6 beta,xsharpen + swscaler = green problem.It was a bug caused xsharpen that did not call EMMS by h_yamagata,libsamplerate 0.1.2.
http://prdownloads.sourceforge.net/ffdshow-tryout/FFdshow-20060828-rev2546.exe?download

[Egh]

Ver. from Aug 23 2006 13:32:09 (icl 9, x86, unicode)...

not all decoders in ffdshow support motion vector visualization.

iirc if you enable decoding thru xvid, you won't see any.

Just tested on last clsid build: motion vectors are properly shown from xvid and divx content decoded by libavc.

[_xxl]

http://prdownloads.sourceforge.net/ffdshow-tryout/FFdshow-Tryouts-20060828-rev100.exe?download

[jffulcrum]

Something wrong with Sorenson SVQ decoder in libavcodec since rev. 91 of ffdshow-tryout:

Application Failure zplayer.exe 4.5.1.0 in libavcodec.dll 0.0.0.0 at offset 000bafba

Same error with MPC. Tried different builds.

[haruhiko_yamagata]

Something wrong with Sorenson SVQ decoder in libavcodec since rev. 91 of ffdshow-tryout:

Application Failure zplayer.exe 4.5.1.0 in libavcodec.dll 0.0.0.0 at offset 000bafba

Same error with MPC. Tried different builds.
Could you upload a sample?

[foxyshadis]

http://samples.mplayerhq.hu/V-codecs/SVQ3/

Great resource for many codec samples. Since svq3 hasn't changed in some time, in either ffmpeg or ffdshow, it seems the bitstream must have changed... unless it is a r30 problem, not r91.

[pandy]

not all decoders in ffdshow support motion vector visualization.

iirc if you enable decoding thru xvid, you won't see any.

Just tested on last clsid build: motion vectors are properly shown from xvid and divx content decoded by libavc.

Yes i know - but for MPEG-2 they should be visible (hm... in the past i use libavcodec but at this moment decode mpeg-2 and mpeg-1 by the libmpeg - maybe this is a problem that motion vector is not showed at all...?)

[haruhiko_yamagata]

Yes i know - but for MPEG-2 they should be visible (hm... in the past i use libavcodec but at this moment decode mpeg-2 and mpeg-1 by the libmpeg - maybe this is a problem that motion vector is not showed at all...?)
You are right, libmpeg2 doesn't show arrows.
clsid's installer overwrite the setting of mpeg1&2 decoder.

[jffulcrum]

haruhiko_yamagata
http://64.207.133.64/upload/Director%20File/Jonas%20Akerlund/high%20Res./madonna_BbandHi.mov (27 mb)

[haruhiko_yamagata]

haruhiko_yamagata
http://64.207.133.64/upload/Director%20File/Jonas%20Akerlund/high%20Res./madonna_BbandHi.mov (27 mb)
Thank you, but I can play your file on my PC. I can't reproduce the problem.
Is the problem specific to zoom player? I use MPC.

[jffulcrum]

haruhiko_yamagata
Same error with MPC

But now i found source of problem. Damn Haali Media Splitter. It`s time to connect them with Norton Wipeinfo :) . With Gabest and Nero splitters all fine. Interesting, that ffdshow had time to decode about 20 frames before player crashed

[_xxl]

SSE:
http://prdownloads.sourceforge.net/ffdshow-tryout/ffdshow-tryouts-rev100-sse.exe?download
SSE2:
http://prdownloads.sourceforge.net/ffdshow-tryout/ffdshow-tryouts-rev100-sse2.exe?download
All files are by GCC 4.0.3.

[haruhiko_yamagata]

@foxyshadis
I can't compile rev 102 by GCC 4.0.3.
gcc -c -I. -I.. -Ilibavcodec -Ilibavutil -I../codecs -I../imgFilters -I../zlib - DHAVE_AV_CONFIG_H -std=gnu99 -DALLOW_INTERLACE -mno-cygwin -mdll -mthreads -pipe -DNDEBUG -UDEBUG -DWIN32 -D_WIN32 -O2 -march=i686 -mtune=i686 -fomit-frame-poi nter -finline-functions -finline -MMD -o libavcodec/i386/dsputil_mmx.o libavcode c/i386/dsputil_mmx.c
In file included from libavcodec/i386/dsputil_mmx.c:150:
libavcodec/i386/dsputil_mmx_rnd.h: In function 'avg_no_rnd_pixels4_mmx':
libavcodec/i386/dsputil_mmx_rnd.h:302: error: syntax error before 'ASMALIGN'
libavcodec/i386/dsputil_mmx_rnd.h: In function 'avg_no_rnd_pixels8_mmx':
libavcodec/i386/dsputil_mmx_rnd.h:322: error: syntax error before 'ASMALIGN'
libavcodec/i386/dsputil_mmx_rnd.h: In function 'avg_no_rnd_pixels16_mmx':
libavcodec/i386/dsputil_mmx_rnd.h:341: error: syntax error before 'ASMALIGN'
libavcodec/i386/dsputil_mmx_rnd.h: In function 'avg_no_rnd_pixels8_x2_mmx':
libavcodec/i386/dsputil_mmx_rnd.h:364: error: syntax error before 'ASMALIGN'
libavcodec/i386/dsputil_mmx_rnd.h: In function 'avg_no_rnd_pixels8_l2_mmx':
libavcodec/i386/dsputil_mmx_rnd.h:384: error: syntax error before 'ASMALIGN'
libavcodec/i386/dsputil_mmx_rnd.h: In function 'avg_no_rnd_pixels16_x2_mmx':
libavcodec/i386/dsputil_mmx_rnd.h:405: error: syntax error before 'ASMALIGN'
libavcodec/i386/dsputil_mmx_rnd.h: In function 'avg_no_rnd_pixels16_l2_mmx':
libavcodec/i386/dsputil_mmx_rnd.h:431: error: syntax error before 'ASMALIGN'
In file included from libavcodec/i386/dsputil_mmx.c:164:
libavcodec/i386/dsputil_mmx_rnd.h: In function 'avg_pixels4_mmx':
libavcodec/i386/dsputil_mmx_rnd.h:302: error: syntax error before 'ASMALIGN'
libavcodec/i386/dsputil_mmx_rnd.h: In function 'avg_pixels8_mmx':
libavcodec/i386/dsputil_mmx_rnd.h:322: error: syntax error before 'ASMALIGN'
libavcodec/i386/dsputil_mmx_rnd.h: In function 'avg_pixels16_mmx':
libavcodec/i386/dsputil_mmx_rnd.h:341: error: syntax error before 'ASMALIGN'
libavcodec/i386/dsputil_mmx_rnd.h: In function 'avg_pixels8_x2_mmx':
libavcodec/i386/dsputil_mmx_rnd.h:364: error: syntax error before 'ASMALIGN'
libavcodec/i386/dsputil_mmx_rnd.h: In function 'avg_pixels8_l2_mmx':
libavcodec/i386/dsputil_mmx_rnd.h:384: error: syntax error before 'ASMALIGN'
libavcodec/i386/dsputil_mmx_rnd.h: In function 'avg_pixels16_x2_mmx':
libavcodec/i386/dsputil_mmx_rnd.h:405: error: syntax error before 'ASMALIGN'
libavcodec/i386/dsputil_mmx_rnd.h: In function 'avg_pixels16_l2_mmx':
libavcodec/i386/dsputil_mmx_rnd.h:431: error: syntax error before 'ASMALIGN'
libavcodec/i386/dsputil_mmx.c: In function 'get_pixels_mmx':
libavcodec/i386/dsputil_mmx.c:210: error: syntax error before 'ASMALIGN'
libavcodec/i386/dsputil_mmx.c: In function 'diff_pixels_mmx':
libavcodec/i386/dsputil_mmx.c:238: error: syntax error before 'ASMALIGN'
libavcodec/i386/dsputil_mmx.c: In function 'put_pixels4_mmx':
libavcodec/i386/dsputil_mmx.c:381: error: syntax error before 'ASMALIGN'
libavcodec/i386/dsputil_mmx.c: In function 'put_pixels8_mmx':
libavcodec/i386/dsputil_mmx.c:407: error: syntax error before 'ASMALIGN'
libavcodec/i386/dsputil_mmx.c: In function 'put_pixels16_mmx':
libavcodec/i386/dsputil_mmx.c:433: error: syntax error before 'ASMALIGN'
libavcodec/i386/dsputil_mmx.c: In function 'vector_fmul_add_add_3dnow':
libavcodec/i386/dsputil_mmx.c:2898: warning: implicit declaration of function 'f f_vector_fmul_add_add_c'
libavcodec/i386/dsputil_mmx.c: In function 'dsputil_init_mmx':
libavcodec/i386/dsputil_mmx.c:3418: error: 'struct DSPContext' has no member nam ed 'vector_fmul'
libavcodec/i386/dsputil_mmx.c:3420: error: 'struct DSPContext' has no member nam ed 'float_to_int16'
libavcodec/i386/dsputil_mmx.c:3423: error: 'struct DSPContext' has no member nam ed 'vector_fmul_reverse'
libavcodec/i386/dsputil_mmx.c:3426: error: 'struct DSPContext' has no member nam ed 'vector_fmul'
libavcodec/i386/dsputil_mmx.c:3427: error: 'struct DSPContext' has no member nam ed 'float_to_int16'
libavcodec/i386/dsputil_mmx.c:3428: error: 'struct DSPContext' has no member nam ed 'vector_fmul_reverse'
libavcodec/i386/dsputil_mmx.c:3429: error: 'struct DSPContext' has no member nam ed 'vector_fmul_add_add'
libavcodec/i386/dsputil_mmx.c:3432: error: 'struct DSPContext' has no member nam ed 'vector_fmul_add_add'
make: *** [libavcodec/i386/dsputil_mmx.o] Error 1

[haruhiko_yamagata]

haruhiko_yamagata


But now i found source of problem. Damn Haali Media Splitter. It`s time to connect them with Norton Wipeinfo :) . With Gabest and Nero splitters all fine. Interesting, that ffdshow had time to decode about 20 frames before player crashed
Confirmed. When MPC's internal filter-"MP4/MOV" turned off, ffdshow-20060730-Q.exe or ffdshow-20051115.exe does not crash, but the picture is blocky and there's no sound. So the culplit is *likely* the parser filter.

[foxyshadis]

Sorry, it compiled with 3.4.x so I assumed it would elsewhere. I reimported a little more carefully and I'm pretty sure I fixed it now, if you could recompile that'd be great.

[clsid]

rev103 build (http://prdownloads.sourceforge.net/ffdshow-tryout/ffdshow_rev2546-103_20060830.exe?download)

[haruhiko_yamagata]

Rev 107

Minor fix : _mm_empty() instead of emms
Minor fix : Use of #if __STDC_VERSION >= 199901L
Bug fix : slow decoder(MPEG4-ASP,etc) : idct_algo
Bug fix : libavcodec : Ignored "Info & debug" CPU setting

Now known bugs of libavcodec decoder have been cleaned.

[clsid]

Nice work Haruhiko!

rev107 build (http://prdownloads.sourceforge.net/ffdshow-tryout/ffdshow_rev2546-107_20060830.exe?download)

[Egh]

Nice work Haruhiko!

rev107 build (http://prdownloads.sourceforge.net/ffdshow-tryout/ffdshow_rev2546-107_20060830.exe?download)

Your builds are now the best imo.

But is it bug or feature? :P Your installer always puts "Use overlay mixer" and "Allow format ..." into intermediate state, regardless of the previous setting (one is off and one is checked in my case).

Also, just noticed: your installer at least last build now specifically sets libmpeg2 as IDCT (regardless of any previous setting as well). Is it better than auto or Xvid MMX? iirc some time ago XVid idct and Skal's one were considered the best.

[devaster]

at weekend (maybe) I release a ffdshow with hw idct on gpu
please check : http://sourceforge.net/projects/x264gpu/

[Egh]

at weekend (maybe) I release a ffdshow with hw idct on gpu
please check : http://sourceforge.net/projects/x264gpu/

I long to try this :) But I'm quite sure it will be slower on older cards compared to pure software? Or does it give better speed difference compared to bicubic resizing?

[devaster]

i use a autodetect because im using a native cg not a dxva...
100% running on nvidia 6600 ...
for now change only and add some dlls...

[videomixer9]

Hm, sounds interesting but I wondered a bit about what I read on the CoreAVC thread with the nvidia api needed SSE2, is that true?

Other than that I did my own build again after returning, with old nsis installer:
http://prdownloads.sourceforge.net/ffdshow-tryout/ffdshow-tryouts-sse-rev110.exe?download

[devaster]

no the CG compiler is not in nvidia api you must downolad full redistributable (http://developer.nvidia.com/object/cg-redistributable-binaries.html)
and a copy a dir for your system from this package . Iwould pack it with ...
and i thing it is for all cpus (in manual isnt restrictions )

[clsid]

But is it bug or feature? :P Your installer always puts "Use overlay mixer" and "Allow format ..." into intermediate state, regardless of the previous setting (one is off and one is checked in my case).

Also, just noticed: your installer at least last build now specifically sets libmpeg2 as IDCT (regardless of any previous setting as well). Is it better than auto or Xvid MMX? iirc some time ago XVid idct and Skal's one were considered the best.I will update the script to respect the previous settings for those options. Afaik these settings give the highest compatibility with the various players. I once read somewhere that the libmpeg2 IDCT is most compatible. Does anyone have some more info on the difference and pro/cons of the IDCTs?

[Egh]

I will update the script to respect the previous settings for those options. Afaik these settings give the highest compatibility with the various players. I once read somewhere that the libmpeg2 IDCT is most compatible. Does anyone have some more info on the difference and pro/cons of the IDCTs?

Hmmm... Wonder how can idct be compatible with anything :)
Usually they differ in speed/quality.

As for quality, i dont' see difference between xvid and reference one. For descriptions of various types of IDCT (including couple not present in ffdshow) refer to dgdecode manual.

I think main question here is speed, not quality. If someone is able to run some tests and see if there's any difference, that would be handy.

[clsid]

DGDecoe Manual APPENDIX B: iDCT Algorithm Notes "The video information inside MPEG files is stored in the frequency domain rather than in the spatial domain (the images we see). That way, the information gets compacted and that compaction can be used to compress (reduce) the amount of information you have to send over the transmission channel. MPEG uses the DCT (Discrete Cosine Transform) to translate spatial information into frequency information. To bring back the spatial information from the MPEG stream you have to apply the iDCT, that is, the Inverse Discrete Cosine Transform, that undoes the DCT that was used during encoding."

"Although MPEG is almost deterministic (given a MPEG stream the output should be identical in all decoders), the standard has a degree of freedom when choosing the iDCT to use. That way, the decoder can be more easily implemented depending on the hardware below it. What the standard requires from the decoder is that the iDCT meets IEEE-1180 specs, or in plain words, that the error from the iDCT doesn't go beyond that the ones pointed out in the IEEE-1180."

Which iDCT you should use depends primarily on what CPU you have and to a lesser degree, on how accurate an iDCT you desire. Most people will not be able to tell the difference in quality between these algorithms but they can be easily observed by combining the AviSynth filters Subtract() and Levels(). All of the available options are IEEE-1180 compliant, except for SSE/MMX (Skal).

Qualitywise: IEEE-1180 Reference > 64-bit Floating Point > Simple MMX (XviD) > Remaining iDCTs.

Speedwise: SSE2/MMX and SSE/MMX (Skal) are usually the fastest. The IEEE-1180 Reference is easily the slowest.
The question becomes how big are the quality/speed differences? If quality difference is insignificant, then the fastest should be used. I guess that is Skal or XviD MMX?

[KoD]

When the DCT coeffs used by the encoder do not match those used during iDCT by the decoder The MPEG1/2/4 and H.261/2/3 IDCT (http://guru.multimedia.cx/category/dct/).

One more plea for people to use h264 since the coefficients are standardized so all h264 decoders will give bit identical output.

[clsid]

rev117 build (http://prdownloads.sourceforge.net/ffdshow-tryout/ffdshow_rev2546-117_20060830.exe?download)

[_xxl]

sse:
http://prdownloads.sourceforge.net/ffdshow-tryout/ffdshow-tryouts-rev111-sse.exe?download
sse2:
http://prdownloads.sourceforge.net/ffdshow-tryout/ffdshow-tryouts-rev111-sse2.exe?download

[Egh]

DGDecoe Manual APPENDIX B: iDCT Algorithm Notes
The question becomes how big are the quality/speed differences? If quality difference is insignificant, then the fastest should be used. I guess that is Skal or XviD MMX?

Yeah, exactly my thought :)

The question which is still open, though, is what kind of idct is the one from libmpeg2?

As for the fastest one, I'll take a stab and say the one with most advanced CPU optimisation is the fastest :) Does any of those idcts in ffdshow have SSE/SSE2 implementation?

[KoD]

If it was encoded with xvid, then simply use the idct meant for xvid encodes.

If it was encoded with h264, then don't worry, those idct options are simply ignored and the standardized way of doing idct is used by the decoder.

If it's something else, god knows ! Consider that even the encoder might have not used a proper way of doing DCT, so even if you use the reference iDCT you may get strange artifacts. (which makes me wonder how does the "auto" iDCT option in ffdshow really works - it should choose an iDCT based on the detected encoder info used to make the video stream, but my guess is it simply chooses one of the available implementations based on what mmx/sse instructons sets the cpu has even if they're not appropriate for the video stream they're used on)

Edit later: I found out that milan wrote what those options mean, actually: Libavcodec options (http://ffdshow.sourceforge.net/tikiwiki/tiki-index.php?page=Miscellaneus+settings). It seems like this option selects a certain implementation to be used by libavcodec. The safest option (and best) should be "auto" (letting libavcodec select a certain idct to be used). It looks like for xvid this auto selection works quite well since xvid code is freely available. For other closed fomats (like divx), the above "god knows what they used" is true so auto selection might work well or not.

Some of the changes made in libavcodec regarding idct can be found here (http://svn.mplayerhq.hu/ffmpeg/trunk/libavcodec/h263dec.c?view=log).

Any changes in xvid regarding idct are here (http://cvs.xvid.org/cvs/chora/cvs.php/xvidcore/src/dct/idct.h) as well as in the specific extended instructions set folders in that directory.

[Egh]

If it was encoded with xvid, then simply use the idct meant for xvid encodes.



Actually as I can see XVid idct was done by Skal himself :) And that one also has SSE2 optimisation. But that code was commented out when it was added in 2005 :)

libmpeg2 could be slower, i guess. So XVid or Skal are top candidates right now. Can someone measure the speed difference, if any, in practice with ffdshow?

[igor1st]

libmpeg2 could be slower, i guess. So XVid or Skal are top candidates right now. Can someone measure the speed difference, if any, in practice with ffdshow?
I have video file (fourcc=divx, bitsream reports divx 4.0) which plays without color artifacts only with libmpeg2 or xvid mmx.

my speed test with ASP:
Athlon XP, libmpeg2 = 199.4 fps
Athlon XP, XviD MMX = 198.3 fps
Pentium-M, libmpeg = 176.7 fps
Pentium-M, XviD MMX = 178.5 fps

[KoD]

The difference in speed is so small that it doesn't really matter. And current xvid default idct method is the one used by libmpeg2 actually. As I've said, either auto or libmpeg2 should be used. It's not worth gaining 1fps when there are already more than 100 fps but in exchange get color artifacts, is it ?

[igor1st]

It's not worth gaining 1fps when there are already more than 100 fps but in exchange get color artifacts, is it ?
Exactly. That's why I choose libmpeg2 as default idct for my builds.

[videomixer9]

rev 119
http://prdownloads.sourceforge.net/ffdshow-tryout/ffdshow-tryouts-rev119.exe?download

imo quality on divx6 videos is slightly worse in ffdshow than with the divx decoder (and yes postprocessing is disabled in divx decoder)

[Bathrone]

Can advise that Vista builds 5536 and 5552 are working happily with build 107 of ffdshow. Will try 119 tonight.

[_xxl]

Take a look!
http://sourceforge.net/project/stats/detail.php?type=prdownload&group_id=53761&ugn=ffdshow&package_id=48261&release_id=95213&file_id=180315
http://sourceforge.net/project/stats/detail.php?type=prdownload&group_id=53761&ugn=ffdshow&package_id=59355&release_id=274595&file_id=600104
Many people download old versions.

[Peuj]

Take a look!
http://sourceforge.net/project/stats/detail.php?type=prdownload&group_id=53761&ugn=ffdshow&package_id=48261&release_id=95213&file_id=180315
http://sourceforge.net/project/stats/detail.php?type=prdownload&group_id=53761&ugn=ffdshow&package_id=59355&release_id=274595&file_id=600104
Many people download old versions.

I'm not surprised, it's still the official versions. ;)

[LoRd_MuldeR]

Take a look!
http://sourceforge.net/project/stats/detail.php?type=prdownload&group_id=53761&ugn=ffdshow&package_id=48261&release_id=95213&file_id=180315
http://sourceforge.net/project/stats/detail.php?type=prdownload&group_id=53761&ugn=ffdshow&package_id=59355&release_id=274595&file_id=600104
Many people download old versions.

Seems to me like the "official" SF site is dead:
http://sourceforge.net/project/showfiles.php?group_id=53761

Last release is October 12, 2004.
Latest "daily build" dates back to 2005 ;)

If they don't want to update their site (at the moment) they should at least support the people that still work on this project!
Why not mark the old builds as "historical" and link to the ffdshow-tryout site ?!

[Egh]

Seems to me like the "official" SF site is dead:

If they don't want to update their site (at the moment) they should at least support the people that still work on this project!
Why not mark the old builds as "historical" and link to the ffdshow-tryout site ?!

IIRC Milan long time ago offered to take control of the ffdshow homepage. I guess you can still get it if you ask him (though first you'd need to find him:P)

[LoRd_MuldeR]

though first you'd need to find him :P

hehe :D

[Egh]

rev121 is already out from VM9 :0 Very high speed indeed :)

http://prdownloads.sourceforge.net/ffdshow-tryout/ffdshow-tryouts-rev121.exe?download

[Bathrone]

Where is the changelog for all these releases?

[Eragon4ever]

http://svn.sourceforge.net/viewvc/ffdshow-tryout/?view=log

[Egh]

BTW, from the point of GUI design at least, options for h264 deblocking need to be moved out of Codecs page.

Since we have decoder's option tab, I think it would be better to move those checkboxes here (or better redone them).

[toby77jo]

http://svn.sourceforge.net/viewvc/ff...yout/?view=log

Revision 122 - Directory Listing
Modified Fri Sep 1 14:02:31 2006 UTC (11 minutes, 8 seconds ago) by h_yamagata

Bug fix of rev 121.

It's nice with a changelog, but where to find the download link to revision 122?

[clsid]

rev122 build (http://prdownloads.sourceforge.net/ffdshow-tryout/ffdshow_rev2546-122_20060901.exe?download)

[foxyshadis]

Changelogs are just internal code, until someone builds them and posts here about them there won't be a download link. :p But most of the time it only means a few hours of waiting until someone does so.

[videomixer9]

It'd be great if we could also nominate something as a stable release, I put up a new frontpage for the ffdshow tryouts page and don't redirect to the tryouts forums directly anymore, and I'd like to link there at least directly to a stable build people can use.

so the tryouts forums is reachable via http://forum.ffdshow.info/ and the new frontpage that's not the hottest of the hottest stuff but at least something easy at http://ffdshow-tryout.sourceforge.net/ ...

better contributions are always welcome.

[Eragon4ever]

Nice,
but could you place the revision number in the file name in the future?
And the link to the forum is not woking for me. I get a 404.

[videomixer9]

sometimes you may hate that the names are case sensitives. fixed the link. and the filename lacks the build number on purpose.

[LoRd_MuldeR]

and the filename lacks the build number on purpose.

This way you don't now what build you are downloading and it's bad to keep various ffdshow installers on your HDD. So please add build number to download. Thx.

[_xxl]

Autodetecting best optimizations:
http://prdownloads.sourceforge.net/ffdshow-tryout/FFdshow-Tryouts-20060901-rev122.exe?download
MMX:
http://prdownloads.sourceforge.net/ffdshow-tryout/ffdshow-tryouts-rev122-mmx.exe?download
SSE:
http://prdownloads.sourceforge.net/ffdshow-tryout/ffdshow-tryouts-rev122-sse.exe?download
SSE2:
http://prdownloads.sourceforge.net/ffdshow-tryout/ffdshow-tryouts-rev122-sse2.exe?download

[KoD]

Wasn't there a consensus a few weeks ago that the best thing is to let ffdshow decide at runtime what optimizations to use and not to force compiling for a specific set of extended instructions ?

[Egh]

Lol but it seems in the last build from clsid decoding options page is disabled :)

[_xxl]

Wasn't there a consensus a few weeks ago that the best thing is to let ffdshow decide at runtime what optimizations to use and not to force compiling for a specific set of extended instructions ?
Why not?
SSE & SSE2 versions are compiled by GCC 4.0.3.

[Kostarum Rex Persia]

What's going on with Video Mixer?

He doesn't make ffdshow builds anymore, or what?

[jffulcrum]

It`s tricky: VM9 build`s show usage of MMXEXT CPU instructions, but clsid build`s does not. Double tricky: Pentium 3 CPU not support MMXEXT, this is AMD feature, right ? :)

in the last build from clsid decoding options page is disabled
VM9 build too.

[foxyshadis]

What's going on with Video Mixer?

He doesn't make ffdshow builds anymore, or what?
What, yesterday's build isn't good enough?

[LoRd_MuldeR]

@drevil_xxl build:
Thanks for SSE build. Works fine here :D

[Zep]

Lol but it seems in the last build from clsid decoding options page is disabled :)


works for me

[_xxl]

Waiting for h264 GPU support!
http://www.bit-tech.net/news/2006/01/07/nvidia_decode_h264/
http://www.nvidia.com/page/purevideo_support.html

[Reino]

Does anyone know when DVD-support for MPEG-2 in FFDShow will be ready?

[KoD]

Does anyone know when DVD-support for MPEG-2 in FFDShow will be ready?

You can use ffdshow to decode mpeg2 video from dvds right now. ffdshow video decoder configuration -> Codecs -> mpeg2 set to libmpeg2 and check DVD decoding. Libavcodec is not usable presently. Furthermore, you can select Deinterlacing and use the "5-tap lowpass" method which seemed to me to give best results. And in ffdshow audio decoder configuration -> Codecs -> enable AC3 and DTS and eventually MP2 and LPCM (LPCM only in rev124).

[LoRd_MuldeR]

You can use ffdshow to decode mpeg2 video from dvds right now. ffdshow video decoder configuration -> Codecs -> mpeg2 set to libmpeg2 and check DVD decoding. Libavcodec is not usable presently. Furthermore, you can select Deinterlacing and use the "5-tap lowpass" method which seemed to me to give best results. And in ffdshow audio decoder configuration -> Codecs -> enable AC3 and DTS and eventually MP2 and LPCM (LPCM only in rev124).

Yes, but ffdshow will only do the audio/video decoding. You'll need additional filters to handle the DVD stuff, like menus and so on. I guess it will work when PowerDVD or alike is installed.

[breez]

I think Media Player Classic has DVD menu navigation.

[videomixer9]

Windows comes with all neccessary navigation stuff.

[LoRd_MuldeR]

I think Media Player Classic has DVD menu navigation.
I doubt that ;)
Of course MPC will use external filters for DVD Navigation, if installed...

[DSP8000]

MPEG2 decoding in drevil's build is broken,mpc & wmp.That is rev 122, when selected libmpg2 as decoder.
It is grashing mpc instantly & wmp output is b0rked.

I'll test vm9's & clsid's build.
edit: clsid's build working.
DSP8000

[LoRd_MuldeR]

MPEG2 decoding in drevil's build is broken,mpc & wmp.That is rev 122, when selected libmpg2 as decoder.
It is grashing mpc instantly & wmp output is b0rked.

I'll test vm9's & clsid's build.
edit: clsid's build working.
DSP8000

ffdshow-tryouts-rev122-sse.exe works fine on my machine (AthlonXP), even on MEPG-2 files (libmpeg2 decoder)...

[clsid]

Wasn't there a consensus a few weeks ago that the best thing is to let ffdshow decide at runtime what optimizations to use and not to force compiling for a specific set of extended instructions ?Correct. There is no significant performance gain in optimized builds. Mainly because all time consuming functions are already hand-optimized. The only file that can potentially benefit from compiler optimizations is ffdshow.ax. However usually some things get broken in it or even get slower with optimized builds.

[clsid]

rev127 build (http://prdownloads.sourceforge.net/ffdshow-tryout/ffdshow_rev2546-127_20060902.exe?download)

[KoD]

LoRd_MuldeR, you should start doubting less and start testing more. Just do as I said and try to use WMP or MPC to play a DVD Video... surprise, surprise, it works.

[khagaroth]

Originally Posted by breez
I think Media Player Classic has DVD menu navigation.

I doubt that
Of course MPC will use external filters for DVD Navigation, if installed...

Actualy, you don't need any external filters for DVD playback with MPC, it does have an internal navigator and mpeg2 decoder.

[LoRd_MuldeR]

Actualy, you don't need any external filters for DVD playback with MPC, it does have an internal navigator and mpeg2 decoder.

Are you 100% sure about that?

I never got DVD Playback working with MPC here since the last time I did a clean re-install of Windows. No matter whether ffdshow is installed or not. I think it was working before, because I had PowerDVD installed and MPC "stole" some filters form it. I use VLC for DVD Playback now and it does a good job on that...

[videomixer9]

As I stated earlier Windows comes with the Microsoft DVD Navigator (MPC uses this too per default if none other is found) filter that software will use for DVD navigation. MPC has an inbuilt MPEG2 decoder which makes it able to play DVDs and it works just fine. For ffdshow you have to enable DVD decoding option in codecs after clicking MPEG2 and it will work just fine. You prolly broke your system quite fast there. There is no need to install another DVD software really, once you have an MPEG2 decoder Windows can play DVDs if your player allows you to. I'm playing all DVDs with ffdshow and MPC.

[LoRd_MuldeR]

As I stated earlier Windows comes with the Microsoft DVD Navigator (MPC uses this too per default if none other is found) filter that software will use for DVD navigation. MPC has an inbuilt MPEG2 decoder which makes it able to play DVDs and it works just fine. For ffdshow you have to enable DVD decoding option in codecs after clicking MPEG2 and it will work just fine. You prolly broke your system quite fast there. There is no need to install another DVD software really, once you have an MPEG2 decoder Windows can play DVDs if your player allows you to. I'm playing all DVDs with ffdshow and MPC.

Sure this "DVD Navigator" comes with Windows and not with some Windows Media Player update/installer ???

[videomixer9]

it at least comes for sure with Windows XP.

[Reino]

You can use ffdshow to decode mpeg2 video from dvds right now. ffdshow video decoder configuration -> Codecs -> mpeg2 set to libmpeg2 and check DVD decoding. Libavcodec is not usable presently. Furthermore, you can select Deinterlacing and use the "5-tap lowpass" method which seemed to me to give best results. And in ffdshow audio decoder configuration -> Codecs -> enable AC3 and DTS and eventually MP2 and LPCM (LPCM only in rev124).But of course, but the problem is not with MPC, because MPC is an exception with it's integrated ds-filters compared to every other DirectShow based media player. With MPC I can play MPEG2 files and DVD's quite well, the problem is with DirectShow self, and so with the media player I use besides MPC that is; Zoom Player.
http://img397.imageshack.us/img397/4730/zpdvdsetupki4.jpg
There never was a problem with the opensource MPEG2 Decoder from Gabest, but somehow there must be something wrong with the way ffdshow registers this MPEG2 Decoder, because otherwise I would be able to see it in the Video Decoder list as you can see in the picture above.

[videomixer9]

In some versions of Zoomplayer ffdshow actually appeared, but it seems to block it on purpose for some reason. However it's not confirmable as Zoomplayer is closed source.

[KoD]

This is something that should probably be reported to Blight (the ZoomPlayer guy).

Microsoft has this small utility called Windows XP Video Decoder Chekup Utility [link (http://www.microsoft.com/downloads/details.aspx?FamilyID=DE1491AC-0AB6-4990-943D-627E6ADE9FCB&displaylang=en)], which lists the mpeg2 video decoders present on the system. In fact, it checks some registry keys. Indeed, ffdshow doesn't appear in the list when set to decode mpeg2 while gabest's mpv decoder does appear.

[haruhiko_yamagata]

Sorry for the lack of specifics.

a) AMD x2 4400+, Abit AN8 SLI Fatal1ty, 2GB Corsair XMS3500-LL, NVIDIA 7800GT running Windows XP SP2

b) ffdshow-20060722-rev2546 either XXL AutoDetect (SSE2) or videomixer9 (SSE)

c) H.264 disabled, Raw Video set to YV12

Resize: 1920x1080 Lanczos4
Avisynth: LimitedSharpenFaster (no supersampling)
Overlay: YV12, Queued Output Samples

Putting Resize after LimitedSharpen worked sometimes without crashing, but I noticed resizing often didnt work (I don't recall a single time where it played without crashing, that resizing worked, but I didn't check always as my interest was in resizing before LSF)

I don't use the Administrator account in XP (recall posts regarding this mentioning resizing this around that time)

d) Zoomplayer Pro 4.51, NVIDIA PureVideo Decoder, any DVD

have i forgotten anything?
I tested LimitedSharpenFaster and it works for me.
What is the calling script written in ffdshow avisynth edit box?
If you are still here...

[Egh]

...


So what do you think about my previous suggestion? Regarding h264 checkboxes being moved to "Decoders option" page.

I would also recommend to revise the deblocking options and make it similar to the options in CoreAVC. i.e. combobox with "deblocking turned off, deblocking off on b-frames, standard deblocking". Thus it will take less space in GUI and be moar comprehensive for a user.

[pdanpdan]

1) I have some problems with the DCT option on all ffdshow versions, even before the tryouts. When I check/activate the DCT (no matter what kind - coeficients/h263/mpeg) the image turns dark, and I can only see some red shades. When I turn it off, the image reverts to visible.
This happens on xvid content and mpeg2 content (the mpeg2 content has all dimensions divisible by 8, the xvid not allways).
2) I have some encoded files (mkv with xvid, aac, ssa - the timestamp from the DV tape). Ffdshow doesn't show the subtitle at all - directvobsub shows it perfectly. Other subtitles (ssa and anything else) is working with ffdshow if not embebed.

[_xxl]

MPEG2 decoding in drevil's build is broken,mpc & wmp.That is rev 122, when selected libmpg2 as decoder.
DSP8000
ffdshow-tryouts-rev122 works fine for me.

[KoD]

Don't use ffdshow to display subtitles, it's not suitable for that.

What do you need the DCT option for ?

[DSP8000]

Quote:
Originally Posted by DSP8000
MPEG2 decoding in drevil's build is broken,mpc & wmp.That is rev 122, when selected libmpg2 as decoder.
DSP8000
ffdshow-tryouts-rev122 works fine for me.

The PC that I use for the net & testing, etc. is P4 with 512 mb ram & that build was crashing mpc instantly.
On my other PC AMD was working fine.

Anyway, rev127 VM9's build works the fastest on my both machines.

DSP8000

[vlada]

LoRd_MuldeR> You can check in Graphedit if you have DVD Navigator filter installed. I think it comes with DirectX, but I'm not sure.

[_xxl]

The PC that I use for the net & testing, etc. is P4 with 512 mb ram & that build was crashing mpc instantly.
DSP8000
What version of ffdshow have you tested?
What part of ffdshow is crashing?
Libavcodec.dll, libmpeg2.dll, ffdshow.ax?
Please test FFdshow-Tryouts-20060901-rev122
Generic, SSE & SSE2.

[_xxl]

Know bugs:
1).ffdshow avc h.264 mobile
http://www.elecard.ru/movies/Olesya/OLESY_SIF_90kbps.mp4
2).snow libavcodec decoder 100% CPU usage
3).WME & WMP 10,11 unsolved problems
http://forum.doom9.org/showthread.php?t=114617
4).ffdshow avisynth crash.

[videomixer9]

tryouts rev 135: http://prdownloads.sourceforge.net/ffdshow-tryout/ffdshow-tryouts-rev135.exe?download (ICL 9.1.028 + GCC 3.4.5)

VC-1 Advanced Profile files should work now, only progressive encoded ones though. I'm not really sure myself if VC-1 even works at all.

[haruhiko_yamagata]

Know bugs:
1).ffdshow avc h.264 mobile
http://www.elecard.ru/movies/Olesya/OLESY_SIF_90kbps.mp4
mplayer have the same problem with that file, so let's wait.

3).WME & WMP 10,11 unsolved problems
http://forum.doom9.org/showthread.php?t=114617
I'll go and read.
4).ffdshow avisynth crash.
What is it? Avisynth plugin for direct show video decoder works for me.

[DSP8000]

What version of ffdshow have you tested?
What part of ffdshow is crashing?
Libavcodec.dll, libmpeg2.dll, ffdshow.ax?
Please test FFdshow-Tryouts-20060901-rev122
Generic, SSE & SSE2.

1.FFdshow-Tryouts-20060901-rev122
2.libmpeg2/DVD Decoding NOT checked.
3.SSE2

VM9's build rev127,
reset button on the resize page not working after first usage.Is this a bug?

[videomixer9]

ffdshow-tryouts-rev135 core2duo special edition
http://prdownloads.sourceforge.net/ffdshow-tryout/ffdshow-tryouts-rev135-core2duo.exe?download
alternative libmplayer, compare speeds if this one is faster: http://ffdshow.pytalhost.eu/libmplayer.dll

maybe works, maybe not, I have no Core2 Duo but maybe it'll give you better speeds ... any benchmarks by Core2 Duo owners appreciated.

[_xxl]

The PC that I use for the net & testing, etc. is P4 with 512 mb ram & that build was crashing mpc instantly.
On my other PC AMD was working fine.
DSP8000
1.FFdshow-Tryouts-20060901-rev122
2.libmpeg2/DVD Decoding NOT checked.
3.SSE2
1.FFdshow-Tryouts-20060901-rev122-sse2 crashes on P4
when decoding MPEG-2 using libmpeg2?
2.Libavcodec works fine?
3.FFdshow-Tryouts-20060901-rev122-sse & mmx are working fine?

[clsid]

rev135 build (http://prdownloads.sourceforge.net/ffdshow-tryout/ffdshow_rev2546-135_20060904.exe?download)
rev135 build (ICL9.1) (http://prdownloads.sourceforge.net/ffdshow-tryout/ffdshow_rev135_20060904_icl91.exe?download)

[_xxl]

Avisynth plugin for direct show video decoder works for me.
Windows Media Player 2 crashes when exiting.http://i7.tinypic.com/287hcg9.jpg

[_xxl]

FFdshow is crashing when using subtitles stereoscopic.
Tested with .srt .sub .txt files.
http://i7.tinypic.com/287hlzr.jpg

[_xxl]

Automatic preset loading does not work.
http://forum.doom9.org/showthread.php?t=111829

[videomixer9]

It only seems to not work with 4CC thing. I randomly tested application name, size and filename matching and it works fine.

Stereoscopic crashes in Tfont.cpp on the line "__m64 mm1=*(__m64*)(dstLn[2]+x);" with a bad pointer on dstLn[2]. What does stereoscopic do anyways?

[Egh]

Revision 136 - Directory Listing
Modified Mon Sep 4 13:37:21 2006 UTC (12 hours, 37 minutes ago) by matmaul

"wmv9" decoder (use ff_wmv9.dll) seems to works only with vfw mode (in original ffdshow, milan don't propose it in directshow decoder)


Hmm... Any clarification about this one? Has it been fixed now?

[_xxl]

ffdshow-tryouts-rev135 error
http://i2.tinypic.com/315g41w.jpg
ICL9 + unicode build.
Tested with XP SP2, media player2.

[foxyshadis]

drevil, doing what? Just loading the config screen?

[_xxl]

drevil, doing what? Just loading the config screen?
Loading version details.

[_xxl]

Leadtools h264 libavcodec decoder?
fourcc:l264,L264,lx64,LX64?
Samples:
http://www.leadcodecs.com/Streaming/H264/CarelessEnglish_720x352_Q23.avi

[MatMaul]

Hmm... Any clarification about this one? Has it been fixed now?

Yes, it is the fixe, rev 136 is ok (you now have just "libavcodec" in the list to decode wmv9 video)

[_xxl]

http://s6.simpleupload.de/f89dd70b8/ffdshow-tryouts-rev137-sse.exe.html

[hellfred]

Leadtools h264 libavcodec decoder?
fourcc:l264,L264,lx64,LX64?
Samples:
http://www.leadcodecs.com/Streaming/H264/CarelessEnglish_720x352_Q23.avi
That is a strange clip. MediaPlayerClassic is able to detect and play the audio part. MPlayer segfaults with FFMPEGs vorbis decoder, plays sound only when one forces the use of libvorbis:mplayer -afm libvorbisThe avi is detected as ogg container (mplayer, coming from the unix/linux world, examines the header of the container, not the file extension), so I can not verify the quality. Did anybody succeed in playing the video?

Hellfred

EDIT: OK, I need a build of ffdshow tryout from revision 137 or later. I will dl, install and try.

[videomixer9]

So which tard told you that on Windows the media playback routines examine only the file extension. It seems to me that already shows you cluelessness. Besides MPC uses a fully different architecture and MPlayer not even one single codec DirectShow codec. As this is about ffdshow not about mplayer I suggest you post that problem to an mplayer thread or the ffmpeg mailinglists. In ffdshow the vorbis part in libavcodec shouldn't be top up2date due to some errors iirc.

[LoRd_MuldeR]

Leadtools h264 libavcodec decoder?
fourcc:l264,L264,lx64,LX64?
Samples:
http://www.leadcodecs.com/Streaming/H264/CarelessEnglish_720x352_Q23.avi

That definitely is *not* an AVI file :sly:

http://img204.imageshack.us/img204/3651/noavibd9.gif

Who ever made this, used a wrong extension. Rename to .ogg please!

[videomixer9]

well vorbis hardly fits into avi container without dirty acm hacks. leadcodecs are total tards?

[LoRd_MuldeR]

well vorbis hardly fits into avi container without dirty acm hacks. leadcodecs are total tards?

I know. But if they want to use Vorbis - wich is 100% okay - then they should use the right container for that. And if they use the OGG container, they should make that clear and not try to make use believe that this is an AVI file!

[hellfred]

So which tard told you that on Windows the media playback routines examine only the file extension. It seems to me that already shows you cluelessness. Besides MPC uses a fully different architecture and MPlayer not even one single codec DirectShow codec. As this is about ffdshow not about mplayer I suggest you post that problem to an mplayer thread or the ffmpeg mailinglists. In ffdshow the vorbis part in libavcodec shouldn't be top up2date due to some errors iirc.
I think you totaly missunderstood my post. My concern is to see the quality of the H264 clip encoded with leads MPEG4 AVC codec, and I wanted to ask for advice on how to play the clip.
I did not state that MediaPlayerClassic or the underlying DirectShow-Framework uses only the file extension choose a splitter and build up the filter chain. I only gave a overview of what I have tryed to watch the video clip and hoped that someone, who succeeded to do so, suggests me how to manage that. In order to watch the clip I have tryed playing it with MediaPlayerClassic and MPlayer. MPlayer is usually able to play anything you throw at it, no matter how broken. So I tryed it in addition to the DirectShow based players and faild with the given error. Nevertheless I am well aware that MPlayer does not use the DirectShow-framework at all, but comes with it own set of buil in splitters, filters and codecs.
So what DirectShow splitter/parser is recommendet to split that video clip?
With the latest ffdshow from drevil_xxl and somewhat aged Haali Media Splitter I only get sound.
Using GraphEdit -> Render Media File I only get Haali Media Splitter v.1.6.46.4 connected to ffdshow Audio decode connetcted to the default Direct Sound device.
Which filters do you have in your filter graph when playing the clip in question?
I reckon that at least foxyshadis - being the developer that added support for the new fourcc - was able to actually watch the clip to verify, that decoding is working. So I asked here for advice. Maybe I should have marked my post as off topic, though.
About filling a bug-report for ffmpeg/mplayer: I wanted to learn more about the file before deciding to do so.
Hellfred

[foxyshadis]

It uses LX64 too? Fer crying out loud, how many extensions do they need. And if it isn't the same, it shouldn't be advertised as H.264.

I'm not even going to think about supporting that misnamed file. (You're forgiven because I love George Michael <3, even if that rendition is evil.) I assumed that it was just yet another fourcc overkill (sadly common with avi) and all it needed was adding another fourcc to h264 support, I was obviously wrong. Audio works fine in vlc (100% lavc) and ffdshow, so not I'm sure what's up with mplayer, but nothing so far can get the video out of it, even stuff that plays theora-in-ogg. Nothing seems to be downloadable on the site, so I doubt it could be further supported short of reverse engineering whatever they hacked into ogg.

[asasadad_1]

Which filters do you have in your filter graph when playing the clip in question?
Hellfred
lead ogg splitter
lead vorbis decoder
lead h264 decoder

[LoRd_MuldeR]

I think you totaly missunderstood my post. My concern is to see the quality of the H264 clip encoded with leads MPEG4 AVC codec, and I wanted to ask for advice on how to play the clip.
I did not state that MediaPlayerClassic or the underlying DirectShow-Framework uses only the file extension choose a splitter and build up the filter chain. I only gave a overview of what I have tryed to watch the video clip and hoped that someone, who succeeded to do so, suggests me how to manage that. In order to watch the clip I have tryed playing it with MediaPlayerClassic and MPlayer. MPlayer is usually able to play anything you throw at it, no matter how broken. So I tryed it in addition to the DirectShow based players and faild with the given error. Nevertheless I am well aware that MPlayer does not use the DirectShow-framework at all, but comes with it own set of buil in splitters, filters and codecs.
So what DirectShow splitter/parser is recommendet to split that video clip?
With the latest ffdshow from drevil_xxl and somewhat aged Haali Media Splitter I only get sound.
Using GraphEdit -> Render Media File I only get Haali Media Splitter v.1.6.46.4 connected to ffdshow Audio decode connetcted to the default Direct Sound device.
Which filters do you have in your filter graph when playing the clip in question?
I reckon that at least foxyshadis - being the developer that added support for the new fourcc - was able to actually watch the clip to verify, that decoding is working. So I asked here for advice. Maybe I should have marked my post as off topic, though.
About filling a bug-report for ffmpeg/mplayer: I wanted to learn more about the file before deciding to do so.
Hellfred

VLC Players and Nero ShowTime with it's own OGG Splitter fails too! MPlayer reports "Ogg stream 0 has a header marker but is of an unknown type". So if you ask me, that file is either damaged or invalid.

[LoRd_MuldeR]

lead ogg splitter
lead vorbis decoder
lead h264 decoder

If it doesn't play with normal H.264 decoders than it's not H.264, but some broked format that breaks the H.264 standards!

[videomixer9]

just wait for them to go bankrupt with their lousy super propiertary crap and the problem is solved. It'll just die the same way that useless CAVS codec will die, cause it's not easy playable anywhere and modifies standards into absurdity.

[Liisachan]

Who ever made this, used a wrong extension. Rename to .ogg please! Technically, it should be .ogm

Apparently LEAD Ogg Splitter is based on Tobias' OggDS.

[hellfred]

I have dl and installed the free LEAD Vorbis Audio Codec. It comes with vorbis decoder filter, vorbis encoder filter, an ogg splitter filter and finally an ogg multiplexer filter.
Then I have tryed to render the files with graphedit once more.
One time, I left the extension unchanged (.avi), once I changed it (.ogg).

.avi:
Errror Message:
Could not construct a graph from this file.
-Have you installed all necessary filters?
-Note that the 'Render File' menu option cannot render *.GRF files.
Es wurden ein ungültiger Medientyp angegeben. (Return code: 0x80040200)
Translation: An insufficient media type was given.
No graph is being created.

.ogg:
Message:
Das Format einiger Daten in diesem Video wird nicht unterstüzt
Translation: The format of some Data in this video are not supported

A graph is being created that can play audio only.

Unknow source filter - pin out
connected to
pin in - LEAD Ogg Splitter, which has two pins:
pin Video, not connected to any filter
pin Audio
connected to
pin XFrom In - LEAD Vorbis Decoder - pin XFromOut
connected to
pin Audio Input - Default Sound Device

Tryed to manually add ffdshwo MPEG-4 Video Decodec and connect it
Message:
These filters cannot agree on a connection. Verify type compatibility of input pin and output pin.
Es wurde keine Zwischenfilter für die Herstellung der Verbindung gefunde. (Return code: 0x80040217)
Translation: No Inter-Filter could be found to establish the connection

Properties of LEAD Ogg Splitter pin Video:
Major Type: Video - Sub Type: {3436324C-0000-0010-8000-00AA00389B71} - Format: L264 720x352, 24 bits
rcSrc=(0,0,0,0)
rcDst=(0,0,0,0)

As usually on win32, the file extension does matter, videomixer9 :devil:

[_xxl]

http://www.leadcodecs.com/Streaming/h264-Streaming-Main.htm
http://www.leadcodecs.com/Download/H264-Videos.htm
You need:
lead ogg splitter,lead vorbis decoder,lead h264 decoder.

[haruhiko_yamagata]

I would also recommend to revise the deblocking options and make it similar to the options in CoreAVC. i.e. combobox with "deblocking turned off, deblocking off on b-frames, standard deblocking". Thus it will take less space in GUI and be moar comprehensive for a user.
Basically I agree with you. One problem is the page is too small to add anything. "Decoder options" page is not the only one that is too small, so I think GUI frame has to be expanded.

[_xxl]

h.264 samples:
http://www.vsofts.com/h264/videoclips.html
libavcodec h.264 decoder is not 100% working.
coreavc works fine.

[Egh]

Basically I agree with you. One problem is the page is too small to add anything. "Decoder options" page is not the only one that is too small, so I think GUI frame has to be expanded.

Well, the frame itself is 450px vertical, so I guess it's possible to increase vertical size quite easily. No need to increase it too much, I'd suggest 550px to try first.

[plane]

Not sure if I should ask at here. Could anyone tell me the default TomsMoComp_ff.dll field order in the ffdshow? The TomsMoComp.dll used in Avisynth is BFF. Are they share the same settings?

[Jeremy Duncan]

http://img297.imageshack.us/img297/4975/1ze8.jpg

This is from the August 21 2006 version of FFdshow.

Using the "Map" tab, the picture gets a white snow outline about images onscreen.
Swap Fields does nothing.
And Linked Luma and Chroma motion masks does nothing.

[Amour]

http://s6.simpleupload.de/f89dd70b8/ffdshow-tryouts-rev137-sse.exe.html
Thanks, but could you make it available on SourceForge?

And I'm waiting for a working SSE2 build from you, as the rev122 one is buggy. Thanks for your job.

[LoRd_MuldeR]

http://img297.imageshack.us/img297/4975/1ze8.jpg

This is from the August 21 2006 version of FFdshow.

Using the "Map" tab, the picture gets a white snow outline about images onscreen.
Swap Fields does nothing.
And Linked Luma and Chroma motion masks does nothing.

Those are options for the deinterlacer and they work as expected!

"Map" shows the motions map. Those are the areas that have motion and need to be deinterlaced. You can enable "Map" to find the right thrashold value. But of course you'll disable it once you found a good value.

"Swap Fileds" will swap the fields, which is needed for some videos to be deinterlaced properly. Set it accordingly to your video's field order and everything is fine.

[foxyshadis]

Swap fields is a really badly named way of selecting BFF, I can't believe that it just does nothing. The other arguments you have to understand the meanings of:

map: presents the pixels that would be deinterlaced in white. Used to refine the threshold.
linked: Instead of analyzing all planes separately, it reuses the map built on the luma for the chroma, which is usually a good idea. Unlinked can sometimes cause ghost chroma.
two way: Use forward and backward deinterlacing instead of backward-only.

Drat, mulder answered. :p

[Jeremy Duncan]

Thank you both for explaining that.

[Episode]

Hmm, I'm little bit confused about all of these builds.. which one should I pick if I want to have an build that has been build with ICL9.1 (I'm using old Northwood P4 and ICL builds are usually the fastest for me) AND SSE2 optimizations? clsid's and videomixer9's builds look very promising, but I couldn't find any information about their optimizations. Thanks in advance!

[_xxl]

http://s21.simpleupload.de/f5f07cdf9/ffdshow-tryouts-rev138-sse.exe.html
http://s21.simpleupload.de/f449a79e8/ffdshow-tryouts-rev138-sse2.exe.html

[clsid]

Hmm, I'm little bit of confused about all of these builds.. which one should I pick if I want to have an build that has been build with ICL9.1 (I'm using old Northwood P4 and ICL builds are usually the fastest for me) AND SSE2 optimizations? clsid's and videomixer9's builds look very promising, but I couldn't find any information about their optimizations. Thanks in advance!
For pure decoding speed the generic builds are just as fast as 'compiler optimized' builds. That is because all the vital parts are already hand-optimized in the source code.

If you make heavy use of the filters in ffdshow, then you could try an ICL9.1 build. Either by me or VM9. Those builds are still generic and thus also work on older CPUs, but they also contain compiler optimizations. My last ICL9.1 build contains MMX, SSE, SSE2 and SSE3 optimizations. Compiler has been patched so that AMD CPUs can also benefit from those optimizations.

[Kostarum Rex Persia]

What about new SSE4 support? Can ffdshow make use of them in near future?

[videomixer9]

libavcodec not really but I just released a Core2 Duo version which is the only CPU that makes use of SSE4 currently. Who cares anyways for that shit, you're just nagging for shit again anyways and don't even bother reading the rest as usual :O

[NULUSIOS]

why all this tension?

[Jeremy Duncan]

http://img297.imageshack.us/img297/4975/1ze8.jpg

Using The Map button and the slider to get the least amount of snow on my High definition DVD "Over Beautiful British Columbia".
The result is a really sharp-ridged picture.
But if I put the slider to a lower number, the snow is bad, but there's no sharp-ridges on the movie.

On the other hand.
On regular dvd movies.
Setting the slider so there's very little snow improves the picture compared to no deinterlacing.

So there's a bug that prevents HD movies using the slider and being deinterlaced.

I'm using the August 21 2006 version.

[Jeremy Duncan]

What about new SSE4 support? Can ffdshow make use of them in near future?

Here's one.
Link (http://prdownloads.sourceforge.net/ffdshow-tryout/ffdshow-tryouts-rev135-core2duo.exe?download)

[foxyshadis]

The SSE4 build is awesome. You guys have no idea what you're missing, it's like finally opening your eyes after years of sleep! I get an awesome .1 fps more! Can you believe it, folks?

Jeremy, the fact that Deinterlacing isn't checked in your screenshots makes me wonder what else is different between them and your actual config. You do have map unchecked while watching, right? It's a debugging option. And for oversharpness, maybe unchecking sharp could help.

[Rgeek]

When I use seek bar of Crystal Player 1.96 and 1.97 several times, the player freezes.

Using build 20060526-rev2546 this error doesn't happens...

Sorry for the bad English.

[Jeremy Duncan]

Deinterlacing is checked when I use it.
Map is unchecked except for when I'm adjusting the slider.
I've tried unchecking sharp, but it didn't help.

[Jeremy Duncan]

When I use seek bar of Crystal Player 1.96 and 1.97 several times, the player freezes.

Using build 20060526-rev2546 this error doesn't happens...

Sorry for the bad English.

Uncheck "Queue output samples" in the Queue & Misc tab.

[cc979]

did a quick test:

tested using elephant dream_hd xvid/ac3 (8mbs stream mostly)
using 07-07-06 hali splitter

drevil_xxl

ffdshow-tryouts-rev138-sse.exe
null: User: 8s, kernel: 0s, total: 8s, real: 9s, fps: 78.7, dfps: 70.7
vmr9: User: 8s, kernel: 0s, total: 9s, real: 10s, fps: 69.4, dfps: 59.7
ovrl: User: 8s, kernel: 0s, total: 8s, real: 9s, fps: 75.3, dfps: 72.0

with que samples off
null: User: 8s, kernel: 0s, total: 8s, real: 8s, fps: 77.8, dfps: 77.1
vmr9: User: 8s, kernel: 0s, total: 8s, real: 10s, fps: 77.1, dfps: 59.9
ovrl: User: 8s, kernel: 0s, total: 8s, real: 9s, fps: 75.6, dfps: 72.4


ffdshow-tryouts-rev138-sse2.exe
null: User: 9s, kernel: 0s, total: 10s, real: 10s, fps: 65.4, dfps: 64.7
vmr9: User: 10s, kernel: 0s, total: 11s, real: 12s, fps: 58.3, dfps: 51.3
ovrl: User: 10s, kernel: 0s, total: 10s, real: 10s, fps: 63.1, dfps: 61.2

:thanks: keep up the good work guys

[haruhiko_yamagata]

When I use seek bar of Crystal Player 1.96 and 1.97 several times, the player freezes.

Using build 20060526-rev2546 this error doesn't happens...

Sorry for the bad English.
Which version of ffdshow has the problem?

[haruhiko_yamagata]

did a quick test:

tested using elephant dream_hd xvid/ac3 (8mbs stream mostly)
using 07-07-06 hali splitter

drevil_xxl

ffdshow-tryouts-rev138-sse.exe
null: User: 8s, kernel: 0s, total: 8s, real: 9s, fps: 78.7, dfps: 70.7
vmr9: User: 8s, kernel: 0s, total: 9s, real: 10s, fps: 69.4, dfps: 59.7
ovrl: User: 8s, kernel: 0s, total: 8s, real: 9s, fps: 75.3, dfps: 72.0

with que samples off
null: User: 8s, kernel: 0s, total: 8s, real: 8s, fps: 77.8, dfps: 77.1
vmr9: User: 8s, kernel: 0s, total: 8s, real: 10s, fps: 77.1, dfps: 59.9
ovrl: User: 8s, kernel: 0s, total: 8s, real: 9s, fps: 75.6, dfps: 72.4


ffdshow-tryouts-rev138-sse2.exe
null: User: 9s, kernel: 0s, total: 10s, real: 10s, fps: 65.4, dfps: 64.7
vmr9: User: 10s, kernel: 0s, total: 11s, real: 12s, fps: 58.3, dfps: 51.3
ovrl: User: 10s, kernel: 0s, total: 10s, real: 10s, fps: 63.1, dfps: 61.2

:thanks: keep up the good work guys
If you are comparing queue on and off, your way of benchmarking is not good. Read this (http://forum.doom9.org/showthread.php?p=832691#post832691).
Timecodec's video renderers don't wait the time stamp and the work load of itself does not chage with the queue. So as far as you are using single core CPU, the result will be the same.

[clsid]

did a quick test:

ffdshow-tryouts-rev138-sse.exe
null: User: 8s, kernel: 0s, total: 8s, real: 9s, fps: 78.7, dfps: 70.7
vmr9: User: 8s, kernel: 0s, total: 9s, real: 10s, fps: 69.4, dfps: 59.7
ovrl: User: 8s, kernel: 0s, total: 8s, real: 9s, fps: 75.3, dfps: 72.0

ffdshow-tryouts-rev138-sse2.exe
null: User: 9s, kernel: 0s, total: 10s, real: 10s, fps: 65.4, dfps: 64.7
vmr9: User: 10s, kernel: 0s, total: 11s, real: 12s, fps: 58.3, dfps: 51.3
ovrl: User: 10s, kernel: 0s, total: 10s, real: 10s, fps: 63.1, dfps: 61.2

Conclusion: SSE build is faster than SSE2 build :devil:

Hopefully the n00bs will now start using generic builds. Because by logic reasoning and interpolation of the above testresults MMX will be faster than SSE.

[LoRd_MuldeR]

http://img297.imageshack.us/img297/4975/1ze8.jpg

Using The Map button and the slider to get the least amount of snow on my High definition DVD "Over Beautiful British Columbia".
The result is a really sharp-ridged picture.

The task is not to reduce the amount of "snow". The Map option is intended for finding the correct Threshold. A good value for threshold is when all moving areas are deinterlaced (this means all Mouseteeth artifacts are removed) while static areas are passed through (this means to keep the full vertical resolution in static areas). So if you enable the Map option, you should see "snow" on all moving objects, but none on static areas. If you see too much "snow", you'll have to use a higher threshold for better quality. And If there are moving objects without "snow" you'll need a lower threshold, otherwise there will be visible artifacts!

IMO a value of 10 is a good starting point. Then you lower it in 1-steps until all artifacts are gone. 32 seems extremely high...

[haruhiko_yamagata]

@clsid
Could you support the language packs by your installer? There are a lot of effort by many people, and many people need them.

[videomixer9]

Conclusion: SSE build is faster than SSE2 build :devil:

Hopefully the n00bs will now start using generic builds. Because by logic reasoning and interpolation of the above testresults MMX will be faster than SSE.

Well just check the remarks ICL gives you, I think most vectorizations are made for the subtitle renderer and the audio mixer. I think that not in a single filter that actually processes pictures any of the vectorizations are made. libmplayer this may be more but the hand optimized code there is way faster too and ICL doesn't compile that code. Postprocessing and resizing mostly is code from libmplayer ...

[Rgeek]

Uncheck "Queue output samples" in the Queue & Misc tab.

Works! Thank you!

Which version of ffdshow has the problem?

All builds after 20060526-rev2546 tested by me.

I'm currently using rev2546-135_20060904.

[haruhiko_yamagata]

Bug fix : wmav2 : revert wmadec.c to rev 112

In some builds it is OK, but in some builds it crashes.
wmav2 file that I tested was created by WMP10(or maybe 11), from CD, with its internal lossless encoder.

@foxyshadis
Please debug and re-commit.

[cc979]

@haruhiko_yamagata i was mainly testing difference of sse/sse2 just added 'que samples off' test because of being curious

@clsid if you recommend what mmx build, i'll test that with same file (my gramma bad when wake up)

[videomixer9]

wma2 samples i tried work fine on my own builds. And WMP10/11 do WMV2? I always thought they create wma9 audio?

[haruhiko_yamagata]

wma2 samples i tried work fine on my own builds.
Yes, your build works. clsids' crashes. Some of my private build works, but some crashes. I can't remember what the difference is. Now I can't make working libavcodec-138.

//EDIT
videomixer9, your rev127 works, rev 138 crashes.

[haruhiko_yamagata]

All builds after 20060526-rev2546 tested by me.

I'm currently using rev2546-135_20060904.
I tested Crystal Player v1.97. It works for me with queue on. Is the problem reproducible with the default setting of Crystal Player?

[videomixer9]

could you post the wma2 samples you use, cause my samples also work in rev138.

[haruhiko_yamagata]

could you post the wma2 samples you use, cause my samples also work in rev138.
Excuse me, it was not lossless. Windows Media Audio 9.1, 128 kbps, 44KHz, stereo 1-pass CBR.
Because of copyright issue, I can't post it to public space.

[videomixer9]

Well wmadec is not really supposed to handle that. As ffdshow already says wmav1 and wmav2

edit: your sample plays fine on my pc with 138 build

[clsid]

@cc979, try this one:

rev142 build (http://prdownloads.sourceforge.net/ffdshow-tryout/ffdshow_rev142_20060906.exe?download)

[Rgeek]

I tested Crystal Player v1.97. It works for me with queue on. Is the problem reproducible with the default setting of Crystal Player?

Yes, default setting as well. With queue checked, seek bar always freezes the player.

Edit: With queue off the problem is rare.

[foxyshadis]

When ffdshow starts randomly crashing for no reason I always remove every build except the one I'm testing, because ffdshow (maybe only on xp) has a very bad habit of picking up another install's libavcodec.dll or whatever. Caused me a lot of grief until I figured out why.

Still, if I can get it to crash I'll look at it. It might be the compiler's fault, not the code's.

[cc979]

another test


tested using elephant dream_hd xvid/ac3 (8mbs stream mostly)
using 07-07-06 hali splitter

clsid
ffdshow_rev142_20060906.exe
null: User: 8s, kernel: 0s, total: 8s, real: 8s, fps: 78.8, dfps: 77.7
vmr9: User: 8s, kernel: 1s, total: 9s, real: 10s, fps: 65.5, dfps: 59.8
ovrl: User: 8s, kernel: 0s, total: 8s, real: 9s, fps: 75.1, dfps: 72.5

drevil_xxl
ffdshow-tryouts-rev138-sse.exe
null: User: 8s, kernel: 0s, total: 8s, real: 9s, fps: 78.7, dfps: 70.7
vmr9: User: 8s, kernel: 0s, total: 9s, real: 10s, fps: 69.4, dfps: 59.7
ovrl: User: 8s, kernel: 0s, total: 8s, real: 9s, fps: 75.3, dfps: 72.0

ffdshow-tryouts-rev138-sse2.exe
null: User: 9s, kernel: 0s, total: 10s, real: 10s, fps: 65.4, dfps: 64.7
vmr9: User: 10s, kernel: 0s, total: 11s, real: 12s, fps: 58.3, dfps: 51.3
ovrl: User: 10s, kernel: 0s, total: 10s, real: 10s, fps: 63.1, dfps: 61.2

[_xxl]

Conclusion: SSE build is faster than SSE2 build

>>SSE performance of Athlon64 is the same as of P4, SSE2 performance lags a little
It's a bad thing to say. You can't generalize about this sort of thing. There are many different tasks that can be performed and in some of them A64 is faster, in others - P4.
Coop
Yes actually AMD K8 CPUs have an extended version of SSE2 - more registers. It can run somewhat 30% faster due to that over regular SSE2. However, don't believe it until you see it.

http://forums.amd.com/index.php?showtopic=5232
http://www.theinquirer.net/default.aspx?article=9278
True or not?

[clsid]

from wikipedia
Differences between MMX and SSE2

SSE2 supports almost every integer operation that MMX supports. Therefore, it is possible to convert all existing MMX code to SSE2 equivalent. Since an XMM register are two times as long as an MMX register, loop counters and memory access may need to be changed to accommodate this.

Although one SSE2 instruction can operate on twice as much data as an MMX instruction, performance might not increase significantly. Two major reasons are: accessing SSE2 data in memory not aligned to a 16-byte boundary will incur significant penalty, and the throughput of SSE2 instructions in most x86 implementations is usually smaller than MMX instructions. Intel has recently addresses the first problem by adding an instruction in SSE3 to reduce the overhead of accessing unaligned data, and the last problem by widening the execution engine in their Core microarchitecture.

[videomixer9]

VC1 decoding still doesn't work for me for some reason and the new VMnc codec also won't run after I added, wonder what I oversaw ... tried various combinations so far already but both won't decode no matter what ...

http://prdownloads.sourceforge.net/ffdshow-tryout/ffdshow-tryouts-rev146.exe?download

[_xxl]

AMD-XP:
http://s21.simpleupload.de/fa18a296e/ffdshow-tryouts-rev145-amd-xp.exe.html
AMD-K8
http://s21.simpleupload.de/fdb8eddb3/ffdshow-tryouts-rev145-k8.exe.html
INTEL-P4
http://s21.simpleupload.de/ffeef6e7a/ffdshow-tryouts-rev145-p4.exe.html

[videomixer9]

Lol bad day today, fucked up too often with that VMnc thing ... works now but it always crashes ... oh well, need to check if ffmpeg itself crashes too or if it's introduced through porting. Was so busy searching for the error that I oversaw the most obvious. It's early stage so I guess some errors are still in there and it hasn't hit mplayer either for now.

http://prdownloads.sourceforge.net/ffdshow-tryout/ffdshow-tryouts-rev148.exe?download

[Amour]

... works now but it always crashes ...
crash ≠ working

By the way, is there a way to rotate or flip the image when watching (decoding) a video with ffdshow?

[Amour]

clsid and videomixer9, can you please fill the Release Date field when adding files to SourceForge (I mean on this page (http://sourceforge.net/project/showfiles.php?group_id=173941))? Like that, the files would be listed according to release date, and the package date wouldn't be August 3, 2006 anymore. Or you could keep filenames with the same logic name, so it becomes easy to spot the latest release.

Thanks a lot!

[cc979]

tested using elephant dream_hd xvid/ac3 (8mbs stream mostly)
using 07-07-06 hali splitter

videomixer9
ffdshow-tryouts-rev148.exe
null: User: 8s, kernel: 0s, total: 8s, real: 8s, fps: 81.4, dfps: 79.0
vmr9: User: 8s, kernel: 0s, total: 9s, real: 10s, fps: 68.4, dfps: 60.8
ovrl: User: 8s, kernel: 0s, total: 8s, real: 8s, fps: 77.2, dfps: 74.6

drevil_xxl
ffdshow-tryouts-rev145-amd-xp.exe
null: User: 8s, kernel: 0s, total: 8s, real: 8s, fps: 79.3, dfps: 76.5
vmr9: User: 8s, kernel: 0s, total: 9s, real: 10s, fps: 67.1, dfps: 59.5
ovrl: User: 8s, kernel: 0s, total: 8s, real: 9s, fps: 74.9, dfps: 72.2
...
ffdshow-tryouts-rev145-k8.exe
null: User: 8s, kernel: 0s, total: 8s, real: 8s, fps: 78.5, dfps: 75.7
vmr9: User: 8s, kernel: 1s, total: 9s, real: 10s, fps: 66.2, dfps: 59.9
ovrl: User: 8s, kernel: 0s, total: 8s, real: 8s, fps: 75.3, dfps: 72.8
...
ffdshow-tryouts-rev145-p4.exe
null: User: 8s, kernel: 0s, total: 8s, real: 8s, fps: 79.7, dfps: 76.8
vmr9: User: 8s, kernel: 1s, total: 9s, real: 10s, fps: 70.2, dfps: 59.9
ovrl: User: 8s, kernel: 0s, total: 8s, real: 9s, fps: 74.5, dfps: 72.7

@videomixer9 your rev148 looks a bit faster, what instruction set is it using?

@drevil_xxl and @clsid if you do a rev148 i'll check them too

:thanks: guys

[cc979]

@videomixer9: using your rev148 virtualdudmod stated it could'nt find msvcr80.dll so i copied the one from ffdshow directory into system32 and vdm shows this error

http://img100.imageshack.us/img100/9916/clipboard1og8.png (http://imageshack.us)

[Amour]

1.FFdshow-Tryouts-20060901-rev122-sse2 crashes on P4
when decoding MPEG-2 using libmpeg2?
2.Libavcodec works fine?
3.FFdshow-Tryouts-20060901-rev122-sse & mmx are working fine?
Don't know, but, here are my tests:

ffdshow-tryouts-rev100-sse2 = working
ffdshow-tryouts-rev111-sse2 = crash
ffdshow-tryouts-rev118-sse2 = crash
ffdshow-tryouts-rev122-sse2 = crash
ffdshow-tryouts-rev138-sse2 = crash
ffdshow-tryouts-rev145-p4 = crash

I made a thread about it on http://ffdshow-tryout.sourceforge.net/phpBB2/

[videomixer9]

@videomixer9: using your rev148 virtualdudmod stated it could'nt find msvcr80.dll so i copied the one from ffdshow directory into system32 and vdm shows this error

http://img100.imageshack.us/img100/9916/clipboard1og8.png (http://imageshack.us)

try copying it to where the vdub plugin is located only, there should be no copy of it in the windows system32 dir, also note even if you have to also copy the manifest file!

If you want the runtime working everywhere please install vcredist_x86.exe instead of copying the file, the redist doesn't copy the file to the system dir and also comes with the manifests.

I tested the plugin and it works fine as always :P


@videomixer9 your rev148 looks a bit faster, what instruction set is it using?


none really, it's a plain generic build with no single instructionset enforced. It should technically also work on CPUs without MMX. Maybe you should also do some tests with filters enabled. Then my build might be slower ...

[Amour]

more tests:

ffdshow-tryouts-rev122-sse = crash
ffdshow-tryouts-rev122-mmx = crash

drevil_xxl, I found where is the issue! If you uncheck Enable output queuing during the installation, it will work fine. So you might consider to disable this option by default, or fix the issue.

[_xxl]

drevil_xxl, I found where is the issue! If you uncheck Enable output queuing during the installation, it will work fine. So you might consider to disable this option by default, or fix the issue.
Output queuing is NOT working and it causes ffdshow to crash?

[Amour]

Ah, sorry, sorry...

Output queuing enabled + BSPlayer = crash
Output queuing enabled + Windows Media Player = working
Output queuing disabled + BSPlayer = working
Output queuing disabled + Windows Media Player = working

So the problem is a combination of Output queuing and BSPlayer.

Same problem with VM9 releases. clsid is ok because it doesn't have Output queuing by default.

[videomixer9]

BSPlayer is crap anyways, the free variant is just a spyware throwing cashmachine and the pro version just crap only without the spyware. I guess everyone knows what bs means too ;O No sense working around bsplayer, guess just adding it to the blacklist for output queuing is adequate.

[_xxl]

Stable version should work with BSPlayer.Adding it to the blacklist is NOT a good idea.

[haruhiko_yamagata]

I think blacklist is not a good idea. Bsplayer, Crystal player and WMP...

Btw, what is working fine with queue?
MPC mostly works, except for (VMR9 renderless+RGB32+specific video card+pause) = blackout problem.

List working application and check by default
Use queue only in "..."
looks better.

What should be included in the list? I would like to add MPC first.

[haruhiko_yamagata]

@videomixer9: using your rev148 virtualdudmod stated it could'nt find msvcr80.dll so i copied the one from ffdshow directory into system32 and vdm shows this error

http://img100.imageshack.us/img100/9916/clipboard1og8.png (http://imageshack.us)
To say good bye to the msvcr80.dll problem, the installer have to deal with it. I'm trying to include vcredist_x86.exe for NSIS installer. vcredist_x86.exe has to be executed by administrator though.

[haruhiko_yamagata]

none really, it's a plain generic build with no single instructionset enforced. It should technically also work on CPUs without MMX. Maybe you should also do some tests with filters enabled. Then my build might be slower ...
Which filter would be affected?
As for MSVC8, the compiler make autodetecting code, doesn't it? If I'm not wrong, MSVC8(for .ax) generic build is not slow on new CPUs and works on old CPUs (maybe wrong).

[videomixer9]

MSVC8 doesn't use any autodetection, per default it generates generic running code for any CPU (of course it can detect if certain instruction sets are available and integrate special assembly for that, but it won't generate code making full use of the instruction sets). There are many variants to compile into machine code and every compiler is differently good on this task. Also some add extra checks or produce check routines etc. MSVC is generally producing stable code that is still faster than most of GCCs often unstable code. Intel Compiler uses many advanced compiling technologies to the fullest even without auto vectorization it reaches decently faster speeds that way. However C/C++ is quite slow compared to real well hand written assembler code (you can see this when compiling the plain C code of libavcodec with the compilers). As a human you still know the best which things you can leave out that a compiler would generate like it does for everything even if it may not be really neccessary. The list of optimization techniques is nowadays quite endless, and even more endless the variations, e.g. simple things like loop unrolling, if improperly done it makes things slow (partly that may also fault of the coder and his code), if properly used it speeds things up (with libavcodec and GCC it's usually slower). Same goes for many things. MSVC is a compromise of stable and fast code. ICL is the speedmachine and GCC is mostly portable but not fast. ICLs vectorization that tries converting things automatically to SSE1-SSE4 doesn't know if the vectorizations it does make sense, so it vectorizes quite useless loops and blocks too which can lead to crashes or speed things down as SSE units get stuffed with things to handle while they could just use the regular instruction units.

So as said often, the decoding speed is thanks to hand written code and the rest of the code is better executed the regular way except for some filters maybe. But maybe someone can test that to see how it compares. I'd think though that full GCC builds end up worse than MSVC and ICL builds will prolly be another tick faster, on some filters maybe special SSE/SSE2/SSE3/SSE4 builds may be faster. I released a Core2 Duo version before that is still quite much up2date with the decoding stuff that is usually tested and filters didn't change much either. I'm still wonder if someone could compare the Core2Duo version with a generic build.

As for the filters I'm usually trying filter speed differences with the rather odd combination of xsharpen + Gradual Denoise + Denoise3D. So far ICL generic managed to be 10 fps ahead of MSVC generic there.

[haruhiko_yamagata]

So MCVC8 creates code for SSE and use it by autodetect, but not so fast. Thank you very much.

[videomixer9]

it only does generates SSE code if /arch:SSE is set. The .ax. file will crash on non-SSE CPUs then once a SSE instruction will be actually used. Only few things will be replaced with SSE instructions when this function is used though.

[haruhiko_yamagata]

Bug fix(? alpha test) : Raw video :
When Divx5=disabled and Raw video=all supported, the Divx5 video flips. The player is MPC.
Please see if it does not have any side effects.

[videomixer9]

Does that even have to really do sth. with ffdshow and not more like funny avi decompressor kicking in before or so? Not even asking which tard uses that kind of configuration ... :O

[haruhiko_yamagata]

it only does generates SSE code if /arch:SSE is set. The .ax. file will crash on non-SSE CPUs then once a SSE instruction will be actually used. Only few things will be replaced with SSE instructions when this function is used though.
In that case, ffdshow.ax should be autodetected rather than libavcodec. I'm talking about installer.

[videomixer9]

Only the Innosetup has an autodetection routine to deploy it with several special compiles ... that installer is done by drevil_xxl and clsid, I have no need for such detections as special versions are useless anyways imo. All important things are hand optimized.

Is this another of those brought to us by 2ch tardness questions you're asking cause the cowards over there aren't lazy enough to try ask themselves? They seem to be totally keen on this optimizations shit even if it's worthless. Just tell them there's a magic detection component build in that automagically makes use of anything available :P

http://prdownloads.sourceforge.net/ffdshow-tryout/ffdshow-tryouts-rev150.exe?download

[Mc Onyx]

I was just wondering what happened with the "special" H.264 GPU accelerated ffdshow that was mentioned few pages back?

[Egh]

I was just wondering what happened with the "special" H.264 GPU accelerated ffdshow that was mentioned few pages back?

looks like that project (along with x264gpu from same author) was born dead :) Not that I actually expected anything soon, especially that devaster mentioned 3rd September for the release :P

Might be that the dev just seriously underestimated the difficulties coding that (take a look at Haali renderer for instance).

[_xxl]

In that case, ffdshow.ax should be autodetected rather than libavcodec. I'm talking about installer.
FFdshow.ax is autodetected.All files are.
GCC 4.0.3 compiler can compile SSE and SSE2 versions of ffdshow.ax.

[clsid]

clsid and videomixer9, can you please fill the Release Date field when adding files to SourceForge (I mean on this page (http://sourceforge.net/project/showfiles.php?group_id=173941))? Like that, the files would be listed according to release date, and the package date wouldn't be August 3, 2006 anymore. Or you could keep filenames with the same logic name, so it becomes easy to spot the latest release.

Thanks a lot!The release date is filled in for each file. However, SF keeps listing them alphabetically.

[clsid]

Lol bad day today, fucked up too often with that VMnc thing ... works now but it always crashes ... oh well, need to check if ffmpeg itself crashes too or if it's introduced through porting. Was so busy searching for the error that I oversaw the most obvious. It's early stage so I guess some errors are still in there and it hasn't hit mplayer either for now.
I tried a nightly build of VLC yesterday and it could play both VC-1 and VMnc samples that I got from the mplayer sample site. VLC also uses code based on ffmpeg, right? So in theory it should work.

[videomixer9]

yeah ... now just need to find out why it doesn't work in ffdshow so far.

[clsid]

Sample file with FourCC "WMVA" crashes. I also got VC-1 sample files with FourCC "WMVP" and "WVP2". MPC reports no filters found for those. It also doesn't seem to want to use ffdshow for VMnc.

[bond]

Sample file with FourCC "WMVA" crashes. I also got VC-1 sample files with FourCC "WMVP" and "WVP2". MPC reports no filters found for those. It also doesn't seem to want to use ffdshow for VMnc."WMVA", "WMVP" and "WVP2" are not vc-1 compliant. therefore ffdshow of course doesnt handle them

[_xxl]

rev 152 crashes:
http://i3.tinypic.com/2mwfcsx.jpg
rev 151 works.

[affter333]

Hi,
What's the last version of ffdshow that
supports Win98se ? The newest version (20060828)
doesn't seem to support it anymore..

Thanks for your help..

[clsid]

My builds support Windows 98.


The info&debug page looks a bit messy since the addition of SSE3/SSE4. Also MMX2 should be named MMXext. MMX2 is incorrect and confusing.

[Anima123]

I have an old P3 core which does not support MMXext while ffdshow defaultly set MMXext optimization on. Is it a bug with cpu detection routine?

[clsid]

As far as I know MMXext is a subset of SSE. All P3 support SSE.

[videomixer9]

ffdshow-tryouts revision 154r2
Download: here (http://prdownloads.sourceforge.net/ffdshow-tryout/ffdshow-tryouts-rev154r2.exe?download)

compilers: ICL 9.1.030 and GCC 3.4.5
changes: igor1st audio normalization tweak

I scraped the new instruction set things for now. Still didn't find the reason for VC-1 and VMnc not properly working, would be great if someone could check those things for stuff I oversaw really :O edit: 5% faster :P

[Eragon4ever]

You mean r154 not 127?

[Egh]

You mean r154 not 127?
prdownloads.sourceforge.net/ffdshow-tryout/ffdshow-tryouts-rev154.exe?download

announced 127, link leads to 154 :)

[videomixer9]

ups copy & paste devil striked

[LoRd_MuldeR]

Found a bug: The "perspective correction" filter makes ffdshow crash when interpolation-mode is set to "cubic", while "linear" and "none" seems to work. Tested with ffdshow-tryouts-rev150-AMD-XP-K8 (drevil) and ffdshow-tryouts-rev154.exe (vm9). My CPU is Athlon-XP.

[videomixer9]

hm works just fine on my athlon xp

[LoRd_MuldeR]

hm works just fine on my athlon xp
hmmm, that's strange. I tested with MPC. But that filter isn't very important anyways...

[foxyshadis]

Probably has something to do with the source being fed into it. Dimensions probably.

[Egh]

Probably has something to do with the source being fed into it. Dimensions probably.

Exactly. Did quick testing with ffdshow resizer put prior the perspective correction in "Cubic interpolation" mode.

ffdshow crashes reproducibly here on Athlon XP if horizontal size (vertical was autodefined from 16:9 sauce AR) hits 768. In my tests, even 766 works, but just 2 pixels increment crashes ffdshow.

[videomixer9]

yeah it indeed does with resizer on

[affter333]

Hi:
There seems to be a huge memory leak When using
kernel deinterlacer on NTSC Mpeg2 (720x480i).
the free physical memory is dropping fast. (-2M/sec)
After playing for about 2 min, free memory down to zero.
But it does release all the used mem when closing player.
(Win98se,wmp 6.4 & 9.0)

My observation is that when there's more motion, it
consume more memory.

It doesn't seem to have the issue when playing
with PAL mpeg2 (720x576i) (or with other deinterlacers)

I'm using ffdshow-tryouts-rev150-INTEL-P4.exe
, don't know about other builds..

Thanks..

[_xxl]

I'm using ffdshow-tryouts-rev150-INTEL-P4.exe, don't know about other builds..
Thanks..
Please test other builds!
videomixer9 and clsid ffdshow builds.

[cc979]

tested using elephant dream_hd xvid/ac3 (8mbs stream mostly)
using 07-07-06 hali splitter

drevil_xxl
ffdshow-tryouts-rev155-AMD-XP-K8.exe
null: User: 8s, kernel: 0s, total: 8s, real: 8s, fps: 79.6, dfps: 78.7
vmr9: User: 8s, kernel: 1s, total: 9s, real: 10s, fps: 68.3, dfps: 59.8
ovrl: User: 8s, kernel: 0s, total: 8s, real: 8s, fps: 76.8, dfps: 73.0

just playing some vids everything working as it should, only one problem using vmr9 output on mpc, i've not used the ffdshow brightness controls for a while but when i do now, i've got weird colour only way it works properly is to use either nv21 output on ffdshow or use ordinary overlay output - have i missed something using nvidia driver 91.45 so maybe its them will try early driver see if that fixes it.

or if anybody have another way to fix - let me know thanks

[cc979]

i've found i get crash playing svq3 files using rev155 build

[_xxl]

i've found i get crash playing svq3 files using rev155 build
Using rev150 you get same crash?
just playing some vids everything working as it should, only one problem using vmr9output on mpc, i've not used the ffdshow brightness controls for a while but when i do now, i've got weird colour only way it works properly is to use either nv21 output on ffdshow or use ordinary overlay output - have i missed something using nvidia driver 91.45 so maybe its them will try early driver see if that fixes it.
or if anybody have another way to fix - let me know thanks
Works with 91.31 nvidia driver.

[videomixer9]

http://prdownloads.sourceforge.net/ffdshow-tryout/ffdshow-tryouts-rev155.exe?download

i place a bet on this rev155 generic that it is faster than the AMD version ;P

[Rash]

Isn't NV21 reserved for DXVA? Sorry if that sounded stupid.

[cc979]

@drevil_xxl rev150 crashes too with same svq3 file
rev150 libavcodec.dll offset 000bbe3a crash shown

videomixer9 rev155 crashes on the svq3 file too
...

tested using elephant dream_hd xvid/ac3 (8mbs stream mostly)
using 07-07-06 hali splitter

videomixer9
ffdshow-tryouts-rev155.exe
null: User: 8s, kernel: 0s, total: 8s, real: 8s, fps: 81.6, dfps: 80.7
vmr9: User: 7s, kernel: 1s, total: 9s, real: 10s, fps: 72.0, dfps: 61.2
User: 8s, kernel: 0s, total: 8s, real: 8s, fps: 78.2, dfps: 75.0

ps. ask if you want anymore crash details

the file in question is here i think
http://files.filefront.com/startreklegacy+om+060706+qtzip/;5127512;/fileinfo.html

guys could you add a little builder tag on the end off yours builds so its easier to kept track of what builds are what, cheers - keep the good work

[cc979]

after further testing found that window media players it ok using ffdshow rev150 .. maybe a bug in mpc

sorry guys i should have tested this earlier

[haruhiko_yamagata]

@drevil_xxl rev150 crashes too with same svq3 file
rev150 libavcodec.dll offset 000bbe3a crash shown

videomixer9 rev155 crashes on the svq3 file too
...

ffdshow may be responsible for it, but mainly it's a problem with splitters, I think.
http://forum.doom9.org/showthread.php?p=869291#post869291

[videomixer9]

ffdshow-tryouts revision 162
Download: here (http://prdownloads.sourceforge.net/ffdshow-tryout/ffdshow-tryouts-rev162.exe?download)

compilers: ICL 9.1.030 and GCC 3.4.5
changes: VP5 & VP6 decoder added

[Liisachan]

VP5 and 6 don't play so far via ffdshow above, however, now VP62, both in AVI and FLV (ie FLV4) plays by celtic_druid's ffplay.

[videomixer9]

any link to some VP6 flvs?

[Liisachan]

glider_flv4.flv (http://minkymomo.info/~meroko_deathnote/tmp/codec_test/glider/)

[yangxi]

i have the AMD X2 4200+ cpu, which built is best encode speed (avi -> mpg)?

[foxyshadis]

They'd all give the same speed, the encoding library isn't dependant on compiler optimization.

[haruhiko_yamagata]

A feature freeze is a good idea. Imho there are already too many formats in ffdshow that are incomplete. First make it stable and then add new features one at a time.

How about this version numbering scheme:
stable: 1.0.0.revision beta
dev: 1.0.0.revision alpha

I agree that a feature freeze is a good idea.

Btw, why beta for stable build?
I suggest
stable: 1.0.0.revision
dev: 1.0.0.revision alpha

Thank you for the list of known problems.

[_xxl]

A list of known/reported problems:
* VC-1 does not work
* VMnc does not work
* VP5 does not work
* VP62 does not work
* The "perspective correction" filter makes ffdshow crash when interpolation-mode is set to "cubic". Happens only if horizontal video size is 768 or greater.
* SVQ3 crashes (splitter related?)
* kernel deinterlacer memory leak on NTSC Mpeg2 (720x480i) video (uncomfirmed)
* Output queue enabled + BSplayer = crash
* Output queue enabled + VMR9 renderless + RGB32 + specific video card + pause = blackout problem
* FFdshow is crashing when using subtitles stereoscopic
I probably missed a few.
* WMP 10,11 crash
* WME crash

[haruhiko_yamagata]

* WMP 10,11 crash

When does WMP crash?

[thuan]

I have a weird problem with some icl9 builds, any build with ffdshow.ax that is about 5MB not the one with around 3MB ax like the 162 one from vm9 or the gcc one compiled by drevil will crash (not exactly it slows my com down to a state that unusable, have to kill player process) when I use denoise3dhq and seek a lot with 5 secs jump time in one direction or sometimes randomly when I seek with denoise3dhq (haven't checked with normal denoise3d). Will happen with all avi files randomly (use MS splitter), IIRC haven't got this with any file that is watched with Haali splitter (I don't check avi here just mkv, ogm, and mp4). I guess it has something to do with compiler optimization. WinXPSP2 T-bred Sempron CPU no SSE2 or above, use MPC rev.611.

[_xxl]

#define IDFF_MOVIE_AUDX 20
#define IDFF_MOVIE_MAX 20
conflicted with IDFF_AUDX ?

[Reino]

* SVQ3 crashes (splitter related?)
In what kind of circumstances?

[clsid]

In what kind of circumstances?See a few pages back in this thread.

I myself don't have any issues with SVQ3.

[Jeremy Duncan]

I think having a official release with links to daily builds would be nice.
The daily builds can have change logs and anybody with a problem can use a daily built with a fix.

The main designers of FFdshow should colaborate on the 1 release version and then work individually on their daily builds.

The main release version can have it's own webpage with a nice page layout, like Mozilla has.

[Peuj]

See a few pages back in this thread.

I myself don't have any issues with SVQ3.

Tested with MPC rev611 (celtic_druid's build) with QuickTime set as DirectShow
FFDShow: ffdshow_rev2546-127_20060902.exe

ftp://mplayerhq.hu/MPlayer/samples/V-codecs/SVQ3/finalfantasy_cc.mov
Sound ok but the image is not good (just grey with some colors and artifacts)

ftp://mplayerhq.hu/MPlayer/samples/V-codecs/SVQ3/gt4_tgs2k3.mov
MPC freezed

ftp://mplayerhq.hu/MPlayer/samples/V-codecs/helldeskmaskcable.mov
Sound ok but the image is not good (just grey with some colors and artifacts)

ftp://mplayerhq.hu/MPlayer/samples/V-codecs/rev_theatre_0x3839_640_dl.mov
all ok but there is a line green and red at the bottom of the image.

ftp://mplayerhq.hu/MPlayer/samples/V-codecs/starfox2.mov
Sound ok but the image is not good (just grey with some colors and artifacts)

ftp://mplayerhq.hu/MPlayer/samples/V-codecs/Vertical400kbit.sorenson3.mov
MPC freezed

ftp://mplayerhq.hu/MPlayer/samples/V-codecs/SVQ1/spec-tampax-red-(frames_loss).mov
MPC freezed

All the mov files are well played with QuickTime or MPC and QuickTime codec

hope it helps.

[Amour]

BTW: Why ffdshow-tryout and not just continue in the original ffdshow sf project?
They said it was because the original admin (Milan) was nowhere to be found, impossible to contact him for a very very very long time (2 years?).

[haruhiko_yamagata]

They said it was because the original admin (Milan) was nowhere to be found, impossible to contact him for a very very very long time (2 years?).
It was two months. Now it is 3.5 months. If it was 2 years, We'd have taken over the original project. Two months was too short to take over. I still hope milan comes back someday.

[clsid]

ftp://mplayerhq.hu/MPlayer/samples/V-codecs/SVQ3/finalfantasy_cc.mov
Sound ok but the image is not good (just grey with some colors and artifacts)Confirmed. If I use Haali splitter instead of Gabest, then it even crashes.

I can confirm the problems with the other samples as well. Exvept for the freezes.

Perhaps SVQ3 should be labeled as "incomplete".

[Peuj]

I can confirm the problems with the other samples as well. Exvept for the freezes.

You don't have the freezes and it works ? because on a clean machine with just MPC and ffdshow I get these messages from MPC:

Vertical400kbit.sorenson3.mov::Apple Sound Media Handler

Media Type 0:
--------------------------
Audio: 0x0000 44100Hz mono

AM_MEDIA_TYPE:
majortype: MEDIATYPE_Audio {73647561-0000-0010-8000-00AA00389B71}
subtype: Unknown GUID Name {34616D69-0000-0010-8000-00AA00389B71}
formattype: FORMAT_WaveFormatEx {05589F81-C356-11CE-BF01-00AA0055595A}
bFixedSizeSamples: 1
bTemporalCompression: 0
lSampleSize: 1
cbFormat: 70

WAVEFORMATEX:
wFormatTag: 0x0000
nChannels: 1
nSamplesPerSec: 44100
nAvgBytesPerSec: 0
nBlockAlign: 34
wBitsPerSample: 16
cbSize: 52 (extra bytes)

pbFormat:
0000: 00 00 01 00 44 ac 00 00 00 00 00 00 22 00 10 00 ....D¬......"...
0010: 34 00|00 00 00 34 69 6d 61 34 00 00 00 00 00 00 4....4ima4......
0020: 00 01 00 01 00 00 00 00 00 00 00 01 00 10 00 00 ................
0030: 00 00 ac 44 00 00 00 00 00 40 00 00 00 22 00 00 ..¬D.....@..."..
0040: 00 22 00 00 00 02 ."....

==> No sound but good image


gt4_tgs2k3.mov::Apple Sound Media Handler

Media Type 0:
--------------------------
Audio: 0x0000 44100Hz stereo

AM_MEDIA_TYPE:
majortype: MEDIATYPE_Audio {73647561-0000-0010-8000-00AA00389B71}
subtype: Unknown GUID Name {34616D69-0000-0010-8000-00AA00389B71}
formattype: FORMAT_WaveFormatEx {05589F81-C356-11CE-BF01-00AA0055595A}
bFixedSizeSamples: 1
bTemporalCompression: 0
lSampleSize: 1
cbFormat: 70

WAVEFORMATEX:
wFormatTag: 0x0000
nChannels: 2
nSamplesPerSec: 44100
nAvgBytesPerSec: 0
nBlockAlign: 68
wBitsPerSample: 16
cbSize: 52 (extra bytes)

pbFormat:
0000: 00 00 02 00 44 ac 00 00 00 00 00 00 44 00 10 00 ....D¬......D...
0010: 34 00|00 00 00 34 69 6d 61 34 00 00 00 00 00 00 4....4ima4......
0020: 00 01 00 01 00 00 00 00 00 00 00 02 00 10 00 00 ................
0030: 00 00 ac 44 00 00 00 00 00 40 00 00 00 22 00 00 ..¬D.....@..."..
0040: 00 44 00 00 00 02 .D....

==> No sound and bad image same as the others mov (just grey with some colors and artifacts)

[clsid]

Enable IMA ADPCM in ffdshow audio decoder.

[Peuj]

Enable IMA ADPCM in ffdshow audio decoder.

Oohh thanks, I miss this one. :thanks:


Anyway for me the image is not good on ftp://mplayerhq.hu/MPlayer/samples/V-codecs/SVQ3/gt4_tgs2k3.mov

[Egh]

It was two months. Now it is 3.5 months. If it was 2 years, We'd have taken over the original project. Two months was too short to take over. I still hope milan comes back someday.


Revision 2546
Author: milan_cutka
Date: Fri May 5 08:33:42 2006 UTC (4 months, 1 week ago)


Time is moving fast :P

[clsid]

Here are some screenshots of SVQ3 files that give bad picture:



http://img18.imagevenue.com/loc451/th_75648_gt4_tgs2k3.mov_122_451lo.jpg (http://img18.imagevenue.com/img.php?image=75648_gt4_tgs2k3.mov_122_451lo.jpg)http://img141.imagevenue.com/loc564/th_75648_finalfantasy_cc.mov_122_564lo.jpg (http://img141.imagevenue.com/img.php?image=75648_finalfantasy_cc.mov_122_564lo.jpg)http://img121.imagevenue.com/loc500/th_75649_helldeskmaskcable.mov_122_500lo.jpg (http://img121.imagevenue.com/img.php?image=75649_helldeskmaskcable.mov_122_500lo.jpg)http://img147.imagevenue.com/loc576/th_75650_rev_theatre_0x3839_640_dl.mov_122_576lo.jpg (http://img147.imagevenue.com/img.php?image=75650_rev_theatre_0x3839_640_dl.mov_122_576lo.jpg)

[Amour]

Weird bug.
I'm using BSPlayer + ffdshow build from the 6th September (could be rev142 from clsid, not 100% sure).

I watch a movie (xvid), I pause it, I switch Windows user, I switch back, and now the movie is upside-down! The video is working fine, I can play, stop, fast-forward, pause,... but everything is upside-down. I had to restart the player to have it back to normal.

[clsid]

Updated list of known issues:

1) VC-1, VMnc, VP5 and VP6 do not work. ffdshow isn't even placed in the DirectShow graph.
2) The "perspective correction" filter makes ffdshow crash when interpolation-mode is set to "cubic". Happens only if horizontal video size is 768 or greater.
3) SVQ3 crashes when used in combination with Haali's MP4 splitter. Possibly not a ffdshow issue.
4) Some SVQ3 files play with artifacts / color shifts / totally messed up picture.
5) kernel deinterlacer memory leak on NTSC Mpeg2 (720x480i) video (unconfirmed)
6) Output queue enabled + BSplayer = crash. May happen on some other players as well. Possibly related to specific renderers used in those players?
7) Output queue enabled + VMR9 renderless + RGB32 + specific video cards + pause = blackout problem
8) ffdshow crashes when enabling stereoscopic option on subtitles page.

Of this list only 1, 6 and 7 are specific to the ffdshow tryouts. The rest was (probably) already present in revision 2543 of ffdshow.

[clsid]

I watch a movie (xvid), I pause it, I switch Windows user, I switch back, and now the movie is upside-down! The video is working fine, I can play, stop, fast-forward, pause,... but everything is upside-down. I had to restart the player to have it back to normal.Could be some weird Windows bug. What happens if you switch back and forth twice?

[haruhiko_yamagata]

Weird bug.
I'm using BSPlayer + ffdshow build from the 6th September (could be rev145 from drevil_xxl or rev142 from clsid, not sure).

I watch a movie (xvid), I pause it, I switch Windows user, I switch back, and now the movie is upside-down! The video is working fine, I can play, stop, fast-forward, pause,... but everything is upside-down. I had to restart the player to have it back to normal.
Revision 150 - Directory Listing
Modified Thu Sep 7 13:15:17 2006 UTC (5 days, 2 hours ago) by h_yamagata
Bug fix(? alpha test) : Raw video :
When Divx5=disabled and Raw video=all supported, the Divx5 video flips. The player is MPC.
Please see if it does not have any side effects.
Does anything change before and after rev 150? Please test.

[cc979]

Confirmed. If I use Haali splitter instead of Gabest, then it even crashes.

I can confirm the problems with the other samples as well. Exvept for the freezes.

Perhaps SVQ3 should be labeled as "incomplete".

that was my problem - cheers

[foxyshadis]

Btw guys, this is the new thread where thoughts on versioning, stable releases, future plans, etc should be discussed:
http://forum.doom9.org/showthread.php?t=115869

[NoX1911]

PostProcessing doesn't work with generic build on AMD K7. Or is it just me? :)
What about a changelog in the installer? Maybe at least the last 5 builds?

[Egh]

Aha.

I found out what was the problem with "decoding page" being disabled.

The page is disabled whenever ffdshow receives uncompressed input (in my case YV12 straight from CoreAVC). Interstingly enough, h264 quality related controls are still enabled on the same page, but they don't contain any values (they can be entered though).

[nibbles]

Somebody may find this interesting. Those CLSIDs in Peuj's output are reserved by Microsoft
to specify FourCC codes. He listed these:

AM_MEDIA_TYPE:
majortype: MEDIATYPE_Audio {73647561-0000-0010-8000-00AA00389B71}
subtype: Unknown GUID Name {34616D69-0000-0010-8000-00AA00389B71}

and it also had this in the hex output:

pbFormat:
0000: 00 00 01 00 44 ac 00 00 00 00 00 00 22 00 10 00 ....D¬......"...
0010: 34 00|00 00 00 34 69 6d 61 34 00 00 00 00 00 00 4....4ima4......



According to these documents at MSDN, (http://windowssdk.msdn.microsoft.com/en-us/library/ms783788.aspx) any CLSID of the form
XXXXXXXX-0000-0010-8000-00AA00389B71
is reserved, and the Xs are 4 hex numbers backwards because
the registry is little endian, that spell out the FourCC code when
converted back to ASCII and reversed.

Soooo.... following this ASCII chart, (http://everything2.com/index.pl?node_id=159986)

This CLSID 34616D69-0000-0010-8000-00AA00389B71
has hex digits 34 61 6D 69
which spells 4 a m i
so the FourCC is: ima4
I came across this while trying to determine whether HuffYUV should
have its ClassGUID specified in its inf or just use CLASS=Media, a
nasty little puzzle still for me.

[haruhiko_yamagata]

Revision 173
"Use queue only in:" "mplayerc.exe;" by default.
Revision 171
msvcr80.dll and msvcr71.dll installer.

If vcredist_x86 is not packed or the installation of vcredist_x86 failed,
and if MSVCR8 is defined,
the installer try to install msvcr80.dll as private assembly.

Installation of private assembly have version check, reference count controls, etc.
The version of VC80.CRT is checked before executing vcredist_x86.exe.

MSVCR8 is defined by default.
Please read the header of ffdshow.nsi carefully.
Good bye runtime library errors!?
Rev 173 should contribute to stability.

[haruhiko_yamagata]

The change to resource.h causes weird thing, if MSVC's intermediate files aren't cleaned. Please clean and rebuild.

[DSP8000]

Good work :)

Any other changes?
I just finished reading/browsing the SVN and it seems to me that again ffdshow it is just getting more and more features.

Now, how about first fixing/removing the bugs that have been reported by clsid and others.IMHO it is better to have one stable build to refer to than 100's revisions updated daily:) , and confusing the "average Joe" which one should I get.
It's been a long time since we had a stable (nominated) build.

Anyways, great to see that the development of ffdshow is in full speed ahead:)

DSP8000

[Egh]

....


Another minor issue with ffdshow (just like the previous one reported by me, it's more amusing than actually serious).

If there's an error in avisynth script in ffdshow, the error message is displayed on top of video. That's ok, but interestingly enough, even if I uncheck the whole avisynth page in ffdshow, the error message still remains.

[Amour]

Does anything change before and after rev 150? Please test.
Sorry, I'm not able to reproduce the flip bug again (any ffdshow revision). When I switched the Windows user, I wasn't on the computer anymore to see what was done by the other user.

[Amour]

Today's tests (with an xvid file):

[drevil_xxl] rev 155 (default settings = Output queueing ON)
BSPlayer: crash
WMP: ok
MPC: using the seekbar, the video will freeze a few seconds (but not the sound)

[VM9] rev162 (default settings = Output queueing ON)
BSPlayer: crash
WMP: ok
MPC: using the seekbar, the video will freeze a few seconds (but not the sound)

[clsid] rev164 (default settings = Output queueing OFF)
BSPlayer: ok
WMP: ok
MPC: ok

[h-yamagata] rev173 (default settings = Output queueing ON)
BSPlayer: ok
WMP: ok
MPC: using the seekbar, the video will freeze a few seconds (but not the sound)

We only have problems with this Output queueing feature: is it really usefull and user friendly? Instead of applying haruhiko_yamagata's rev173 MPC-only-by-default solution, I advice to use clsid's feature-disabled-by-default solution.

[KoD]

I still wonder how can something that was not designed to be multithreaded from the beginning can be made to be multithreading-safe ? I mean, how feasible is this ?

[LoRd_MuldeR]

I still wonder how can something that was not designed to be multithreaded from the beginning can be made to be multithreaded ? I mean, how feasible is this ?

They made the filter-chain multi-threaded, so each filter can run in it's own thread. And the resizer also has it's own thread. Furthermore some of the decoders used by ffdshow are multi-threaded now too. Except some bugs, that always can slip in, there should be nothing that should prevent multi-threaded ffdshow from working properly...

[clsid]

Updated list of known issues:

1) VC-1, VMnc, VP5 and VP6 do not work. ffdshow isn't even placed in the DirectShow graph.
2) The "perspective correction" filter makes ffdshow crash when interpolation-mode is set to "cubic". Happens only if horizontal video size is 768 or greater.
3) SVQ3 crashes when used in combination with Haali's MP4 splitter. Possibly not a ffdshow issue. (reported by Peuj)
4) Some SVQ3 files play with artifacts / color shifts / totally messed up picture. (reported by Peuj)
5) Kernel deinterlacer memory leak on NTSC Mpeg2 (720x480i) video. (unconfirmed)
6) Output queue enabled + BSplayer = crash. May happen on some other players as well. Possibly related to specific renderers used in those players? Current workaround: queue is only used in MPC.
7) Output queue enabled + VMR9 renderless + RGB32 + specific video cards + pause = blackout problem
8) ffdshow crashes when enabling stereoscopic option on subtitles page. (reported by drevil_xxl)
9) If ffdshow gets uncompressed input, then the decoding options page is disabled. However the H.264 quality controls are not disabled. (reported by Egh)
10) If there's an error in avisynth script in ffdshow, the error message is displayed on top of video. That's ok, but interestingly enough, even if I uncheck the whole avisynth page in ffdshow, the error message still remains. (reported by Egh)
11) Video may freeze for a few seconds in MPC when seeking if queue is enabled. (reported by Amour)

Of this list only 1, 6, 7, 9 and 11 are specific to the ffdshow tryouts. The rest was (probably) already present in revision 2543 of ffdshow.

[Amour]

Current workaround: queue is only used in MPC.
As I said earlier today, queueing is causing problems even with MPC: when using the seekbar to change the position, the image will often freeze for a few seconds, but not the sound. And that's very annoying.

What is the meaning, the purpose of queueing? I do not understand this functionality/feature.

[KoD]

The problem in multithreading is when all those threads have to communicate with one another.

[clsid]

As I said earlier today, queueing is causing problems even with MPC: when using the seekbar to change the position, the image will often freeze for a few seconds, but not the sound. And that's very annoying.

What is the meaning, the purpose of queueing? I do not understand this functionality/feature.
What the queue does is queue video frames before they get send to the renderer. It also executes the video renderer on worker thread. The queue can help decrease the number of frame drops when running at near 100% cpu usage. It is only useful on multicore cpus and perhaps also on P4s with hyperthreading. Haruhiko can give you a better explanation.

Just disable the option if it gives you troubles. I consider it both an expert and experimental option that should be disabled by default.

[Px]

Revision 150 - Directory Listing
Modified Thu Sep 7 13:15:17 2006 UTC (5 days, 2 hours ago) by h_yamagata
Bug fix(? alpha test) : Raw video :
When Divx5=disabled and Raw video=all supported, the Divx5 video flips. The player is MPC.
Please see if it does not have any side effects.
Does anything change before and after rev 150? Please test.
That is old bug, it appears even if Raw video is also disabled. Reason - when system don't find needed DSH-codec, it search for suitable VfW-codec, and if find it - use him through DSH/VfW layer. Disable support for DivX5 in ffdshow VfW codec, and bug will dissapear....

[Jeremy Duncan]

Queue output samples, when checked, freezes Media Player Classic when you use the seek bar.

I'm using the August 21, 2006 version of FFdshow, and Media Player Classic 6.4.9.0 and Dscaler 5008 with Reclock.

[deadfones]

Today's tests (with an xvid file):

<snip>

[clsid] rev164 (default settings = Output queueing OFF)
BSPlayer: ok
WMP: ok
MPC: ok

<snip>

We only have problems with this Output queueing feature: is it really usefull and user friendly? Instead of applying haruhiko_yamagata's rev173 MPC-only-by-default solution, I advice to use clsid's feature-disabled-by-default solution.

clsid build 164 is good. Instant seeks with the seekbar in MPC and where I used to get video stalling using the seekbar, at least now it waits a little, stutters, and catches up.

Without output queuing, I get frame drops (more like play a second then stutter) on divx3, divx5, and xvid, at least. This is with a P4c 3.0ghz and ddr400. My CPU isn't maxed out, it's actually at about 15%, but the frames still drop. It may or may not have something to do with my multiple display configuration (running latest nvidia drivers).

Edit: 15% with HT, so about 30% actually.

[DSP8000]

clsid build 164 is good. Instant seeks with the seekbar in MPC and where I used to get video stalling using the seekbar, at least now it waits a little, stutters, and catches up.

I can confirm this, build 164 works good.
However, build 181 clsid instantly crashes MPC and ffdshow video decoder config.Even after reinstall & reset settings.I think Output queueing is causing this.

[haruhiko_yamagata]

We only have problems with this Output queueing feature: is it really usefull and user friendly? Instead of applying haruhiko_yamagata's rev173 MPC-only-by-default solution, I advice to use clsid's feature-disabled-by-default solution.
Don't forget ffdshow is a free software. Be polite.

It's a bug of MPC, not queue's. Even old ffdshow without queue the symptom is here, though the frequency is lower.

Simply uncheck MPC's internal source filter "AVI" and everything goes fine.

[clsid]

I don't have any seeking problems at all when the queue is turned on. Tried with both the internal AVI splitter on and off in MPC (rev 611).

Perhaps the issue appears on multicore or hyperthreading cpus only?

[haruhiko_yamagata]

I don't have any seeking problems at all when the queue is turned on. Tried with both the internal AVI splitter on and off in MPC (rev 611).

Perhaps the issue appears on multicore or hyperthreading cpus only?
It's reproducible on sigle CPU, too.
The video will freeze a few seconds. It's not a crash nor hang up. I know it's time stamp issue, I'll try to work around the bug when I have time.

[haruhiko_yamagata]

I can confirm this, build 164 works good.
However, build 181 clsid instantly crashes MPC and ffdshow video decoder config.Even after reinstall & reset settings.I think Output queueing is causing this.
It's not reproducible for me. Why do you think queue is the culplit?

[haruhiko_yamagata]

Queue output samples, when checked, freezes Media Player Classic when you use the seek bar.

I'm using the August 21, 2006 version of FFdshow, and Media Player Classic 6.4.9.0 and Dscaler 5008 with Reclock.
Such a complicated setting with one line report. What am I supposed to do?

[Amour]

We only have problems with this Output queueing feature: is it really usefull and user friendly? Instead of applying haruhiko_yamagata's rev173 MPC-only-by-default solution, I advice to use clsid's feature-disabled-by-default solution.Don't forget ffdshow is a free software. Be polite.
Sorry.

Was giving an advice, reporting a problem, didn't mean to hurt you.

[Amour]

I would like to update the French translation of ffdshow (some phrases are displayed in English). Where/What/How should I do?

[haruhiko_yamagata]

I would like to update the French translation of ffdshow (some phrases are displayed in English). Where/What/How should I do?
Create a patch and post it here (http://sourceforge.net/tracker/?group_id=173941&atid=867362).

[KoD]

This might be or not related to some audio issues when resuming playback after pausing on WinXpSP2: KB 920872 (http://support.microsoft.com/?kbid=920872).

[n3w813]

On these 2 ffdshow versions.....(and maybe on other new builds, haven't tried them all yet)

H-Yamagata's
ffdshow-20060913-rev173-Q.exe

CLSID's
ffdshow_rev181_20060913_clsid.exe

When you enable 'Automatic Preset Loading' in the Image Settings menu, and exit, you will get an exception error when you try to start the ffdshow video configuration program again.

To disable auto preset, I go in the registry and change the value for...
HKCU\Software\GNU\ffdshow\autoPresets from 1 to 0
then ffdshow configuration utility will open correctly.

Can somebody else verify this?

[clsid]

Confirmed.

[_xxl]

http://minkymomo.info/~meroko_deathnote/tmp/codec_test/glider/glider_vp6.2.0.10.avi
please test!

[_xxl]

On these 2 ffdshow versions.....(and maybe on other new builds, haven't tried them all yet)

H-Yamagata's
ffdshow-20060913-rev173-Q.exe

CLSID's
ffdshow_rev181_20060913_clsid.exe

When you enable 'Automatic Preset Loading' in the Image Settings menu, and exit, you will get an exception error when you try to start the ffdshow video configuration program again.

To disable auto preset, I go in the registry and change the value for...
HKCU\Software\GNU\ffdshow\autoPresets from 1 to 0
then ffdshow configuration utility will open correctly.

Can somebody else verify this?

ffdshow_rev164_20060911_clsid and ffdshow-tryouts-rev155-AMD-XP-K8 are working for me.

[shae]

Btw guys, this is the new thread where thoughts on versioning, stable releases, future plans, etc should be discussed:
http://forum.doom9.org/showthread.php?t=115869
foxyshadis! The next thing, you might suggest having a dedicated sub-forum! :) (Does anyone really think it's NOT better? Do the admins object?)

Updated list of known issues:Why not use the SF bug tracker? Much more organized, and maybe easier, though I'm not a fan of the SF structure in general.

---

BTW, how come ffdshow doesn't have a dynamic range compessor? It's very much needed for AC3 and the likes, but also for stereo in cases this was converted from AC3 without adjustments.

[KoD]

BTW, how come ffdshow doesn't have a dynamic range compessor? It's very much needed for AC3 and the likes, but also for stereo in cases this was converted from AC3 without adjustments.
What do you think Normalize does ?

And you can always enable drc (better than how normalize does its thing) on AC3 and DTS streams since they're meant to work with drc anyway. (it's on the codecs page when you select either ac3 or dts)

[foxyshadis]

Normalize and DRC solve totally different problems, so neither can substitute for the other. (Though they should be used together.) DRC is very helpful for AAC/MP3 as well, but you have to use a winamp2 filter for it, and those make ffdshow very unstable. There's nothing special about AC3/DTS concerning DRC, other than having in-stream values, just that their decoders include the capability. (The in-stream values tend to suck if the encoding was done cheaply anyway.)

[shae]

What do you think Normalize does ?Normalizes? :) I.e., maximizes peaks while keeping the wave proportions. That's not very helpful in most cases.

And you can always enable drc
It's on, but doesn't appear to do much (for that reason or another?). The sound shifts from whisper silent parts to unbearable loudness, so I'm constantly playing with the volume.

What's needed is something configurable.

[Jeremy Duncan]

Such a complicated setting with one line report. What am I supposed to do?

I'm using Media Player Classic 6.4.9.0
FFdshow August 21, 2006 found at afterdawn.com
I'm using Dscaler 5008 audio and video decoder, and reclock in Media Player Classic External Filters.

Dscaler is using YV12
FFdshow is inputting all supported, outputing YV12.
Blur & NR, Denoise3d: Luma: 0.00, Chroma: 2.00, Time: 4.00, HQ checked.
FFdshow is using Resize & Aspect:
Multiply by Resize 2.000, No aspect ratio correction, Lanczos, No luma sharpening, accurate rounding unchecked.
My monitor is set to a refresh rate of 75 Hz.
Reclock is doing a custom adaption set to 60.000 fps, the reclock light is green.
When Reclock is set to 25fps and the refresh rate is 75 Hz I get the same problem.
And lastly, when queue output samples is checked in FFdshow, when I use the seek bar in Media Player Classic, it freezes and I need to use the task manager to end the Media Player Classic process.

Using no setting in FFdshow but queue output samples. Input raw video all supported, output YV12. It still freezes when I use the seek bar in Media Player Classic.

The internal filter in Media Player Classic: "AVI" is unchecked.
The freezing problem is there with it checked or unchecked.

[_xxl]

On these 2 ffdshow versions.....(and maybe on other new builds, haven't tried them all yet)

H-Yamagata's
ffdshow-20060913-rev173-Q.exe

CLSID's
ffdshow_rev181_20060913_clsid.exe

When you enable 'Automatic Preset Loading' in the Image Settings menu, and exit, you will get an exception error when you try to start the ffdshow video configuration program again.

To disable auto preset, I go in the registry and change the value for...
HKCU\Software\GNU\ffdshow\autoPresets from 1 to 0
then ffdshow configuration utility will open correctly.

Can somebody else verify this?
http://rapidshare.de/files/33170803/FFdshow-Tryouts-200609015-rev172.exe.html
FFdshow-Tryouts-200609015-rev172 works ok.
rev 173 is not working?

[clsid]

rev173 contains a small change to the presets code. Haruhiko should be able to fix the bug.

In rev184 ffdshow finally gets used in the DirectShow graph for a VP6 file. However, the decoded video is b0rked with weird colors (mostly green) and artifacts. Same for VP5.
I haven't managed to play FLV files containing VP6 video yet. I even tried adding FLV4 FourCC which MPC uses to play such files.

[haruhiko_yamagata]

I'm using Media Player Classic 6.4.9.0
FFdshow August 21, 2006 found at afterdawn.com
I'm using Dscaler 5008 audio and video decoder, and reclock in Media Player Classic External Filters.

Dscaler is using YV12
FFdshow is inputting all supported, outputing YV12.
Blur & NR, Denoise3d: Luma: 0.00, Chroma: 2.00, Time: 4.00, HQ checked.
FFdshow is using Resize & Aspect:
Multiply by Resize 2.000, No aspect ratio correction, Lanczos, No luma sharpening, accurate rounding unchecked.
My monitor is set to a refresh rate of 75 Hz.
Reclock is doing a custom adaption set to 60.000 fps, the reclock light is green.
When Reclock is set to 25fps and the refresh rate is 75 Hz I get the same problem.
And lastly, when queue output samples is checked in FFdshow, when I use the seek bar in Media Player Classic, it freezes and I need to use the task manager to end the Media Player Classic process.

Using no setting in FFdshow but queue output samples. Input raw video all supported, output YV12. It still freezes when I use the seek bar in Media Player Classic.

The internal filter in Media Player Classic: "AVI" is unchecked.
The freezing problem is there with it checked or unchecked.
I found a bug related to queue and seek. The internal filter in Media Player Classic "AVI" does not call Run at the expected timing. It may not be good, but a bug seems to be in ffdshow side. The bug should not be too difficult to fix, though I'm not sure if the bug is related to your problem.

I have not tryed a reproduce test. Please keep wating here and when the bug is fixed, please test again and report.

[haruhiko_yamagata]

Recent builds including clsid's and videomixer9's genelic builds do not work on Pentium-MMX.

My "SSE" build ffdshow-20060730-Q.exe works. My "SSE" build simply use "official" makefile and vcproj. I label my build "SSE" just because they are not tested on non SSE CPUs.

So it's not a build problem nor installer's. Perhaps some code in ffdshow-tryouts. Could you confirm this?

[igor1st]

Recent builds including clsid's and videomixer9's genelic builds do not work on Pentium-MMX.

So it's not a build problem nor installer's. Perhaps some code in ffdshow-tryouts. Could you confirm this?
Revision 71:
"-O3 -march=i586" -> "-O2 -march=i686"

From manual:
i586 - Intel Pentium CPU with no MMX support.
i686 - Intel PentiumPro CPU.
pentium-mmx - Intel PentiumMMX CPU based on Pentium core with MMX instruction set support.

[clsid]

I will make a test build with "-march=pentium-mmx -mtune=pentium2".

test build (http://www.mytempdir.com/930756)

[KoD]

The end result of Normalize is range compression. And it's dynamic if it keeps autoadjusting itself during playback to avoid clipping. Which means the quality is not that great since you'll already have clipping when these adjustments have to be made. Normalization done during playback is quite bad because of this and very dependent on the audio material it has to work with. If the audio stream has a wide dynamic range already which varies a lot during playback, normalization is quite useless. There can be artifacts like in this case: "two people are speaking and suddenly a phone rings somewhere in the room while the people keep on speaking - if clipping happens, normalization will reduce the amplification gain and suddenly the voices of those people are much softer than they were. And if the ring tone is really loud it can happen you won't hear the people speaking at all".

DRC in AC3 and DTS is great if the audio stream was compressed with proper hints. Some of them aren't. DRC hints are much better than normalization during playback for obvious reasons (no real-time analysis for adjustments constraint). One can also use those hints and implement various degrees of range compression, like those "quiet/normal/noisy environment" profiles most commercial decoders have.

[Egh]

Normalization done during playback is quite bad because of this and very dependent on the audio material it has to work with. If the audio stream has a wide dynamic range already which varies a lot during playback, normalization is quite useless.



I still think that AC3Filter dynamic normalization is superior. I never use ffdshow's one, on my system I make AC3Filter accept decoded raw audio streams from ffdshow.

[KoD]

AC3filter and ffdshow normalization work in the same way since they are both based on the same source code. Here is a more in-depth explanation: ac3filter audio theory docs (http://ac3filter.net/doc/ac3filter/ac3filter_eng.html#theory).

[haruhiko_yamagata]

I will make a test build with "-march=pentium-mmx -mtune=pentium2".

test build (http://www.mytempdir.com/930756)
Great. It works on Pentium-MMX.

[cc979]

just quick test


tested using elephant dream_hd xvid/ac3 (8mbs stream mostly)
using 07-07-06 hali splitter

clsid
ffdshow_rev187_20060915_clsid.exe
null: User: 8s, kernel: 0s, total: 8s, real: 8s, fps: 76.4, dfps: 73.7
vmr9: User: 8s, kernel: 0s, total: 8s, real: 10s, fps: 77.8, dfps: 59.8
ovrl: User: 8s, kernel: 0s, total: 9s, real: 9s, fps: 72.4, dfps: 70.3

clsid
ffdshow_rev187_20060915_clsid_generic.exe
null: User: 8s, kernel: 0s, total: 8s, real: 9s, fps: 74.5, dfps: 70.1
vmr9: User: 8s, kernel: 0s, total: 8s, real: 11s, fps: 75.0, dfps: 58.1
ovrl: User: 8s, kernel: 0s, total: 9s, real: 9s, fps: 72.7, dfps: 70.1

clsid
ffdshow_rev181_20060913_clsid.exe
null: User: 8s, kernel: 0s, total: 8s, real: 8s, fps: 75.7, dfps: 74.6
vmr9: User: 8s, kernel: 0s, total: 8s, real: 11s, fps: 75.4, dfps: 59.0
ovrl: User: 9s, kernel: 0s, total: 9s, real: 9s, fps: 70.2, dfps: 68.7

videomixer9
ffdshow-tryouts-rev155.exe
null: User: 8s, kernel: 0s, total: 8s, real: 8s, fps: 81.6, dfps: 80.7
vmr9: User: 7s, kernel: 1s, total: 9s, real: 10s, fps: 72.0, dfps: 61.2
User: 8s, kernel: 0s, total: 8s, real: 8s, fps: 78.2, dfps: 75.0

drevil_xxl
ffdshow-tryouts-rev155-AMD-XP-K8.exe
null: User: 8s, kernel: 0s, total: 8s, real: 8s, fps: 79.6, dfps: 78.7
vmr9: User: 8s, kernel: 1s, total: 9s, real: 10s, fps: 68.3, dfps: 59.8
ovrl: User: 8s, kernel: 0s, total: 8s, real: 8s, fps: 76.8, dfps: 73.0

[videomixer9]

for GCC 3.4.5 add -frename-registers -fweb and -funit-at-a-time. ICL should have global optimizations turned on (/Qip and /Og) and be set to O3 optimization, also for the .ax turn of loop unrolling (/Qunroll0). Any kind of automatic vectorization will make things usually slower, so leave off /Qx or /Qax switches and also don't use /arch. Link Time Code Generation is not really worth it for the .ax file.

[foxyshadis]

The end result of Normalize is range compression. And it's dynamic if it keeps autoadjusting itself during playback to avoid clipping. Which means the quality is not that great since you'll already have clipping when these adjustments have to be made. Normalization done during playback is quite bad because of this and very dependent on the audio material it has to work with. If the audio stream has a wide dynamic range already which varies a lot during playback, normalization is quite useless. There can be artifacts like in this case: "two people are speaking and suddenly a phone rings somewhere in the room while the people keep on speaking - if clipping happens, normalization will reduce the amplification gain and suddenly the voices of those people are much softer than they were. And if the ring tone is really loud it can happen you won't hear the people speaking at all".

DRC in AC3 and DTS is great if the audio stream was compressed with proper hints. Some of them aren't. DRC hints are much better than normalization during playback for obvious reasons (no real-time analysis for adjustments constraint). One can also use those hints and implement various degrees of range compression, like those "quiet/normal/noisy environment" profiles most commercial decoders have.
That's not DRC, that's dynamic (windowed) peak normalization. DRC is peak amplitude reduction, narrowing the range between soft and loud sounds on a much smaller timescale (limiters are almost instantaneous on/off), usually including some gain/peak normalization to bring the total volume back to maximum. Reference (http://en.wikipedia.org/wiki/Audio_level_compression). No special information is needed in the bitstream, although it can help prevent over- or under-compression, especially is the player is too slow to pick up peaks in time. That's why DTS and AC3 include the DRC hints in the bitstream, along with normalization hints.

I'd like to find some code that can be included in ffdshow, because I don't fancy writing my own. Most of the open source compressors utilize bitstream hints so they don't need a full implementation, which won't be any help to processing raw audio.

[igor1st]

AC3filter and ffdshow normalization work in the same way since they are both based on the same source code. Here is a more in-depth explanation: ac3filter audio theory docs (http://ac3filter.net/doc/ac3filter/ac3filter_eng.html#theory).
ffdshow uses method2 from mplayer (use several samples to smooth the variations via the standard weighted mean over past samples). And this is not the same as DRC or one-pass normalize in AC3Filter.

[KoD]

Range compression is still range compression no matter what algorithm you use to achieve it, people.

"use several samples to smooth the variations via the standard weighted mean over past samples". Now I understand why it does so badly on recovering from loud sounds... it treats them as too widely dispersed variations.

[TFM_TheMask]

I think I found a bug in the latest tryouts.

It seems like when your source is *.MOV with AAC 6 channel audio (for example the HD trailers from Apple (http://www.apple.com/trailers/)) and you want to encode the audio to AC3 with the ffdshow AC3 encoder there are a lot of downmix overflows. The sound seems to be distorted as if the volume is to high.

Now I have installed back an old version of ffdshow (dated 24-11-2005) and the sound is Ok now when doing the AC3 encoding. No downmix overflows.

[Jeremy Duncan]

I found a bug related to queue and seek. The internal filter in Media Player Classic "AVI" does not call Run at the expected timing. It may not be good, but a bug seems to be in ffdshow side. The bug should not be too difficult to fix, though I'm not sure if the bug is related to your problem.

I have not tryed a reproduce test. Please keep wating here and when the bug is fixed, please test again and report.

When you release the patched version of FFdshow. Please post the download link too.

[LoRd_MuldeR]

When you release the patched version of FFdshow. Please post the download link too.

Look at the Changelog (http://svn.sourceforge.net/viewvc/ffdshow-tryout/?view=log) to see what has been fixed recently and then get the latest build here (http://sourceforge.net/project/showfiles.php?group_id=173941).

[Egh]

Range compression is still range compression no matter what algorithm you use to achieve it, people.


Yes, but exact implementation is different. I prefer AC3Filter's one to ffdshow. I think AC3F recovers quicker from loud sounds. And overall does more balanced job, i think. (of course it's largely subjective, but I asked Milan last year to use AC3Filter's one in ffdshow :P)

And i don't use DRC or one-pass btw.

[foxyshadis]

AC3Filter does seem to have a much better normalizer and compressor, igor1st pointed it out to me and is working with the code. It would be quite nice to have that integrated instead of mplayer's mediocre version.

[ckjnigel]

I searched to no avail for new trunk (?) builds optimized for 64-bit Windows. Is anybody making them?

[pdanpdan]

I also have the same problems with autoloading presets on ffdshow video configure on all versions after rev 173

"When you enable 'Automatic Preset Loading' in the Image Settings menu, and exit, you will get an exception error when you try to start the ffdshow video configuration program again."

[Kostarum Rex Persia]

@

videomixer9 , haruhiko_yamagata, clsid and drevil_xxl

Can you guys, join your forces and finally make some decent 64-bit optimized FFDSHOW. Only celtic_druid made one 64-bit build, long time ago.

This is a big feature request.

[Px]

As I remember, the 64-bit ffdshow needs 64-bit player, do you know such players?

[thuan]

There's one as fas as I know graphedit64

[foxyshadis]

KRP, file an enhancement request on the sourceforge page, so that it's not forgotten, but I can tell you now that it's going to be a low priority request for some time, at least until vista comes out and the rest of the world wakes up to 64-bit.

[videomixer9]

GCC cannot output win64 binary format, as long as it cannot do that you won't get any real stable x64 builds. However there is a binutils that should be able to output it now, need to make GCC use that stuff though now. A actual build will be still far away though.

[Px]

What about ICL or MSVC?

[LoRd_MuldeR]

What about ICL or MSVC?

AFAIK at least the libavcodec needs to be build with GCC...

[Px]

AFAIK at least the libavcodec needs to be build with GCC...

Revision 183 - Directory Listing
Modified Thu Sep 14 17:33:13 2006 UTC (2 days, 20 hours ago) by drevil_xxl

libavcodec.dll can be compiled with MINGW 3.4.x, 4.0.x, ICL9 and MSVC80.
:rolleyes:

[videomixer9]

well then compile yourself if you know everything better ... maybe you'll see it then ...

I guess I quit building ffdshow with this too, too many tards involved.

[LoRd_MuldeR]

:rolleyes:

Even if it can be done, that doesn't mean it works fast and/or stable...
At the moment all ffdshow builds use GCC for libavcodec/libmplayer and I think they have a reason for this ;)
VM9 even uses good old GCC 3.4.5, because this seems to be most reliable...


BTW: Is there any assembler code in ffdshow/libavcodec for AMD64 architecture yet? If not so, then speed improvements for x64 builds won't be that big, eh? Maybe even slower ???

[foxyshadis]

If you build lavc with MSVC or ICL you get none of the assembly optimizations, and speed plummets. It's just a non-starter. (Good for debugging as long as the problem isn't in the asm, though.) The only thing drevil_xxl did was fix some broken compatibility issues in the C side.

I'm not sure about x64 assembly; I thought there was but I don't see it in ffmpeg.

[akupenguin]

gcc can compile the same inline asm to either ia32 or x86_64. Yes, you'd need separate asm to take full advantage of the extra registers, but the limit is usually the number of mmx registers not the gprs. So you won't see x86_64 specific asm in ffmpeg until one of the developers gets an intel x86_64 box and starts adding sse2.

[LoRd_MuldeR]

gcc can compile the same inline asm to either ia32 or x86_64. Yes, you'd need separate asm to take full advantage of the extra registers, but the limit is usually the number of mmx registers not the gprs. So you won't see x86_64 specific asm in ffmpeg until one of the developers gets an intel x86_64 box and starts adding sse2.

To sum up: x64 builds won't be any faster until new asm code with x64 specific optimizations is added. And this will presumably not be in the near future...

[Kostarum Rex Persia]

Ok, thank you guys. It's pity that ffdshow doesn't have decent 64-bit version.

Can anyone start working on changing asm code for 64-bit? It's no matter when that hard job will be completed, it's very important to start.

[Episode]

Why is it so important to you have everything as 64-bit -versions? I bet it won't have any affect on speeds even if someone would make such build. If you have followed this thread for couple of months, I'm sure you'll have noticed how little different compiling optimizations affect on overall speed.

Also, it's quite annoying to see you whine in every codec/program related thread about these 64-bit versions and say things like "you need to do this for me right away because it's the most important thing in the world for me". Why don't you learn to code yourself and make your own 64-bit versions if it's so goddamn important to you. People are developing fine things like ffdshow on their own freetime without any payment and if they don't see any benefit on starting such a huge project for so minor speed increase, they won't do it, no matter how hard you try to order them to do so.

[akupenguin]

To sum up: x64 builds won't be any faster until new asm code with x64 specific optimizations is added.
Quite the reverse. I'm saying that we don't need any new asm for amd64 (only for intel). linux64 builds of ffmpeg are already ~10% faster than linux32. win64 will get the same benefit as soon as gcc can compile for win64.

[LoRd_MuldeR]

Quite the reverse. I'm saying that we don't need any new asm for amd64 (only for intel).
Hmm, interesting! Why does the old ASM code, which was written for x86 and thus doesn't use any x64 specific features, run 10% faster when compiled for x64? And why do we need new ASM code for Intel's 64-bit CPUs while the old code works fine on AMD's 64-Bit CPUs? Shouldn't Intel's implementation of x64 (EM64T) be compatible with AMD64 ???

:confused:

[clsid]

Perhaps because in some cases less operations are required to move the same amount of data to/from memory since the registers are twice as big?

[LoRd_MuldeR]

Perhaps because in some cases less operations are required to move the same amount of data to/from memory since the registers are twice as big?

But if we use the *unchanged* old ASM code (and *not* new ASM code with x64 optimizations), then it will still use the same operations that were used on x86. So how can it then benefit form the larger registers? Or will the compiler automatically "change" the ASM code to use x64 operations?

[akupenguin]

Why does the old ASM code, which was written for x86 and thus doesn't use any x64 specific features, run 10% faster when compiled for x64?
* gcc inline asm, as used in ffmpeg, lets the compiler do register allocation. In order to be compatible with ia32 w/ PIC, we refrain from directly using more than 6 registers at once. But to free up even that many, the compiler has to spill any other values it might be using before the asm block, whereas in x64 it can keep them in the other 9 regs. Also, some arguments (usually any scalar value except pointers) can work either as registers or directly from memory, so we can tell gcc to allocate a register for that if there is one available.

* Calling convention. The x64 abi uses fastcall(6), i.e. passes the first 6 function arguments in registers, whereas ia32's cdecl uses the stack. (Yes, one could use fastcall on ia32 also, but that's only 2 registers, and libavcodec is a library so it has to match the applications' abi.)
btw, win64 won't gain as much as linux64 does from the calling convention, since win64 only uses fastcall(4) and also imposes some other overhead on function calls.

And why do we need new ASM code for Intel's 64-bit CPUs while the old code works fine on AMD's 64-Bit CPUs? Shouldn't Intel's implementation of x64 (EM64T) be compatible with AMD64 ?
It's compatible, but not the fastest way.
(This applies to both ia32 and x64) On amd, mmx is usually faster than sse2 unless your algorithm specifically needs to fit 16 bytes in one xmm register (e.g. 8x8 dct benefits from sse2, but it's one of the few such functions in video codecs.) Intel has optimized for sse2 rather than mmx. But most of the ffmpeg developers use amd, so there isn't much sse2 code in ffmpeg.
x64 provides extra xmm registers, but not extra mmx registers. If you're on intel and using sse2, then x64 allows you to unroll some more, or avoid spilling. ia32 sse2 asm will still work, and it will still be faster than on ia32, but not by as much as if you rewrite the asm to take advantage of that. (Well, vector instrinsics could do that automatically, but so far they're only used on ppc/altivec, and don't work as well as inline asm on x86.) Amd64 of course also has the extra xmm registers, but mmx is still usually faster.

The other factor that would require new asm is sse4.

[Px]

well then compile yourself if you know everything better
If i could do that, then I don't ask about such builds in forum :)

[Px]

Even if it can be done, that doesn't mean it works fast and/or stable...
I agree with that, but someone tests this?

BTW: Is there any assembler code in ffdshow/libavcodec for AMD64 architecture yet? If not so, then speed improvements for x64 builds won't be that big, eh? Maybe even slower ???
It can be slower in x64 build because of REX prefix in such mode....

[Px]

If you build lavc with MSVC or ICL you get none of the assembly optimizations, and speed plummets.

Is there any such build to make a speed comparsion? For ICL interests not only native build, but also optimised. Preffered -QxP or -QxB build.
And why assembly optimisations must depend on compiler?

[LoRd_MuldeR]

@akupenguin:
Think I understood. Thanks for explanation.

[Kostarum Rex Persia]

akupenguin, can you explain how celtic_druid made his x64 build. Did he used some changed code, or not?

[BlindWanderer]

There was a change between rev164 & rev181 that has resulted in ffdshow not working at all (and subsequent builds). Cannot even bring up the video decoder config box (via the start menu), crashes instead. When trying to play video it crashes (video does not load at all). The audio config page loads no prob.

running xpsp2, amd athlon 64 X2 4200+

EDIT:
for grins, i backed up the dll's from the 195 revision; installed the 164 revision, and dropped in the new dll's from 195; haven't had any problems yet...
So what ever the problem is, it's in ffdshow.ax

[haruhiko_yamagata]

8) ffdshow crashes when enabling stereoscopic option on subtitles page. (reported by drevil_xxl)
I think it should have been working in old ffdshow.
I have ffdshow-20040806.exe and ffdshow-20050303-sse.exe.
ffdshow-20040806.exe don't have strereoscopic option.
ffdshow-20050303-sse.exe have stereoscopic and does not crash, but it's not viewable.
Because building from old svn is hard, could someone help me?

[haruhiko_yamagata]

btw, what is stereoscopic subtitle?
With old ffdshow, with stereoscopic subtitle on, I've got this.

http://i9.tinypic.com/2aglufd.jpg

Is this what it supposed to be?

[Egh]

btw, what is stereoscopic subtitle?


Thats has been mystery for me since I first discovered this option.

I'd suggest to remove that feature completely and thus rectify this bug :P (Same probably with cubic interpolation in the perspective correction section).


There was a change between rev164 & rev181 that has resulted in ffdshow not working at all (and subsequent builds).
...
running xpsp2, amd athlon 64 X2 4200+


Thats not quite true. In spite of several recent reports, I still have no serious problems whatsoever with any of the last clsid builds. (181, 187 and 195 to be precise) I'm on XP SP2 too, just single-core A64.

[haruhiko_yamagata]

Thats has been mystery for me since I first discovered this option.

I'd suggest to remove that feature completely and thus rectify this bug.
Yes, if there's no objection, I will delete the option.

Same probably with cubic interpolation in the perspective correction section
I'm trying to debug it now, but with low motivation. Considering the time, it may be wise to hide the option. Everybody agree?

[Jeremy Duncan]

Works fine now using Queue output samples in Media Player Classic 6.4.9.0 using FFdshow Avisynth with MT and without MT and without Avisynth.
Input All Supported, Output YV12
VMR7 Windowed renderer.
Reclock.
PowerDVD 6.0 Video Render, using forced weave.
DScaler 5008 Audio Render.
Avisynth 2.5.7 Alpha 2
FFdshow Avisynth script:
SPresso(limit=8, limitC=8, bias=100, biasC=100, RGmode=1, RGmodeC=17)
or
MT("SPresso(limit=8, limitC=8, bias=100, biasC=100, RGmode=1, RGmodeC=17)",3)

[Jeremy Duncan]

ZoomPlayer 451pro
Overlay Mixer Video Render
Reclock.
PowerDVD 6.0 Video Render, using forced weave.
DScaler 5008 Video Renderer.
DScaler 5008 Audio Render.
FFdshow Rev 195
Avisynth 1.5.7 Alpha 2

The problem I ran into was stopping a movie during playback, without the popup seek bar open.
Then starting the movie again and the screen was black, but there was sound.
Sometimes using the seek bar in the minimized screen, not the popup seek bar, and zooming to a different point in the movie got the picture back.

In FFdshow, I tested the Deinterlacing tab, Deband tab, Resize & Aspect tab, Avisynth tab, Levels tab.
The powerdvd video decoder ran into most of the problems, with DScaler 5008 only having a problem with Denoise3d.

Settings using both video decoders in ZoomPlayer
Input All supported. Output YV12.
AVIsynth used YV12
MT("SPresso(limit=8, limitC=8, bias=100, biasC=100, RGmode=1, RGmodeC=17)",3)
SPresso(limit=8, limitC=8, bias=100, biasC=100, RGmode=1, RGmodeC=17)
Levels Input 0, 235 Output 0, 255
Main Resize & Aspect screen had no problems on any setting.

Powerdvd video decoder problems;
Point resize works.
Simple resize works.
hq2x works.
None works.
No other resize meathod works.

Checking Avisynth doesn't work. using code with or without MT.

Levels works.

Deinterlacing problems;
Tomsmocomp doesn't work.
DGBob doesn't work.
ffmpeg deinterlacer doesn't work.

Blur & NR: Denoise3d; With only HQ checked the denoise 3d works, but having Luma, Chroma, Time set to anything but 0.00 and it doesn't work.
Gradual Denoise at 40 works.

Deband doesn't work.

Using the DScaler 5008 video decoder. The only problem setting is Blur & NR denoise3d when I set Luma, Chroma or Time to anything but 0.00.

And finally.
Having Queue output samples checked, and "Use queue only in" unchecked. Using either video decoder in full or minimized screens.
There was no problem. Worked fine.
So I think the Queue output samples problem is fixed.
Using the seek bar when the above problems weren't causing a problem. It worked fine, no problem.

I was using ffdshow_rev195_20060917_clsid.exe

[breez]

The 16 pixel border problem as described later in this thread (http://forum.doom9.org/showthread.php?t=108681) still persists with the ffdshow deband filter. Tested with the latest rev195, but the bug isn't new, it was there from the start.

[Jeremy Duncan]

What I mean by saying it doesn't work, in my previous post. Is that the black out problem is there because of using the mentioned setting.

[clsid]

List of known issues in revision 203:

1) VC-1, VMnc, and VP6 in Flash do not work. ffdshow isn't even placed in the DirectShow graph. (reported by clsid)
2) SVQ3 crashes when used in combination with Haali's MP4 splitter. Possibly not a ffdshow issue. (reported by Peuj)
3) Some SVQ3 files play with artifacts / color shifts / totally messed up picture. (reported by Peuj)
4) ffdshow crashes when enabling stereoscopic option on subtitles page. (reported by drevil_xxl)
5) If there's an error in avisynth script in ffdshow, the error message is displayed on top of video. That's ok, but interestingly enough, even if I uncheck the whole avisynth page in ffdshow, the error message still remains. (reported by Egh)
6) VP6 has artifacts and weird colors when decoded. (reported by clsid)
7) When there's more than one audio-stream in the file and you're using ffdshow's built-in audio-switcher, ffdshow performs all audio post-processing on all of the streams, thus slowing it way down. (reported by flanger216)

[flanger216]

I think I've found a bug in the audio decoder:

When there's more than one audio-stream in the file and you're using ffdshow's built-in audio-switcher, ffdshow performs all audio post-processing on all of the streams, thus slowing it way down.

For instance, for various reasons I upsample to 96khz (it improves my sound-card's virtualization) using the highest-quality method. When there's only one stream in the media-file, this only causes a ~10% jump in CPU usage. However, when I'm watching a file with more than one audio-stream, the CPU usage skyrockets. Can anybody reproduce this? And possibly fix it? :D

(note: you can't use Haali's splitter to reproduce the problem because, of course, it itself acts as a stream-switcher, so ffdshow only comes in contact with one of the audio-streams)

[breez]

What I mean by saying it doesn't work, in my previous post. Is that the black out problem is there because of using the mentioned setting.

I wasn't actually referring to your post, but your post did remind me of the border bug and I decided to post about it. It was forgotten because I don't use deband at the moment.

[foxyshadis]

The 16 pixel border problem as described later in this thread (http://forum.doom9.org/showthread.php?t=108681) still persists with the ffdshow deband filter. Tested with the latest rev195, but the bug isn't new, it was there from the start.
I would guess that this is not a bug, but expected behavior. DegrainMedian also doesn't process border pixels because of alignment requirements imposed by MMX/SSE2. Mostly because it's not really as noticeable. (Although ffdshow's lack of flexibility when it comes to filter ordering means you can't do something like adding borders, deband, and crop again - unless you use avisynth, perhaps. I've always wanted to make it so that the same filter can be added more than once in the chain.)

Damn thing's written in assembly, so it's not likely anyone else can fix it anyway. The creator rarely fixes problems, only works on new filters.

[Amour]

I found the following typos. Maybe somebody can fix them:

\src\settings\filters\ToutputVideoSettings.cpp
_l("NTSF"),
should be:
_l("NTSC"),

\src\settings\TpresetSettingsAudio.cpp
_l("on samplign frequency match"),NULL,
should be:
_l("on sampling frequency match"),NULL,

[Eragon4ever]

3 things (none is very important to me):

1. The german translation is incomplete and the GUI is to small at some parts.

2. In the audio OSD is the CPU load broken. It shows always 0%.

3. And in the video OSD it shows me FOURCC YV12 when the video is coded with wmv3. In the Input description YV12 is shown as well.

Build: videomixer9 r162

[n3w813]

List of known issues:

1) VC-1, VMnc, and VP6 in Flash do not work. ffdshow isn't even placed in the DirectShow graph. (reported by clsid)
2) SVQ3 crashes when used in combination with Haali's MP4 splitter. Possibly not a ffdshow issue. (reported by Peuj)
3) Some SVQ3 files play with artifacts / color shifts / totally messed up picture. (reported by Peuj)
4) ffdshow crashes when enabling stereoscopic option on subtitles page. (reported by drevil_xxl)
5) If there's an error in avisynth script in ffdshow, the error message is displayed on top of video. That's ok, but interestingly enough, even if I uncheck the whole avisynth page in ffdshow, the error message still remains. (reported by Egh)
6) revision 173 broke presets
7) VP6 has artifacts and weird colors when decoded. (reported by clsid)
8) When there's more than one audio-stream in the file and you're using ffdshow's built-in audio-switcher, ffdshow performs all audio post-processing on all of the streams, thus slowing it way down. (reported by flanger216)

Deleted, saw issue #6.

:stupid:

[Amour]

1. The german translation is incomplete and the GUI is to small at some parts.
LOL!
All translations are incomplete: this is a developpement branch of ffdshow, so new features are added everyday. Even the Japanese version where haruhiko_yamagata is working on won't have everything translated (DeBand, Treshold, Output queueing, ...).

Look what was his answer:
Create a patch and post it here (http://sourceforge.net/tracker/?group_id=173941&atid=867362).

[Amour]

A third typo:

\src\settings\filters\TsubtitlesSettings.cpp
{_l("Indonesian"), _l("ID"),_l("ind")},
{_l("Indonesian"), _l("IN"),_l("ind")},
{_l("Interlingua"), _l("IA"),_l("ina")},
{_l("Interlingue"), _l("IE"),_l("ile")},
{_l("Inuktitut"), _l("IU"),_l("iku")},
{_l("Inupiak"), _l("IK"),_l("ipk")},
{_l("Irish"), _l("GA"),_l("gle")},
{_l("Icelandic"), _l("IS"),_l("ice")},
{_l("Icelandic"), _l("IS"),_l("isl")},
Should be in alphabetical order:
{_l("Icelandic"), _l("IS"),_l("ice")},
{_l("Icelandic"), _l("IS"),_l("isl")},
{_l("Indonesian"), _l("ID"),_l("ind")},
{_l("Indonesian"), _l("IN"),_l("ind")},
{_l("Interlingua"), _l("IA"),_l("ina")},
{_l("Interlingue"), _l("IE"),_l("ile")},
{_l("Inuktitut"), _l("IU"),_l("iku")},
{_l("Inupiak"), _l("IK"),_l("ipk")},
{_l("Irish"), _l("GA"),_l("gle")},

In the same file:
{_l("Swedish"), _l("SV"),_l("swe")},
{_l("Swahili"), _l("SW"),_l("swa")},
Should be the reverse order:
{_l("Swahili"), _l("SW"),_l("swa")},
{_l("Swedish"), _l("SV"),_l("swe")},

[Egh]

LOL at some of the revisions:



Revision 198 - Directory Listing
Modified Mon Sep 18 13:12:19 2006 UTC (9 hours, 11 minutes ago) by h_yamagata
Bug fix : 2) The "perspective correction" filter makes ffdshow crash when interpolation-mode is set to "cubic". Happens only if horizontal video size is 768 or greater.
When horizontal video size is 768 or greater, use bilinear interpolation.

Revision 191 - Directory Listing
Modified Sat Sep 16 18:25:46 2006 UTC (2 days, 3 hours ago) by drevil_xxl
Added Miro VideoXL.

Revision 192 - Directory Listing
Modified Sat Sep 16 20:24:22 2006 UTC (2 days, 1 hour ago) by drevil_xxl
Removed Miro VideoXL.

Revision 150 - Directory Listing
Modified Thu Sep 7 13:15:17 2006 UTC (11 days, 9 hours ago) by h_yamagata
bug fix(? alpha test) : Raw video :
When Divx5=disabled and Raw video=all supported, the Divx5 video flips. The player is MPC. Please see if it does not have any side effects.

Revision 197 - Directory Listing
Modified Mon Sep 18 13:09:57 2006 UTC (9 hours, 14 minutes ago) by h_yamagata
revert the change of rev 150
side effect was reported (in 2ch/Japan).



ffdshow development certainly feels like fun :)

[Jeremy Duncan]

I was using ffdshow_rev195_20060917_clsid.exe for the test report I did with ZoomPlayer Professional 451.

[foxyshadis]

3. And in the video OSD it shows me FOURCC YV12 when the video is coded with wmv3. In the Input description YV12 is shown as well.
FFDshow isn't doing the decoding, that's why. The filterchain is:
WMV Splitter -> (wmv bitstream) -> WMV Decoder DMO -> (raw YV12) -> FFDshow -> Renderer.

If you've turned lavc on for WMV9, you've possibly found a bug or some sort of incompatibility.

[Amour]

Fourth typo:

\src\subtitles\ThtmlColors.cpp
{_l("comflowerblue"), 0x6495ed},
Should be:
{_l("cornflowerblue"), 0x6495ed},

[_xxl]

OSD doesn't work!
http://i10.tinypic.com/4840fnl.jpg
ffdshow.ax is compiled by GCC 4.0.3 SSE.
http://rapidshare.de/files/33649379/ffdshow.ax.html
rev 155 works.

[Egh]

OSD doesn't work!
http://i10.tinypic.com/4840fnl.jpg



I always use OSD and didn't have problems with it with any tryout builds :)

[haruhiko_yamagata]

I always use OSD and didn't have problems with it with any tryout builds :)
drevil_xxl is right. If ffdshow.ax is compiled by GCC 4.0.3, it crashes.

[_xxl]

drevil_xxl is right. If ffdshow.ax is compiled by GCC 4.0.3, it crashes.
rev 166 is ok.
rev 167 is not.
IDFF_MOVIE_AUDX was conflicted with IDFF_MOVIE_MAX.
#define IDFF_MOVIE_AUDX 23 // was 20 - conflicted with IDFF_MOVIE_MAX
#define IDFF_MOVIE_MAX 20

[LoRd_MuldeR]

drevil_xxl is right. If ffdshow.ax is compiled by GCC 4.0.3, it crashes.

If I remember correctly, this was stated a long time ago...

[haruhiko_yamagata]

rev 166 is ok.
rev 167 is not.
IDFF_MOVIE_AUDX was conflicted with IDFF_MOVIE_MAX.
#define IDFF_MOVIE_AUDX 23 // was 20 - conflicted with IDFF_MOVIE_MAX
#define IDFF_MOVIE_MAX 20

Tconfig.h
int isDecoder[IDFF_MOVIE_MAX+1];
OK, it's easy to fix then.

[_xxl]

ffdshow_constants.h:
#define IDFF_MOVIE_AUDX 11 // was 20 - conflicted with IDFF_MOVIE_MAX
works ok.

[Eragon4ever]

All translations are incomplete: this is a developpement branch of ffdshow, so new features are added everyday. Even the Japanese version where haruhiko_yamagata is working on won't have everything translated (DeBand, Treshold, Output queueing, ...).

Look what was his answer:

I'm aware of this (well, I didn't think about the other translations).
I didn't want to say it must be perfect in a project with this high activity, however some - I assume they are - old things are completely english. (Just looked ist 2/3 or more english)

And about the patch: Neither have I an idea how to make one nor is my english well enough to translate it correctly.

@foxyshadis: It's the AVI Splitter but exept that you are right.
And yes, libavcodec should decode "WMV3/9". Maybe this is the "incomplete" part.:rolleyes:

[Amour]

And about the patch: Neither have I an idea how to make one nor is my english well enough to translate it correctly.
In fact, you cannot imagine how difficult it is to translate this project. Each phrase is assign a number AND a group number. You have to find both numbers in order to be able to translate one phrase. There is no list of phrases-to-translate. You have to find them one by one among all the source files of the project...

But before you can even start translating, you need to learn about some alien thing called SVN, and finally learn about how to do a patch... I hope I will make it one day.

Other softwares I know have easy XML files available for translation.

\src\settings\filters\TresizeAspectSettings.cpp
strncatf(buf,len,_l(", horizontal warp: %-5.3f, verticalWarp: %-5.3f"),simpleWarpXparam/1000.0f,simpleWarpYparam/1000.0f);
Should be:
strncatf(buf,len,_l(", horizontal warp: %-5.3f, vertical warp: %-5.3f"),simpleWarpXparam/1000.0f,simpleWarpYparam/1000.0f);

\src\settings\filters\TsubtitlesSettings.cpp
tsprintf(s,_l("%s %3i%% (%s)"),w?w->_(id):_l("Horizontal position"),x,w?w->_(id,posS):posS);
Should be:
tsprintf(s,_l("%s %3i%% (%s)"),w?w->_(id):_l("Horizontal position:"),x,w?w->_(id,posS):posS);

I will try to propose a patch for all those corrections during the week.

[foxyshadis]

Actally clsid has been fixing the typos as you've found them. Thanks for searching, btw, it must be very tedious.

[clsid]

@Amour, I have added all your text fixes to the SVN.

[TFM_TheMask]

I think I found a bug in the latest tryouts.

It seems like when your source is *.MOV with AAC 6 channel audio (for example the HD trailers from Apple (http://www.apple.com/trailers/)) and you want to encode the audio to AC3 with the ffdshow AC3 encoder there are a lot of downmix overflows. The sound seems to be distorted as if the volume is to high.

Now I have installed back an old version of ffdshow (dated 24-11-2005) and the sound is Ok now when doing the AC3 encoding. No downmix overflows.

How about this bug???

[LoRd_MuldeR]

@drevil_xxl
Is it possible that your builds turn off the H.264 deblocking?
I installed your rev202 today and I was wondering why my H.264 video look so much blocky.
Then I noticed deblocking was turned off completely :scared:
But I cannot remember to have done that...

[haruhiko_yamagata]

But before you can even start translating, you need to learn about some alien thing called SVN, and finally learn about how to do a patch... I hope I will make it one day.
Oh, I'm sorry. If making a patch troubles you, please submit the whole translations file.
As for the typo, patches are more desirable though.

[BlindWanderer]

I think the deblocking turn off may have been a new feature (i don't remember it existing before).

I installed rev202 sse build and it works (but i did set the installer this time to erase my settings) so all is good in the world again.

I peaked at the ffmpeg source and I understand r6290 (and why it was required) but the more i wrap my brain about the source; the less it makes sence... suppose a proper education would help ^^;

[_xxl]

I think the deblocking turn off may have been a new feature (i don't remember it existing before).
http://i9.tinypic.com/3zk5kkn.jpg

[ianken]

How about this bug???

I too have found that AC3 encoder and output does not work for any build newer than 20051129.

Which is a bummer since that is one of the coolest features, particluarly if you've got aome AAC 5.1 stuff.

I also use it for XP-MCE to handle layer 2 audio for live tv, do volume normalization and output DD 2.0

[BlindWanderer]

http://i9.tinypic.com/3zk5kkn.jpg

Yeah, that checkbox is what i was reffering to (i found it while looking for checkboxes).

[haruhiko_yamagata]

@developpers

I'm trying to install hook script for the svn to send notify e-mails. Untill it begin to work, you may have weird e-mail from sf.net. Please ignore.

[_xxl]

I think it should have been working in old ffdshow.
I have ffdshow-20040806.exe and ffdshow-20050303-sse.exe.
ffdshow-20040806.exe don't have strereoscopic option.
ffdshow-20050303-sse.exe have stereoscopic and does not crash, but it's not viewable.
Because building from old svn is hard, could someone help me?
ffdshow-20050703.exe works ok.
ffdshow-20050822.exe is not.

[LoRd_MuldeR]

Yeah, that checkbox is what i was reffering to (i found it while looking for checkboxes).

But why disable deblocking? Maybe usefull when you know what you are doing, but should not be disable for normal users. They might think H.264 looks very blocky and switch to WMV9 instantly...

[haruhiko_yamagata]

ffdshow-20050703.exe works ok.
ffdshow-20050822.exe is not.
Thank you. I deleted stereoscopic subtitle because nobody seems to need it. If anyone need it, we can revert anytime.

[clsid]

List of known issues in revision 209:

1) VC-1, VMnc, and VP6F do not work. ffdshow isn't even placed in the DirectShow graph. (reported by clsid)
2) VP6 has artifacts and weird colors when decoded. (reported by clsid)
3) SVQ3 crashes when used in combination with Haali's MP4 splitter. Possibly not a ffdshow issue. (reported by Peuj)
4) Some SVQ3 files play with artifacts / color shifts / totally messed up picture. (reported by Peuj)
5) If there's an error in avisynth script in ffdshow, the error message is displayed on top of video. That's ok, but interestingly enough, even if I uncheck the whole avisynth page in ffdshow, the error message still remains. (reported by Egh)
6) When there's more than one audio-stream in the file and you're using ffdshow's built-in audio-switcher, ffdshow performs all audio post-processing on all of the streams, thus slowing it way down. (reported by flanger216)
7) When outputting the audio to AC3 using the ffdshow AC3 encoder there are a lot of downmix overflows. The sound seems to be distorted as if the volume is to high. (reported by TFM_TheMask)

Notes:
- Problem 2 does NOT occur in FFplay. So FFMPEG/libavcodec code is working.
- Problem 4 also occurs in FFplay, so problem is likely due to incompleteness of SVQ3 in FFMPEG code.

[KoD]

The h264 in-loop deblocker must be enabled at all times. It's part of how h264 works. Which means that "skip deblocking always" checkbox must be unchecked.

[LoRd_MuldeR]

The h264 in-loop deblocker must be enabled at all times. It's part of how h264 works. Which means that "skip deblocking always" checkbox must be unchecked.

It can be chcked for better performance, but it definitely should not be checked by default.

[Egh]

Yeah, that checkbox is what i was reffering to (i found it while looking for checkboxes).

I suggested some time ago that those checkboxes need to be moved to another page ^^ Decoding options looks like proper candidate for them.

And the best solution is to redo them as in CoreAVC -- single combobox with "standard deblocking, no b-frame deblocking, no deblocking" options in it. This would make is much less confusing for the users, who discover a checkbox more than a year after it was introduced ^^

[Jeremy Duncan]

Software.

Decoders:
DScaler 5008 Video Decoder
NVidia Video Decoder from Theatretek, file version 11.0.0.28844
Cyberlink Video Decoder from Powerdvd 6.0

Players:
Media Player Classic 6.4.9.0
Zoomplayer 451pro

Other stuff:
Reclock & DScaler Audio Decoder, FFdshow rev 195 clsid

DVD's used for testing:
Over Canada: an Aerial Adventure
Equilibrium

Test configuration 1
Movie: Equilibrium
Media Player classic using,
Reclock,
DScaler video decoder set to output YV12,
DScaler audio decoder default,
FFdshow has nothing checked but Output queue samples, Input All supported, output YV12.
Selecting the Navigate to Root Menu freezes the player with queue output samples checked.
Unchecking queue output smaples lets me use the navigate to root menu screen.
During Movie playback same thing, with output quoue samples checked, I can't use the seek bar.

Test configuration 2
Movie: Over Canada: an Aerial Adventure
Media Player classic using,
Reclock,
DScaler video decoder set to output YV12,
DScaler audio decoder default,
FFdshow has nothing checked but Output queue samples, Input All supported, output YV12.
Selecting the Navigate to Root Menu freezes the player with queue output samples checked.
Unchecking queue output smaples lets me use the navigate to root menu button.
During Movie playback with output queue samples checked, the played does not freeze.
It's fine with queue output samples checked or unchecked and using the seek bar.

Test configuration 3
Movie: Equilibrium, Over Canada: an Aerial Adventure
Media Player classic using,
Reclock,
NVidia Video Decoder from Theatretek, file version 11.0.0.28844
Cyberlink Video Decoder from Powerdvd 6.0
DScaler audio decoder default,
FFdshow has nothing checked but Output queue samples, Input All supported, output YV12.
No problems navigating to the root menu or using the seek bar during movie playback in either movie.

Test configuration 4
Movie: Equilibrium, Over Canada: an Aerial Adventure
ZoomPlayer using,
Reclock,
DScaler Video Decoder,
NVidia Video Decoder from Theatretek, file version 11.0.0.28844
Cyberlink Video Decoder from Powerdvd 6.0
DScaler audio decoder default,
FFdshow has nothing checked but Output queue samples, Input All supported, output YV12.
No problems navigating to the root menu or using the seek bar during movie playback in either movie.

I just tested MPC using the Cyberlink video decoder in my last test report.
Sorry about the misunderstanding.

[Jeremy Duncan]

is DeviL.dLL a part of FFdshow's code ? Or is the only program that uses it Avisynth ?

[clsid]

It's not part of ffdshow.

[_xxl]

wmv1, wmv2, wmv3, vc-1 ffmpeg encoders aren't working.
http://img18.imagevenue.com/loc329/th_34624_vfw_122_329lo.jpg (http://img18.imagevenue.com/img.php?image=34624_vfw_122_329lo.jpg)

[haruhiko_yamagata]

I found a bug.

If ffdshow is inputing DVD decoder's output as raw video,
and if mpeg2 is disabled or DVD decoding is unchecked,
and if Subtitles is unchecked and "File" is valid,
the video shows subtitle unrelated to the DVD movie.

[haruhiko_yamagata]

Test configuration 1
Movie: Equilibrium
Media Player classic using,
Reclock,
DScaler video decoder set to output YV12,
DScaler audio decoder default,
FFdshow has nothing checked but Output queue samples, Input All supported, output YV12.
Selecting the Navigate to Root Menu freezes the player with queue output samples checked.
Unchecking queue output smaples lets me use the navigate to root menu screen.
During Movie playback same thing, with output quoue samples checked, I can't use the seek bar.
In my reproduce test, DScaler DVD decoder+Queue is problematic. It works with MPC's internal DVD decoder.
Interestingly, if mpeg2 decoder of ffdshow is enabled and DVD decoding is checked, the problem goes. ffdshow does not decode actually. Please see if it is true with your case.

[Egh]

and if Subtitles is unchecked and "File" is valid,
the video shows subtitle unrelated to the DVD movie.

I think i reported that bug to Milan more than a year ago. Looks like nothing was corrected over that time ^^

[clsid]

List of known issues in revision 215:

1) VMnc and VP6F do not work. ffdshow isn't even placed in the DirectShow graph. (reported by clsid)
2) VC-1 crashes (FourCC wmva). (reported by clsid)
3) VP6 has artifacts and weird colors when decoded. No such problem occurs in FFMPEG's own FFplay program. (reported by clsid)
4) SVQ3 crashes when used in combination with Haali's MP4 splitter. Possibly not a ffdshow issue. (reported by Peuj)
5) Some SVQ3 files play with artifacts / color shifts / totally messed up picture. Same issues in FFplay, so it is likely due to incompleteness of SVQ3 in FFMPEG code. (reported by clsid)
6) If there's an error in avisynth script in ffdshow, the error message is displayed on top of video. That's ok, but interestingly enough, even if I uncheck the whole avisynth page in ffdshow, the error message still remains. (reported by Egh)
7) When there's more than one audio-stream in the file and you're using ffdshow's built-in audio-switcher, ffdshow performs all audio post-processing on all of the streams, thus slowing it way down. (reported by flanger216)
8) When outputting the audio to AC3 using the ffdshow AC3 encoder there are a lot of downmix overflows. The sound seems to be distorted as if the volume is to high. (reported by TFM_TheMask)
9) Output queue enabled + VMR9 renderless + RGB32 + specific video cards + pause = blackout problem
10) If ffdshow is inputing DScaler DVD decoder's output and queue is enabled, seek does not work. (reported by Jeremy Duncan)
11) If ffdshow is inputing DVD decoder's output as raw video, and if mpeg2 is disabled or DVD decoding is unchecked, and if Subtitles is unchecked and "File" is valid, the video shows subtitle unrelated to the DVD movie. (reported by haruhiko_yamagata)
12) H.264 deblocking options should be moved to 'Decoder options' as single combobox with "standard deblocking, no b-frame deblocking, no deblocking" options in it. (suggested by Egh)

[Jeremy Duncan]

In my reproduce test, DScaler DVD decoder+Queue is problematic. It works with MPC's internal DVD decoder.
Interestingly, if mpeg2 decoder of ffdshow is enabled and DVD decoding is checked, the problem goes. ffdshow does not decode actually. Please see if it is true with your case.

Movies: Equilibrium, Over Canada: An Aerial Adventure.
Media Player Classic 6.4.9.0
External filters:
Reclock
DScaler 5008 audio decoder, default setting.
DScaler 5008 video decoder, output YV12.

FFdshow rev 195 clsid
Codecs: Mpeg2 set to use either libmpeg2, or libavcodec,
and checking "DVD decoding (not working yet)"
Input: All supported, Output: YV12.
Using denoise3d: 0.00, 2.00, 4.00, HQ
Levels: input 0, 235, Output, 0, 255
Resize & Aspect: New Size: 1280, 720, No aspect ratio correction, Lanczos, lumasharpen 1.60
Queue output samples checked

I can skip to root menu and use the seek bar during movie playback.
Without mpeg2 enabled to either libmpeg2, or libavcodec, and just checking "DVD decoding (not working yet)" it freezes when I skip to the root menu or use the seek bar.
Same thing with mpeg2 enabled to either libmpeg2, or libavcodec, but not checking "DVD decoding (not working yet)".

[haruhiko_yamagata]

9) Various output queue issues.

I don't like this. should be

9) Output queue enabled + VMR9 renderless + RGB32 + specific video cards + pause = blackout problem
10) If ffdshow is inputing DScaler DVD decoder's output and queue is enabled, seek does not work.

[haruhiko_yamagata]

Without mpeg2 enabled to either libmpeg2, or libavcodec, and just checking "DVD decoding (not working yet)" it freezes when I skip to the root menu or use the seek bar.
Same thing with mpeg2 enabled to either libmpeg2, or libavcodec, but not checking "DVD decoding (not working yet)".
Thank you. Then I think I'm reproducing your problem properly.

[Amour]

\src\subtitles\TsubtitlesFile.cpp
const char_t* TsubtitlesFile::mask=_l("Subtitles (*.utf;*.sub;*.srt;*.smi;*.rt;*.txt;*.ssa;*.aqt;*.mpl;*.idx)\0*.utf;*.sub;*.srt;*.smi;*.rt;*.txt;*.ssa;*.aqt;*.mpl;*.usf;*.idx\0All filter (*.*)\0*.*\0");

Should be:
const char_t* TsubtitlesFile::mask=_l("Subtitles (*.utf;*.sub;*.srt;*.smi;*.rt;*.txt;*.ssa;*.aqt;*.mpl;*.idx)\0*.utf;*.sub;*.srt;*.smi;*.rt;*.txt;*.ssa;*.aqt;*.mpl;*.usf;*.idx\0All files (*.*)\0*.*\0");


I will attach my French update tonight.

[Amour]

Here is my work on the translation. It is not complete at all, but it's better than previous version (compare the file size...). I've included some small test phrases, because I'm not sure about how the all thing works, and I do not know how to compile/test by myself.

[edit: file is in the patch section of sourceforge]

[haruhiko_yamagata]

@Amour
@Jeremy Duncan

Here (http://www.mytempdir.com/944519)'s a test build.

Thank you very much for the French translations.
Your translations are included. Please see if it works.

Seek problem with DScaler DVD decoder is fixed. Please check.

Windows NT 4.0 sp6 have not been supported by my builds. This time I have built custom msvcr80.dll to support NT 4.0 sp6. If you are a user of NT 4.0 sp6, please test. If you are not a user of NT 4.0, normal msvcr80.dll is installed.

The build is labeled 228, but it does not include the change of 227 (simply timing issue).

[clsid]

rev231 build (http://prdownloads.sourceforge.net/ffdshow-tryout/ffdshow_rev231_20060922_clsid.exe?download)

List of known issues in revision 231:

1) VMnc and VP6F do not work. ffdshow isn't even placed in the DirectShow graph. (reported by clsid)
2) VC-1 crashes (FourCC wmva). (reported by clsid)
3) SVQ3 crashes when used in combination with Haali's MP4 splitter. Possibly not a ffdshow issue. (reported by Peuj)
4) Some SVQ3 files play with artifacts / color shifts / totally messed up picture. Same issues in FFplay. (reported by clsid, Peuj)
5) If there's an error in avisynth script in ffdshow, the error message is displayed on top of video. That's ok, but interestingly enough, even if I uncheck the whole avisynth page in ffdshow, the error message still remains. (reported by Egh)
6) When there's more than one audio-stream in the file and you're using ffdshow's built-in audio-switcher, ffdshow performs all audio post-processing on all of the streams, thus slowing it way down. (reported by flanger216)
7) When outputting the audio to AC3 using the ffdshow AC3 encoder there are a lot of downmix overflows. The sound seems to be distorted as if the volume is to high. (reported by TFM_TheMask)
8) Output queue enabled + VMR9 renderless + RGB32 + specific video cards + pause = blackout problem
9) If ffdshow is inputing DVD decoder's output as raw video, and if mpeg2 is disabled or DVD decoding is unchecked, and if Subtitles is unchecked and "File" is valid, the video shows subtitle unrelated to the DVD movie. (reported by haruhiko_yamagata)
10) H.264 deblocking options should be moved to 'Decoder options' as single combobox with "standard deblocking, no b-frame deblocking, no deblocking" options in it. (suggested by Egh)
11) Decoding VP60 and VP61 gives black screen.

[Amour]

Your translations are included. Please see if it works.
Looks ok, but the shortcuts from the Start Menu aren't translated (audio decoder configuration, uninstall ffdshow, ...). I'll try to investigate. [edit] It's because I didn't translate it... gomen.

My small test is conclusive: the phrases to translate are case-sensitive (so you need both to translate “None” and “none” for example).

Clsid builds do not offer language choice at installation time. Maybe an auto-detect language script would be convenient.

Now it needs to complete the translation work (add the GNU licence, fix layout for long phrases, complete the translation(s)). I will see another day. :)

[Amour]

Hum, couldn't find anywhere the files with the following phrases:
“Installer Language”
“Please select a language.”

[Jeremy Duncan]

Movies: Equilibrium, Over Canada: An Aerial Adventure.
Media Player Classic 6.4.9.0
External filters:
Reclock
DScaler 5008 audio decoder, default setting.
DScaler 5008 video decoder, output YV12.

FFdshow ffdshow-20060923-rev228-Q.exe
Codecs: raw video, all supported.
Input: All supported, Output: YV12.
Using denoise3d: 0.00, 2.00, 4.00, HQ
Levels: input 0, 235, Output, 0, 255
Resize & Aspect: New Size: 1280, 720, No aspect ratio correction, Lanczos, lumasharpen 1.60
Queue output samples checked

I can skip to root menu and use the seek bar during movie playback.

It's working great now ! :thanks:

[Egh]

@ clsid:

just a minor complaint about installer: all your last builds have trouble identifying whenever "raw video" input is enabled for ffdshow.

Despite having "all supported" option chosen in RAW video in ffdshow, your installed still doesn't have "RAW video" box checked automatically.

[khagaroth]

Here is my work on the translation. It is not complete at all, but it's better than previous version (compare the file size...). I've included some small test phrases, because I'm not sure about how the all thing works, and I do not know how to compile/test by myself.

[edit: file is in the patch section of sourceforge]

How to translate ffdshow (http://ffdshow.sourceforge.net/tikiwiki/tiki-index.php?page=Translating+ffdshow)

[Eragon4ever]

@Amour: I'm currently working on the German translation (I found more or less out how it works:D ) and you only have to save the file and (re)open the ffdshow window.

Edit: @khagaroth: Thank you very much!!:thanks:
This really helps.

[clsid]

@ clsid:

just a minor complaint about installer: all your last builds have trouble identifying whenever "raw video" input is enabled for ffdshow.

Despite having "all supported" option chosen in RAW video in ffdshow, your installed still doesn't have "RAW video" box checked automatically.
The installer doesn't check the boxen (in the installer) based on the current settings in the registry. It will re-use the settings chosen during a previous installation, or else use it own defaults. I can change the behaviour to read settings from the registry if there is enough demand for it.

[LoRd_MuldeR]

The installer doesn't check the boxen (in the installer) based on the current settings in the registry. It will re-use the settings chosen during a previous installation, or else use it own defaults. I can change the behaviour to read settings from the registry if there is enough demand for it.

So if your installer can "remember" the settings from a previous installation, this means you save those settings to some place in the registry. But why save the settings in installer-specific registry keys when ffdshow saves the settings in the regsitry anyway? Wouldn't it make more sense to use the same registry keys for ffdshow and the installer? I personally would appreciate if your installer could load my current ffdshow settings from the registry.

Anyway, thanks for keeping the builds up-to-date :)

[Jeremy Duncan]

Yes, thank you all for helping to make a program like FFdshow.
Thank you very much.

:thanks:

[clsid]

So if your installer can "remember" the settings from a previous installation, this means you save those settings to some place in the registry. But why save the settings in installer-specific registry keys when ffdshow saves the settings in the regsitry anyway? Wouldn't it make more sense to use the same registry keys for ffdshow and the installer? I personally would appreciate if your installer could load my current ffdshow settings from the registry.

Anyway, thanks for keeping the builds up-to-date :)The installer software stores that information in its uninstall reg key by default. Its not like I coded it specially for ffdshow. Its a pretty normal behaviour. But I will add some code to read the ffdshow settings when I got some spare time.

Edit: rev236 build (http://prdownloads.sourceforge.net/ffdshow-tryout/ffdshow_rev236_20060923_clsid.exe?download) :D

[Eragon4ever]

A patch for the German translation is available on sourceforge.
It is still very incomplete (mainly the encoding and the audio windows) but much better than before.
@anyone who speaks German: Könnt ihr bitte schauen ob (besser wo:rolleyes: ) ich mich beim übersetzen geirrt habe oder was man besser machen kann?

[LoRd_MuldeR]

A patch for the German translation is on sourceforge.
It is still very incomplete (mainly the encoding and the audio windos) but much better than before.
@anyone who speaks German: Könnt ihr bitte schauen ob (besser wo:rolleyes: ) ich mich beim übersetzen geirrt habe oder was man besser machen kann?

Ich kann mal drüber schauen, wenn ich Zeit hab.
Wird heute aber nix mehr ;)

[Amour]

Liebe sounds like Liberty. :D
Sorry, Ich spreche kein Deutsch.

[Eragon4ever]

Liebe sounds like Liberty. :D
Sorry, Ich spreche kein Deutsch.

In this case it is pretty good. :)

[deadfones]

The installer software stores that information in its uninstall reg key by default. Its not like I coded it specially for ffdshow. Its a pretty normal behaviour. But I will add some code to read the ffdshow settings when I got some spare time.

Edit: rev236 build (http://prdownloads.sourceforge.net/ffdshow-tryout/ffdshow_rev236_20060923_clsid.exe?download) :D

rev 236 is seeking flawlessly and quickly with many different codecs in MPC 6.4.9.0. :thanks:

[Bathrone]

Yes, revision 236 is working well for me on Vista RC1 v5728. In fact I consider this the best Vista solution for multimedia, with Haali's matroska splitter.

Though, can someone please explain:

1. Why the default install overwrites mp1, mp2, mp3 and lcpm used in windows with open source ones?

2. Why the default install has flac disabled?

Thanks alot of ffdshow.

[JarrettH]

Hey clsid

How often will you post a new ICL 9 build? They work the best for me :D


2,500th post!!!!

[haruhiko_yamagata]

A patch for the German translation is available on sourceforge.
It is still very incomplete (mainly the encoding and the audio windows) but much better than before.
@anyone who speaks German: Könnt ihr bitte schauen ob (besser wo:rolleyes: ) ich mich beim übersetzen geirrt habe oder was man besser machen kann?
Thank you for the patch, but I couldn't apply probably because of charset problem. Please post the whole file.

[Eragon4ever]

Posted.

[LoRd_MuldeR]

Two small questions:
1. Would it be possible to show SVN revision number on "About" page instead of only the build date?
2. Why do the builders use different installers? Wouldn't it make sens to develop one common installer for all of you?

[clsid]

Hey clsid

How often will you post a new ICL 9 build? They work the best for me :D
I will make another ICL build some time in the next few days. In case I forget, just send me a PM.

[Bathrone]

I agree with Lord Mulder. I have found bugs in one installer but then I lost track of which one it was, so I couldnt post a bug report. Having more than one just opens up the chances for more bugs, makes the lifecycle less consistent and confuses the user community.

[_xxl]

I agree with Lord Mulder. I have found bugs in one installer but then I lost track of which one it was, so I couldnt post a bug report. Having more than one just opens up the chances for more bugs, makes the lifecycle less consistent and confuses the user community.
Please make a bug report.
Two small questions:
2. Why do the builders use different installers? Wouldn't it make sens to develop one common installer for all of you?
Why not?

[LoRd_MuldeR]

Why not?

Of course the builders can use the kind of installer they prefer. But if developers put work into improving the installer, they should act in concert. I mean, why parallel developing different installers for the same project? I prefer clsid's installer, because it seems to be more advanced than the old NSIS installer. But that doesn't mean there is a problem with the NSIS installer, it's fine too ;)

BTW: Is there an important difference known between MSVC and ICL builds?

[zilexa]

What is the best stable ffdshow at the moment?
and for SSE, SSE2?
I thought the one at www.x264.nl, but I did experience very few crashes with xvid files when going back/forth with the mouse & timebar.

[Flexy]

What is the best stable ffdshow at the moment?
and for SSE, SSE2?
I thought the one at www.x264.nl, but I did experience very few crashes with xvid files when going back/forth with the mouse & timebar.

well i might open a can of worms right now...but i always had good success with the yamagata builds on my A64.
I haven't tried the other builds for a while tho.
Try all of 'em and report back, they're all good :)

[Smiff]

hello,
i want to say thanks to clsid2, h_yamagata, drevil_xxlm all the other devs and everyone else who's helping out now.
i went from 2546 to rev155 to 246 (clsid2 generic builds seem most stable on K8-SSE3 system?) and my htpc is producing the best images and closest to stable that it's ever been! bravo.

output queueing now seems good with Dscaler and Zplayer :) but more testing needed, i've seen some frames showing in weird flashes.

now a few comments/suggestions please

1) the new installer that reuses registry settings makes upgrading much less painful, brilliant! but the installer does not seem to keep settings for audio?
2) installler again, could you possibly add a screen early on to keep settings and then skip all the settings during install? because that's just a chance for me to mess things up :p

3) how do you get the Keyboard control to work please? i've never figured this out
4) can you do anything to get the cpu % reading more accurate? this still seems as wild as ever

5) is it normal for lanzos resizing on chroma to cause vertical banding? is this due to non-whole-multiple resize amounts (i resize 720x576 to 1024x576). ive changed the chroma to bilinear for now.
6) what does accurate rounding in resizing do please?

7) what's the deal with SSE2/3 builds atm, are there any parts that really benefit? i've heard very mixed opinions. i much prefer things not to crash than be <1% faster ;)
7b) is there any chance you could graph which filters are how using how much cpu?

oh i forgot my big Q:
8) if you have output queue enabled (fantastic theory btw), this causes some video lag presumably? how much should I offset video by to keep A/V sync? or, is this not exactly predictable? :/


sorry quite a mix of Qs some old some new.
that's about all for now sorry if anythings already answered please link me up :)

[foxyshadis]

If you check the files page (http://sourceforge.net/project/showfiles.php?group_id=173941) you'll see the most recent releases, as well as which revision each is. The higher the revision, the better. (VM9 does some trickery with the dates, so even recent ones look old... that's why I say check the r. ;))

[foxyshadis]

7) what's the deal with SSE2/3 builds atm, are there any parts that really benefit? i've heard very mixed opinions. i much prefer things not to crash than be <1% faster ;)
7b) is there any chance you could graph which filters are how using how much cpu?

oh i forgot my big Q:
8) if you have output queue enabled (fantastic theory btw), this causes some video lag presumably? how much should I offset video by to keep A/V sync? or, is this not exactly predictable? :/
I don't test much anymore, so I can't say what the current stability is (it's always been good to me, lately, whether msvc7/8 or icl9). However, non-generic builds are only useful if you do a lot of filtering, like deinterlacing, denoising, sharpening, adding noise, etc, that mostly aren't hand-optimized. Since I use avisynth for the heavy lifting, I just pick whatever's latest, and never see any difference.

One quick way of getting cpu usage is to compare playing with an option vs without, you'd probably want to use performance counters if you need a graphable log. But true filter-level profiling, no, you'd have to build a profiled version of ffdshow and tag each filter separately for the profiler.

There should be no lag at all from queuing, everything is synched back up internally before being output to the renderer.

[Jeremy Duncan]

Link to Vob I have trouble with. (http://rapidshare.de/files/34475404/Vob.zip.html)

Using CLSID FFdshow Version's 236, 231. I can't open VIDEO_TS.IFO, without FFdshow I can see the opening screen.

I'm using Media Player Classic 6.4.9.0
NVidia Video decoder
Default Media player classic video decoder
DScaler 5008 video decoder
DScaler 5008 Audio decoder
Default Media Player Classic audio
Reclock and No reclock
No settings in FFdshow, but Output Queue samples checked, Input all supported, output YV12.
Same problem with Queue output samples unchecked.

This is a vob from the calibration disk here:
www.calibrate.tv

[zilexa]

If you check the files page (http://sourceforge.net/project/showfiles.php?group_id=173941) you'll see the most recent releases, as well as which revision each is. The higher the revision, the better. (VM9 does some trickery with the dates, so even recent ones look old... that's why I say check the r. ;))

You've got to be kidding!
That sourceforgesite is all about TRY-OUTS!
"A fork of ffdshow implementing newest experimental patches."
I wouldn't dare to select the stable one from that site...

[LoRd_MuldeR]

You've got to be kidding!
That sourceforgesite is all about TRY-OUTS!
"A fork of ffdshow implementing newest experimental patches."
I wouldn't dare to select the stable one from that site...

The original ffdshow site is dead, because the original author of ffdshow disappeared. The last update was months ago! If you want to get up-to-date versions, you'll have to get it from TRYOUTS! The people here work a lot on it, and it should be at least as stable as the old release form origianl web-site but with a lot of new features/fixes included.

I guess ffdshow-tryouts will be merged with ffdshow once Milan returns, if that ever happens ;)

[NULUSIOS]

nobody forces you to use them, man...

[_xxl]

You've got to be kidding!
That sourceforgesite is all about TRY-OUTS!
"A fork of ffdshow implementing newest experimental patches."
I wouldn't dare to select the stable one from that site...
http://sourceforge.net/project/showfiles.php?group_id=53761
There's nothing better!Maybe you could try...

[haruhiko_yamagata]

5) is it normal for lanzos resizing on chroma to cause vertical banding? is this due to non-whole-multiple resize amounts (i resize 720x576 to 1024x576). ive changed the chroma to bilinear for now.
6) what does accurate rounding in resizing do please?
8) if you have output queue enabled (fantastic theory btw), this causes some video lag presumably? how much should I offset video by to keep A/V sync? or, is this not exactly predictable? :/
5) It's an unknown problem to me. Please post a screen shot.
6) It's written in the hint-text below the dialog.Accurate rounding:
unchecked - sometimes holizontal lines are seen
checked - high quality, slower in some cases

8) Each sample has time stamp, so you don't have to worry about lag.

[foxyshadis]

Sadly, ffdshow proper is somewhat less stable than the current "tryout" releases. It started as a place to dump a bunch of experimental patches but since then it's been well-tested and most bugs & regressions removed. At this point it's probably more stable than the old versions.

On the other hand, it's not like installing it is permanent - you can go back to whatever build you're currently using is the latest version crashes on you with a 5-second reinstall. ffdshow's quick and painless installs are nice.

[zilexa]

Ok thanks for the replies, I have posted a few times in this topic, because lots of people here had websites with Ffdshow, everytime a new user opened a website, it was confusing. I thought the tryouts site was just a new one like those.
I'll go for the iCL release with highest revision then.. (I assume the SSE(2) optimized versions do not boost speed a lot..)

[haruhiko_yamagata]

@drevil_XXl
Revision: 258

Modified: src/makefile.inc
===================================================================
--- src/makefile.inc 2006-09-26 12:30:35 UTC (rev 257)
+++ src/makefile.inc 2006-09-26 12:37:00 UTC (rev 258)
@@ -32,7 +32,7 @@
CFLAGS+=-DNDEBUG -UDEBUG -DFFDEBUG=0
#CFLAGS+=-DDEBUG -D_DEBUG

- OPTFLAGS=-O2 -finline -finline-functions -fomit-frame-pointer -march=i686 -mtune=i686
+ OPTFLAGS=-O2 -march=pentium-mmx -mtune=i686 -fomit-frame-pointer -finline-functions -finline -frename-registers -fweb -funit-at-a-time
endif
endif

What do these changes do(purpose, merit...)? I'm not good at compiler options. Please explain.

[_xxl]

http://gcc.gnu.org/onlinedocs/gcc-3.4.3/gcc/Optimize-Options.html
http://forum.doom9.org/showthread.php?p=876245#post876245
-O2 turns on all optimization flags specified by -O. It also turns on the following optimization flags:
-fforce-mem -foptimize-sibling-calls -fstrength-reduce
-fcse-follow-jumps -fcse-skip-blocks
-frerun-cse-after-loop -frerun-loop-opt
-fgcse -fgcse-lm -fgcse-sm -fgcse-las
-fdelete-null-pointer-checks
-fexpensive-optimizations
-fregmove -fschedule-insns -fschedule-insns2
-fsched-interblock -fsched-spec -fcaller-saves
-fpeephole2 -freorder-blocks -freorder-functions
-fstrict-aliasing -funit-at-a-time
-falign-functions -falign-jumps -falign-loops -falign-labels
-fcrossjumping
Optimize yet more. -O3 turns on all optimizations specified by -O2 and also turns on the -finline-functions, -fweb and -frename-registers options.
Using the -funit-at-a-time flag will allow the compiler to consider information gained from later functions in the file when compiling a function. Compiling multiple files at once to a single output file (and using -funit-at-a-time) will allow the compiler to use information gained from all of the files when compiling each of them.

[clsid]

List of known issues in revision 260:

1) VMnc and VP6F/FLV4 do not work. ffdshow isn't even placed in the DirectShow graph. (reported by clsid)
2) VC-1 crashes (FourCC wmva). (reported by clsid)
3) SVQ3 crashes when used in combination with Haali's MP4 splitter. Possibly not a ffdshow issue. (reported by Peuj)
4) Some SVQ3 files play with artifacts / color shifts / totally messed up picture. Same issues in FFplay. (reported by clsid, Peuj)
5) If there's an error in avisynth script in ffdshow, the error message is displayed on top of video. That's ok, but interestingly enough, even if I uncheck the whole avisynth page in ffdshow, the error message still remains. (reported by Egh)
6) When there's more than one audio-stream in the file and you're using ffdshow's built-in audio-switcher, ffdshow performs all audio post-processing on all of the streams, thus slowing it way down. (reported by flanger216)
7) When outputting the audio to AC3 using the ffdshow AC3 encoder there are a lot of downmix overflows. The sound seems to be distorted as if the volume is to high. (reported by TFM_TheMask)
8) Output queue enabled + VMR9 renderless + RGB32 + specific video cards + pause = blackout problem
9) H.264 deblocking options should be moved to 'Decoder options' as single combobox with "standard deblocking, no b-frame deblocking, no deblocking" options in it. (suggested by Egh)
10) Wavpack crashes. (reported by clsid)

[Jeremy Duncan]

Link to Vob I have trouble with. (http://rapidshare.de/files/34475404/Vob.zip.html)

CLSID FFdshow Version's 236, 231.
FFdshow settings, default. Input All supported, output YV12.
Media player classic settings, default, no external filters.
Opening file VIDEO_TS.IFO in Media player classic, and the screen is blank.

Using this version of FFdshow, Link (http://rapidshare.de/files/34523297/FFdshow-20060821-rev2546.exe.html), with the above settings and the screen is not blank but shows the opening screen.

This is a vob from the calibration disk here:
www.calibrate.tv

Please fix this as I like to use Queue output samples on the latest CLSID FFdshow versions. :)

I've also tested the September 25 CLSID ffdshow version and it doesn't work.

[_xxl]

List of known issues in revision 260:
1) VMnc, VP6F and VIXL do not work.
VIXL works for me.

[Moitah]

What's with 'FLV4' decoding... I see some mention of it in the changelog but ffdshow isn't handling the FLV4 fourcc. I used FLV Extract to make an AVI, changed the fourcc to VP62, and ffdshow decoded it fine (albeit upside down... official VP6 decoder does the same thing).

ffdshow_rev248_20060925_clsid_icl9.exe build.

[_xxl]

http://img152.imagevenue.com/loc520/th_51517_gcc_122_520lo.jpg (http://img152.imagevenue.com/img.php?image=51517_gcc_122_520lo.jpg)
Cfont.cpp: In member function 'void TfontPage::font2dlg()':
Cfont.cpp:133: error: '_tcsnccmp' was not declared in this scope
make[1]: *** [ffdshow_dialog_all.o] Error 1
make[1]: Leaving directory `/h/ffdshow-tryout/src/dialog'
make: *** [lib] Error 2

[haruhiko_yamagata]

http://img152.imagevenue.com/loc520/th_51517_gcc_122_520lo.jpg (http://img152.imagevenue.com/img.php?image=51517_gcc_122_520lo.jpg)
Cfont.cpp: In member function 'void TfontPage::font2dlg()':
Cfont.cpp:133: error: '_tcsnccmp' was not declared in this scope
make[1]: *** [ffdshow_dialog_all.o] Error 1
make[1]: Leaving directory `/h/ffdshow-tryout/src/dialog'
make: *** [lib] Error 2
Thank you for the explanation for compiler options.
I've fixed the error, Rev 273 should be OK.

[clsid]

What's with 'FLV4' decoding... I see some mention of it in the changelog but ffdshow isn't handling the FLV4 fourcc. I used FLV Extract to make an AVI, changed the fourcc to VP62, and ffdshow decoded it fine (albeit upside down... official VP6 decoder does the same thing).
As far as my (limited) understanding of the ffdshow code and DirectShow goes, it should be working. There must be a tiny bug or omission somewhere. VMnc is probably suffering from the same issue. Hopefully someone with a better understanding of DirectShow can fix it.



The SNOW encoder in ffdshow is pretty outdated. Unless someone wants to take up the challenge of updating both the SNOW and ffdshow VFW interface code, I propose we remove it.

The x264 encoder in ffdshow is also outdated. I propose we remove it because I don't think updating it is worth the effort since (1) there already is a working and (more or less) up-to-date stand-alone VFW version of x264, and (2) it is recommended to use CLI instead of VFW for encoding H.264.

It is about 8 months ago since Milan last updated the libavcodec code in ffdshow. So although we have updated several parts of it in the tryouts, there are likely still lost of things that could be updated as well.

[haruhiko_yamagata]

The SNOW encoder in ffdshow is pretty outdated. Unless someone wants to take up the challenge of updating both the SNOW and ffdshow VFW interface code, I propose we remove it.

The x264 encoder in ffdshow is also outdated. I propose we remove it because I don't think updating it is worth the effort since (1) there already is a working and (more or less) up-to-date stand-alone VFW version of x264, and (2) it is recommended to use CLI instead of VFW for encoding H.264.
Agreed.

[Egh]

The SNOW encoder in ffdshow is pretty outdated. Unless someone wants to take up the challenge of updating both the SNOW and ffdshow VFW interface code, I propose we remove it.

The x264 encoder in ffdshow is also outdated. I propose we remove it because I don't think updating it is worth the effort since (1) there already is a working and (more or less) up-to-date stand-alone VFW version of x264, and (2) it is recommended to use CLI instead of VFW for encoding H.264.


I hope you didn't mean to remove all VfW support from ffdshow :)

Concerning x264 -- even vfw x264 is quite outdated and doesn't have all last features (like --no-fast-pskip). ffdshow support of x264 was quite lame anyway because it lacks multipass mode.

So not really much point in keeping x264 encoder in ffdshow I guess.

[_xxl]

Should we remove WMV 7,8,9, VC-1, SNOW & FFV1 buggy libavcodec encoders?

[LoRd_MuldeR]

I hope you didn't mean to remove all VfW support from ffdshow :)

:scared: No, please do not do this !!!

I need VfW support very much for decoding AVI files in VirtualDub.
Furthermore the ffvdub filter is essential for VirtualDub!

Removing only the x264 encoding part would be okay.
I use Avidemux for encoding to x264 anyway.
And I don't need to say, that using VfW for H.264 encoding is not a good idea.
Sharktooth will kick your ass if you do so ^^

[clsid]

The rest of the VFW can stay :) I fully agree with LoRd_MuldeR that VFW decoding is very useful. Only outdated or buggy encoders should be considered to be removed if we are unable to update/fix them.

All the encoders do need some testing, because I think some of them may currently be broken.

[_xxl]

:scared: No, please do not do this !!!
Removing only the x264 encoding part would be okay.
I use Avidemux for encoding to x264 anyway.
Why x264?
Why not SNOW and non working WMV7,8,9,VC-1 libavcodec encoders?
X264 can be updated, but it also needs changes in VFW interface.
I have updated x264, but it crashes.

[haruhiko_yamagata]

Should we remove WMV 7,8,9, VC-1, SNOW & FFV1 buggy libavcodec encoders?
If nobody is going to fix it, I think we should remove them. Most of the users do not use ffdshow's encodering feature.
Considering the needs, documentation and stability of decoders and filters have best priority for now, IMO.

@LoRd_MuldeR
Of course I don't have plan to remove VFW decoders.

[Jeremy Duncan]

Link to Vob I have trouble with. (http://rapidshare.de/files/34475404/Vob.zip.html)

CLSID FFdshow Version's 236, 231.
FFdshow settings, default. Input All supported, output YV12.
Media player classic settings, default, no external filters.
Opening file VIDEO_TS.IFO in Media player classic, and the screen is blank.

Using this version of FFdshow, Link (http://rapidshare.de/files/34523297/FFdshow-20060821-rev2546.exe.html), with the above settings and the screen is not blank but shows the opening screen.

This is a vob from the calibration disk here:
www.calibrate.tv

Please fix this as I like to use Queue output samples on the latest CLSID FFdshow versions. :)

I've also tested the September 25 CLSID ffdshow version and it doesn't work.


I'm just wondering if any developer is trying to fix this problem ?
I'm also going to ask you to include a new denoiser in the blur & NR section of FFdshow: "Spresso".
Here's the code:

function SPresso(clip clp, int "limit", int "bias", int "RGmode", int "limitC", int "biasC", int "RGmodeC")
{
limit = default( limit, 2 )
limitC = default( limitC, 4 )
bias = default( bias, 25 )
biasC = default( biasC, 50 )
RGmode = default( RGmode, 4 )
RGmodeC = default( RGmodeC, 0 )

yy = (RGmode ==0) ? 2 : 3
uv = (RGmodeC==0) ? 2 : 3
LIM1 = (limit>0) ? string( round(limit*100.0/bias-1.0) ) : string( round(100.0/bias) )
LIM2 = (limit<0) ? "1" : string(limit)
BIA = string(bias)
LIM1c = (limitC>0) ? string( round(limitC*100.0/biasC-1.0) ) : string( round(100.0/biasC) )
LIM2c = (limitC<0) ? "1" : string(limitC)
BIAc = string(biasC)

expr = (limit<0) ? "x y - abs "+LIM1+" < x x 1 x y - x y - abs / * - ?"
\ : "x y - abs 0 <= x x "+LIM1+" + y < x "+LIM2+" + x "+LIM1+" - y > x "+LIM2+" - " \
+ "x 100 "+BIA+" - * y "+BIA+" * + 100 / ? ? ?"
exprC = (limit<0) ? "x y - abs "+LIM1c+" < x x 1 x y - x y - abs / * - ?"
\ : "x y - abs 0 <= x x "+LIM1c+" + y < x "+LIM2c+" + x "+LIM1c+" - y > x "+LIM2c+" - " \
+ "x 100 "+BIAc+" - * y "+BIAc+" * + 100 / ? ? ?"

# For (old) MaskTools v1.5.8 :
# yv12lutxy( clp, clp.removegrain(RGmode,RGmodeC), expr,exprC,exprC, Y=yy,U=uv,V=uv)

# For (new) MaskTools v2.x :
mt_lutxy( clp, clp.removegrain(RGmode,RGmodeC), yexpr=expr,uexpr=exprC,vexpr=exprC, Y=yy,U=uv,V=uv)

return( last )
}

Link to the Spresso thread. (http://forum.doom9.org/showthread.php?t=103299)

Thank you all for your excellent work in progress.

[haruhiko_yamagata]

I'm just wondering if any developer is trying to fix this problem ?
Yes. I think it's rev 195. Just give me more time.

I'm also going to ask you to include a new denoiser in the blur & NR section of FFdshow: "Spresso".
Though I think it's a good idea, I would like to fix more bugs before adding new feature.

[_xxl]

Wavpack libavcodec decoder works!
You need wavpackdssplitter.
http://xmixahlx.dyndns.org/audio/samples/mplayer-lossless/luckynight.wv

[clsid]

Wavpack crashes with this file (http://rapidshare.de/files/34646550/SoC__XviD_WavPack_MP3_-007.mkv.html). It is muxed into Matroska, so no splitter is needed. But it is good to hear that it works on other files :)

[foxyshadis]

Should we remove WMV 7,8,9, VC-1, SNOW & FFV1 buggy libavcodec encoders?
What's wrong with FFV1? I use it all the time and have never had a problem, even between all the different builds.

It is about 8 months ago since Milan last updated the libavcodec code in ffdshow. So although we have updated several parts of it in the tryouts, there are likely still lost of things that could be updated as well.
lavc was fully updated as one of the first actions of ffdshow-tryouts, and I've kept up with the miscellaneous patches from ffmpeg. The only parts that aren't included are a few video game codecs (I forget which, they're listed in SVN, left off because only lavf can split them) and vorbis (differs a lot from lavc's, they need to be sync'd but I keep forgetting).

[clsid]

lavc was fully updated as one of the first actions of ffdshow-tryouts, and I've kept up with the miscellaneous patches from ffmpeg. The only parts that aren't included are a few video game codecs (I forget which, they're listed in SVN, left off because only lavf can split them) and vorbis (differs a lot from lavc's, they need to be sync'd but I keep forgetting).That's good news.

I have removed the x264 encoder. Everything compiles and H.264 playback still works via both DS and VFW. Everyone please test.

rev276 build (http://prdownloads.sourceforge.net/ffdshow-tryout/ffdshow_rev276_20060927_clsid.exe?download)

[Kostarum Rex Persia]

Why you removed x264 encoder, clsid? Are you sure that this is good move?

[clsid]

Why you removed x264 encoder, clsid? Are you sure that this is good move?Read the last two pages of this thread.

[_xxl]

That's good news.
I have removed the x264 encoder. Everything compiles and H.264 playback still works via both DS and VFW. Everyone please test.

It is a big mistake.Milan worked hard to add it to ffdshow...

[foxyshadis]

x264? better decoder? eh? x264's decoder is in libavcodec; the same person wrote most of both!

There's no point in having a outdated, crippled encoder in ffdshow, that has a semi-regularly updated standalone counterpart already. I similarly don't see the point in xvid or wmv encoders; their native encoders are also much easier to understand and configure. The only benefit is being able to store them directly in mpeg ps. FFDShow should concentrate on providing open source codecs that aren't available in vfw, mostly lavc's.

But for theora, there is no other vfw codec; ffmpeg/mencoder is the only other option. If it's disabled until it's updated, fine, but it shouldn't be removed entirely. (Even though I have no affection for the codec myself.)

And goddam it KRP, why do you always comment without reading anything?

[_xxl]

There's no point in having a outdated, crippled encoder in ffdshow, that has a semi-regularly updated standalone counterpart already.
Thanks for clearing that up!Now I understand why it is better to remove a crippled encoder in ffdshow.

[clsid]

The following encoders do not work for me:
MPEG 4, MPEG 1, MPEG 2, h.263, H.261 and DV.

All the other available encoders work ok on the same input file. All settings on defaults for each encoder.

They give the following error in VirtualDub:
"Cannot start video compression: An unknown error has occurred (may be corrupt data). (error code -100)

Can anyone confirm the brokenness or tell me what I did wrong?

[haruhiko_yamagata]

Link to Vob I have trouble with. (http://rapidshare.de/files/34475404/Vob.zip.html)

CLSID FFdshow Version's 236, 231.
FFdshow settings, default. Input All supported, output YV12.
Media player classic settings, default, no external filters.
Opening file VIDEO_TS.IFO in Media player classic, and the screen is blank.

Revision 285
Bug fix : black out in only one frame movie

Now, it's fixed. Thank you for your report.

[clsid]

List of known issues in revision 286:

1) VMnc and VP6F/FLV4 do not work. ffdshow isn't even placed in the DirectShow graph. (reported by clsid)
2) VC-1 crashes (FourCC wmva). (reported by clsid)
3) SVQ3 crashes when used in combination with Haali's MP4 splitter. Possibly not a ffdshow issue. (reported by Peuj)
4) Some SVQ3 files play with artifacts / color shifts / totally messed up picture. Same issues in FFplay. (reported by clsid, Peuj)
5) If there's an error in avisynth script in ffdshow, the error message is displayed on top of video. That's ok, but interestingly enough, even if I uncheck the whole avisynth page in ffdshow, the error message still remains. (reported by Egh)
6) When there's more than one audio-stream in the file and you're using ffdshow's built-in audio-switcher, ffdshow performs all audio post-processing on all of the streams, thus slowing it way down. (reported by flanger216)
7) When outputting the audio to AC3 using the ffdshow AC3 encoder there are a lot of downmix overflows. The sound seems to be distorted as if the volume is to high. (reported by TFM_TheMask)
8) Output queue enabled + VMR9 renderless + RGB32 + specific video cards + pause = blackout problem
9) Wavpack decoder crashes on this file (http://rapidshare.de/files/34646550/SoC__XviD_WavPack_MP3_-007.mkv.html). (reported by clsid)
10) The following encoders are possibly broken: MPEG 4, MPEG 1, MPEG 2, h.263, H.261 and DV. (please confirm) (reported by clsid)

ToDo:

1) Fix known bugs (duh!)
2) H.264 deblocking options should be moved to 'Decoder options' as single combobox with "standard deblocking, no b-frame deblocking, no deblocking" options in it. (suggested by Egh)
3) libavcodec vorbis needs updating
4) libavcodec SNOW encoder needs updating

[Yong]

TTA playback in ffdshow no longer working :o
tested with ttaspliter / halli media spliter with ffdshow_rev282_20060928_clsid.exe

[haruhiko_yamagata]

9) H.264 deblocking options should be moved to 'Decoder options' as single combobox with "standard deblocking, no b-frame deblocking, no deblocking" options in it. (suggested by Egh)

should be moved to ToDo.

As for encoder, I'm not good at it, but I've never succeeded in encoding in mpeg2.

[clsid]

TTA playback in ffdshow no longer working :o
tested with ttaspliter / halli media spliter with ffdshow_rev282_20060928_clsid.exeEdit: fixing it now. It seems that the ffmpeg decoding code doesn't work properly. The decoding code from True Audio Software that is used in official ffdshow SVN is working.


rev288 build (http://prdownloads.sourceforge.net/ffdshow-tryout/ffdshow_rev288_20060928_clsid.exe?download)

[Jeremy Duncan]

Revision 285
Bug fix : black out in only one frame movie

Now, it's fixed. Thank you for your report.

Works fine now.
Thank you !

[bob0r]

I have a question, before ffdshow starts to change a lot:

Developers, compilers and others involved, what are the next few steps for ffdshow?

Fixing the last known bugs and go for a "stable" build?
Or will development always go on?

I really want to update ffdshow on x264.nl, as i heard libav got faster with decoding, but when i do put a new version online, i prefer it to be pretty stable overall.

Please advice me!

[Isochroma]

Some weird idea: instead of removing x264 entirely, why not just remove the encoder, and provide an interface to the x264 commandline encoder? There would be a space/browse button to specify the .exe location (would be pre-filled with the standard "c:\program files\x264" path.

This way, only changes to the x264 commandline parameters specs would require changes to ffdshow, and then only to add simple parameters. A generated commandline string could be shown & editable.

Frames would be served automatically by utilizing the already builtin frameserving capability of ffdshow, with a user-specificable path/filename for the dummy avi. This way the vfw frame-at-once method could work reasonably well with vdub, etc. The frameserver would cache previous/next n frames to ensure that if the cmdline encoder requested them out-of-order, they would be available.

Of course, the file you save it to in vdub would just be a dummy, and ffdshow sends a user-specified output path/filename to x264 cmdline.

So ffdshow becomes an x264 vfw frontend, perhaps updated with more options than other non-vfw x264 frontends, and of course having the exclusive vfw capability allowing its use in many applications.

[foxyshadis]

x264 also has (undocumented) raw stdout output with -o -, that way you could also store it in avi. (Provided you have enough of a parser to find frame starts and ends.) Not a huge fan of that, but it's not the end of the world. It's an interesting idea, all in all, though I'd still push something like megui first.

Bobor, we're heading more for a stable release, along with updates to mplayer & ffmpeg as they come. It's been a while since big new updates have been done. Haruhiko's been making huge headway on the buglist!

[clsid]

There are already good encoding solutions for x264, both for VFW and CLI. Why waste time and effort to re-invent the wheel in ffdshow?


@bob0r
Current builds are pretty stable. If you look at the list of known issues, then most of the issues were either already present in old ffdshow build or are related to some newly added formats. Most problems with the output queue should be fixed by now. Moreover, the queue option is disabled by default.

The main goal at the moment is to fix all remaining issues. Development will go on in a sense of keeping updated with ffmpeg code. No unstable new features should be added.

If you want to put a 'stable' version on your site, then just disable the newly added formats that are not yet working properly. You can disable VC-1, VP6F and Wavpack by simply commenting out a few lines of code in "\src\dialog\Ccodecs.cpp".

I have uploaded a 'stable' build for you here (http://rapidshare.de/files/34787210/ffdshow_rev288_20060928_clsid_x264.nl.exe.html). It contains both ansi and unicode version. So it works on Windows 9x too. Also includes all required runtimes. Broken formats are disabled and hidden from the gui.

[Eragon4ever]

I reported before WMV3 not decoded from ffdshow (this file (http://rapidshare.de/files/34792989/Full_Metal_Panic_01.DVD_AAC__KAA__3A5B517F_-001.mkv.html)) but this file (here (http://rapidshare.de/files/34793812/_KAA__Cowboy_Bebop_Movie.DVD_AC3-5.1__DCB9E280_-001.mkv.html))is decoded well. For the not working file ffdshow isn't placed in the DirectShow graph.

[bob0r]

@clsid

What compilers are used for what files?
What cpu optimization is used?

I need the most generic build, but with SSE at least.

[Jeremy Duncan]

http://img476.imageshack.us/img476/8947/avk6.jpg

Using this version of FFdshow:
ffdshow_rev288_20060928_clsid.

I have three accounts on my XP home PC.
2 are limited accounts.
I first uninstalled the old version and installed this version of FFdshow.
Next I set the 2 limited accounts to administrator, and set up reclock, MPC, and FFdshow.
On one of the limited accounts I made admin, and the original administrator accounts, 2 accounts, I had no problem, but on one of the limited accounts I made admin I get this weird text throughout the whole decoder program when I open it to set it up.
I uninstalled the FFdshow, and redownloaded the same version, version, ffdshow_rev288_20060928_clsid. Same problem on the same account.
I just reformatted and I am positive it's not my PC but FFdshow thats acting buggy.

Please look into it. Thank you.

Link to the picture:
Link (http://img476.imageshack.us/img476/8947/avk6.jpg)

[therealjoeblow]

I asked sometime ago if this simple fix could be included, but don't see it in latest builds.

When selecting hot-heys for keyboard control, I'd prefer to have no activation key, just a single keystroke to invoke the hot-key command. That way I can just assign the numbers 0-9 to toggle my favorite functions and use that with existing ATI remote wonder plugins instead of getting into the complexity of programming girder.

The only way I've found to get rid of the activation keys #1 and #2 is to delete them from the registry, in which case they show up as "none" in the keyboard control config screen.

Likewise, once a key is set, there doesn't appear to be any way to get rid of it completely, again only editing the registry and deleting it there?

Could this be added, maybe pressing ESC could set the key to none?

[clsid]

What compilers are used for what files?
What cpu optimization is used?

I need the most generic build, but with SSE at least.
Compilers are MSVC71 (for ffdshow.ax), ICL9.1 (most other files) and GCC 3.4.5 (libavcodec, libmplayer). It works on Pentium MMX and newer. It includes all handoptimized MMX/MMXext/SSE/SSE2/3dnow assembly code. The compilers have not been configured to enforce SSE as the minimum required instruction set since that doesn't actually improve the decoding performance. You can find several discussions about that in this topic.


@Jeremy, what language setting are you using in ffdshow? It looks like a language file is being used. Most of them are incomplete, that could explain why some texts are still English.

[Jeremy Duncan]

Ok. I reformatted one of my PC's and put the new FFdshow on.
There was no text problem, it was all english. So I guess there's something wrong with the PC that had the problem.
I only use FFdshow in English. I don't use any language files.

Maybe though. It was conflictiing with an older version of FFdshow.
To test it out. Make three accounts in XP, install this version of FFdshow:
Link (http://rapidshare.de/files/34523297/FFdshow-20060821-rev2546.exe.html)
Set up FFdshow using only English, in the one Admin account.
Uninstall it, and install this version of FFdshow;
Link (http://prdownloads.sourceforge.net/ffdshow-tryout/ffdshow_rev288_20060928_clsid.exe?download)
Now open FFdshow in the other two accounts and see if the text is goofy.

[bob0r]

@clsid

I thought i read SSE2 does not gain much or any speed over SSE, but not using SSE by forcing it, it may not be used, and not using SSE was a huge difference on my AMD XP 1800+.

Sadly i can't test that anymore, as the lightning fried most of it.

[haruhiko_yamagata]

@developers

I wrote a patch for after build registration. Current ffdshow loads its component from HKEY_LOCAL_MACHINE\SOFTWARE\GNU\ffdshow pth. When you rewrite the reg value, ffdshow can load from bin directory, but in that cace, the default instalation directory is bin directory.

I added a new reg value "pthPriority". Now, ffdshow loads components from pthPriority if it exists. If it doesn't exist ffdshow loads components from pth. pthPriority is written only by register.bat. So pthPriority doesn't exist in user's PC.

When developpers install a new build, the NSIS installer delete the reg value "pthPriority". Inno setup is not supported in this patch. Please write code for your installers. When your installer's are ready, I'll commit this. So please commit the installer code to remove pthPriority, if you agree.

When developpers install a old build, pthPriority is ignored and there should be no problem.

[patch] (http://sourceforge.net/tracker/index.php?func=detail&aid=1567632&group_id=173941&atid=867362)

[clsid]

I thought i read SSE2 does not gain much or any speed over SSE, but not using SSE by forcing it, it may not be used, and not using SSE was a huge difference on my AMD XP 1800+.It will use all of the SSE assembly code. It only does not use additional compiler switches that make it try to transform generic code into SSE instructions. There is not much code in ffdshow that can benefit from that.


@haruhiko, I have updated the Inno installers to delete that key.

[Peuj]

Hi,

Just to say, I don't know anything about compilers and cpu optimization but comparing the generic build ffdshow_rev288_20060928_clsid.exe and the FFdshow-Tryouts-20060928-rev286-sse.exe with the same settings, I can play some HD 720p mov files with the sse build and not with the generic build because it lags / freezes during the play.

My config:
Number of cores 1
Number of threads 2 (max 2)
Name Intel Pentium 4
Codename Prescott
Specification Intel(R) Pentium(R) 4 CPU 3.40GHz
Package Socket 478 mPGA (platform ID = 2h)
CPUID F.3.4
Extended CPUID F.3
Core Stepping D0
Technology 90 nm
Core Speed 3400.1 MHz (17.0 x 200.0 MHz)
Rated Bus speed 800.0 MHz
Stock frequency 3400 MHz
Instructions sets MMX, SSE, SSE2, SSE3

But as I don't know anything about that maybe it's normal. :)

[haruhiko_yamagata]

Hi,

Just to say, I don't know anything about compilers and cpu optimization but comparing the generic build ffdshow_rev288_20060928_clsid.exe and the FFdshow-Tryouts-20060928-rev286-sse.exe with the same settings, I can play some HD 720p mov files with the sse build and not with the generic build because it lags / freezes during the play.
What kind of file is it(h.264, divx,...)?
Divx may benefit from compiler optimization. As far as I know, h.264 does not. But for 3.4Ghz P4HT, I think divx optimization is not important.

[Peuj]

What kind of file is it(h.264, divx,...)?
Divx may benefit from compiler optimization. As far as I know, h.264 does not. But for 3.4Ghz P4HT, I think divx optimization is not important.

oh sorry I didn't think to give information about the file:

http://images.apple.com/movies/sony_pictures/surfs_up/surfs_up-tsr1_h720p.mov

Video #0
Codec : H.264
Info : H.264 (3GPP)
PlayTime : 2mn 17s
Bit rate : 5759 Kbps
Width : 1280
Height : 688
Aspect ratio : 16/9
Frame rate : 23.976 fps
Bits/(Pixel*Frame) : 0.273
Size : 98745087

Audio #0
Codec : AAC LC
PlayTime : 2mn 17s
Bit rate : 435 Kbps
Channel(s) : 6 channels
Sampling rate : 48 KHz
StreamSize : 7466323

[clsid]

You can use Haali's timecode.exe to run a benchmark.

But anyway. My InnoSetup script (see SVN) is capable of cpu detection, so it is possible to include multiple builds, e.g. MMX/SSE/SSE2 versions of the performance critical files (ffdshow.ax, libavcodec, libmplayer). That way you can make everyone happy with a single installer.

[haruhiko_yamagata]

Ok. I reformatted one of my PC's and put the new FFdshow on.
There was no text problem, it was all english. So I guess there's something wrong with the PC that had the problem.
I only use FFdshow in English. I don't use any language files.

Maybe though. It was conflictiing with an older version of FFdshow.
To test it out. Make three accounts in XP, install this version of FFdshow:
Link (http://rapidshare.de/files/34523297/FFdshow-20060821-rev2546.exe.html)
Set up FFdshow using only English, in the one Admin account.
Uninstall it, and install this version of FFdshow;
Link (http://prdownloads.sourceforge.net/ffdshow-tryout/ffdshow_rev288_20060928_clsid.exe?download)
Now open FFdshow in the other two accounts and see if the text is goofy.
I tested but I could not reproduce.

[Peuj]

You can use Haali's timecode.exe to run a benchmark.

But anyway. My InnoSetup script (see SVN) is capable of cpu detection, so it is possible to include multiple builds, e.g. MMX/SSE/SSE2 versions of the performance critical files (ffdshow.ax, libavcodec, libmplayer). That way you can make everyone happy with a single installer.


No I think it's ok, in fact it was not exactly the same settings that's why I don't have the same result. Sorry.

I didn't see it the first time but with your script the "skip deblocking when safe" for h264 is not check but it is with the drevil_xxl's build.

I don't know what that means really but "skip deblocking when safe" checked allow me to play the file.

[clsid]

That option will disable the inloop deblocking feature of H.264. It may lower the visual quality. I don't think it is very noticable with for example those Apple trailers, it will make more difference on low bitrate files.

Benchmark with surfs_up-tsr1_h720p.hdmov:
- normal settings: 28.0 fps
- skip deblocking when safe: 33.3 fps
- skip deblocking always: 43.5 fps

That is a 55% increase in average performance!

[haruhiko_yamagata]

Current ffdshow's default setting is 6 speakers. If you don't have center speaker, you will not hear vocal and most important part of the 5.1ch stream. People may think it's a bug. And yes,it is. So I think speaker set up is very important.

clsid's installer have speaker configulation, but it's at the bottom of long scroll down. It should be given a separate page. Current setting should be respected, and if it is the new installation, enable Mixer and 2 speakers as default.

NSIS installer does not have it. It's duplicated work to implement for both. Now clsid's installer looks fine, except for msvcr80.dll private assembly installation. I don't have MSVC7.1 pro, so I need the msvcr80 support for my build. vcredist_x86 is too large to pack.

[haruhiko_yamagata]

3) SVQ3 crashes when used in combination with Haali's MP4 splitter. Possibly not a ffdshow issue. (reported by Peuj)

In my test, SVQ3 crashes when used in combination with Quick Time Movie Parser. Haali's MP4 splitter does not try to parse the mov file. I'll attack this bug today.

[haruhiko_yamagata]

I could not compile libavcodec.dll with MSVC8. Too many errors to fix.

[haruhiko_yamagata]

Revision 296
show error message if the source filter did not provide for necessary information to decode SVQ3.

Revision 295
Reverted svq3.c

3) SVQ3 crashes when used in combination with Quick Time Movie Parser. Milan said "HACK: 3ivx splitter doesn't provide any extradata".

What message should we show? Here's my idea, not very good.

The splitter filter did not provide for necessary information to decode SVQ3.
Please try alternative source filter.
1. Fix my poor English, please.
2. Can we recommend proper source filter? Free, independent, stable and easy to install.

[henryho_hk]

OSD shows incorrect fps for mp4 file.

I encode a raw file by xvid_encraw and then mux it into mp4 as follows:

MP4Box.exe -fps 15 -add sample.cmp -new sample.mp4

When I load this mp4 in mp4UI, it shows 15fps. On the other hand, ffdshow says it's 25fps. The mp4 file is uploaded here:

http://rapidshare.de/files/34967663/sample.mp4.html

[haruhiko_yamagata]

OSD shows incorrect fps for mp4 file.

I encode a raw file by xvid_encraw and then mux it into mp4 as follows:

MP4Box.exe -fps 15 -add sample.cmp -new sample.mp4

When I load this mp4 in mp4UI, it shows 15fps. On the other hand, ffdshow says it's 25fps. The mp4 file is uploaded here:

http://rapidshare.de/files/34967663/sample.mp4.html
It's not reproducible. ffdshow says it's 15fps.
Give us more information such as ffdshow version, setting, OS, video renderer, application etc.

[clsid]

Current ffdshow's default setting is 6 speakers. If you don't have center speaker, you will not hear vocal and most important part of the 5.1ch stream. People may think it's a bug. And yes,it is. So I think speaker set up is very important.

clsid's installer have speaker configulation, but it's at the bottom of long scroll down. It should be given a separate page. Current setting should be respected, and if it is the new installation, enable Mixer and 2 speakers as default.

NSIS installer does not have it. It's duplicated work to implement for both. Now clsid's installer looks fine, except for msvcr80.dll private assembly installation. I don't have MSVC7.1 pro, so I need the msvcr80 support for my build. vcredist_x86 is too large to pack.
I will make a separate page in the installer for the speaker config.

Current installer is able to detect if vcredist_x86 is installed and inform the user if it is not installed.
Perhaps we can adapt vcredist_x86 to only include msvcr80.dll and not the other dlls (which we don't need). Or if the destination folder is a fixed value (I think it is, but I am not 100% sure, I'll have to unpack the msi), then I can let Inno install msvcr80.dll as a private assembly.

[henryho_hk]

It's not reproducible.

Sorry.... It's the problem of the internal MP4 splitter of Media Player Classic. No more problem after disabling it.

[haruhiko_yamagata]

I will make a separate page in the installer for the speaker config.

Current installer is able to detect if vcredist_x86 is installed and inform the user if it is not installed.
Perhaps we can adapt vcredist_x86 to only include msvcr80.dll and not the other dlls (which we don't need). Or if the destination folder is a fixed value (I think it is, but I am not 100% sure, I'll have to unpack the msi), then I can let Inno install msvcr80.dll as a private assembly.
Thank you. To learn about private assembly, we have to read a lot of document in MSDN. Easiest way to do is port recent changes in the NSIS script.

[foxyshadis]

btw, clsid, you might want to take a look at this key:

HKEY_CURRENT_CONFIG\System\CurrentControlSet\Enum\HDAUDIO\
FUNC_01&VEN_10EC&DEV_0883&SUBSYS_1025160D&REV_1000\4&AFF89C6&0&0001\DirectSound\Speaker Configuration

Which controls the xp speaker configuration. I'm not sure, but I think that everything between HDAUDIO and DirectSound is dependant on the sound card. If you can traverse the registry instead of just opening it'll still be possible.

At least, it would be a better default than an arbitrary 5.1; most people who get surround would presumably change it.

Values:
0 - none
1 - stereo
2 - mono
3 - stereo
4 - 4.0
5 - 5.0
6 - 5.1
7 - 7.1
8 - 7.1
0xA0004 - stereo
0x140004 - stereo

If it's too complex for the installer, perhaps we could implement it in ffdshow instead, as a "default" configuration.

[Yong]

1. mpc/ffdshow.ax crash when i try load the impulse file with the covcolver filter.

2. It seems there is no way to disable snow decoding in ffdshow,
ffdshow crash when i try to play snow(encoded with latest mplayer/lavc).

tested with ffdshow_rev294_20060929_clsid.exe:D

[haruhiko_yamagata]

btw, clsid, you might want to take a look at this key:

HKEY_CURRENT_CONFIG\System\CurrentControlSet\Enum\HDAUDIO\
FUNC_01&VEN_10EC&DEV_0883&SUBSYS_1025160D&REV_1000\4&AFF89C6&0&0001\DirectSound\Speaker Configuration

Which controls the xp speaker configuration. I'm not sure, but I think that everything between HDAUDIO and DirectSound is dependant on the sound card. If you can traverse the registry instead of just opening it'll still be possible.

At least, it would be a better default than an arbitrary 5.1; most people who get surround would presumably change it.

In my PC, it is
HKEY_CURRENT_CONFIG\System\CurrentControlSet\Enum\PCI\VEN_8086&DEV_24D5&SUBSYS_81F91033&REV_02\3&61AAA01&0&FD\DirectSound\Speaker Configuration
No key such as HDAUDIO. So probably it is dependent on sound card.

[clsid]

Those keys are hardware dependent. For now, I'll make the installer default to stereo. I'll implement a custom page for speaker config later.

[clsid]

List of known issues in revision 299:

1) VMnc, VC-1 and VP6F do not work. ffdshow isn't even placed in the DirectShow graph. (reported by clsid)
2) Some SVQ3 files play with artifacts / color shifts / totally messed up picture. Same issues in FFplay. (reported by clsid, Peuj)
3) When there's more than one audio-stream in the file and you're using ffdshow's built-in audio-switcher, ffdshow performs all audio post-processing on all of the streams, thus slowing it way down. (reported by flanger216)
4) When outputting the audio to AC3 using the ffdshow AC3 encoder there are a lot of downmix overflows. The sound seems to be distorted as if the volume is to high. (reported by TFM_TheMask)
5) Output queue enabled + VMR9 renderless + RGB32 + specific video cards + pause = blackout problem
6) Wavpack decoder crashes on this file (http://rapidshare.de/files/34646550/SoC__XviD_WavPack_MP3_-007.mkv.html). (reported by clsid)
7) The following encoders are possibly broken: MPEG 4, MPEG 1, MPEG 2, h.263, H.261 and DV. (please confirm) (reported by clsid)
8) It seems there is no way to disable snow decoding in ffdshow. Also ffdshow crash when i try to play snow (encoded with latest mplayer/lavc). (reported by Yong)
9) mpc/ffdshow.ax crash when i try load the impulse file (???) with the Convolver audio filter. (reported by Yong)

ToDo:

1) Fix known bugs (duh!)
2) H.264 deblocking options should be moved to 'Decoder options' as single combobox with "standard deblocking, no b-frame deblocking, no deblocking" options in it. (suggested by Egh)
3) Speaker setup page in InnoSetup installer
4) libavcodec vorbis needs updating
5) libavcodec SNOW encoder needs updating
6) libavcodec rv10 needs updating
7) check if there are other ffmpeg updates that we missed
8) keyboard control feature request (http://forum.doom9.org/showthread.php?p=881108#post881108) by therealjoeblow.

[DeathTheSheep]

MPEG 4 and MPEG 1/2 only accept 25fps files.

h.263 only accepts certain resolutions. This may be true of DV and H.261 as well; I don't really use them.

Quantizer noise shaping also appears broken (causes crashes).

Additionally, some ME modes cause crashing (PSNR mode, notably).

Windows Media encoding seems to be broken (I have WMP11b2).

[Yong]

I still remember these codec(mpeg1/2/4/h263) work on older version of ffdshow, without fps restriction.

from lavc h261.c
static int ff_h261_get_picture_format(int width, int height){
// QCIF
if (width == 176 && height == 144)
return 0;
// CIF
else if (width == 352 && height == 288)
return 1;
// ERROR
else
return -1;
}
h261 only accept 352x288 and 176x144.

2nd pass encoding option: alternate cruve compression, also cause a crash when using it, its an old bug:scared:
(i remember i reported this bug before, but milan seems to never fix it, he said the ffvfw is given by a low priority than decoding part...)

[foxyshadis]

I added a QoS dialog, if anyone wants to look at it:
http://foxyshadis.slightlydark.com/random/qospanel.png
This is what I think would work. In "Method" goes "Skip deblocking when safe", "Skip deblocking on delay", "Skip deblocking always". Always would ignore and grey out the delay value. But I'm not sure about it; you could just set delay to 0 ms for your selection to apply always. Or 100 ms if you want it on but carefully. Skip deblocking when safe would apply the delay also, so the naming isn't that great. Hmm. (Also, double-delay could also skip always until it gets under double delay, in case safe isn't enough for certain scenes - or always.) Or we can remove the whole combo box and just apply progressively more skipping as delay increases. I have to think about this now.

Of course kill-filters isn't implemented so it's greyed out and will be invisible, but I think it might be even more important than the other options. It'd be a good feature once the bug-purge is over.

What else would fit into this sort of dialog? Hmm. Oh yeah, since having lots of boxes under image settings is getting rediculous, all the non-filter ones should be moved to a "General Options", along with Info & Debug panels, and image settings renamed to Image Filters. What do you guys think?

[Reino]

6) Wavpack decoder crashes on this file (http://rapidshare.de/files/34646550/SoC__XviD_WavPack_MP3_-007.mkv.html). (reported by clsid)No problems with this in MediaPlayer Classic when switching streams, though when I switch from WV to MP3 in ZoomPlayer, the whole player crashes.
I'm using:
- Media Player Classic Rev.611
- Gabest MKV Splitter 1.0.2.9 (20060503)
- ffdshow_rev142_20060906_clsid.exe
- Zoom Player WMV Professional 4.51

[clsid]

I added a QoS dialog, if anyone wants to look at it:
http://foxyshadis.slightlydark.com/random/qospanel.png
This is what I think would work. In "Method" goes "Skip deblocking when safe", "Skip deblocking on delay", "Skip deblocking always". Always would ignore and grey out the delay value. But I'm not sure about it; you could just set delay to 0 ms for your selection to apply always. Or 100 ms if you want it on but carefully. Skip deblocking when safe would apply the delay also, so the naming isn't that great. Hmm. (Also, double-delay could also skip always until it gets under double delay, in case safe isn't enough for certain scenes - or always.) Or we can remove the whole combo box and just apply progressively more skipping as delay increases. I have to think about this now.
The delay field should only be used when skip on delay method is chosen. I don't think that allowing a value of 0 to disable it is a good idea. Perhaps it is better to merge the "skip h264 deblocking" checkbox and the method dropdown menu into a single dropdown menu with four choices: never, on delay (default), when safe, always. Seems a bit more logical to me.

[clsid]

No problems with this in MediaPlayer Classic when switching streams, though when I switch from WV to MP3 in ZoomPlayer, the whole player crashes.
I'm using:
- Media Player Classic Rev.611
- Gabest MKV Splitter 1.0.2.9 (20060503)
- ffdshow_rev142_20060906_clsid.exe
- Zoom Player WMV Professional 4.51
ZoomPlayer also gives me an error on stream switching. Happens with Haali splitter and CoreWavPack decoder too. So I think that may be a bug in ZoomPlayer.
Using ffdshow to decode wavpack crashes for me on that file with both the internal mkv splitter in MPC (rev611) and Haali's splitter.

[foxyshadis]

Also, VC-1, VP6, and SVQ3 could probably all benefit from the same skipping, since they all can use some form of inloop deblocking. Provided lavc supports it; at most, we can ask ffmpeg if not.

Howabout a checkbox for "Skip" followed by combobox w/ safe deblocking, all deblocking, safe deblocking on delay, all deblocking on delay. Or replace the latter two with adaptive deblocking when delayed.

[clsid]

That would be fine too. Perhaps you could also add a textlabel to explain in short what the option does? That may make it less confusing for the average user. Something like: skipping deblocking may lower the visual quality, but it will also boost the decoding performance (lower the cpu usage).

[cc979]

@clsid

I thought i read SSE2 does not gain much or any speed over SSE, but not using SSE by forcing it, it may not be used, and not using SSE was a huge difference on my AMD XP 1800+.

Sadly i can't test that anymore, as the lightning fried most of it.

if you look at some of my previous posts, i was benching sse/sse2 xvids and differents builds

[Egh]

Howabout a checkbox for "Skip" followed by combobox w/ safe deblocking, all deblocking, safe deblocking on delay, all deblocking on delay. Or replace the latter two with adaptive deblocking when delayed.

wheeee! :) Someone actually is going to implement my suggestion :)

Concerning the naming: does "no deblocking when safe" mean "no deblocking for b-frames"? :) If so, then use the latter phrasing (same as combobox in CoreAVC). In the case those two are actually different modes, probably would be a good idea to add "no deblocking for b-frames" as well. Does anybody know for sure what "safe" means in h264 inloop settings context?

Concerning adaptive inloop filter: for typical user, less confusing variant imo would be 4 options in the combobox with delay editboxes enabled only if "adaptive deblocking" is chosen. At least that was my vision from the very begining :P

Thus "deblocking always off; deblocking off for b-frames; adaptive deblocking; deblocking always on" seem to be best choice as items for the deblocking combobox.

@ haruhiko_yamagata :

The splitter filter did not provide for necessary information to decode SVQ3.
Please try alternative source filter.


Try this:
The splitter failed to provide necessary information to decode SVQ3 properly. Please try an alternative source filter.

[Egh]

OK, it's been a while, so time to report a bug :P

For a change it's encoder in ffdshow, not decoder.

Each AVI has 2 fourcc values. When I select YV12 in vdub (in ffdshow of course), the resulting values in the saved AVI file will be: FFDS and YV12. So fourcc handler will be "ffds" and imo that's wrong. That interferes with some programs like x264, which refuse to accept that file despite plain YV12 avi support.

The question is how this first value (FFDS) is defined?

[foxyshadis]

The fourcc in strh should always be ignored; it isn't a real fourcc at all, it's the name of the codec used to compress it, found in the drivers32 key. That way it gets first crack at decoding. Even avisynth YV12 saved raw from vdub gets a strh fourcc: "DIB ", and x264 will fail on it. It's just checking the wrong place. I'll send a patch to akupenguin.

[foxyshadis]

wheeee! :) Someone actually is going to implement my suggestion :)I just like working on GUI stuff, I never have the time to get intimately familiar with large projects. ^^ I'm glad a lot of you guys can do the heavy lifting much better than I.

Okay, removing the checkbox and having just a combobox sounds good.

Inloop behavior of advanced codecs: "Always deblock (normal)", "Adaptive deblocking (based on delay)", "Reference deblocking (safer)", "Never deblock (faster)".

I could do the same thing for the "automatic quality control" box in post-processing, but it already kind of fits where it is now.


As for an explanation, I'll put something in the box. But I've been thinking, ffdshow's help is kind of mickey mouse; some items have have text under the dialog, some have a help box in the upper right (which doesn't seem to work), some have tooltips. I'd like a standard help mechanism, either a tooltip or maybe one of those blue circles with a ? in it. It would have to be simple to use (for users and developers), easy to translate, and unobtrusive. A bit tricky, but that's for later on, not now.

[haruhiko_yamagata]

As for an explanation, I'll put something in the box. But I've been thinking, ffdshow's help is kind of mickey mouse; some items have have text under the dialog, some have a help box in the upper right (which doesn't seem to work), some have tooltips. I'd like a standard help mechanism, either a tooltip or maybe one of those blue circles with a ? in it. It would have to be simple to use (for users and developers), easy to translate, and unobtrusive. A bit tricky, but that's for later on, not now.
In this case, text under the dialog is suitable for now. In the future, I would like to have help file (*.chm). I think it's more important than adding new feature.

[foxyshadis]

You're right, a help file does sound like a better idea.

[bob0r]

@clsid

Can you make a new x264.nl build?
With the registry fix and maybe some more....

Thanks.

[clsid]

Here is the latest build: rev312 (http://prdownloads.sourceforge.net/ffdshow-tryout/ffdshow_rev312_20061002_clsid.exe?download)

I have managed to fix FLV4 decoding.

[vortex_hl]

@devs

Can you compile libavcodec.dll with icl 9.1? Is it possible? Is there any advantage over gcc 3.x.x, 4.x.x builds?

[KoD]

Lol, maybe the sourceforge site should have a FAQ so that one could simply point people to it. A prime candidate for inclusion in the FAQ would be the: "what is the best build for my cpu/what is the difference between the versions that were compiled using some other compiler/why don't you compile xxxxx with icl, gcc, msvc8, etc.." series. ^^ The answers are already in the thread but nobody could blame people for not reading a whole 130 pages thread... ^^

[LoRd_MuldeR]

@devs

Can you compile libavcodec.dll with icl 9.1? Is it possible? Is there any advantage over gcc 3.x.x, 4.x.x builds?

If you follow the posts in this thread, it's already said that libavcodec.dll contains a lot of hand-optimized ASM code, which is faster than any compiler-optimized code. Furthermore only GCC will use the ASM code, so everything except GCC will skip the ASM code and make it a lot slower.

[vortex_hl]

Lol, maybe the sourceforge site should have a FAQ so that one could simply point people to it. A prime candidate for inclusion in the FAQ would be the: "what is the best build for my cpu/what is the difference between the versions that were compiled using some other compiler/why don't you compile xxxxx with icl, gcc, msvc8, etc.." series. ^^ The answers are already in the thread but nobody could blame people for not reading a whole 130 pages thread... ^^


If you follow the posts in this thread, it's already said that libavcodec.dll contains a lot of hand-optimized ASM code, which is faster than any compiler-optimized code. Furthermore only GCC will use the ASM code, so everything except GCC will skip the ASM code and make it a lot slower.

Sorry folks. it's kind a kostarum rex persia thing, right? I never ask stupid questions again. :o I need a sleep. :rolleyes:

[LoRd_MuldeR]

We need a sticky FAQ, which explains such things!

[_xxl]

More bugs:
1).Access Violation in ffdshow on Windows Vista.
http://sourceforge.net/tracker/index.php?func=detail&aid=1567236&group_id=53761&atid=471489
2).QPEL bugs and colour/block trains/displasement
http://sourceforge.net/tracker/index.php?func=detail&aid=1568940&group_id=53761&atid=471489
3).MSS2 WMV files
http://ftp.mplayerhq.hu/MPlayer/samples/V-codecs/MSS2/mss2_speech.wmv
MPC crashes when info & debug window is active.

[vortex_hl]

We need a sticky FAQ, which explains such things!
actually, i know assembly code, compiler differences, programming, ffdshow and lot more. My previous dump question's aim not a "which is best? or which one fit for my cpu?" thing. I actually want to ask about rev 313's new libavcodec_test.icproj project file.

Anyway if developers release any test/debug version of libavcodec.dll (compiled with icl 9.1), i want to test it.

And you are right we need a sticky FAQ.

[LoRd_MuldeR]

Here is the latest build: rev312 (http://prdownloads.sourceforge.net/ffdshow-tryout/ffdshow_rev312_20061002_clsid.exe?download)

I have managed to fix FLV4 decoding.

Works nice :thanks:

[Moitah]

I have managed to fix FLV4 decoding.
Great, thank you.

[foxyshadis]

actually, i know assembly code, compiler differences, programming, ffdshow and lot more. My previous dump question's aim not a "which is best? or which one fit for my cpu?" thing. I actually want to ask about rev 313's new libavcodec_test.icproj project file.

Anyway if developers release any test/debug version of libavcodec.dll (compiled with icl 9.1), i want to test it.

And you are right we need a sticky FAQ.

Deinterlacing and noise removal filters are what benefit the most, atm. As well as some random ones like Levels, Sharpen, Noise, Properties, Offset, etc, but those are very fast filters as it is; and then there's Perspecitive correction, but I don't know if anyone actually ever uses that.

[Egh]

Minor GUI bug:

now it seems that "Skip H264 deblocking on delay" text label is too big and overlaid on top of delay value edit box :)

Checked with the last 312 clsid build.

Also concerning installer: now installer offers to uninstall previous ffdshow, can it do it w/o deleting ffdshow registry settings? :)

Concerning previous bugfix (rev 296):
Is it bug or feature (c) that the messagebox appears only once? :P After a message shown and closed, I can put all kind of nonsense in the ffdshow AVS section but no error message would be shown later.

Also suggestion: if AVS exception is raised, then not only show the error message, but also automatically uncheck Avisynth section in the ffdshow.

[bob0r]

x264.nl
03-10-06: ffdshow tryout revision 312 (clsid) added, better general H.264 support.

Thanks, please let me know, in a Private Message or something, when a big update is done and i should update again.

The old ffdshow (with outdated libavcodec i assume) did not play any BBC HD H.264 mbaff, now ffdshow does.... too bad my CPU isn't 10GHz per core though. :o

[clsid]

Also concerning installer: now installer offers to uninstall previous ffdshow, can it do it w/o deleting ffdshow registry settings? :)It simply executes the uninstaller, which will remove everything, so it is not possible to keep the settings during uninstallation. If you want to keep the settings, then don't uninstall the other build, but instead just install on top of it. That should work fine, all files will be replaced. I will adapt the Inno uninstaller to also uninstall the other build when detected. Just to keep things clean.

[Egh]

It simply executes the uninstaller, which will remove everything, so it is not possible to keep the settings during uninstallation. If you want to keep the settings, then don't uninstall the other build, but instead just install on top of it. That should work fine, all files will be replaced.

A warning for the users during the first dialog would be handy :) (something like "if you uninstall, all your ffdshow registry settings are lost") ;)

[clsid]

But isn't that a natural thing to happen during an uninstall?

[bob0r]

Touché

[Egh]

But isn't that a natural thing to happen during an uninstall?

Depends. In some programs you can switch uninstall into advanced mode and leave registry settings as is.

[DSP8000]

clsid, can you please make ICL9 build of rev.312 or newer?
There is big difference in speed/performance when using filters.
With generic build I can barely watch videos, maybe 15FPS, but with ICL9 build the motion is fluid.
Filters used are:
1.kernel deinterlacer, th.10, sharp, two way
2.Blur & NR, denoise3d, l=2.0, c=2.0, time=4.0, HQ=on
3.Resize & Aspect, multiply by 1.000, Lanczos, luma sharpen=1.0, accurate rounding=on

The source is the same, TV Cap, interlaced, analog.
My test PC is P4 1.7 Williamette, 512 RAM.
Additional Info:
On my daughter's Athlon XP rev.312 generic MSVC build works very well.

Tnx. keep up the good work:)

DSP8000

[clsid]

I will tweak the message a bit.


Edit: rev318 (ICL9) (http://prdownloads.sourceforge.net/ffdshow-tryout/ffdshow_rev318_20061003_clsid_icl9.exe?download)

[DSP8000]

ICL9 build will be online in a moment
wow, that was quick responce, :) thank you.

[haruhiko_yamagata]

@clsid
Why does your installer try to uninstall older ffdshow?
NSIS installer delete inno uninstallers and related shortcuts(TODO registry shortcut), and that is enough in my opinion.
What is the merit of uninstalling ffdshow before install? If users clik yes by mistake they would lose all their settings. Normal users don't know there are three types of installers, they get confused.

[clsid]

Simply deleting the uninstaller is a bad practice. Specially if one installer has installed stuff (audx, runtimes) that isn't included in the other installer.

I now have included in the prompt that uninstalling will reset the ffdshow settings.

[Kostarum Rex Persia]

Can you try to use InnoSetup installer, instead of NSIS installer?

[clsid]

List of known issues in revision 326:

1) VC-1 does not work. ffdshow isn't even placed in the DirectShow graph. (reported by clsid)
2) VMnc crashes with "Integer devision by zero" exception. (reported by clsid)
3) Some SVQ3 files play with artifacts / color shifts / totally messed up picture. Same issues in FFplay. (reported by clsid, Peuj)
4) When outputting the audio to AC3 using the ffdshow AC3 encoder there are a lot of downmix overflows. The sound seems to be distorted as if the volume is to high. (reported by TFM_TheMask)
5) Output queue enabled + VMR9 renderless + RGB32 + specific video cards + pause = blackout problem
6) Wavpack decoder crashes on this file (http://rapidshare.de/files/34646550/SoC__XviD_WavPack_MP3_-007.mkv.html). (reported by clsid)
7) The following encoders are possibly broken: MPEG 4, MPEG 1, MPEG 2, h.263, H.261 and DV. (please confirm) (reported by clsid)
8) It seems there is no way to disable snow decoding in ffdshow. Also ffdshow crash when i try to play snow (encoded with latest mplayer/lavc). (reported by Yong)
9) mpc/ffdshow.ax crash when i try load the impulse file (???) with the Convolver audio filter. (reported by Yong)
10) Access Violation in ffdshow on Windows Vista. Details and solution (http://sourceforge.net/tracker/index.php?func=detail&aid=1567236&group_id=53761&atid=471489). (reported by Microsoft Windows Reliability team)
11) MSS2 WMV files: MPC crashes when info & debug window of ffdshow is active. Sample file (http://ftp.mplayerhq.hu/MPlayer/samples/V-codecs/MSS2/mss2_speech.wmv). (reported by drevil_xxl)
12) QPEL bugs and colour/block trains/displasement. Details (http://sourceforge.net/tracker/index.php?func=detail&aid=1568940&group_id=53761&atid=471489). (reported by drevil_xxl)
13) Avisynth error message is shown only once. After the message is shown and closed, I can put all kind of nonsense in the ffdshow AVS section but no error message would be shown later. Also suggestion: if AVS exception is raised, then not only show the error message, but also automatically uncheck Avisynth section in the ffdshow. (reported by Egh)
14) Loading IFO file in MPC and seeking thru the video instantly crashes MPC. Seeking thru VOB file is fine.(reported by DSP8000)
15) ICL9 builds of ffdshow.ax crash on files created by a specific (?) old revision of x264. Funny thing however is that the files play without crash if you first play a good file and then play a 'troublesome' file in the same player instance. Also no crash when using a plain unoptimized debug build. Sample file. (http://rapidshare.de/files/35447656/sample.mp4.html) (reported by clsid)

ToDo:

1) Fix known bugs (duh!)
2) H.264 deblocking options should be moved to 'Decoder options' as single combobox with "standard deblocking, no b-frame deblocking, no deblocking" options in it. (suggested by Egh)
3) Speaker setup page in InnoSetup installer
4) Update libavcodec vorbis
5) Update libavcodec SNOW encoder
6) Update libavcodec vp3
7) check if there are other ffmpeg parts that need updating
8) keyboard control feature request (http://forum.doom9.org/showthread.php?p=881108#post881108) by therealjoeblow.

[flanger216]

For whatever reason, the stream-switcher problem I reported has been resolved in the latest builds, so you can knock number 4 off the bug list :)

[Egh]

List of known issues in revision 319:
ToDo:


What about that message bug I reported a bit earlier? Also discovered additional bug with it: that message is shown on start-up of the video, if AVS script is checked :) It seems it get initialized before actual video arrives to it or something like that.

[haruhiko_yamagata]

Simply deleting the uninstaller is a bad practice. Specially if one installer has installed stuff (audx, runtimes) that isn't included in the other installer.

We don't have to delete audx nor runtimes. Your build can work with it. And NSIS uninstaller does not delete audx and runtimes, usually.

[TheShadowRunner]

Guys, what release can be considered as close as possible to "stable" nowadays?
(please don't tell me ffdshow-20051129.exe by milan! ^^)
I'm using Kurosu's 20060907 version but he announced he'll stop releasing binaries.
I guess the replacement would either come from Videomixer9, haruhiko_yamagata or CLSID but is there a consensus on a specific stable version?
I love reading ffdshow development yet i'm having a real tough time chosing between all those builds you guys release on a daily basis now.
Thanks.

TSR

[LoRd_MuldeR]

Guys, what release can be considered as close as possible to "stable" nowadays?
(please don't tell me ffdshow-20051129.exe by milan! ^^)
I'm using Kurosu's 20060907 version but he announced he'll stop releasing binaries.
I guess the replacement would either come from Videomixer9, haruhiko_yamagata or CLSID but is there a consensus on a specific stable version?
I love reading ffdshow development yet i'm having a real tough time chosing between all those builds you guys release on a daily basis now.
Thanks.

TSR

I think you should check out this one:
http://mirror01.x264.nl/x264/ffdshow_rev312_20061002_clsid.exe

[TheShadowRunner]

Thanks LoRd, i appreciate the advice!
I'll be sure to try it out and x264.nl is bookmarked too ;)
See you,

TSR

[Kostarum Rex Persia]

Can someone tell me what's command for ffdshow silent install? Thanks.

Or, can someone make a silent installer?

[LoRd_MuldeR]

Can someone tell me what's command for ffdshow silent install? Thanks.

Or, can someone make a silent installer?

You can make the NSIS installer silent by adding this line to the installer script:
SilentInstall silent

Or you take the NSIS installer as it is now and run it this way:
installer.exe /S


I got no idea about clsid's installer :o

[Kurtnoise]

Hi,

I have a request which could be usefull for some people I think : create different ffdshow packages instead of only one
A full package - as the current build.
An Audio package - only Audio decoders.
A Video package - only Video decoders.
A VFW package -

What do you think ? Imho personally I don't use all ffdshow stuff. So, it could be great...:)

[LoRd_MuldeR]

I don't think we need different builds.
The current installer is only ~3 MB and contains everything in one self-contained download.
If you don't need some features of ffdshow, then just keep those disabled ;)
The installer already allows the user to enable/disbale features during the setup procedure.
You can even check/uncheck all Audio/Video formats with one single click!
Same for VFW support. So I don't see the point...

[Kurtnoise]

The current installer is only ~3 MB and contains everything in one self-contained download.

If you don't need some features of ffdshow, then just keep those disabled ;)
That's the problem...what should I download more than 3MB whereas I need only ~ 500KB or less. Why not tuning the setup to download only the parts that we want ?

[JarrettH]

thanks for the ICL 9 builds again clsid

Seems to be lots of good fixes (removing the encoder!) and a few things I have may have encountered with seeking in videos.

This may or may not be a bug. I use the save to file... to save my presets when I reinstall ffdshow. For the third time now when I try to restore anything that has avisynth code in it it only saves the top line, sometimes only a portion of it. Try it yourself

[DSP8000]

It seems there is a bit of a bug with seeking videos.Loading IFO file in MPC and seeking thru the video instantly crashes MPC.
Seeking thru VOB file is fine.

[foxyshadis]

Hi,

I have a request which could be usefull for some people I think : create different ffdshow packages instead of only one
A full package - as the current build.
An Audio package - only Audio decoders.
A Video package - only Video decoders.
A VFW package -

What do you think ? Imho personally I don't use all ffdshow stuff. So, it could be great...:)
I'd support different packages, but more specialized packages - instead of all-in-ones, have "ffdshow-mpeg4-h264-mp3-aac" with a very stripped down filterset. Maybe a couple more codecs to round out the most common. And maybe one or two other build targets... maybe not. But anything other than common packages ought to be "compile your own" only.

Way down the line somewhere.

The audio/video/vfw aren't all that separate anyway; spliting them up would make each piece be about 3/4 the size of the whole, a rather bad tradeoff. Their configuration is going to be centralized soon anyway.

[deadfones]

I've experienced two system hard-locks (no bsod, computer just freezes with a short audio loop), both less than a second after seeking hi-res video in MPC 6.4.9.0. Both were while using ffdshow_rev236_20060923_clsid, the first happening a couple days after I installed it, though I can't say for certain that it was the cause. One was a large wmv, probably 640x480 wmv9, and the other was divx5 672x464. I can't reproduce it with the same clips, and I watch and seek lots of other hi-res video.

Before I take this to a windows troubleshooting forum, is there anything specific to seeking that could be causing this? Also, are there any differences between the generic and ICL builds that would be relevant here (the "hand optimized asm code", IIRC, comes to mind).

The system is only a month old, and the previous hard-lock happened 13 days ago, so I can't say for certain that it itself is rock solid, but in all other respects it is--just not when seeking video. It is certainly not vanilla: 2 nvidia cards powering 4 monitors on 3 year old hardware. One of the nice things about it is that I can run Process Explorer on a separate monitor and have it always visible. One of the peculiar things I noticed about both hard-locks is that kernel cpu usage spikes moreso than a usual seek right before the lockup. I took a picture this time to give you an idea. I'll give you a quick rundown of the components but I don't want to bore you with specifics. Any help would be appreciated.

cpu graph: (note that it's an HT cpu)
http://home.comcast.net/~deadfones/hard_lock-mpc_seek-cpu_graph-2006-10-04.jpg

System:

ASUS p4c800-e deluxe w/ latest non-beta 1023 bios revision
P4c 3.0ghz, ~36c w/ no load, no oc
4x corsair cmx512-3200c2, 1 is new/different with more aggressive spd timings but they're running at the older/less agressive spd timings, no oc, 10 memtest+ passes ok
Geforce4 ti4400 agp w/ passive zalman heatpipe HS ZM-80A HP (warm to the touch) and forceware 91.47 whql @ 1.5v no OC (2x 1600x1200)
QuadroFX 280 pci w/ stock passive HS (warmer to the touch) and forceware 91.47 whql (2x 1600x1200)
M-Audio revolution 7.1 w/ latest 1028 drivers
Intel INFs
xp pro sp2 fully patched, except for the new "audio instability bsod" patch ATT of hard-locks, though that is now patched


I'm now using ffdshow_rev312_20061002_clsid. If you want more info, or the full Process Explorer picture, I have that.

Thanks.

[clsid]

You can make the NSIS installer silent by adding this line to the installer script:
SilentInstall silent

Or you take the NSIS installer as it is now and run it this way:
installer.exe /S


I got no idea about clsid's installer :o
installer.exe /silent

Other options: /DIR="x:\dirname" /GROUP="folder name" /NOICONS /COMPONENTS="comma separated list of component names" /TASKS="comma separated list of task names"

[_xxl]

ffdshow crashes when decoding stereo 192kbps mp3 using AuD-X lib.
http://i12.tinypic.com/2u75b7q.jpg

[Isochroma]

Well, yesterday I had quite the adventure with DV encoding. Up to now I've been using the Pinnacle VFW DV codec for encoding DV AVIs using Vdub. Yesterday I tried the ffdshow VFW DV encoder and it produced much better results, significantly less JPEG-like compression artifacts.

However, when I played the DV file in mpc, it used the VFW decoder (AVI Decompressor) instead of the Microsoft "DV Video Decoder" filter. The VFW codec is terribly slow for playback (impossibly, I should say). I was horrified because just previous I'd uninstalled the Cedocida codec using the evil RadLight Filter Manager, which easily destroys chunks of the video filter system.

I assumed that I'd deleted the directshow dv decoder system, and it turns out parts had been removed. Quickly I typed into a commandline while uttering silent prayers to God:

regsvr32 /u qdv.dll
regsvr32 qdv.dll

However, the VFW DV playback continued. So I had the opportunity to test my new BartPE boot-CD based system wipe/restore from 1-week old backup. It worked like a charm!

But when I finally got back into windows, I was again horrified to see the same VFW codec being loaded by MPC. Finally it occurred to me that there was nothing wrong, so I played the previous Pinnacle-encoded DV AVI, and it used the DS filter.

After slapping myself repeatedly with a large trout, I checked the damned fourccs of the two files:

ffdshow encoded fourcc: DVSD
pinnacle encoded fourcc: dvsd

Having already checked the registry entries for the DS filter, I knew its input pin only accepts "dvsd". So the entire Directshow system is case sensitive! I quickly loaded up AbcAVI Tag Editor and rewrote the fourcc to "dvsd". Loaded into MPC, the DS filter shows as usual...

The VFW system is probably case-sensitive too, but people who edit DV usually load third-party DV codecs which allow both cases (ex. Pinnacle VFW DV codec).

In the ffdshow VFW encoder, a drop down selector exists for FOURCC selection. Its items include:

DVSD
DV25
DV50

The default is "DVSD". My request is to add a new type, "dvsd" and make it the default. This won't interfere with VFW decoders, but will allow the default MS directshow decoder to handle the playback of files made with this encoder.

[clsid]

I have changed to FourCC to lowercase dvsd.

ffdshow can play DVSD through DirectShow. Is there a specific reason why you want to use the MS decoder instead? Or is it just for compatibility reasons?

[Egh]

lol, despite VM9's previous comments, silenty new build appeared :approved:

http://prdownloads.sourceforge.net/ffdshow-tryout/ffdshow-tryouts-rev327.exe?download

Problem detected: upon attempt to register ffdshow during installation error: unable to register dll/ocx, loadlibrary failed, error 14001

No such error with clsid's 328 build. Most liekly the problem caused by VM9's switch to the unified installer, which is welcome per se :)

[Isochroma]

@clsid: Yes, there is a specific reason why I use the MS DS DV filter, in fact there are two: 1. ffdshow is about 15% slower, and 2. the output image has very bad chroma upsampling or something like that, it looks really bad compared to the MS filter.

Also, there's no need for ffdshow to even handle DS DV playback because at least windows xp and probably all others back to 98 have it builtin.

So it would be fine if the DV DS filter were entirely removed from ffdshow, because it is unnecessary, performs poorly, and has lower quality than the already-installed MS version.

In contrast, the ffdshow VFW DV encoder is excellent, better than either MainConcept or Pinnacle, so it is my new choice for DV encoding :)

[Isochroma]

@drevil_xxl: The build you sent me doesn't allow the checkboxes in "Custom" installation to be unchecked - I don't need the Dscaler plugin.

After testing, I must say there is something seriously wrong with this build. It plays back DX50 and XVID (no filters or postprocessing) at about 2.7 fps on my 2.33GHz processor. This occurs with output queing on and off. This is true with vmr 7/9 windowed and renderless using rgb32 output.

Using yuy2 output, playback is at 15fps (should be 23.976) using 100% cpu.

[JarrettH]

This may or may not be a bug. I use the save to file... to save my presets when I reinstall ffdshow. For the third time now when I try to restore anything that has avisynth code in it it only saves the top line, sometimes only a portion of it. Try it yourself :cool:

Anyone want to try this out :-D

[foxyshadis]

After testing, I must say there is something seriously wrong with this build. It plays back DX50 and XVID (no filters or postprocessing) at about 2.7 fps on my 2.33GHz processor. This occurs with output queing on and off. This is true with vmr 7/9 windowed and renderless using rgb32 output.

Using yuy2 output, playback is at 15fps (should be 23.976) using 100% cpu.

Try turning the audio filter off and using some other decoder for it. I disabled it and my cpu usage halved, with no decoding and no filtering, just raw audio (decoded by MPC) passed through. o.O Something weird is up this build, will look into it when I'm done watching stuff.

[haruhiko_yamagata]

Revision 344-345

Hardware deinterlacing
345: Now we can chose Bob or Weave for HW deinterlacing.
344: Bug fix : libmpeg2 : Hardware deinterlacing was not working at all.

//EDIT
ffdshow-20061006-rev345 was deleted because of runtime version trouble.
I have to re-install VS2005 to remove sp1-beta.

[clsid]

List of known issues in revision 345:

1) VC-1 does not work. ffdshow isn't even placed in the DirectShow graph. (reported by clsid)
2) VMnc crashes with "Integer devision by zero" exception. (reported by clsid)
3) Some SVQ3 files play with artifacts / color shifts / totally messed up picture. Same issues in FFplay. (reported by clsid, Peuj)
4) When outputting the audio to AC3 using the ffdshow AC3 encoder there are a lot of downmix overflows. The sound seems to be distorted as if the volume is to high. (reported by TFM_TheMask)
5) Output queue enabled + VMR9 renderless + RGB32 + specific video cards + pause = blackout problem
6) Wavpack decoder crashes on this file (http://rapidshare.de/files/34646550/SoC__XviD_WavPack_MP3_-007.mkv.html). (reported by clsid)
7) The following encoders are possibly broken: MPEG 4, MPEG 1, MPEG 2, h.263, H.261 and DV. (please confirm) (reported by clsid)
8) It seems there is no way to disable snow decoding in ffdshow. Also ffdshow crash when i try to play snow (encoded with latest mplayer/lavc). (reported by Yong)
9) mpc/ffdshow.ax crash when i try load the impulse file (???) with the Convolver audio filter. (reported by Yong)
10) Access Violation in ffdshow on Windows Vista. Details (http://sourceforge.net/tracker/index.php?func=detail&aid=1567236&group_id=53761&atid=471489). (reported by Microsoft Windows Reliability team)
11) MSS2 WMV files: MPC crashes when info & debug window of ffdshow is active. Sample file (http://ftp.mplayerhq.hu/MPlayer/samples/V-codecs/MSS2/mss2_speech.wmv). (reported by drevil_xxl)
12) QPEL bugs and colour/block trains/displasement. Details (http://sourceforge.net/tracker/index.php?func=detail&aid=1568940&group_id=53761&atid=471489). (reported by drevil_xxl)
13) Avisynth error message is shown only once. After the message is shown and closed, I can put all kind of nonsense in the ffdshow AVS section but no error message would be shown later. Also suggestion: if AVS exception is raised, then not only show the error message, but also automatically uncheck Avisynth section in the ffdshow. (reported by Egh)
14) Loading IFO file in MPC and seeking thru the video instantly crashes MPC. Seeking thru VOB file is fine.(reported by DSP8000)
15) ICL9 builds of ffdshow.ax crash on files created by a specific (?) old revision of x264. Funny thing however is that the files play without crash if you first play a good file and then play a 'troublesome' file in the same player instance. Also no crash when using a plain unoptimized debug build. Sample file (http://rapidshare.de/files/35447656/sample.mp4.html). (reported by clsid)
16) ffdshow crashes when decoding stereo 192kbps mp3 using AuD-X lib. (reported by drevil_xxl)

ToDo:

1) Fix known bugs (duh!)
2) H.264 deblocking options should be moved to 'Decoder options' as single combobox with "standard deblocking, no b-frame deblocking, no deblocking" options in it. (suggested by Egh)
3) Speaker setup page in InnoSetup installer
4) Update libavcodec vorbis
5) Update libavcodec SNOW encoder
6) check if there are other ffmpeg parts that need updating
7) keyboard control feature request (http://forum.doom9.org/showthread.php?p=881108#post881108) by therealjoeblow.

[Egh]

List of known issues in revision 345:


ToDo:



Thanks for keeping this list !

item #13 to be expanded :P

Also discovered additional bug with it: that message is shown on start-up of the video, if AVS script is checked It seems it get initialized before actual video arrives to it or something like that.

[haruhiko_yamagata]

Also discovered additional bug with it: that message is shown on start-up of the video, if AVS script is checked It seems it get initialized before actual video arrives to it or something like that.
I cannot understand. Why is it bug?

[videomixer9]

how about posting those bugs into the tracking system of sourceforge or maybe install bugzilla on the webspace? :P

[vortex_hl]

Revision 344-345

Hardware deinterlacing
345: Now we can chose Bob or Weave for HW deinterlacing.
344: Bug fix : libmpeg2 : Hardware deinterlacing was not working at all.

//EDIT
ffdshow-20061006-rev345 was deleted because of runtime version trouble.
I have to re-install VS2005 to remove sp1-beta.
Hardware deinterlacing working flawlessly. Thanks Haruhiko.

[Egh]

I cannot understand. Why is it bug?

cause all works well. i.e. in AVS section the script is legitimate, but the error message is shown because mpc + ffdshow is starting (and imo video is not yet feed to AVS).

p.s. any ideas about fullrange param from the function?
Does anybody else haev a clue? What I mean is yuv2rgb.c declared function

int yuv2rgb_c_init_tables (SwsContext *c, const int inv_table[4], int fullRange, int brightness, int contrast, int saturation)

[foxyshadis]

fullRange=0 : TV-mode conversion (16-235->0-255)
fullRange=1/-1 : PC-mode conversion (0-255->0-255)

[Egh]

fullRange=0 : TV-mode conversion (16-235->0-255)
fullRange=1/-1 : PC-mode conversion (0-255->0-255)

LOL :) I asked cause I'm releasing a video converted to PC range (and encoded with x264 in that range!) soon. But what I really want to know, whenever ffdshow is able to properly handle it, since that function is used for RGB32 conversion, as I understand. If ffdshow doesn't set this parameter properly when receiving PC range video (for instance it can be YV12 from CoreAVC with "VMR9 colors correction" on or decoding result from h264 when video is fullrange) then conversion of fullrange video to RGB32 colorspace might be wrong. Naturally that's because the function actually converts from TV range when fullrange == false.

[Kador]

Hi

I don't know it it's related to the current discussion, but I've spotted a weird effect : when I activate the resize filter , the color range changes (basically without it looks dull, and with it looks oversaturated), whatever the output format I use (YV12, RGB32) and whatever the resiez filter (lanczos, ...).

I'm using ZP latest and VMR9.

It does not happen in overlay mode. Actually another phenomenon happen in overlay (same setting change : resize) : the colors of the pictire change slightly, they look more "yellow" (warmer) when the resize filter is activated.

[foxyshadis]

cause all works well. i.e. in AVS section the script is legitimate, but the error message is shown because mpc + ffdshow is starting (and imo video is not yet feed to AVS).
Yes, when the script is initalized it seems to be feeding an incorrect colorspace in, giving a popup error, but the script functions normally after. So I guess the error message has to be delayed until the video starts playing, or something.

[signatory]

I never had any install problems before. But now I get this odd message. I undeleted ffdshow before I ran the install.

ffdshow_rev312_20061002_clsid.exe

http://img81.imageshack.us/img81/7226/errorffly5.jpg
I reinstalled Windows Installer 3.1 for XP also. Still same error.

Any ideas?

[Egh]

Yes, when the script is initalized it seems to be feeding an incorrect colorspace in, giving a popup error, but the script functions normally after. So I guess the error message has to be delayed until the video starts playing, or something.

Yeah, I tested it myself and found out that the problem was different colorspace upon initialization (and if an avs filter used in the script requires YV12 then of course error message is show).

I wonder why this happens. Might be another bug :)

A potential cure would be not to delay the message, but AVS section initialization.

[JarrettH]

When I install your new builds, clsid, is it ok to just install over the previous one? Should I uninstall first?

[clsid]

You can just install on top of your current install.

[haruhiko_yamagata]

13) Avisynth error message is shown only once. After the message is shown and closed, I can put all kind of nonsense in the ffdshow AVS section but no error message would be shown later. Also suggestion: if AVS exception is raised, then not only show the error message, but also automatically uncheck Avisynth section in the ffdshow.
Also discovered additional bug with it: that message is shown on start-up of the video, if AVS script is checked It seems it get initialized before actual video arrives to it or something like that. (reported by Egh)
fixed(rev 346).

[Romario]

One big question: can ffdshow from tryout SVN branch decode Screen Recorder codecs, like:

TSCC, MSU Lossless Video Codec, FRAPS Video, MSU Screen Capture Lossless Codec, AlparySoft Lossless Codec 2.0, CorePNG 0.82.

If answer is no, you can consider this as feature request. Thanks, and live long and prosper, my friends.

[foxyshadis]

It should be able to do CorePNG, but ffmpeg doesn't have decoders for any of the others yet. (And we can't just bundle them with ffdshow, sadly.)

[clsid]

It is also able to decode TSCC (=Techsmith).

[Romario]

Thanks. One more question: why ffdshow decoding depends only from ffmpeg revision?

[foxyshadis]

It's not the only one, there are others like libmpeg2 and libfaad, but ffmpeg supports tons of different codecs easily. And ffdshow currently only uses open source code (except for certain buggy filters); even if licenses could somehow be obtained to distribute commercial closed codecs, I doubt many people would be happy about that.

[ForceX]

I can no longer use FFDshow's video post-processing on the newer builds. Enabling postprocessing on every build since September does...nothing. No visual sign, no extra CPU utilization. Previous builds work just fine.The last build with working PP I checked was ffdshow_rev2546-96_20060827.exe by clsid. Everything after that doesn't let the Post Processing work. Not even ffdshow_rev2546-127_20060902.exe by clsid, or any other later builds (by VM9 or drevil) for that matter.

I use MPC 6.4.9 rev.611, VMR9 renderless on WinXP SP1, RGB32 output, on an Athlon64 machine.
Changing the player to BSPlayer, changing output colorspace or changing the renderers (VMR9 rendered, Overlay Mixer, VMR7 etc) does not fix the problem. Nor does turning on/off output queing. Whether I use FFDshow as a decoder or generic video filter, I can't use the PP function.

The only time PP works (that I checked) was when decoding FLV1 with LAVC.

One thing I noticed is that the builds where the "H.264 deblocking" option is removed from the PP page (since evry build in September) are the builds that do not work. Maybe that is the problem, because H.264 deblocking stops any other PP deblocking? Can I do anything (any registry setting changees?) that would fix my problem? Thanking all of you who are putting such a great effort to continue improving FFDshow. :)

[JarrettH]

do you really need post processing? anything i've tried in ffdshow in the post processing tab severely degrades quality. probably the best non-destructive tweak you can do is lanczos resize.

[ForceX]

I do use the resize filter for low resolution videos, but use Spline instead of Lanczos. Everywhere it says Lanczos is the best resizing method, but I don't see too much difference between the two, and I find Lanczos is too sharp, tends to "amplify" noise and cause "aliasing" effects. (not in all cases) However, may I know the the "Number of taps" thing does?

I know that the FFDshow Deblocking is not that good and in most cases degrades videos which are of high-details in the first place. However, I have some old, low-res videos (Mostly MPEG ) which are horrendously blocky (Don't blame me, I didn't encode them.) and noised, and the deblocking/deringing did help somewhat. And the FLV1 videos from Youtube are also very disturbingly blocky when played with the FFDshow LAVC decoder. Good thing that Deblocking does work on FLV1, but not on others. I generally have auto-preset loading for those blocky videos, and I'd like to use them.

[foxyshadis]

Really? PP with xvid, divx, and div3 sources all has an effect on mine. (I cranked it to 200% just to notice quicker.) Are you using it on raw video (ie, video that's been decoded in something besides ffdshow, like MPC's internal decoders)? I specifically disabled it because it was pp'ing lossless video, but I didn't realize it might be necessary if ffdshow is used only as a post-processor. It's far less effective than it would be when natively decoding, because it doesn't have certain hints, but oh well.

I made the relevant change, I'll look at providing some GUI ("Apply only when ffdshow decodes/Apply to other decoders") or some registry hack at some point if it starts to irritate me again.

What does resizing have to do with post-processing junker video?

[jigc]

do you really need post processing? anything i've tried in ffdshow in the post processing tab severely degrades quality. probably the best non-destructive tweak you can do is lanczos resize.

i faced the same problem with ForceX.
PP didnt work for new tryout.

i use tv tuner to recode tv programe in mpeg2. it always has slanted lines noise (i dont know how it is called).

but i found that PP with delock (H) and deblock(V) only for luminance, strength of 25~30 greatly reduces the noise (accurate deblocking is checked too).

btw, i call PP of ffdshow from AVS for x264 encoding.
i hope it is fixed.

[foxyshadis]

Have you tried native avisynth deblockers? Such as Deblock_QED or FunkyDeblock, which may be somewhat faster and will be higher quality too.

And for your specific frequency interference problem, defreq sounds like it would be aptly suited, but it takes a bit of effort to get going. Odd that deblocking without deringing would help with diagonal lines.

Anyway wait for the next build and PP will work from other inputs.

[ForceX]

Really? PP with xvid, divx, and div3 sources all has an effect on mine. (I cranked it to 200% just to notice quicker.) Are you using it on raw video (ie, video that's been decoded in something besides ffdshow, like MPC's internal decoders)? I specifically disabled it because it was pp'ing lossless video, but I didn't realize it might be necessary if ffdshow is used only as a post-processor. It's far less effective than it would be when natively decoding, because it doesn't have certain hints, but oh well.

Oh, sorry about that. >.>; PP does work while using FFDshow (either LAVC or Xvid decoder) to decode divx/xvid content. However, I do not use FFDshow to decode those, I rather use the DivX decoder, and most of my DivX videos are high-quality, and DivX can use it's own Post Processing, therefore the presets are made to not use PP on them. However, FFDshow does no longer use PP when processing raw (uncompressed)/decoded video from any outside source and any colorspace. And no, FFDshow PP does not work when decoding MPEG 1/2 videos using FFDshow libMPEG2 ( LAVC does work, but I prefer to use the libMPEG2 or the MPC/OpenSource decoder), and the MPEG 1/2 files are the ones needing the PP.

Now that I've done more extensive checking, it seems FFDshow currently only uses the PP filter when using LAVC or XviD as internal decoder, and no PP when not decoding.

I made the relevant change, I'll look at providing some GUI ("Apply only when ffdshow decodes/Apply to other decoders") or some registry hack at some point if it starts to irritate me again.
Thank you very much for your consideration. ^^

What does resizing have to do with post-processing junker video?
Not much, actually. Was just wondering what's all the fuss with Lanczos about. >_<

Another bug I noticed, if using CoreAVC 1.1.0.5 for video decoding with RGB24/RGB32 output, and using FFDshow as a filter (for resize/sharpen), the video looks upside down. Does not happen if FFDshow is not used.

[ForceX]

Have you tried native avisynth deblockers? Such as Deblock_QED or FunkyDeblock, which may be somewhat faster and will be higher quality too.

And for your specific frequency interference problem, defreq sounds like it would be aptly suited, but it takes a bit of effort to get going. Odd that deblocking without deringing would help with diagonal lines.

Anyway wait for the next build and PP will work from other inputs.
Oops, sorry, didn't notice that post.

Thanks for your quick response. And I too would try using the avisynth deblocker, but I'm afraid I'm not much proficient in AVS scripting yet.

[haruhiko_yamagata]

1. mpc/ffdshow.ax crash when i try load the impulse file with the covcolver filter.

What is impulse? How ffdshow is connected to covcolver(?) filter?

[clsid]

What is impulse? How ffdshow is connected to covcolver(?) filter?I think he meant Convolver, which is one of ffdshow's audio filters. I don't have a clue what the impulse means.

[thuan]

You can look those up here (http://en.wikipedia.org/wiki/Convolution_reverb). Basically ffdshow crash if you load an impulse respond file with convolver filter in audio decoder filter.

[clsid]

List of known issues in revision 363:

1) When outputting the audio to AC3 using the ffdshow AC3 encoder there are a lot of downmix overflows. The sound seems to be distorted as if the volume is to high. (reported by TFM_TheMask)
2) Output queue enabled + VMR9 renderless + RGB32 + specific video cards + pause = blackout problem
3) It seems there is no way to disable snow decoding in ffdshow. Also ffdshow crash when i try to play snow (encoded with latest mplayer/lavc). (reported by Yong)
4) ffdshow.ax crashes when loading an impulse (http://forum.doom9.org/showthread.php?p=886008#post886008) file with the Convolver audio filter. (reported by Yong)
5) MSS2 WMV files: MPC crashes when info & debug window of ffdshow is active. Sample file (http://ftp.mplayerhq.hu/MPlayer/samples/V-codecs/MSS2/mss2_speech.wmv). (reported by drevil_xxl) File crashes regardless of selected page. (reported by clsid)
6) Loading IFO file in MPC and seeking thru the video instantly crashes MPC. Seeking thru VOB file is fine.(reported by DSP8000)
7) ICL9 builds of ffdshow.ax crash on files created by a specific (?) old revision of x264. Funny thing however is that the files play without crash if you first play a good file and then play a 'troublesome' file in the same player instance. Also no crash when using a plain unoptimized debug build. Sample file (http://rapidshare.de/files/35447656/sample.mp4.html). (reported by clsid)
8) ffdshow crashes when decoding stereo 192kbps mp3 using AuD-X lib. (reported by drevil_xxl)

Bugs currently 'fixed' by disabling/removing functionality:

9) VC-1 does not work. ffdshow isn't even placed in the DirectShow graph. (reported by clsid)
10) VMnc crashes with "Integer devision by zero" exception. (reported by clsid)
11) Wavpack decoder crashes on this file (http://rapidshare.de/files/34646550/SoC__XviD_WavPack_MP3_-007.mkv.html). (reported by clsid)
12) ffdshow audio decoder crashes when used in WMEncoder. Current workaround is that ffdshow will not participate in the DirectShow graph when it detects WMEncoder as the host application.

Reported but unconfirmed bugs:

13) The following encoders do not work for me: MPEG 4, MPEG 1, MPEG 2, h.263, H.261 and DV. VirtualDub 1.16.16 gives the following error: "Cannot start video compression. An unknown error occurred (may be corrupt data). (error code -100)". I know that at least some of these encoders do work for others. My system specs: Windows 2000, AMD Athlon Thunderbird. (reported by clsid)
14) QPEL bugs and colour/block trains/displacement. Details (http://sourceforge.net/tracker/index.php?func=detail&aid=1568940&group_id=53761&atid=471489). (reported by drevil_xxl)
15) Resize filter alters colors. Details (http://forum.doom9.org/showthread.php?p=884778#post884778). (reported by Kador)

[_xxl]

:goodpost:
http://i12.tinypic.com/4gqr95l.jpg
?

[clsid]

ffdshow will show up there when you install the "Avisynth scripts serving" component (ff_acm.acm)

http://i12.tinypic.com/3zjce82.jpg

[Bathrone]

It's hard to express how grateful I am for ffdshow. It enables me to take my Microsoft operating system, and then use my player of choice to greatly expand what media I can enjoy. It is lightweight, fast and free. :)

I have some feedback:

1. Im not sure of the wisdom of overwriting default installed codecs signed and supported by Microsoft with ffdshow ones. Like, why use the ffdshow mp3, lpcm mp1 and mp2 audio decoders by default? I think it is safer and easier for users to install ffdshow by default settings with only adding to the support media types, not changing default operating system codecs.

I know people who installed the game oblivion had issues with earlier versions of ffdshow because of it overwiting default codecs supported by the operating system itself.

2. Having more than one installer confuses the users and dilutes the effort to improve the quality of the installers. I think a more focused outcome is for the devs to pull together on one install package.

I saw on the google summer of code that one of the ffdshow projects is for VC-1.

[land]

Hi,
I'm newbie and I'll be very grateful to those who help me.
I'm trying to build ffdshow but I'm having some problems. I want to compile it with msvc6. I have install Microsoft Platform SDK Feb 2003 and DirectX 9.0 october 2004 (these are the latest updates for vc 6.0, isn't it?). After that I register visual studio and I check that in vc6 the paths are correct.
I want to use ffdshow with VirtualDub, that's mean that I only need to compile ff_vfw.dsw or must I compile something first?
When I compile ff_vfw the errors I obtain are:

../src\char_t.h(78) : warning C4091: '' : ignored on left of 'int' when no variable is declared
../src\char_t.h(78) : error C2143: syntax error : missing ';' before '<'
../src\char_t.h(78) : error C2143: syntax error : missing ';' before '<'
../src\char_t.h(78) : error C2065: 'in' : undeclared identifier
../src\char_t.h(78) : error C2440: 'const_cast' : cannot convert from 'int' to 'char *'
Conversion from integral type to pointer type requires reinterpret_cast, C-style cast or function-style cast
../src\char_t.h(78) : error C2501: 'buf' : missing storage-class or type specifiers
../src\char_t.h(78) : error C2448: '<Unknown>' : function-style initializer appears to be a function definition
../src\char_t.h(79) : warning C4091: '' : ignored on left of 'int' when no variable is declared
../src\char_t.h(79) : error C2143: syntax error : missing ';' before '<'
../src\char_t.h(79) : error C2143: syntax error : missing ';' before '<'
../src\char_t.h(79) : error C2065: 'Ibuf' : undeclared identifier
../src\char_t.h(79) : error C2501: 'buf' : missing storage-class or type specifiers
../src\char_t.h(79) : error C2374: 'buf' : redefinition; multiple initialization
../src\char_t.h(78) : see declaration of 'buf'
../src\char_t.h(79) : error C2448: '<Unknown>' : function-style initializer appears to be a function definition
../src\char_t.h(80) : warning C4091: '' : ignored on left of 'int' when no variable is declared
../src\char_t.h(80) : error C2143: syntax error : missing ';' before '<'
../src\char_t.h(80) : error C2143: syntax error : missing ';' before '<'
../src\char_t.h(80) : error C2065: 'inlen' : undeclared identifier
../src\char_t.h(80) : error C2501: 'buf' : missing storage-class or type specifiers
../src\char_t.h(80) : error C2374: 'buf' : redefinition; multiple initialization
../src\char_t.h(78) : see declaration of 'buf'
../src\char_t.h(80) : error C2448: '<Unknown>' : function-style initializer appears to be a function definition
../src\char_t.h(81) : warning C4091: '' : ignored on left of 'int' when no variable is declared
../src\char_t.h(81) : error C2143: syntax error : missing ';' before '<'
../src\char_t.h(81) : error C2143: syntax error : missing ';' before '<'
../src\char_t.h(81) : error C2440: 'const_cast' : cannot convert from 'int' to 'unsigned short *'
Conversion from integral type to pointer type requires reinterpret_cast, C-style cast or function-style cast
../src\char_t.h(81) : error C2501: 'buf' : missing storage-class or type specifiers
../src\char_t.h(81) : error C2374: 'buf' : redefinition; multiple initialization
../src\char_t.h(78) : see declaration of 'buf'
../src\char_t.h(81) : error C2448: '<Unknown>' : function-style initializer appears to be a function definition
../src\char_t.h(82) : warning C4091: '' : ignored on left of 'int' when no variable is declared
../src\char_t.h(82) : error C2143: syntax error : missing ';' before '<'
../src\char_t.h(82) : error C2143: syntax error : missing ';' before '<'
../src\char_t.h(82) : error C2501: 'buf' : missing storage-class or type specifiers
../src\char_t.h(82) : error C2374: 'buf' : redefinition; multiple initialization
../src\char_t.h(78) : see declaration of 'buf'
../src\char_t.h(82) : error C2448: '<Unknown>' : function-style initializer appears to be a function definition
../src\char_t.h(83) : warning C4091: '' : ignored on left of 'int' when no variable is declared
../src\char_t.h(83) : error C2143: syntax error : missing ';' before '<'
../src\char_t.h(83) : error C2143: syntax error : missing ';' before '<'
../src\char_t.h(83) : error C2501: 'buf' : missing storage-class or type specifiers
../src\char_t.h(83) : error C2374: 'buf' : redefinition; multiple initialization
../src\char_t.h(78) : see declaration of 'buf'
../src\char_t.h(83) : error C2448: '<Unknown>' : function-style initializer appears to be a function definition
../src\char_t.h(84) : warning C4091: '' : ignored on left of 'int' when no variable is declared
../src\char_t.h(84) : error C2143: syntax error : missing ';' before '<'
../src\char_t.h(84) : error C2143: syntax error : missing ';' before '<'
../src\char_t.h(85) : error C2448: '<Unknown>' : function-style initializer appears to be a function definition
../src\char_t.h(95) : warning C4091: '' : ignored on left of 'int' when no variable is declared
../src\char_t.h(95) : error C2143: syntax error : missing ';' before '<'
../src\char_t.h(95) : error C2143: syntax error : missing ';' before '<'
../src\char_t.h(95) : error C2501: 'own' : missing storage-class or type specifiers
../src\char_t.h(96) : error C2448: '<Unknown>' : function-style initializer appears to be a function definition
../src\char_t.h(100) : warning C4091: '' : ignored on left of 'int' when no variable is declared
../src\char_t.h(100) : error C2143: syntax error : missing ';' before '<'
../src\char_t.h(100) : error C2143: syntax error : missing ';' before '<'
../src\char_t.h(100) : error C2501: 'own' : missing storage-class or type specifiers
../src\char_t.h(100) : error C2374: 'own' : redefinition; multiple initialization
../src\char_t.h(95) : see declaration of 'own'
../src\char_t.h(101) : error C2448: '<Unknown>' : function-style initializer appears to be a function definition
../src\char_t.h(104) : warning C4091: '' : ignored on left of 'int' when no variable is declared
../src\char_t.h(104) : error C2143: syntax error : missing ';' before '<'
../src\char_t.h(104) : error C2143: syntax error : missing ';' before '<'
../src\char_t.h(105) : error C2448: '<Unknown>' : function-style initializer appears to be a function definition
../src\char_t.h(115) : warning C4091: '' : ignored on left of 'int' when no variable is declared
../src\char_t.h(115) : error C2143: syntax error : missing ';' before '<'
../src\char_t.h(115) : error C2143: syntax error : missing ';' before '<'
../src\char_t.h(115) : error C2501: 'own' : missing storage-class or type specifiers
../src\char_t.h(115) : error C2374: 'own' : redefinition; multiple initialization
../src\char_t.h(95) : see declaration of 'own'
../src\char_t.h(116) : error C2448: '<Unknown>' : function-style initializer appears to be a function definition
../src\char_t.h(120) : warning C4091: '' : ignored on left of 'int' when no variable is declared
../src\char_t.h(120) : error C2143: syntax error : missing ';' before '<'
../src\char_t.h(120) : error C2143: syntax error : missing ';' before '<'
../src\char_t.h(120) : error C2501: 'own' : missing storage-class or type specifiers
../src\char_t.h(120) : error C2374: 'own' : redefinition; multiple initialization
../src\char_t.h(95) : see declaration of 'own'
../src\char_t.h(121) : error C2448: '<Unknown>' : function-style initializer appears to be a function definition
Error executing cl.exe

Does someone know how to fix that?

[_xxl]

Use MSVC7.1 (8.0) or GCC 4.0.3

[land]

I can't use msvc7.1. I'm working at university and I can only use free programs and msvc6.0. I have gcc 3.4.5. I tried compile ffdshow with msys but I didn't know how to do it

[_xxl]

I can't use msvc7.1. I'm working at university and I can only use free programs and msvc6.0. I have gcc 3.4.5. I tried compile ffdshow with msys but I didn't know how to do it
ffdshow can be compiled with GCC 4.0.3 - MinGW:
http://gda.utp.edu.co/~ceniza/GCC-4.0.3/
http://sourceforge.net/projects/mingw/
Set up enviroment for MinGW:
http://csjava.occ.cccd.edu/~gilberts/mingw/
You'll also need the following:
binutils, mingw-runtime, nasm-0.98.39, msys & w32api.
How to:
http://forum.doom9.org/showthread.php?p=727041#post727041
Don't forget to convert the DirectX SDK Lib!

[haruhiko_yamagata]

I can't use msvc7.1. I'm working at university and I can only use free programs and msvc6.0. I have gcc 3.4.5. I tried compile ffdshow with msys but I didn't know how to do it
Here (http://forum.doom9.org/showthread.php?p=840193#post840193)'s my post.

[_xxl]

Hi,
I'm newbie and I'll be very grateful to those who help me.
I'm trying to build ffdshow but I'm having some problems. I want to compile it with msvc6. I have install Microsoft Platform SDK Feb 2003 and DirectX 9.0 october 2004 (these are the latest updates for vc 6.0, isn't it?). After that I register visual studio and I check that in vc6 the paths are correct.
I want to use ffdshow with VirtualDub, that's mean that I only need to compile ff_vfw.dsw or must I compile something first?
When I compile ff_vfw the errors I obtain are:
Does someone know how to fix that?
Should we remove vc 6.0 files from ffdshow?

[clsid]

1. Im not sure of the wisdom of overwriting default installed codecs signed and supported by Microsoft with ffdshow ones. Like, why use the ffdshow mp3, lpcm mp1 and mp2 audio decoders by default? I think it is safer and easier for users to install ffdshow by default settings with only adding to the support media types, not changing default operating system codecs.

I know people who installed the game oblivion had issues with earlier versions of ffdshow because of it overwiting default codecs supported by the operating system itself.ffdshow does NOT overwrite any codecs on your system. It only has a higher merit than other filters. So if you enable a format in ffdshow, then ffdshow will be used to decode that format. If you disable it, your player will use another filter.

It is possible in the options to entirely disable the use of ffdshow in specific applications. I'll change the installer to do so for Oblivion.exe

[Yong]

What is impulse? How ffdshow is connected to covcolver(?) filter?

What a lame typo:p
Sorry.
u can download impulse files here:
http://www.sjeng.org/foobar2000.html

Im not sure its compatible with ffdshow internal convolver filter,
but its works with other external dshow convolver filter:
http://convolver.sourceforge.net/

I remember older version doesnt crash when loading impulse file, but it just dont work.

[haruhiko_yamagata]

u can download impulse files here:
http://www.sjeng.org/foobar2000.html

Im not sure its compatible with ffdshow internal convolver filter,
but its works with other external dshow convolver filter:
http://convolver.sourceforge.net/

I remember older version doesnt crash when loading impulse file, but it just dont work.
Thank you. Now I can reproduce the problem.

[Peuj]

A (stupid?) question about the installer:

The "Application plugins" -> "AviSynth" it's a plugin to be able to use AviSynth, it doesn't install AviSynth ?

[foxyshadis]

Actually, it's an avisynth plugin to use ffdshow inside of avisynth.

[Bathrone]

ffdshow does NOT overwrite any codecs on your system. It only has a higher merit than other filters. So if you enable a format in ffdshow, then ffdshow will be used to decode that format. If you disable it, your player will use another filter.

It is possible in the options to entirely disable the use of ffdshow in specific applications. I'll change the installer to do so for Oblivion.exe

Yes sorry I should have been more specific in my wording. What I'm getting at is questioning the wisdom of why media formats that are allready supported natively by Windows should be replaced, by default, by a higher priority filter from ffdshow.

I don't see how ffdshow can be considered as stable as release versions of Microsoft supported codecs. FFdshow is in active development.

I appreciate that problem applications can be ruled out within ffdshow options but that does not mean all the user community is going to know that or be equipped to do it. The default settings should be safe settings that are suitable for novice users.

To me the real value in ffdshow is to provide direct show filters for formats that are not natively supported in Windows like mpeg4-asp and AVC. Including others just increases the chances for people having a negative experience and for newer users to get confused about configuration.

[foxyshadis]

ffdshow defaults to enabling WMV? That's not good. Hmm, they don't appear in the current installer script, however, so it must be an artifact of old installers. Reset all ffdshow settings and try reinstalling.

WMV7/8/9/VC-1 should also all be part of the same checkbox in the installer, I think, they're all about about the same stage of completeness now that WMV9 is mostly fixed. If not, at least WMV7/8 and WMV9/VC-1. That's if they're ever returned to it.

[Bathrone]

I believe sir that currently the installers overwrite MP3, LPCM, MP1 and MP2 codecs by default. Recently the installers were changed not to overwrite MPEG1 and MPEG2 by default, so thanks for that. I don't believe any of the WMV decoders are enabled by default.

[land]

Here (http://forum.doom9.org/showthread.php?p=840193#post840193)'s my post.

Thank's for your response, but I have a problem. In the guide you point me it's necessary to apply the patch baseclasses.diff, can you tell me where can I find this patch?

[haruhiko_yamagata]

Thank's for your response, but I have a problem. In the guide you point me it's necessary to apply the patch baseclasses.diff, can you tell me where can I find this patch?
It's not me who wrote about baseclasses.diff. Perhaps it is written about old ffdshow. Current ffdshow have microsoft baseclasses in its SVN, we don't need the patch. We don't need STL port to compile ffdshow.

[land]

It's not me who wrote about baseclasses.diff. Perhaps it is written about old ffdshow. Current ffdshow have microsoft baseclasses in its SVN, we don't need the patch. We don't need STL port to compile ffdshow.

Ok, then I don't know how can I fix my problem. When I go to ffdshow/src directory and run make command I have the following errors:

.
.
.
.
wxutil.cpp:600: error: expected `;' before '{' token
wxutil.cpp:621: error: `_asm' was not declared in this scope
wxutil.cpp:621: error: expected `;' before '{' token
wxutil.cpp: In function `LONGLONG llMulDiv(LONGLONG, LONGLONG, LONGLONG, LONGLONG)':
wxutil.cpp:745: error: integer constant is too large for "long" type
wxutil.cpp:746: error: integer constant is too large for "long" type
wxutil.cpp: In function `LONGLONG Int64x32Div32(LONGLONG, LONG, LONG, LONG)':
wxutil.cpp:906: error: integer constant is too large for "long" type
wxutil.cpp:907: error: integer constant is too large for "long" type
In file included from strmiids.cpp:1,
from baseclasses_all.cpp:11:
../stdafx.h:76:22: qnetwork.h: No such file or directory
In file included from ../stdafx.h:82,
from strmiids.cpp:1,
from baseclasses_all.cpp:11:
../dwstring.h: At global scope:
../dwstring.h:92: error: ISO C++ forbids declaration of `CUnknown' with no type
../dwstring.h:92: error: expected `;' before '*' token
../dwstring.h:97: error: expected `;' before "static"
make[1]: *** [baseclasses_all.o] Error 1
make[1]: Leaving directory `/c/ffdshow/src/baseclasses'
make: *** [lib] Error 2

Do you think this could happen due to the version of the SDK? I have use the DirectX 9.0 October 2004 and the Microsoft Platform Feb 2003. Maybe I must use newer versions.

[haruhiko_yamagata]

In my wxutil.cpp line 600 is /* Do long multiplication */

line 621,745 does not match as well.
Which version of ffdshow are you trying to compile?

Check up your environmental variables.

[_xxl]

Ok, then I don't know how can I fix my problem. When I go to ffdshow/src directory and run make command I have the following errors:
Do you think this could happen due to the version of the SDK? I have use the DirectX 9.0 October 2004 and the Microsoft Platform Feb 2003. Maybe I must use newer versions.
You need: dxsdk_sum2004 & dxsdk_oct2005.
Don't forget to generate x86 directx9 lib files compatible with gcc.Use that script.(.sh)
http://forum.doom9.org/showthread.php?p=727041#post727041

[clsid]

I believe sir that currently the installers overwrite MP3, LPCM, MP1 and MP2 codecs by default. Recently the installers were changed not to overwrite MPEG1 and MPEG2 by default, so thanks for that. I don't believe any of the WMV decoders are enabled by default.
ffdshow may be in active development, but the decoding code for those formats is stable and rarely changed.
Those formats are enabled by default in ffdshow because most of us prefer using ffdshow above the default MS codecs. We could perhaps do a poll. But unless you have some specific reason (read: found a bug) for not using them, or a good argument why the MS ones are better, I'll leave things as they are now.

[haruhiko_yamagata]

I believe sir that currently the installers overwrite MP3, LPCM, MP1 and MP2 codecs by default. Recently the installers were changed not to overwrite MPEG1 and MPEG2 by default, so thanks for that. I don't believe any of the WMV decoders are enabled by default.
As far as I know, WMV decoders are not enabled by default.
As for the audio codecs, I agree to clsid.

[Egh]

I don't see how ffdshow can be considered as stable as release versions of Microsoft supported codecs. FFdshow is in active development.

To me the real value in ffdshow is to provide direct show filters for formats that are not natively supported in Windows like mpeg4-asp and AVC. Including others just increases the chances for people having a negative experience and for newer users to get confused about configuration.

Let's put it like that. You don't want ffdshow codecs, you don't install it :) Or you can disable the codecs you don't like.

The real value of ffdshow, as I see it, is provision of open-source alternatives for codecs, be those supported natively by windows or not (note there're also different windows versions). Different to all those m$ codecs you mentioned, ffdshow source is freely available.

[foxyshadis]

ffdshow may be in active development, but the decoding code for those formats is stable and rarely changed.
Those formats are enabled by default in ffdshow because most of us prefer using ffdshow above the default MS codecs. We could perhaps do a poll. But unless you have some specific reason (read: found a bug) for not using them, or a good argument why the MS ones are better, I'll leave things as they are now.

Although I believe this is the case, there's one report I'd like to test out:
http://forum.doom9.org/showthread.php?p=881257#post881257
I don't do much wmv encoding, but I'm sure it's not that hard to test a few things, so I'll see what happens.

Oops, I'm too late, it was already fixed by drevil. btw, I'm going to update the homepage for a more recent build, 127 is oooold. I also have QoS working pretty well (except with haali's), but I'm going to sit on it until we start working on disruptive features again. I don't want to introduce more bugs when it's coser to bugfree than it's ever been.

[bob0r]

Added ffdshow_rev382_20061012_clsid.exe to x264.nl
"faster H.264 decoding."

It indeed looks like it's faster, but coreavc is still much much faster in decoding a BBC-HD H.264 1080mbaff stream.

But i guess its good to see there is improvement.

[LoRd_MuldeR]

Added ffdshow_rev382_20061012_clsid.exe to x264.nl
"faster H.264 decoding."

It indeed looks like it's faster, but coreavc is still much much faster in decoding a BBC-HD H.264 1080mbaff stream.

But i guess its good to see there is improvement.

The first time MPC + ffdshow can play that super high-res Superman trailer without heavy frame drop :)
But only with Overlay Mixer and everything else turned off.
Performance in MPlayer is still a bit better...

[clsid]

There is some assembly code for cabac added in FFmpeg recently. That might give an additional few % performance boost when decoding H.264. However, I can't get that code to compile with GCC atm. The code contains a comment that this might happen, so I guess it is still a work in progress. Hopefully this gets fixed soon.

I benchmarked an improvement of ~4% in rev382 compared to rev312.

[Livesms]

What latest STABLE ffdshow build you can recommend?

Now I use CCCP - Combined Community Codec Pack (http://www.cccp-project.net/)) pack with ffdshow dated with Jul 26 2006 compiled by Milan Cutka (milan_cutka@yahoo.com)

Is it possible to replace ffdshow with new stable version?

[_xxl]

What latest STABLE ffdshow build you can recommend?
Now I use CCCP - Combined Community Codec Pack (http://www.cccp-project.net/)) pack with ffdshow dated with Jul 26 2006 compiled by Milan Cutka (milan_cutka@yahoo.com)

Milan?

[Livesms]

[QUOTE=Livesms;887249]
Now I use CCCP - Combined Community Codec Pack (http://www.cccp-project.net/)) pack with ffdshow dated with Jul 26 2006 compiled by Milan Cutka (milan_cutka@yahoo.com)[QUOTE]
Milan?

Maybe Milan. In any case his sign is placed in buttom of about page.

[Episode]

@Livesms: You can install any version you want on top of CCCP without any worry. With clsid's version you just need to be sure that it installs ffdshow on Program Files\Combined Community Codec Pack\Filters directory instead of Program Files\Combined Community Codec Pack\Filters\ffdshow (you need to write this by hand, for some reason just selecting plain Filters directory won't install it on there.) The most current version by clsid is in my opinion at least as stable as the one that came with cccp, so there is no need to worry about stability, just download the newest one.

[Livesms]

@Livesms: You can install any version you want on top of CCCP without any worry. With clsid's version you just need to be sure that it installs ffdshow on Program Files\Combined Community Codec Pack\Filters directory instead of Program Files\Combined Community Codec Pack\Filters\ffdshow (you need to write this by hand, for some reason just selecting plain Filters directory won't install it on there.) The most current version by clsid is in my opinion at least as stable as the one that came with cccp, so there is no need to worry about stability, just download the newest one.

Could you give me a link(or post number) to latest stable clsid's build.

[clsid]

Now I use CCCP - Combined Community Codec Pack (http://www.cccp-project.net/) pack with ffdshow dated with Jul 26 2006 compiled by Milan Cutka (milan_cutka@yahoo.com)
Those builds are made by the CCCP team themselves, not by Milan.

You can find the latest 'stable' build on www.x264.nl

[bob0r]

Please dont use www. its so useless! :mad:

[clsid]

Its just so that vBulletin automatically makes a clickable link out of it. Otherwise I would have left away the www :)

[Liisachan]

it's much easier to config your server by typing a few lines to automaticlly force everyone who use it, than to ask human beings to do what you'd like, no? :)

RewriteEngine on
RewriteCond %{HTTP_HOST} www.x264.nl [NC]
RewriteRule ^(.*)$ http://x264.nl/$1 [R=301,L]

[_xxl]

List of known issues in revision 363:
5) MSS2 WMV files: MPC crashes when info & debug window of ffdshow is active. Sample file (http://ftp.mplayerhq.hu/MPlayer/samples/V-codecs/MSS2/mss2_speech.wmv). (reported by drevil_xxl) File crashes regardless of selected page. (reported by clsid)

Please confirm that rev 382 works.

[Livesms]

Those builds are made by the CCCP team themselves, not by Milan.

You can find the latest 'stable' build on www.x264.nl

And what about SSE2/SSE3/3dNow! Support.

Looks like everyone has CPU at least with SSE2.
My Sempron 3000 supports SSE2/SSE3/3DNow and P4 2.8 supports SSE2/SSE3.

Is there any build with such optimizations.

[bob0r]

@clsid:
Try: http://x264.nl and press submit.

@Liisachan:
Way too lazy for that, i have to message people then :p

@Livesms:
As read in this thread, the ffdshow installer has all optimized files in it.
Some specific optimizes do not speed up at all.
So the build is best for general use.

[clsid]

And what about SSE2/SSE3/3dNow! Support.

Looks like everyone has CPU at least with SSE2.
My Sempron 3000 supports SSE2/SSE3/3DNow and P4 2.8 supports SSE2/SSE3.

Is there any build with such optimizations.
ffdshow contains MMX, MMXext, SSE, SSE2, 3dnow and 3dnow2 code.

And no, not everyone has at least SSE2.

[BlindWanderer]

I've found a new bug, or really just an unhandled one.

ffdshow doesn't want to play wmv1/7 streams if the first frame is not a key frame (or is a corrupt key frame). In media player black is all that is displayed (for the duration of the video). When the user seeks the video, the video is properly displayed.

VLC doesn't choke on this.

(i've been using the latest builds)

example: http://cabbit.home.comcast.net/russians_300k.wmv

[FredThompson]

Looks like everyone has CPU at least with SSE2.That is a...very...(opposite of insightful)...assumption.

[BlindWanderer]

Only chips newer then (and including) Pentium 4's support SSE2. I still have a P2 & P3 still in operation. I imagine there are still allot of people using Athlons too. If you checkout the wikipedia sse2 article, they list chips that do and do not support SSE2.

[haruhiko_yamagata]

WMP11 and mplayer2.exe crashes when decoding MJPEG. Most of the files here (http://samples.mplayerhq.hu/V-codecs/MJPEGs/) crashes. MPC can play them.

The crash occurs in mjpeg_decode_frame in mjpeg.c.
case SOS:
mjpeg_decode_sos(s);

Between these two lines, we have stack over flow exception. I can't find why.

[_xxl]

WMP11 and mplayer2.exe crashes when decoding MJPEG. Most of the files here (http://samples.mplayerhq.hu/V-codecs/MJPEGs/) crashes. MPC can play them.

The crash occurs in mjpeg_decode_frame in mjpeg.c.
case SOS:
mjpeg_decode_sos(s);

Between these two lines, we have stack over flow exception. I can't find why.
Jpeg in .mov files doesn't work.
http://rapidshare.de/files/36834385/P1020240.MOV.html
ffdshow-20060226 is working fine,
ffdshow-20060420 is not.

[clsid]

List of known issues in revision 413:

1) When outputting the audio to AC3 using the ffdshow AC3 encoder there are a lot of downmix overflows. The sound seems to be distorted as if the volume is to high. (reported by TFM_TheMask)
2) Loading IFO file in MPC and seeking thru the video instantly crashes MPC. Seeking thru VOB file is fine.(reported by DSP8000)
3) Wavpack decoder crashes on this file (http://rapidshare.de/files/34646550/SoC__XviD_WavPack_MP3_-007.mkv.html). (reported by clsid)
4) ICL9 builds of ffdshow.ax crash on files created by a specific old revision of x264 (don't know the rev number). Funny thing however is that the files play without crash if you first play a good file and then play a 'troublesome' file in the same player instance. Also no crash when using a plain unoptimized debug build. Sample file (http://rapidshare.de/files/35447656/sample.mp4.html). (reported by clsid)
5) ffdshow crashes when I try to play snow (encoded with latest mplayer/lavc). (reported by Yong) Sample (http://rapidshare.de/files/36857216/SoC__SNOW_Vorbis_.mkv.html) that crashes during playback at time 00:53. More crash details (http://forum.doom9.org/showthread.php?p=888583#post888583).
6) image settings->preset autoload conditions-> "on FOURCC match" ain't working. (reported by juskixxx)

Reported but unconfirmed bugs:

7) Instant crash when trying to play this MSS2 WMV file (http://ftp.mplayerhq.hu/MPlayer/samples/V-codecs/MSS2/mss2_speech.wmv). Happens on Windows 2000. No crash on drevil_xxl's XP system. (reported by clsid)
8) The following encoders do not work for me: MPEG 4, MPEG 1, MPEG 2, h.263, H.261 and DV. VirtualDub 1.16.16 gives the following error: "Cannot start video compression. An unknown error occurred (may be corrupt data). (error code -100)". I know that at least some of these encoders do work for others. My system specs: Windows 2000, AMD Athlon Thunderbird. (reported by clsid)
9) QPEL bugs and colour/block trains/displacement. Details (http://sourceforge.net/tracker/index.php?func=detail&aid=1568940&group_id=53761&atid=471489). (reported by drevil_xxl)
10) Resize filter alters colors. Details (http://forum.doom9.org/showthread.php?p=884778#post884778). (reported by Kador)

Bugs currently 'fixed' by disabling/removing functionality:

11) VC-1 does not work. ffdshow isn't even placed in the DirectShow graph. (reported by clsid)
12) VMnc crashes with "Integer devision by zero" exception. (reported by clsid)


ffdshow doesn't want to play wmv1/7 streams if the first frame is not a key frame (or is a corrupt key frame). In media player black is all that is displayed (for the duration of the video). When the user seeks the video, the video is properly displayed. Sample file (http://cabbit.home.comcast.net/russians_300k.wmv). (reported by BlindWanderer)The MS WMVideo Decoder DMO has the same problem. So not really a 'bug' specific to ffdshow.

ffdshow crashes when decoding stereo 192kbps mp3 using AuD-X lib. (reported by drevil_xxl)Do you have a sample file?

[BlindWanderer]

The MS WMVideo Decoder DMO has the same problem. So not really a 'bug' specific to ffdshow.
That explains why a number of years ago some of my files became broken; must have been when I installed a new version of the WM DMO.

Guess i should be greatful it didn't crash instead :p

[BlindWanderer]

I've been playing with the M$ Vista RC. ffdshow works, only problem is the config interface doesn't quite work properly in one regard. The hack used on the Codecs page to position the dropdown list no longer works. Instead the dropdown list is placed at the top of the column over the column title.

(see attachment)

[haruhiko_yamagata]

6) ffdshow crashes when I try to play snow (encoded with latest mplayer/lavc). (reported by Yong) Sample (http://rapidshare.de/files/36857216/SoC__SNOW_Vorbis_.mkv.html) that crashes during playback.

It works for me.
MPC / Zoom player / Mplayer2.exe
Haali Media splitter / MPC's internal matroska
RGB out / YUY2 out
Resize on / off
OS = Windows Xp sp2
CPU = P4HT

[Peuj]

It works for me.
MPC / Zoom player / Mplayer2.exe
Haali Media splitter / MPC's internal matroska
RGB out / YUY2 out
Resize on / off
OS = Windows Xp sp2
CPU = P4HT

no crash for me too

edit: My config
MPC
Haali Media splitter / MPC's internal matroska
OS = Windows Xp sp2
CPU = P4HT

[clsid]

It crashes during playback at time 00:53

[kurt]

List of known issues in revision 394:...
6) ffdshow crashes when I try to play snow (encoded with latest mplayer/lavc). (reported by Yong) Sample (http://rapidshare.de/files/36857216/SoC__SNOW_Vorbis_.mkv.html) that crashes during playback at time 00:53.
no crash here

haali media splitter/gabest
mpc (latest by c_d)
ffdshow_rev392_20061015_clsid.exe
win xp pro
core duo 2.00ghz

[_xxl]

It crashes for me.
haali media splitter/gabest
mpc
ffdshow_rev392_20061015_clsid.exe
win xp pro
AMD XP

[videomixer9]

works just fine here.

haali media splitter
mpc rev611-2
windows xp sp2
athlon xp 3000+
also installed the clsid build to test

[CruNcher]

Hi, im working @ the moment on my X264 based intermediate Editing Workflow and i found something susspicius in the way and i think it's a ffdshow vfw bug, the problem is with an avi file it's framerate is 29.970 but ffdshow shows this in the host application (Virtualdub) as 29.969 i tried 29.971 then that is shown correctly as 29.971 but 29.969 for example is shown as 29.968 something is seriously wrong here.

http://cruncher.mufflastig.com/hdtv/hdv/intermidiate/huhwtf.png

Edit: ok it seems to be a avc2avi bug thx to seeag1 for the help :)

[_xxl]

http://i10.tinypic.com/35l85f7.jpg
Snow crashes for me.

[Eragon4ever]

I just got an error message while playing the SNOW sample. http://img441.imageshack.us/img441/692/snap0002pq8.png

I played it 2 times before without this message.

haali media splitter
WMP
ffdshow_rev392_20061015_clsid.exe
win xp sp2
P4 HT 3,0

Edit: I noticed an other bug with this sample.
When you play the file without seeking the video stops at 01:46 of 01:47. However if I seek the video restarts as it should (repeat function).

Config is above.

[foxyshadis]

Here's a current svn of mplayer (http://forum.doom9.org/showthread.php?p=839965); if it displays the same behavior it won't do much good to be concerned with it until ffmpeg can fix it. I'm wary of snow bug reports, since it hardly ever works for me in mplayer anyway.

[LigH]

And no, not everyone has at least SSE2.

:o Still AMD Duron 800 here...

celtic-druid (http://celticdruid.no-ip.com/xvid) has a nice poll on his site.

[haruhiko_yamagata]

A bit too late, let me introduce new build of MPC.
celtic_druid has build new MPC with a patch (http://sourceforge.net/tracker/index.php?func=detail&aid=1577522&group_id=82303&atid=565651) posted by me.
Download (http://ffdshow.faireal.net/#mpc)

It works best with newest ffdshow.
rev389
Bug fix : Output queue enabled + VMR9 renderless + RGB32 + specific video cards + pause = blackout problem

Patched Media Player Classic + newest ffdshow can use queue without these bugs.
Patched Media Player Classic + old ffdshow: no merit.
With non patched Media Player Classic, ffdshow does not use queue in the setting above.

With the new MPC, screen shot(F5:Save Image) works in the same setting.

mplayerc.rev611-2.2kxp.7z (http://ffdshow.faireal.net/mirror/Media%20Player%20Classic/mplayerc.rev611-2.2kxp.7z) have updated internal AAC decoder.

[clsid]

Does anyone know of anyone who is actually using the SNOW codec in real life? Afaik it is still an experimental codec. Perhaps it would be best to just remove it altogether.

[_xxl]

Jpeg in .mov files doesn't work.
http://samples.mplayerhq.hu/V-codecs/MJPEGs/Minolta-MJPEG.mov.
ffmpeg (libavcodec) log message:
Subsampling in JPEG-LS is not supported.

[_xxl]

Does anyone know of anyone who is actually using the SNOW codec in real life? Afaik it is still an experimental codec. Perhaps it would be best to just remove it altogether.
Remove it Permanently.

[Episode]

Where can I download rev389?

http://www.ffdshow.info as usual.

[haruhiko_yamagata]

Does anyone know of anyone who is actually using the SNOW codec in real life? Afaik it is still an experimental codec. Perhaps it would be best to just remove it altogether.
Let's remove it temporarily and restore when ready. As far as we borrow code from ffmpeg, it's not good to ignore SNOW.

[clsid]

The InnoSetup installer script now has (optional) localization support. Everyone that speaks languages other than English, feel free to update "custom_messages.iss".

[juskixxx]

clsid builds and h-yamagata builds: image settings->preset autoload conditions-> "on FOURCC match" ain't working (none of your tryouts builds work)
drevils builds work just fine....

would be nice to have individual PP, sharpening, softening settings etc. setting for every FOURCC...



Thanks anyway for making great job with ffdshow!!!

[Bathrone]

Drevil XXL build 411 during setup on Vista RC2 wont allow me to unitck dscaler, vfw and all the other options for a custom install.

[_xxl]

Drevil XXL build 411 during setup on Vista RC2 wont allow me to unitck dscaler, vfw and all the other options for a custom install.
Fixed.

[_xxl]

ffdshow crashes when Avisynth plugin is used.
# Prints out version information

Version()

http://i11.tinypic.com/42x4mqw.jpg
MPlayer2 is using WMR-7 in non-mixing mode.
MPC is working fine.

[clsid]

Are there any known issues with GCC 4.0.3 builds of libavcodec? I can't remember if all of them were resolved or not.

H.264 decoding is a little bit faster for me with GCC 4.0.3. I don't know about the impact on the other formats. If there aren't any known issues, then I am thinking of starting to use 4.0.3 for libavcodec.

[haruhiko_yamagata]

Are there any known issues with GCC 4.0.3 builds of libavcodec? I can't remember if all of them were resolved or not.

H.264 decoding is a little bit faster for me with GCC 4.0.3. I don't know about the impact on the other formats. If there aren't any known issues, then I am thinking of starting to use 4.0.3 for libavcodec.
I have been using GCC 4.0.3 for all of my builds, I have never experienced bugs caused by GCC version.

[haruhiko_yamagata]

clsid builds and h-yamagata builds: image settings->preset autoload conditions-> "on FOURCC match" ain't working (none of your tryouts builds work)
drevils builds work just fine....
It's working for me. Please give us detailed information.

[clsid]

Matching based on filename (filename==presetname) works ok. Matching on FourCC does not work for me either.

[_xxl]

http://img17.imagevenue.com/loc457/th_51966_Capture_122_457lo.jpg (http://img17.imagevenue.com/img.php?image=51966_Capture_122_457lo.jpg)
Mingw GCC 4.0.3 fails.

[ExtraEye]

there's some serious issue with avisynth filter in ffdshow. when enabling avisynth (even without any filters in it), cpu jumps from 30-40 to 100. disabling it wouldn't work so you have to restart the video in order for it to come back to normal cpu values...

[clsid]

I don't have such a cpu usage increase when I enable the AviSynth filter. I am using AviSynth 2.5.6a.

But there is a funny thing when I used the Version() filter to test. The output is mirrored both horizontally and vertically. I suppose that isn't how it is supposed to work?

[igor1st]

Are there any known issues with GCC 4.0.3 builds of libavcodec? I can't remember if all of them were resolved or not.
Always work ok for me.


H.264 decoding is a little bit faster for me with GCC 4.0.3. I don't know about the impact on the other formats. If there aren't any known issues, then I am thinking of starting to use 4.0.3 for libavcodec.
This situation changes with time. For example ~r300 (IIRC) was faster with 4.0.3, but r371 was faster with 3.4.5 again.

But you should change compiler for deinterlacers - kernel/toms are much more faster with 4.0.3.

[videomixer9]

those things weren't updated forever, I usually also compiled them with ICL9 on the compiles I did with ICL, Core2 Duo users may want to try the kerneldeint and tomsmocomp from my Core2 Duo build. To not get your memory eaten away use link time code generation to compile both with ICL, that works way better than the regular way.

[ExtraEye]

I don't have such a cpu usage increase when I enable the AviSynth filter. I am using AviSynth 2.5.6a.


after reading your comment i went to the doom9 main page to download avisynth (according to it it's:AviSynth 2.56).
afterwards tried reanabling it on ffdshow and the effect was the same.
is doom9's link to avisynth bad or is my version o.k?

[Egh]

afterwards tried reanabling it on ffdshow and the effect was the same.


lol do you have anything in the avisynth section? If you have something like FFT3D() there it's no wonder CPU usage increases :)

I don't have any noticable increase for simple filters in avs.

I use last 2.5.7 alpha.

[ExtraEye]

as i said:
when enabling avisynth (even without any filters in it), cpu jumps from 30-40 to 100.

Besides. I also said that even after I disable the avisynth filter the cpu doesn't go down. I have to restart the playback in order for it to go back to normal again.
Maybe im not using the right version of something? Although ffdshow is from 18/10... i tried all the different latest versions from the different builders and this happens in all of them.

[foxyshadis]

Are all input colorspaces checked? Oddly, I've never seen it, but I haven't really checked for differences between on and off (it'd be a lot less noticeable on my system, likely, compared to a 3000+).

[ExtraEye]

yes they are.

[haruhiko_yamagata]

as i said:


Besides. I also said that even after I disable the avisynth filter the cpu doesn't go down. I have to restart the playback in order for it to go back to normal again.
Maybe im not using the right version of something? Although ffdshow is from 18/10... i tried all the different latest versions from the different builders and this happens in all of them.
What is the color space of your input file and output? Color space conversion before and after avisynth may happen. In that case, I guess you'll have to restart the video application after you uncheck avisynth, which I don't think a bug.

[ExtraEye]

It's normal for avisynth to make the video playing process weigh twice as much in memory just for enabling it ?
Are color space conversions supposed to take so much memory?

edit: forgot to answer the question ^^.
input colorspace: YV12
output colorspace: RGB32 high quality conversion (because I use VMR9)

Avisynth is set to recieve any colorspace it offers to handle (YV12, YUY2,RGB24,RGB32). So why should it make any space conversion happen in the first place if the actual script box is empty?

[kneep]

I got an error when build rev420 with gcc4.0.3:
============================================
make[1]: Entering directory `/d/cyghome/src/ffdshow-tryout/src/codecs/wmv9'
gcc -c -I. -I../.. -I../../cygwin -I../../baseclasses -Iinclude -DSUPPORT_INTERLACE -I/dx/Include -L/dx/MingLib -ldx9 -mno-cygwin -mdll -mthreads -pipe -mmmx -msse -mfpmath=sse -msse2 -DNDEBUG -UDEBUG -DWIN32 -D_WIN32 -DUNICODE -D_UNICODE -w -fno-rtti -fno-exceptions -O2 -march=pentium4 -mtune=pentium4 -fomit-frame-pointer -finline-functions -finline -frename-registers -fweb -funit-at-a-time -MMD -o ff_wmv9.o ff_wmv9.cpp
../../array_allocator.h:42: error: expected template-name before '<' token
../../array_allocator.h:42: error: expected `{' before '<' token
../../array_allocator.h:42: error: expected unqualified-id before '<' token
../../imgFilters/ffImgfmt.h:372: error: expected template-name before '<' token
../../imgFilters/ffImgfmt.h:372: error: expected `{' before '<' token
../../imgFilters/ffImgfmt.h:372: error: expected unqualified-id before '<' token
../../imgFilters/ffImgfmt.h: In function 'void csp_yuv_order(int&, unsigned char**, stride_t*)':
../../imgFilters/ffImgfmt.h:468: error: 'swap' is not a member of 'std'
../../imgFilters/ffImgfmt.h:469: error: 'swap' is not a member of 'std'
ff_wmv9.cpp: At global scope:
ff_wmv9.cpp:50: error: using-declaration for non-member at class scope
ff_wmv9.cpp:50: error: expected ';' before '<' token
ff_wmv9.cpp: In constructor 'Tff_wmv9::Tff_wmv9()':
ff_wmv9.cpp:105: error: 'codecs' was not declared in this scope
ff_wmv9.cpp:115: error: 'codecs' was not declared in this scope
ff_wmv9.cpp:115: error: 'sort' was not declared in this scope
ff_wmv9.cpp: In member function 'virtual size_t Tff_wmv9::getCodecCount()':
ff_wmv9.cpp:124: error: 'codecs' was not declared in this scope
ff_wmv9.cpp: In member function 'virtual bool Tff_wmv9::getCodecInfo(size_t, const Tff_wmv9codecInfo**)':
ff_wmv9.cpp:127: error: 'codecs' was not declared in this scope
ff_wmv9.cpp:128: error: 'codecs' was not declared in this scope
ff_wmv9.cpp: In member function 'virtual bool Tff_wmv9::start(const Tff_wmv9cfg&)':
ff_wmv9.cpp:177: error: 'codecs' was not declared in this scope
ff_wmv9.cpp: In member function 'bool Tff_wmv9::startAVI()':
ff_wmv9.cpp:390: error: 'vector' was not declared in this scope
ff_wmv9.cpp:390: error: expected primary-expression before '>' token
ff_wmv9.cpp:390: error: '::const_iterator' has not been declared
ff_wmv9.cpp:390: error: expected `;' before 'c'
ff_wmv9.cpp:390: error: 'c' was not declared in this scope
ff_wmv9.cpp:390: error: 'codecs' was not declared in this scope
ff_wmv9.cpp: In member function 'virtual const Tff_wmv9codecInfo* Tff_wmv9::findCodec(FOURCC)':
ff_wmv9.cpp:653: error: 'vector' was not declared in this scope
ff_wmv9.cpp:653: error: expected primary-expression before '>' token
ff_wmv9.cpp:653: error: '::const_iterator' has not been declared
ff_wmv9.cpp:653: error: expected `;' before 'c'
ff_wmv9.cpp:653: error: 'c' was not declared in this scope
ff_wmv9.cpp:653: error: 'codecs' was not declared in this scope
ff_wmv9.cpp: In member function 'virtual bool Tff_wmv9::decStart(FOURCC, double, unsigned int, unsigned int, const void*, size_t, int*)':
ff_wmv9.cpp:664: error: 'vector' was not declared in this scope
ff_wmv9.cpp:664: error: expected primary-expression before '>' token
ff_wmv9.cpp:664: error: '::const_iterator' has not been declared
ff_wmv9.cpp:664: error: expected `;' before 'c'
ff_wmv9.cpp:665: error: 'c' was not declared in this scope
ff_wmv9.cpp:665: error: 'codecs' was not declared in this scope
make[1]: *** [ff_wmv9.o] Error 1
make[1]: Leaving directory `/d/cyghome/src/ffdshow-tryout/src/codecs/wmv9'
make: *** [WMV9] Error 2
========================================
Can anyone help?
Thank you.

[DSP8000]

Maybe just a cosmetic bug,
in the about tab build rev.420 shows, build on 17 Oct even that the build is from 20th of Oct.
This is ffdshow_rev420_20061020_clsid.

[_xxl]

I got an error when build rev420 with gcc4.0.3:
Thank you.
Make sure the include path is correct for the DX include and the library files.
"-L/dx/MingLib" is "c:/Msys/dx/MingLib" and
"-I/dx/Include " is c:/Msys/dx/Include".

[kneep]

Make sure the include path is correct for the DX include and the library files.
"-L/dx/MingLib" is "c:/Msys/dx/MingLib" and
"-I/dx/Include " is c:/Msys/dx/Include".
Yes. I confirm this.
If the path were not correct, it should not have found the headers and would have failed the bassclasses compilation before this.
Seems like sort of grammar error but I keep every source file untouched.
I'm not an expert in C++ template grammar.

[BlindWanderer]

I think you need to change the "-c" flag to "-cpp"

[sillKotscha]

hello,

since here are most of all devs are around I may ask a question about one of my files...

it is an very old mov file which I can't get to play in the way it should play...

I use ffdshow_rev420_20061020 by clsid and the following infos are shown...

audio:

http://img343.imageshack.us/img343/2365/audioec3.jpg (http://imageshack.us)

video:

http://img343.imageshack.us/img343/4720/videosc4.jpg (http://imageshack.us)

I use latest MPC by celtic_druid and enabled the built-in MP4/MOV source filter

maybe it is a splitter issue but it may also some kind of an audio issue (the way ffdshow handles the mono stream).

it would be very kind of you to have a look at the clip.

it is hosted here:

http://www.verzend.be/v/8468583/Tyson_Bite.mov.html

thank you

EDIT:

S/PDIF output is because of on.the.fly audio->AC3 transcoding

[clsid]

That sample outputs a lot of noise here. Sound is ok with QuickTime as decoder.

[sillKotscha]

1st thx for taking the time...

2nd: I know that sound is ok when played via quicktime OR using other than built-in mp4 splitter used by mpc...

what do you mean by "outputs a lot of noise"? the picture does look acceptable for CVID and a nearly 10 year old clip...

but that is not my question. Are you able to reproduce my probleme with a setup as described above? If yes, would you say it may be a ffdshow issue or you tend to say it is more or less splitter related...

thank you

[_xxl]

S/PDIF output is because of on.the.fly audio->AC3 transcoding
1) When outputting the audio to AC3 using the ffdshow AC3 encoder there are a lot of downmix overflows. The sound seems to be distorted as if the volume is to high. (reported by TFM_TheMask)

It is broken since rev 2517.
EDIT:
I have fixed this bug.

[clsid]

I meant audio noise.

If it works ok with other splitters (which ones?) then it could indeed just be a bug in the internal splitter of MPC.

[sillKotscha]

thank you... if it is released, I'll test it asap :-)

...other splitters (which ones?)

ok, now it gets difficult :D

1st)

IF I disable built-in MP4/MOV splitter, QuickTime Movie Parser is used for the Mike Tyson sample...

http://img105.imageshack.us/img105/1538/splitterfortysonwf6.jpg (http://imageshack.us)

2nd)

I have a mov (svq3) which is decoded very nice & smooth by ffdshow in conjunction with Nero MP4 Splitter

http://img174.imageshack.us/img174/8288/splitterforsvq3mm1.jpg (http://imageshack.us)

so, no problem someone could say... but

3rd)

I have another svq3 encoded mov which won't be decoded by ffdshow with the following error message:

http://img105.imageshack.us/img105/961/erroruu9.jpg (http://imageshack.us)

for testing purposes I have uploaded Tyson as seen above and the svq3 sample here:

http://www.verzend.be/v/6340154/at_the_gay_bar.mov.html

because of all the hazzle I like MPC's built-in splitter very much and the Tyson sample is the only one I got which "won't play"... therefor I thought it could be ffdshow related.

and before someone asks... I do have Haali's media splitter installed :) but it doesn't help much and to get even more OT:

I do like MPC's MP4/MOV splitter because of another nice feature...

it is able to display tagged infos as seen here:

http://img107.imageshack.us/img107/8746/infonz0.jpg (http://imageshack.us)

neither Nero's splitter nor Haali's splitter show up these tags using MPC

[clsid]

ffdshow audio decoder doesn't seem to be used when you use QuickTime Movie Parser in the Mike Tyson sample. What happens if you disable IMA ADPCM in ffdshow and use the internal splitter of MPC?

That ffdshow error is because it isn't able to decode SVQ3 video properly when QuickTime Movie Parser (a standard Windows filter) is used.

[sillKotscha]

ffdshow audio decoder doesn't seem to be used when you use QuickTime Movie Parser in the Mike Tyson sample. What happens if you disable IMA ADPCM in ffdshow and use the internal splitter of MPC?

than I have smooth playback of video but no audio...

That ffdshow error is because it isn't able to decode SVQ3 video properly when QuickTime Movie Parser (a standard Windows filter) is used.

I understand that but as I said I have another svq3 sample which does play nice and Nero's splitter gets used and with the posted svq3 sample it seems QuickTime Mov Parser is choosen automatically by windows - I didn't lowered/ raised any merits.

I hope you can understand me :scared:

[clsid]

Yes, I understand. The QuickTime Movie Parser has a higher merit. But if it fails to parse a file, then Nero will be used. If you prefer the Nero filter, then increase its merit.

[Bathrone]

Drevil thankyou for fixing the installer bug. :)

Some questions for all the installers:

1. Wouldnt it be better if ffdshow used the windows speaker configuration within the OS rather than have an external speaker configuration within ffdshow?

2. Why is flac optional in some installers and not present in others? Whats wrong with the flac decoder in ffdshow anyway for it not to be installed by default?

3. Why is explorer.exe disabled by default? This will prevent thumbnailling within Vista for mpeg4-asp and h.264 encoded video within avi files.

[foxyshadis]

1. Wouldnt it be better if ffdshow used the windows speaker configuration within the OS rather than have an external speaker configuration within ffdshow?
We discussed that a few pages back. (Not that I hold it against you, even a few pages is a long way back. :p) Since it's not held in one constant registry key, it requires enumeration and some heuristics (and probably testing on a lot of different sound drivers), so it isn't in yet.

3. Why is explorer.exe disabled by default? This will prevent thumbnailling within Vista for mpeg4-asp and h.264 encoded video within avi files.
That's exactly why. Thumbnailing consumes obscene resources on XP and has led to at least a few crash and hang reports, notably on broken avis. One can only hope that Vista's avi splitter is smarter than xp's. You can always take it out and test it, we'd be glad to hear how well Vista handles thumbnailing and info-gathering on broken avis, and how well it works with ffdshow in general.

[Bathrone]

I can report on many vista builds, including the latest ones RC1 and RC2 I have not had any problems with enabling thumbnailling for H.264 or MPEG4-ASP encoded video inside avi containers.

I have installed Haali media splitter latest version ontop of Vista, but did not enable the overwrite of the avi splitter from the default Vista one.

With MKV containers thumbnailling doesnt work for Vista using Haali media splitter.

If anyone has any problem avi files they want tested under ffdshow and vista I would be happy to do so.

[DeeGee]

I also have a small problem with the svq3 decoder. I decided to check this video (http://www.gamershell.com/news/33732.html) out. I have quicktime alternative, haali media splitter and ffdshow installed. I renamed the video to .hdmov that causes mpc to use haali + ffdshow to decoding.

The problem is that partly the video plays correctly, but some cuts seem to cause the frame order to change from:
--0-1-2-3-4-5-6-7-8-9
to something like:
1-0-3-2-5-4-7-6-9-8--
So basically the video stutters.

[Bathrone]

Just wondering why raw video and output is not set to YV12 be default?

[_xxl]

I can report on many vista builds, including the latest ones RC1 and RC2 I have not had any problems with enabling thumbnailling for H.264 or MPEG4-ASP encoded video inside avi containers.
Confirmed.

[_xxl]

Access Violation in ffdshow on Windows Vista RC2.(5744)
http://sourceforge.net/tracker/index.php?func=detail&aid=1567236&group_id=53761&atid=471489
FFdshow rev 350 is working ok, no crash.
FFdshow reports 0% cpu usage always.

[WorBry]

Quick question.

Can FFV1 now be considered stable and suitable for archiving or is it still 'experimental' and subject to further development?

[foxyshadis]

Yes, FFV1's format hasn't changed in well over a year and is officially considered permanently stable now.

DeeGee: Didn't see anything like that. There's a hokey slo-mo at the beginning that works by doubling every frame, but it decoded fine. (If ungodly slow.) Neat video, too, but I has hoping for more racing. =p

[BlindWanderer]

I've been using ffdshow in Vista (5744 - 32 bit) without problems (besides the interface glitch previously mentioned).

I watched a mpeg2 movie, and resized the video (decoding & resizing done in ffdshow). Was using clsid's build 382 (couple days ago).

I'll hop back over to the vista build in a couple days to check with new builds. The question that needs to be asked: is it 32-bit or 64 bit vista?

[thuan]

There're maybe 2 new bugs in ffdshow:
First is the automatic preset loading seems to be not working right with on "number of channel match" and AC3 audio on DVD (don't know if it happens with AC3 audio in MKV, AVI or not or any other audio format). I use "3;4;5;6" in "number of channel match" so this should load with audio track with more than 2 channels but with an AC3 2 channels track this is also loaded.

The last is Automatic Quality Control in PostProcessing video filter is broken I think. At least with what I see with the preset slider. Normally the dot should go up and down as the CPU load differs but now it only stuck at where it defaults.

The ffdshow I use is clsid's 420.

[clsid]

Access Violation in ffdshow on Windows Vista RC2.(5744)
http://sourceforge.net/tracker/index.php?func=detail&aid=1567236&group_id=53761&atid=471489
FFdshow rev 350 is working ok, no crash.
FFdshow reports 0% cpu usage always.
I think haruhiko fixed that bug by removing the CPU usage OSD item.

[clsid]

List of known issues in revision 423:

1) Wavpack decoder crashes on this file (http://rapidshare.de/files/34646550/SoC__XviD_WavPack_MP3_-007.mkv.html). (reported by clsid)
2) ICL9 builds of ffdshow.ax crash on files created by a specific old revision of x264 (don't know the rev number). Funny thing however is that the files play without crash if you first play a good file and then play a 'troublesome' file in the same player instance. Also no crash when using a plain unoptimized debug build. Sample file (http://rapidshare.de/files/35447656/sample.mp4.html). (reported by clsid)
3) image settings -> preset autoload conditions -> "on FOURCC match" doesn't work. (reported by juskixxx)
4) Interface bug in Windows Vista. Details and screenshot (http://forum.doom9.org/showthread.php?p=888312#post888312). (reported by BlindWanderer)

Reported but unconfirmed bugs:

5) Loading IFO file in MPC and seeking thru the video instantly crashes MPC. Seeking thru VOB file is fine.(reported by DSP8000)
6) Instant crash when trying to play this MSS2 WMV file (http://ftp.mplayerhq.hu/MPlayer/samples/V-codecs/MSS2/mss2_speech.wmv). Happens on Windows 2000. No crash on drevil_xxl's XP system. (reported by clsid)
7) The following encoders do not work for me: MPEG 4, MPEG 1, MPEG 2, h.263, H.261 and DV. VirtualDub 1.16.16 gives the following error: "Cannot start video compression. An unknown error occurred (may be corrupt data). (error code -100)". I know that at least some of these encoders do work for others. My system specs: Windows 2000, AMD Athlon Thunderbird. (reported by clsid)
8) Resize filter alters colors. Details (http://forum.doom9.org/showthread.php?p=884778#post884778). (reported by Kador)
9) Automatic preset loading doesn't seems to be working with on "number of channel match" and AC3 audio on DVD (don't know if it happens with AC3 audio in MKV, AVI or not or any other audio format). I use "3;4;5;6" in "number of channel match" so this should load with audio track with more than 2 channels but with an AC3 2 channels track this is also loaded. (reported by thuan)
10) Automatic Quality Control in PostProcessing video filter is broken I think. At least with what I see with the preset slider. Normally the dot should go up and down as the CPU load differs but now it only stuck at where it defaults. (reported by thuan)

When outputting the audio to AC3 using the ffdshow AC3 encoder there are a lot of downmix overflows. The sound seems to be distorted as if the volume is to high. (reported by TFM_TheMask)Has this one been fixed or not?

[_xxl]

I think haruhiko fixed that bug by removing the CPU usage OSD item.
Rev 350 is working ok with Vista RC2 x86.
Please confirm.
When outputting the audio to AC3 using the ffdshow AC3 encoder there are a lot of downmix overflows. The sound seems to be distorted as if the volume is to high. (reported by TFM_TheMask)
I'll commit when ready.

[clsid]

The default User no longer has permission to access
the pseudo registry key HKEY_PERFORMANCE_DATA. This
causes the call to RegQueryValueEx to return
ACCESS_DENIED.Please test Vista RC2 with a regular user account, because it probably works ok on admin and poweruser accounts.

[_xxl]

The default User no longer has permission to access
the pseudo registry key HKEY_PERFORMANCE_DATA. This
causes the call to RegQueryValueEx to return
ACCESS_DENIED.
True, but no crash.
Please test Vista RC2 with a regular user account, because it probably works ok on admin and poweruser accounts.
Tested no crash.Cpu usage is 0% always.
EDIT:
The same results on admin and guest accounts.

[foxyshadis]

The last is Automatic Quality Control in PostProcessing video filter is broken I think. At least with what I see with the preset slider. Normally the dot should go up and down as the CPU load differs but now it only stuck at where it defaults.

I think that's actually related to the cpu usage bug. But I can't test now because sourceforge's one old-files site is down and I don't have the motivation to recompile a couple old revs. ^^;

[JarrettH]

anyone know why sourceforge only has one download mirror? (not working)

[foxyshadis]

It's some kind of database mishap, but I found that if you had a preferred mirror selected already, you can click on the link it defaults to and the files will be there, even though the mirror doesn't show up.

[Liisachan]

direct links are usually discouraged, but just for now:
http://surfnet.dl.sourceforge.net/sourceforge/ffdshow-tryout/ffdshow-20061010-rev369-Q.exe
http://surfnet.dl.sourceforge.net/sourceforge/ffdshow-tryout/ffdshow_rev382_20061012_clsid.exe
http://surfnet.dl.sourceforge.net/sourceforge/ffdshow-tryout/ffdshow_rev386_20061014_clsid_icl9.exe
http://surfnet.dl.sourceforge.net/sourceforge/ffdshow-tryout/ffdshow-20061015-rev389-Q.exe
http://surfnet.dl.sourceforge.net/sourceforge/ffdshow-tryout/ffdshow_rev392_20061015_clsid.exe
http://surfnet.dl.sourceforge.net/sourceforge/ffdshow-tryout/ffdshow-20061017-rev401-Q.exe
http://surfnet.dl.sourceforge.net/sourceforge/ffdshow-tryout/ffdshow_rev404_20061017_clsid.exe
http://surfnet.dl.sourceforge.net/sourceforge/ffdshow-tryout/ffdshow_rev404_20061017_clsid_icl9.exe
http://surfnet.dl.sourceforge.net/sourceforge/ffdshow-tryout/FFdshow-Tryouts-20061018-rev411.exe
http://surfnet.dl.sourceforge.net/sourceforge/ffdshow-tryout/FFdshow-Tryouts-20061018-rev415.exe
http://surfnet.dl.sourceforge.net/sourceforge/ffdshow-tryout/ffdshow_rev420_20061020_clsid.exe

[dk75]

anyone know why sourceforge only has one download mirror? (not working)
???
One?
They have bunch of download mirrors. Usualy you use one as preffered but you can:
1. change it any time when downloading;
2. regardles of preffered mirror you could use any other mirror listed at download page by just klicking it.

[Eragon4ever]

Added a patch for German and other languages to SourceForge. Link (http://sourceforge.net/tracker/index.php?func=detail&aid=1582239&group_id=173941&atid=867362)
I hope it works this time.

[haruhiko_yamagata]

Added a patch for German and other languages to SourceForge. Link (http://sourceforge.net/tracker/index.php?func=detail&aid=1582239&group_id=173941&atid=867362)
I hope it works this time.
*.isl files
If you want to overwrite one or two messages, you can do it from custom_messages.iss. Leaving it in compiler directory should be better because Inno set up will be updated. I copied Japanese.isl because Inno setup does not include it.
*.rtf files
*rtf are not included because your svn client thinks they are binary files.

[DonGato]

direct links are usually discouraged, but just for now:
http://surfnet.dl.sourceforge.net/sourceforge/ffdshow-tryout/ffdshow-20061017-rev401-Q.exe
http://surfnet.dl.sourceforge.net/sourceforge/ffdshow-tryout/ffdshow_rev420_20061020_clsid.exe
Strange, those links don't work in Firefox but in IE. -_-;
Now M$ is paying Sourceforge? :P

[LoRd_MuldeR]

Strange, those links don't work in Firefox but in IE. -_-;
Now M$ is paying Sourceforge? :P

Works fine in FreeDownloadManager ;)
Also works when I copy the URL to address-bar in Firefox.
Leftclick on the link don't work. Maybe they don't like the referrer...

Also try this, should work with leftclick:
http://anonym.to/?http://surfnet.dl.sourceforge.net/sourceforge/ffdshow-tryout/ffdshow_rev420_20061020_clsid.exe

[Eragon4ever]

*.isl files
If you want to overwrite one or two messages, you can do it from custom_messages.iss. Leaving it in compiler directory should be better because Inno set up will be updated. I copied Japanese.isl because Inno setup does not include it.
*.rtf files
*rtf are not included because your svn client thinks they are binary files.

Well, I edited 20 - maybe more, maybe less - messeges. Should I really do this in custom_messages.iss?

Anyway, I uploaded the *.rtf files separately.

[haruhiko_yamagata]

Well, I edited 20 - maybe more, maybe less - messeges. Should I really do this in custom_messages.iss?

Well, it's hard but better to do it in custom_messages.iss, I think. It will make clear where you edited and it's important.

[clsid]

If you are modifying german.isl, then maybe it is best to submit those changes to the InnoSetup creators.

[Eragon4ever]

If you are modifying german.isl, then maybe it is best to submit those changes to the InnoSetup creators.

I'll do so sometimes next week due to a lack of time.

P.S. A typo in ffdshow_installer.iss:
Name: "ge"; MessagesFile: "compiler:Languages\German.isl" should be
Name: "de"; MessagesFile: "compiler:Languages\German.isl"
I don't want to make a patch just for this now.

[Amour]

added patch for French InnoDB.

[clsid]

I have updated both German and French translations.

[foxyshadis]

If you don't update custom_messages.iss, they'll most likely get accidentally overwritten with the next inno update anyway. Or at least anyone who updates will always have to keep it in mind.

???
One?
They have bunch of download mirrors. Usualy you use one as preffered but you can:
1. change it any time when downloading;
2. regardles of preffered mirror you could use any other mirror listed at download page by just klicking it.

No, mirror selection was broken the last day or two. Some weirdness related to updating the site with new code. Fixed today though.

[haruhiko_yamagata]

FPS1

FPS1 v0,v1 is OK, not v2(blackness or crash(random and rare)).
Samples (http://samples.mplayerhq.hu/V-codecs/FPS1/)

[_xxl]

FPS1

FPS1 v0,v1 is OK, not v2(blackness or crash(random and rare)).
Samples (http://samples.mplayerhq.hu/V-codecs/FPS1/)
/**
* @file fraps.c
* Lossless Fraps 'FPS1' decoder
* @author Roine Gustafsson <roine at users sf net>
*
* Only decodes version 0 and 1 files.
* Codec algorithm for version 0 is taken from Transcode <www.transcoding.org>
*
* Version 2 files, which are the most commonly found Fraps files, cannot be
* decoded yet.
*/
v2 is not supported.

[Eragon4ever]

Added a patch of the two new lines in custom_messages.iss. Here (http://sourceforge.net/tracker/index.php?func=detail&aid=1582239&group_id=173941&atid=867362)

[regeszter]

Hi,

may I ask for two feature? I like to see on OSD the preset name and girder template for audio part (there is girder template for video part).

Thanks

[BlindWanderer]

Feature Request :D
And a bug :(

Two checkboxs on the Resize module config:

Resize with SAR consideration. (resize the video so the output after SAR is applied, will be what was requested, either the size (not exactly sure how this would work) or aspect ratio).
Aspect ratio resize subtracks (instead of adding blackspace).


Goal:
You see, i have some video I want to play on my desktop (M$ video Screensaver & Vital Desktop (http://vital-desktop.sourceforge.net/). Trouble is the video has a non 1:1 SAR. This means i cannot cut the video to the aspect ratio of my screen (which is 5:4) and maintain the videos SAR. It's driving me nuts.

Bug:
When using the overlay mixer & hw deinterlacing, it sometimes doesn't deinterlace.
I've found changing the aspect ratio on the overlay mixer to anything then back to default gets deinterlacing to work. It would be good if there were a checkbox to do this workaround automaticaly. If you set the aspect ratio to that of the SAR and back to default it wouldn't be noticeable.

[_xxl]

http://www.mplayerhq.hu/design7/news.html

[Xirss]

Where I can find an x64 build of ffdshow with x64 installer? 32-bit installer registers vfw frontend in the 32-bit part of registry :( so it isn't work for VirtualDub.

[Liisachan]

It's not from those recent ffdshow-tryouts, but celtic_druid compiled ffdshow64-rev2546.exe (http://ffdshow.faireal.net/mirror/ffdshow/ffdshow64-rev2546.exe), dated 2006-05-17. 2546 is the newest rev by milan, i.e. original ffdshow.

[Romario]

Well, I am sure that drevil_xxl or haruhiko_yamagata CAN compile 64-bit build from tryout builds, based on revision 454.

[cc979]

I've found something out when playing a svq3 file with ffdshow I get colour corruption which is known, but if use nv21 output the colour corruption is gone

[TFM_TheMask]

New bug:

From version 420 the AC3 output is broken. For example when I want to encode an AAC audio file to AC3 using the Filewriter filter the ffdshow audio decoder filter can't connect to the Filewriter. Older versions work OK.

Attached the graph from the older versions when it still works and the settings used.

This bug was found during the fixing of the AAC to AC3 distortion bug.

[haruhiko_yamagata]

Bug:
When using the overlay mixer & hw deinterlacing, it sometimes doesn't deinterlace.
I've found changing the aspect ratio on the overlay mixer to anything then back to default gets deinterlacing to work. It would be good if there were a checkbox to do this workaround automaticaly. If you set the aspect ratio to that of the SAR and back to default it wouldn't be noticeable.
Is "Use ovelay mixer" checked? Is not it in intermediate state?

[haruhiko_yamagata]

I installed Vista RC1 yesterday.
makeAvis.exe does not start in windows Vista.

// EDIT
Oops! Excuse me.
I just forgot to install avisynth. It works.

[BlindWanderer]

Is "Use ovelay mixer" checked? Is not it in intermediate state?

I have it in the intermediate state (grey/green; depends who's compile i'm using).

I have two nv 7900GT in SLi mode, running 91.47 of the nvidia drivers (i could run 92.91, but i've had some problems with it).

[haruhiko_yamagata]

I have it in the intermediate state (grey/green; depends who's compile i'm using).

I have two nv 7900GT in SLi mode, running 91.47 of the nvidia drivers (i could run 92.91, but i've had some problems with it).
If it is intermediate state, it's expected.
To force hardware deinteracing, check "Use overlay mixer".
The problem is not the code, but the name of the check box and documentation.

[BlindWanderer]

I keep a close eye on the change log, and noticed a "restart replace for libmpeg2_ff.dll" for revision 458.

I hate rebooting my system. I like the "file is locked" dialog box. Usualy when a file is locked with the ffdshow installer, it's because of explorer or some media player.

I checked the "Use overlay mixer" box in "output" (was intermediate); still isn't deinterlacing until i change the overlay aspect ratio (changing any other aspect of the size is enough to trigger deinterlacing as well).

[haruhiko_yamagata]

I keep a close eye on the change log, and noticed a "restart replace for libmpeg2_ff.dll" for revision 458.

I hate rebooting my system. I like the "file is locked" dialog box. Usualy when a file is locked with the ffdshow installer, it's because of explorer or some media player.
Restart replace is flaged for many other dlls and ax too. See rev 430.
Reboot is required only when file is locked. If you check "Don't use ffdshow in"-"explorer.exe", it is rarely required.
If it is locked by explorer.exe, we can't install ffdshow unless we reboot or log off. Or do you instruct beginers to use task manager and quit explorer.exe? Beginers don't know what to do against "file is locked" message.
If it is locked by some media player, the installer will find libmplayer.dll is locked earlier, and warn user to quit the applicatin, thus no problem.

In Windows vista, explorer.exe still lock ffdshow.ax and libmpeg2_ff.dll and never releases. I made "Don't use ffdshow in"-"explorer.exe" default because of this issue. clsid, I don't think it is good to remove explorer.exe from black list in Windows Vista.

[haruhiko_yamagata]

I checked the "Use overlay mixer" box in "output" (was intermediate); still isn't deinterlacing until i change the overlay aspect ratio (changing any other aspect of the size is enough to trigger deinterlacing as well).
It's not reproducible to me. Perhaps it depends on the video card.

[_xxl]

New bug:
From version 420 the AC3 output is broken. For example when I want to encode an AAC audio file to AC3 using the Filewriter filter the ffdshow audio decoder filter can't connect to the Filewriter. Older versions work OK.
Attached the graph from the older versions when it still works and the settings used.
This bug was found during the fixing of the AAC to AC3 distortion bug.
Please tell us more!Last revision that worked?
AAC to AC3 distortion bug is in rev 2517.

[haruhiko_yamagata]

I have it in the intermediate state (grey/green; depends who's compile i'm using).

I have two nv 7900GT in SLi mode, running 91.47 of the nvidia drivers (i could run 92.91, but i've had some problems with it).
It may be bug of nvidia's driver. Please test MPC's internal MPEG2 decoder. Double click the decoder in MPC's dialog and you'll see "Set interlaced flag in output media type". Enable it and try. If failed, please report the bug to nvidia.

[TFM_TheMask]

Please tell us more!Last revision that worked?
AAC to AC3 distortion bug is in rev 2517.

Where can I download rev 416, 417, 418 and 419?

[BlindWanderer]

Restart replace is flaged for many other dlls and ax too. See rev 430.
Reboot is required only when file is locked. If you check "Don't use ffdshow in"-"explorer.exe", it is rarely required.
If it is locked by explorer.exe, we can't install ffdshow unless we reboot or log off. Or do you instruct beginers to use task manager and quit explorer.exe? Beginers don't know what to do against "file is locked" message.
If it is locked by some media player, the installer will find libmplayer.dll is locked earlier, and warn user to quit the applicatin, thus no problem.

In Windows vista, explorer.exe still lock ffdshow.ax and libmpeg2_ff.dll and never releases. I made "Don't use ffdshow in"-"explorer.exe" default because of this issue. clsid, I don't think it is good to remove explorer.exe from black list in Windows Vista.

I've had explorer still lock ffdshow dispite being on the blacklist. I have no fear of killing it. :devil: but then i'm an old windows registry hacker; nothing scares me. I would not recommend a new user kill explorer.

I won't mind being prompted with a box that said something like "the file is locked, do you want to: [replace on reboot] [try again] [ignore] [cancel]"


It's not reproducible to me. Perhaps it depends on the video card.


It doesn't surprise me it's not reproducible.
Maybe a checkbox that sets the aspect ratio of the overlay mixer to that of the SAR on the output tab would be enough?

(i forgot to hit post)

[BlindWanderer]

It may be bug of nvidia's driver. Please test MPC's internal MPEG2 decoder. Double click the decoder in MPC's dialog and you'll see "Set interlaced flag in output media type". Enable it and try. If failed, please report the bug to nvidia.

Will do.
But how do i get it to do the deinterlacing at the hw side?
(i thought MPC's deinterlacing was handled exclusively internaly in software)

[_xxl]

Where can I download rev 416, 417, 418 and 419?
Please test:2546,127,164,195,246,312,345,382,392,404.

[BlindWanderer]

Where can I download rev 416, 417, 418 and 419?

Not all the revisions get built. I don't think any of those got built. 404 is the last i know of before 420.

[haruhiko_yamagata]

I won't mind being prompted with a box that said something like "the file is locked, do you want to: [replace on reboot] [try again] [ignore] [cancel]"

Inno setup does not support the feature. It's too hard to code that logic.

[TFM_TheMask]

Not all the revisions get built. I don't think any of those got built. 404 is the last i know of before 420.

Ok thanks.

[haruhiko_yamagata]

Will do.
But how do i get it to do the deinterlacing at the hw side?
(i thought MPC's deinterlacing was handled exclusively internaly in software)
No, I think MPC supports hardware deinterlacing. Read my prior post again.

[TFM_TheMask]

Please tell us more!Last revision that worked?
AAC to AC3 distortion bug is in rev 2517.

Ok the latest revision that worked for the ac3 encoded output is ffdshow_rev345_20061006_clsid.exe. The newer versions starting from ffdshow_rev382_20061012_clsid.exe don't work.

[BlindWanderer]

No, I think MPC supports hardware deinterlacing. Read my prior post again.

I didn't know it would do that. Checking the check box did not enable deinterlacing (though software deinterlacing works find in MPC). I'll try the newer video driver (it's supposed to be for Nvidia's high def video decoder; which is neither here nor there).

[BlindWanderer]

*slams head into keyboard*
helps to make sure you have overlay mixer set as the video renderer.
Which begs the question, how & why *was* i able to get hardware deinterlacing to work *without* the overlay mixer before? Seems like hw deinterlacing isn't dependent on overlay.

EDIT:
Course the screen saver i use doesn't let you choose the renderer.

[haruhiko_yamagata]

*slams head into keyboard*
helps to make sure you have overlay mixer set as the video renderer.
Which begs the question, how & why *was* i able to get hardware deinterlacing to work *without* the overlay mixer before? Seems like hw deinterlacing isn't dependent on overlay.

EDIT:
Course the screen saver i use doesn't let you choose the renderer.
In my PC, hw deinterlacing is working with overlay mixer. Unless you are very sure that the bug is within ffdshow, please report the bug to nvidia. They'll make sure which is bugy.

[Jeremy Duncan]

Does the FFdshow HW Deinterlacing feature allow Nvidia purevideo to use HW Deinterlacing ?
If so, which is recommended Bob, or Weave in FFdshow ?

[Bathrone]

Motion adapative temporal deinterlacing. Bob and weave are so poor. I reckon ATI and NVIDIA's hw deinterlacers are great - check out the HQV benchmark disc.

[Jeremy Duncan]

Motion adapative temporal deinterlacing.

Does the FFdshow HW Deinterlacing do "Motion adapative temporal deinterlacing" ?

With the FFdshow HW Deinterlacing enabled and if I used the Purevideo codec and hardware, do I get "Purevideo HW Deinterlacing" ?

[foxyshadis]

Er, there is no ffdshow "hardware" deinterlacing; it has its own software deinterlacing, or you can ignore it and run whatever your ATI, Nvidia, or other DXVA drivers do. (So yes, Purevideo or AVIVO.) ffdshow has no control whatsoever over downstream deinterlacing, other than being able to signal that the video's interlaced.

Weave isn't even "deinterlacing", it just displays combined fields. Both ATI & Nvidia's hardware methods have mocomp and pulldown detection, and look nearly as good as tdeint in a fast mode.

[Jeremy Duncan]

Er, there is no ffdshow "hardware" deinterlacing; it has its own software deinterlacing, or you can ignore it and run whatever your ATI, Nvidia, or other DXVA drivers do. (So yes, Purevideo or AVIVO.) ffdshow has no control whatsoever over downstream deinterlacing, other than being able to signal that the video's interlaced.

Weave isn't even "deinterlacing", it just displays combined fields. Both ATI & Nvidia's hardware methods have mocomp and pulldown detection, and look nearly as good as tdeint in a fast mode.

Does this mean that if I enable HW Deinterlacing in FFdshow, that my Purevideo certified video card and purevideo codec will use "Motion adapative temporal deinterlacing" ?
If so, why is there a box to choose from bob or weave when I enable HW Deinterlacing in FFdshow ? What's that for ?

[foxyshadis]

All of that is controlled by the driver's control panel. I only have ATI, so I have only vague recollections about Purevideo, but I know it's in its control panel too. Bob and weave are relics of the stone age of gpu-assists, when it only worked with MPEG2, both flags are probably ignored now.

[Jeremy Duncan]

All of that is controlled by the driver's control panel. I only have ATI, so I have only vague recollections about Purevideo, but I know it's in its control panel too. Bob and weave are relics of the stone age of gpu-assists, when it only worked with MPEG2, both flags are probably ignored now.

Since they are probably ignored.
Does this mean if I enable HW Deinterlacing in FFdshow's output tab that "Motion adapative temporal deinterlacing" is available with the right hardware and codec ?

[Bathrone]

A developer can confirm, but I believe that if you dont enable any deinterlacing within ffdshow and set your driver to do the deinteracling and the method thats what I do. I use ati, there is an applet that let's me configure my deinterlacing preference.

[haruhiko_yamagata]

foxyshadis is right, there is no hardware deinterlacing control in ffdshow. But generally the downstream of ffdshow is video renderer and the check box works as hardware deinterlacing. If the downstream of ffdshow is software deinterlacer, it works as hint for the software deinterlacer.
Bob and Weave is a hint for downstream. In most cases Weave is same as unchecked hardware deinterlacing.
ffdshow can only show hint for down stream. What happens in down stream is down stream's responsibility.

[Jeremy Duncan]

foxyshadis is right, there is no hardware deinterlacing control in ffdshow. But generally the downstream of ffdshow is video renderer and the check box works as hardware deinterlacing. If the downstream of ffdshow is software deinterlacer, it works as hint for the software deinterlacer.
Bob and Weave is a hint for downstream. In most cases Weave is same as unchecked hardware deinterlacing.
ffdshow can only show hint for down stream. What happens in down stream is down stream's responsibility.

In other words. If my video card and Video codec support "Motion adapative temporal deinterlacing" / "Purevideo Deinterlacing".
Running FFdshow won't cancel this and force it into software mode, as long as HW Deinterlacing is checked, and Bob is selected in the Output tab ?

[BlindWanderer]

New bug, when parsing FLV1 (which seems to have no restrictions on resolution) with the OSD turned on, the osd color is overlayed in the wrong (vertical) position when the video resolution is not divisible by 4. This bug does not effect playback (read: ffdshow doesn't crash & no new artifacts).

So a video with a resolution 320x239, the osd color is rendered about 3 pixels higher then the the text.

http://mailerdaemon.home.comcast.net/6SK-bcWCU8w.flv

[haruhiko_yamagata]

In other words. If my video card and Video codec support "Motion adapative temporal deinterlacing" / "Purevideo Deinterlacing".
Running FFdshow won't cancel this and force it into software mode, as long as HW Deinterlacing is checked, and Bob is selected in the Output tab ?
I don't have nvidia cards and don't know what to answer.

[clsid]

List of known issues in revision 483:

1) Interface bug in Windows Vista. Details and screenshot (http://forum.doom9.org/showthread.php?p=888312#post888312). (reported by BlindWanderer)
2) Wavpack decoder doesn't work with this file (http://rapidshare.com/files/1609252/SoC__XviD_WavPack_MP3_-007.mkv.html). (reported by clsid)

Reported but unconfirmed bugs:

3) image settings -> preset autoload conditions -> "on FOURCC match" doesn't work. (reported by juskixxx) (confirmed by clsid) (haruhiko could not reproduce it)
4) Automatic preset loading doesn't seems to be working with on "number of channel match" and AC3 audio on DVD (don't know if it happens with AC3 audio in MKV, AVI or not or any other audio format). I use "3;4;5;6" in "number of channel match" so this should load with audio track with more than 2 channels but with an AC3 2 channels track this is also loaded. (reported by thuan)
5) Instant crash when trying to play this MSS2 WMV file (http://ftp.mplayerhq.hu/MPlayer/samples/V-codecs/MSS2/mss2_speech.wmv). Happens on Windows 2000. No crash on drevil_xxl's XP system. (reported by clsid)
6) The following encoders do not work for me: MPEG 4, MPEG 1, MPEG 2, h.263, H.261 and DV. VirtualDub 1.16.16 gives the following error: "Cannot start video compression. An unknown error occurred (may be corrupt data). (error code -100)". I know that at least some of these encoders do work for others. My system specs: Windows 2000, AMD Athlon Thunderbird. (reported by clsid)
7) Resize filter alters colors. Details (http://forum.doom9.org/showthread.php?p=884778#post884778). (reported by Kador)

Other issues:

8) ICL9 builds of ffdshow.ax crash on files created by a specific old revision of x264 (don't know the rev number). Funny thing however is that the files play without crash if you first play a good file and then play a 'troublesome' file in the same player instance. Also no crash when using an unoptimized debug build. So this seems to be a compiler bug. Sample file (http://rapidshare.com/files/1609924/sample.mp4.html). (reported by clsid)

[Eragon4ever]

What about putting the List of known issues in todo.txt?

[clsid]

This way they are visible to everyone.

[Eragon4ever]

Of cause, I meant additionaly and form time to time you can copy&past it here.
And the developers would not have to search here for it.

[haruhiko_yamagata]

3) image settings -> preset autoload conditions -> "on FOURCC match" doesn't work. (reported by juskixxx)
I still can't reproduce this. Any suggestions how to reproduce is appreciated.

[Egh]

clsid: I notice you've changed your installer script in the rev457 build.

Now it offers to restart at the end of process instead of "Abort, Retry, Ignore" during it.

[clsid]

I still can't reproduce this. Any suggestions how to reproduce is appreciated.
Create two extra presets (besides default).
Enable 'automatic preset loading'.
Preset 1: on FourCC match = DX50;DIVX
Preset 2: on FourCC match = XVID

Now play some DivX and XviD files. The correct preset doesn't get chosen.

clsid: I notice you've changed your installer script in the rev457 build.

Now it offers to restart at the end of process instead of "Abort, Retry, Ignore" during it.
Haruhiko made most of those changes. But do you think it is good or bad?

[Egh]

Haruhiko made most of those changes. But do you think it is good or bad?

Dunno. For casual user it's prooly bettar.

Though for me it's a bit of inconvinience, since I used to kill tc when that dialog was shown :)

But of course I can close it once I've started ffdshow setup, so it's not a major problem. I just wanted to clarify things ;)

Btw it would be great if ppl who do changes would explain them in this thread. One single-line phrase in ffdshow svn is not quite enough to comprehend :)

Would be interested to know what exactly was done in the following revisions:

Revision 458 - Directory Listing
Modified Sat Oct 28 00:13:53 2006 UTC (39 hours, 31 minutes ago) by h_yamagata

restart replace for libmpeg2_ff.dll

Revision 447 - Directory Listing
Modified Wed Oct 25 09:13:04 2006 UTC (4 days, 6 hours ago) by h_yamagata

ANSI/MBCS more security fixes

Revision 444 - Directory Listing
Modified Tue Oct 24 13:54:30 2006 UTC (5 days, 1 hour ago) by h_yamagata

Bug fix (ANSI)
ANSI debug build often crashes when we click task tray icons. This affects release build too, and is security issue.


Revision 439 - Directory Listing
Modified Tue Oct 24 11:02:52 2006 UTC (5 days, 4 hours ago) by h_yamagata

Bug fix : Quality control of post processing. It has been depending on CPU usage(performance monitor). Now that I removed the performance monitor related code, it has to be fixed.

Revision 398 - Directory Listing
Modified Mon Oct 16 14:00:29 2006 UTC (13 days, 1 hour ago) by drevil_xxl

fixed(?) stereoscopic subtitles.

[_xxl]

Would be interested to know what exactly was done in the following revisions:
Revision 398 - Directory Listing
fixed(?) stereoscopic subtitles.

Bug Fix:ffdshow crashes when enabling stereoscopic option on subtitles page.
Stereoscopic Subtitle Guide:
http://www.3dtv.at/Knowhow/Subtitles_en.aspx

[Egh]

Bug Fix:ffdshow crashes when enabling stereoscopic option on subtitles page.
Stereoscopic Subtitle Guide:
http://www.3dtv.at/Knowhow/Subtitles_en.aspx

My confusion was caused by the fact that stereoscopic subtitles were due to be removed from ffdshow :)

[clsid]

That was because it crashed previously. So the option was disabled. Now the function works without crashing. However its usefulness is debatable.

[Jeremy Duncan]

I don't have nvidia cards and don't know what to answer.

"The best de-interlacing methods, yielding the highest quality, are complex. There are no simple ways to do this job to eliminate artifacts in all cases. Old de-interlacing methods, simple and non-adaptive to motion speed changes, are becoming a thing of the past, providing not the best quality. Modern video chips use complex de-interlacing, so called adaptive methods. Adaptive de-interlacing methods selectively cut pixels, based on information about motion in a frame. Before AVIVO, ATI products already used adaptive methods. But it's R520, RV530, RV515 video chips supporting AVIVO, where the company introduced the advanced method - vector adaptive de-interlacing. This improved algorithm resembles the new method in NVIDIA PureVideo - spatial-temporal. In brief, it works in the following way: the algorithm selects data to build a progressive frame from either the raw field data, when motion is detected to be low; in case of fragments with fast motion, it uses video data that is interpolated along several vectors, which provides the greatest amount of details to the progressive frame. Vector adaptive de-interlacing removes the comb effect very well even in traditionally difficult scenes (for example, diagonal lines at a small angle, or flying colors ) without much quality losses and blurring."

http://www.digit-life.com/articles2/video/avivo.html

How about you download a trial of Nvidia's video codec, set it to use hardware acceleration, and see if it's in hardware mode in this tab.

http://img230.imageshack.us/img230/73/nvidiakp7.jpg
Link to Picture (http://img230.imageshack.us/img230/73/nvidiakp7.jpg)

Here's the link to the download.
Link (http://www.nvidia.com/object/dvd_decoder_1.02-223-trial.html)

DXVA is Hardware Acceleration, which includes HW Deinterlacing.
You don't need a Nvidia video card to run Nvidia video decoder in DXVA mode.
See the link
Link (http://www.htpcnews.com/forums/index.php?showtopic=20763)

Using the Software I linked to,
Setting the Nvidia Decoder to use Hardware Acceleration,
and Setting FFdshow to use HW Deinterlacing.
The Screen I linked to will tell you if it's in hardware mode.
Meaning if it's in Software Mode, your HW Deinterlacing in the FFdshow output tab isn't working.

You can test this now, you don't need a Nvidia video card to test if dxva is used.

[haruhiko_yamagata]

Would be interested to know what exactly was done in the following revisions:Revision 458 - Directory Listing
Modified Sat Oct 28 00:13:53 2006 UTC (39 hours, 31 minutes ago) by h_yamagata

restart replace for libmpeg2_ff.dll

Revision 447 - Directory Listing
Modified Wed Oct 25 09:13:04 2006 UTC (4 days, 6 hours ago) by h_yamagata

ANSI/MBCS more security fixes

Revision 444 - Directory Listing
Modified Tue Oct 24 13:54:30 2006 UTC (5 days, 1 hour ago) by h_yamagata

Bug fix (ANSI)
ANSI debug build often crashes when we click task tray icons. This affects release build too, and is security issue.


Revision 439 - Directory Listing
Modified Tue Oct 24 11:02:52 2006 UTC (5 days, 4 hours ago) by h_yamagata

Bug fix : Quality control of post processing. It has been depending on CPU usage(performance monitor). Now that I removed the performance monitor related code, it has to be fixed.

Rev458
Supporting beginers have priority on supporting experts.

Rev 444, 447
Unless you use MBCS languages like Japanese, it's hard to reproduce. For more information, please look at the code.

Rev 439
Postprocessing was depending on CPU usage. Now it depends on IQualityControl (Not "Time on ffdshow"). CPU usage was broken, so this is better IMO. It may not be best though. If you are using queue, "Queued samples<8" may be a good try.

[haruhiko_yamagata]

Create two extra presets (besides default).
Enable 'automatic preset loading'.
Preset 1: on FourCC match = DX50;DIVX
Preset 2: on FourCC match = XVID

Now play some DivX and XviD files. The correct preset doesn't get chosen.
Yes. It works for me.

Windows XP sp2, MPC, Zoom Player, AVI, MKV...all settings I have tried is OK.

[haruhiko_yamagata]

Does this mean that if I enable HW Deinterlacing in FFdshow, that my Purevideo certified video card and purevideo codec will use "Motion adapative temporal deinterlacing" ?
If so, why is there a box to choose from bob or weave when I enable HW Deinterlacing in FFdshow ? What's that for ?
I read your links. ffdshow does not force anything to down stream. All is free for down stream's choice anyways.
Bob or Weave may be ignored by Pure video, but usefull in my GPU(i82865G). In my system hardware deinterlacing checed + Weave is same as hardware deinterlacing uncheckd. But it won't be understood by users. Showing Bob or Weave is better UI, I think.

[_xxl]

New bug, when parsing FLV1 (which seems to have no restrictions on resolution) with the OSD turned on, the osd color is overlayed in the wrong (vertical) position when the video resolution is not divisible by 4. This bug does not effect playback (read: ffdshow doesn't crash & no new artifacts).
So a video with a resolution 320x239, the osd color is rendered about 3 pixels higher then the the text.
http://mailerdaemon.home.comcast.net/6SK-bcWCU8w.flv
http://i13.tinypic.com/4d61744.jpg
http://i14.tinypic.com/2gspriq.jpg
YV12 is on.
http://i14.tinypic.com/2itmr9h.jpg
http://i14.tinypic.com/2qkl3py.jpg
YV12 is off.
EDIT:MPC is running in WMR9 (renderless) mode.

[haruhiko_yamagata]

http://i13.tinypic.com/4d61744.jpg
http://i14.tinypic.com/2gspriq.jpg
YV12 is on.

Many video cards says it supports YV12, but actually it does not. It's likely a bug of the video renderer. The problem is YV12 is set by default because it is often the fastest. Some media players test YV12 on install and ask user if it was played properly. Because ffdshow is not a player, there's no way to know if YV12 is actually supported.

Another problem is that the picture has odd number lines. Are picutures with even number lines played back properly?

Turn on post processing for the picture and you'll see another bug - green line at the bottom.

[BlindWanderer]

Wow cool bug drevil, but thats not the one I'm seeing, thought it does only happen for yv12

Also get a line of random pixels (the color is random) across the bottom row in mpc when using rgb15 but thats probably my computer.

http://mailerdaemon.home.comcast.net/snapshot-yv12.jpg

As you can see in the screen shot the the all of the color is higher then it should be (same frame as your clean frame). So you are probably right, suggesting it being my GC. Especial consider that when i resize up to the next multiple of 4 it aligns. (sorry for wasting time)

[cc979]

Many video cards says it supports YV12, but actually it does not. It's likely a bug of the video renderer. The problem is YV12 is set by default because it is often the fastest. Some media players test YV12 on install and ask user if it was played properly. Because ffdshow is not a player, there's no way to know if YV12 is actually supported.

Another problem is that the picture has odd number lines. Are picutures with even number lines played back properly?

Turn on post processing for the picture and you'll see another bug - green line at the bottom.

i've just tested the clip with mine, i get the colour corruption if yv12 is the only output, but plays fine with yuy2 or nv12

[Jeremy Duncan]

Showing Bob or Weave is better UI, I think.

Not everybody knows that weave is the same as film, and bob is the same as video.
I wasn't sure myself until you made it clear.

Maybe change the ui to this:
Film - Deinterlacing Off
Video - Deinterlacing On

[arfster]

Film is different - that means to use IVTC to extract the original 24 progressive frames from 60 interlaced "half-frames" (aka fields) each second.

Bob and weave are just two different methods of deinterlacing, and you only want to be using this on genuine interlaced video content, or on film content where the flags are so thoroughly messed up that the 24fps stream is not really readable.

Regardless of bob/weave in ffdshow, Purevideo will probably ignore it and do its own thing anyway. If it's in film mode, it's working fine and will be producing the correct 24fps.

[Jeremy Duncan]

"Film - Forces film mode and does not apply de-interlacing.

Video - Forces video mode and applies de-interlacing."

Link (http://www.htpcnews.com/forums/index.php?showtopic=20763)

This is how Nvidia catalogs Deinterlacing in the deinterlacing control. Weave is the same as Film, and Bob is the same as video.
See the screen.

http://img230.imageshack.us/img230/73/nvidiakp7.jpg

"DeInterlacing - Weave:
Weave is the default mode that should be used when viewing progressive DVD data (Movies for example). Using this mode you will see horizontal streaking for non-movie data (NTSC/PAL content such as movie trailers or content recorded using interlaced video cameras). If you plan on viewing DVD movies, you must set your decoding software to weave mode. Weave is the default DeInterlacing mode all decoding software use for Movie playback (unless set to some sort of detection mode).

DeInterlacing - BOB:
T.V. playback works a bit differently compared to a computer monitor. It has two fields, each playing in an interlaced form at either 30 or 25fps (NTSC/PAL). When NTSC/PAL content is played on a computer monitor using the standard Weave DeInterlacing mode, you get a lot of horizontal streaking when the camera pans or objects move quickly within a scene. To combat this, you can use the BOB DeInterlacing mode. What BOB does is play the content at twice it's frame rate and each frame is displayed in only one of the fields. This makes the image appear to BOB up and down a bit, especially when text is displayed, but doing so eliminates the streaking. Using the BOB mode is only useful for NTSC/PAL content such as Movie Trailers. On Progressively encoded Movie content BOB will cause the image to look slightly blurred."

Link (http://terms.inmatrix.com/)

"Weave works perfectly for progressive sourced1080i (from film). But for true interlaced 1080i, only bob is an option, and it only provides 540 lines of vertical resolution from 1080i video source."
Link (http://www.avsforum.com/avs-vb/showthread.php?p=6221665&highlight=weave+bob#post6221665)

Weave is for Film, Bob for Video/T.V.

I wasn't sure if setting FFdshow's HW Deinterlacer to Weave did anything until haruhiko_yamagata made it clear what it did.

I think a hint beside the bob and weave option would make it easier to use properly.

[Liisachan]

@arfster
just to make it clear... I think you are assuming NTSC, while european films (European R2 DVDs) are 25fps (PAL). Many of the ppl here should be europeans...

[arfster]

"Film - Forces film mode and does not apply de-interlacing.


This is the key quote. Weaving is a type of deinterlacing, and in film mode (as above) you don't de-interlace, you IVTC if at all possible.

Certainly you can apply deinterlacing, whether bob or weave, to interlaced content (1080i or DVD) with IVTC flags, but it won't look as good as setting film mode. Remember, the original content is 24fps, which film mode can retrieve losslessly. Deinterlacing will only result in 30 weaved frames, or 60 bobbed ones - neither are the original video, but a processed reconstruction of it with resultant artifacts.

See the telecine graphic here:

http://www.doom9.org/ivtc-tut.htm

Weaving will mix half of frame 2, with half of frame 3, and 3/4 in the same way. Messy if there's any movement in between, and juddery. Bob will scale the half-frames into separate full frames, and thus lower detail. True film mode IVTC will rebuild the original 4 frames perfectly.



Liisachan: yes, this was for NTSC. Obviously PAL is a lot simpler with 50>25fps, rather than 60>24.

[Jeremy Duncan]

In most cases Weave is same as unchecked hardware deinterlacing.

Weaving is a type of deinterlacing...


This is where I was confused.
Because I thought that weaving was deinterlacing, like Arfster said in the quote.
But when haruhiko_yamagata said weave was the same as no hardware deinterlacing, I took it that no hardware deinterlacing meant there was no software deinterlacing either, which meant it was film.
So I suggested that Weave, and Bob should have a hint or be called something else to show that weave was the same as no deinterlacing.

As I understand it, haruhiko_yamagata said that Weave and Bob sent hints that enabled or disabled hardware deinterlacing, that they were called weave and bob wasn't important.
If hardware deinterlacing was enabled by selecting Bob, the hardware deinterlacing wouldn't nessessarily be bob deinterlacing, it could be spatial-temporal deinterlacing.

[haruhiko_yamagata]

This is where I was confused.
Because I thought that weaving was deinterlacing, like Arfster said in the quote.
But when haruhiko_yamagata said weave was the same as no hardware deinterlacing, I took it that no hardware deinterlacing meant there was no software deinterlacing either, which meant it was film.
So I suggested that Weave, and Bob should have a hint or be called something else to show that weave was the same as no deinterlacing.

As I understand it, haruhiko_yamagata said that Weave and Bob sent hints that enabled or disabled hardware deinterlacing, that they were called weave and bob wasn't important.
If hardware deinterlacing was enabled by selecting Bob, the hardware deinterlacing wouldn't nessessarily be bob deinterlacing, it could be spatial-temporal deinterlacing.
Weave is a simple deinterlacing method.
If weave is selected, ffdshow say to the down stream "deinterlace this in Weave mode". But usually this is same as saying nothing(HW deinterlacing is unchecked). If ffdshow says nothing to down stream, I guess, down stream can find the samples are interlaced and try to deinterlace in weave mode. That's why the result is the same, if I'm not wrong.

[Egh]

This is where I was confused.
Because I thought that weaving was deinterlacing, like Arfster said in the quote.
But when haruhiko_yamagata said weave was the same as no hardware deinterlacing, I took it that no hardware deinterlacing meant there was no software deinterlacing either, which meant it was film.
So I suggested that Weave, and Bob should have a hint or be called something else to show that weave was the same as no deinterlacing.


It's just a bit of methodology confusion here :)


Deinterlacing is the process of converting interlaced video (a sequence of fields) into a non-interlaced form (a sequence of frames). This is a fundamentally impossible process that must always produce some image degradation, since it ideally requires "temporal interpolation" which involves guessing the movement of every object in the image and applying motion correction to every object.


Weave is a straightforward conversion from interlaced video stream (i.e. sequence of fiels=halfframes) to progressive one (sequence of full frames). Weave effectively pairs adjacent halfframes into full ones, thus if there's any fast movement you'll see combing artefacts (cause in *real* interlaced stream adjacent fields are taken at slightly different times). But in theoretical cases, for instance if you have 12fps progressive duplicated into 24fps, fastened to 25fps and then interlaced into
50 fields per sec, then weave would fully recover original stream.

To sum up, weave is deinterlacing, but just creates too much motion artefacts in practice.

[vlada]

I think here is a confusion which is caused because of PAL and NTSC differences.

If your movie is PAL, you have 2 possibilities:
25 progressive frames (uses Weave = do nothing) or 50 interlaced fields (deinterlace).

For NTSC you have 3 possible types:
30 frames progressive (probably not very common, Weave), 60 interlaced fields (deinterlace) and 60 telecined fields (inverse telecine -> output 24 fps).

Anyway BOB is not a very good method for deinterlacing. It creates a smooth video, but static objects loose resolution. The best solution is to sombine BOB and WEAVE in each frame. Use BOB for moving parts and WEAVE for static picture. Most good deinterlace filters are doing it this way.

If you're interested some time ago I did a test to compare different deinterlace algorithms. You can find it here (http://jech2.webz.cz/gallery/thumbnails.php?album=4). Have a look at the sign in bottom left corner. It perfectly shows the lost resolution when using BOB.

Also I wouldn't even call Weave deinterlacing. Any PC monitor/PDP/LCD/projector only displays video progressively. So you can't display interlaced video on it. Weave means that you just combine two fields into one frame. You can't separate them back into fields whenever you want. It is just a method of how the fields are stored or displayed. There is no processing of the video.

[FredThompson]

...at least it is for ffdshow_rev457_20061027_clsid...

Here's a short video and frame grab to show the problem. Yellow looks teal with ffdshow's decoder.

This was encoded with the GZip option at level 9 for maximum compression.

As a side note: I'm not sure if CamCodec has proper indexing or not. Loading its files into VirtualDub shows those created by CamStudio don't have indexing data. Sure would be nice if this were added, especially since CamStudio will allow ffdshow to be the encoder. As it is, CamCodec stuff really needs to be saved to another format for editing use.

[Wilbert]

Anyway BOB is not a very good method for deinterlacing. It creates a smooth video, but static objects loose resolution.
Why ...?

[haruhiko_yamagata]

http://i12.tinypic.com/2igg4lu.jpg
I'm trying to expand the dialong.
The upper sample is the original, the next is the new one.
What do you think about the size?

[MacAddict]

Personally I like the new dialog.

[Jeremy Duncan]

The new Dialong looks nice.

[haruhiko_yamagata]

Thank you. I commited resized dialog. Please see if it works.

[MatMaul]

I have made a little test to mesure the h264 decoding speed improvement.

ffdshow rev2546 : 104 fps
ffdshow-tryout rev494 : 130 fps (!!)

+25% speed-up !!

Great job and thanks to ffmpeg guys !

[Jeremy Duncan]

The newest FFdshow on ffdshow.info is rev490 by drevil_xxl

The rev I need to test the new dialong is rev492.
I'm hoping rev494 will be released today, as MatMaul made me curious.

[MatMaul]

The newest FFdshow on ffdshow.info is rev490 by drevil_xxl

The rev I need to test the new dialong is rev492.
I'm hoping rev494 will be released today, as MatMaul made me curious.

Sorry but I have just compiled libavcodec.dll because rev494 have h264 decoding improvement, and not ffdshow.ax to do this test.
So waiting a new build of drevil_xxl or clsid.

[clsid]

Most of the H.264 decoding improvements were already done before rev. 490. But I'll put a new build up in a moment.

@Haruhiko, can you increase the width of the 'Format' and 'Supported ...' columns a bit on the codecs page?

[Jeremy Duncan]

The new dialong looks nice.

And rev494 is faster.

[Bathrone]

How do I check what changes are in the ffmpeg builds. When I look on the ffmpeg svn in changelog.txt it doesnt have the related build numbers that the ffdshow changelog references.

Also, why is flac decoding not present in all the installer builds? And ones that do have it, it's not installed by default?

[Px]

I have made a little test to mesure the h264 decoding speed improvement.

ffdshow rev2546 : 104 fps
ffdshow-tryout rev494 : 130 fps (!!)

+25% speed-up !!

Great job and thanks to ffmpeg guys !
I don't see such big improvement before, is someting dramatically change in rev 494?
H.264 1080p

ffdshow-20060604-rev2546.exe
Overlay VMR9
fps dfps fps dfps
19,6 19,3 16,8 15,6
ffdshow_rev420_20061020_clsid.exe
20,6 20,2 20,3 19,1
Overlay VMR9
fps dfps fps dfps

[Eragon4ever]

How do I check what changes are in the ffmpeg builds. When I look on the ffmpeg svn in changelog.txt it doesnt have the related build numbers that the ffdshow changelog references.

When I checked the svn I can say "show log" in my svn client. (TortioseSVN)
It is not in the changelog file.

[clsid]

I don't see such big improvement before, is someting dramatically change in rev 494?Nope, there have been several revisions with small performance updates.

[Px]

Nope, there have been several revisions with small performance updates.
Ok, I'll test it in 3 or 4 hours...

[clsid]

List of known issues in revision 513:

1) Wavpack decoder only works with lossless wavpack. Lossy and hybrid wavpack is not yet supported.
2) Levels (ylevels, ylevelsG, ylevelsS, ylevelsC) are not implemented correctly. Details about the correct way can be found here (http://forum.doom9.org/showthread.php?p=897854).

Reported but unconfirmed bugs:

3) image settings -> preset autoload conditions -> "on FOURCC match" doesn't work. (reported by juskixxx) (confirmed by clsid) (haruhiko could not reproduce it)
4) Automatic preset loading doesn't seems to be working with on "number of channel match" and AC3 audio on DVD (don't know if it happens with AC3 audio in MKV, AVI or not or any other audio format). I use "3;4;5;6" in "number of channel match" so this should load with audio track with more than 2 channels but with an AC3 2 channels track this is also loaded. (reported by thuan)
5) The following encoders do not work for me: MPEG 4, MPEG 1, MPEG 2, h.263, H.261 and DV. VirtualDub 1.16.16 gives the following error: "Cannot start video compression. An unknown error occurred (may be corrupt data). (error code -100)". I know that at least some of these encoders do work for others. My system specs: Windows 2000, AMD Athlon Thunderbird. (reported by clsid)
6) Resize filter alters colors. Details (http://forum.doom9.org/showthread.php?p=884778#post884778). (reported by Kador)

Other issues:

7) ICL9 builds of ffdshow.ax crash on files created by a specific old revision of x264 (don't know the rev number). Funny thing however is that the files play without crash if you first play a good file and then play a 'troublesome' file in the same player instance. Also no crash when using an unoptimized debug build. So this seems to be a compiler bug. Sample file (http://rapidshare.com/files/1609924/sample.mp4.html). (reported by clsid)

[MatMaul]

1) Wavpack decoder doesn't work with this file (http://rapidshare.com/files/1609252/SoC__XviD_WavPack_MP3_-007.mkv.html). (reported by clsid)
After more testing, it doesn't work with all hybrid wavpack files, but I think it's an ffmpeg issue.

EDIT : You have removed vc1 support because it doesn't work.
Can you give me some links of files wich do not work please ?

[clsid]

http://samples.mplayerhq.hu/V-codecs/WVC1/

[MatMaul]

thanks.
wavpack decoder also doesn't work with 5.1 files.

[Liisachan]

After more testing, it doesn't work with all hybrid wavpack files, but I think it's an ffmpeg issue.? If using Gabest's splitter, hybrid wavpack doesn't work anyway. Haali's newer splitter works ok with CoreWavpack, which supports hybrid wavpack beautifully. foo_packet_decoder_wavpack too, by using codes from CoreWavpack.

[Inventive Software]

Potential bug with "Automatic Quality Control" in "Post-processing" tab, as in it's not automatic.
With any previous ffdshow build (not ffdshow_tryout) it's fine, no problems, but ffdshow tryout builds don't have this.
File in question is DivX encoded, and quite post-processing heavy, but with ffdshow builds it's definitely adjusted well. ffdshow_tryouts just don't do this, so something's broken between ffdshow and ffdshow_tryouts.

ffdshow_tryout builds tested include ffdshow_rev494_20061103_clsid (my current build I use), ffdshow-Tryouts-20061023-rev435. Both exhibited said post-processing behaviour.

ffdshow builds tried include ffdshow_rev2543_20060816, ffdshow-20060123, both were fine with automatic post-processing.

Quick note: post processing settings used were 100% strength, mplayer accurate deblocking (and tested with accurate deblocking off).

[MatMaul]

If using Gabest's splitter, hybrid wavpack doesn't work anyway. Haali's newer splitter works ok with CoreWavpack, which supports hybrid wavpack beautifully. foo_packet_decoder_wavpack too, by using codes from CoreWavpack.
I haven't any problem with splitters, I think it's just ffmpeg wavpack decoder actually doesn't support hybrid and 5.1 wavpack files.

[Liisachan]

Gabest's splitter does split A_WAVPACK4 in mkv, but the "correction" part--if it exists--is ignored and the downstream receives the lossy part only. Note, if the lossy part is transparent (enough high bitrate) you can't tell this problem by just listening to the output. It's not like it doesn't play. It's not bit-identical like it should be. So, if you use Gabest's splitter or MPC's internal splitter, hybrid wv packed in mkv won't play as lossless even if the wavpack decoder knows how to handle the hybrid. More specifically, altho CoreWavpack can decode the hybrid, it plays the hybrid as Lossy if the upstream is Gabest's splitter.

In this case, tho, it seems that the decoder in ffdshow doesn't like the hybrid either.

Anyway, to test that sample file above, the upstream splitter should be Haali's. That's what I meant. As another note, wv+wvc in matroska would be practically pointless. If you'd like to make the audio in mkv lossless, you can just use lossless WV in the 1st place.

On the other hand, if ffdshow's audio decoder can decode foo.wv+foo.wvc (bare 2 files, not packed in matroska), that may come in handy for some users. But then again, that would mean ffdshow has to look for .wvc in the same folder whenever it plays lossy .wv... I'm not sure if its worth even bothering to do so.

[foxyshadis]

Potential bug with "Automatic Quality Control" in "Post-processing" tab, as in it's not automatic.
With any previous ffdshow build (not ffdshow_tryout) it's fine, no problems, but ffdshow tryout builds don't have this.

Yes, this was mentioned before, it's most likely because cpu usage has been broken for some time. Unfortunately so far there's no good way to make it work on x64 XP/Vista, without crashing, so it was disabled. If a better method's found, I'm sure it'll be back, but I don't know all that much about WMI.

(Another method is to work that into the QoS system, but the reason I stalled on finishing that is Haali's renderer doesn't send Quality messages, like the windows ones, since ffdshow used to work very badly with them.)

@Everyone: How about releasing a build considered "Stable" at this point?

[MatMaul]

ok, I understand.
But actually, ffdshow doesn't decode at all the lossy part, so I think ffmpeg library also can't decode a lossy+correction file.

[_xxl]

@Everyone: How about releasing a build considered "Stable" at this point?
There are known bugs that need to be fixed before the first beta is released.

[Inventive Software]

@Everyone: How about releasing a build considered "Stable" at this point?
Keep with revisions in my opinion.

Question: how does CPU usage affect the automatic post-processing? That seems like an ineffective way to adjust it IMO. Surely the best way is analysing the image for blocks and adjusting as appropriate?

[clsid]

That would mean to do some form of post-processing to determine if post-processing is required :P

In old builds auto pp would adjust the amount of pp depending on the current cpu load. The cpu load monitoring code is not used in new builds because it fails on Vista.

[foxyshadis]

There will always be known bugs, that's the nature of software. Currently, most of the bugs since the last big releases of ffdshow are long gone, performance has risen, and the lack of "stable releases" that don't have disruptive new codecs and features will keep it from being used by many who'd benefit from it.

Inventive: This was all designed back when the P3 was common, and cpu usage was a major problem. PP basically hasn't changed since then, aside from the now-removed h.264, and SPP. The deblocking itself will use the quant to estimate blocks (thus its terrible performance on raw sources), but if you want real analysis you need to use SPP, but it will be slow and kind of fuzzy. ffdshow, ffmpeg, mplayer, and so on have nothing similar to Deblock_QED, sadly.

But ffdshow does have avisynth input. ;)

[ExtraEye]

sorry but what is "Deblock_QED"?

[foxyshadis]

An Avisynth filter, you should be able to search the board for it. The wiki isn't up or I'd link to the page for it.

[yesgrey]

I think ffdshow misses one thing in the resize dialog: the option to resize to screen size.

I use two screens for seeing my stuff: a projector and a monitor. Since both screens have different resolutions, I have always to change the resolution by typing the new one. It would be easier to just set it to screen size and then ffdshow get it and resize to it. Even better if we could also set the aspect ratio, then ffdshow should only use the screen width as the reference.

Something like this:

Resize:
- Specify Size (already done)
- Specify Aspect Ratio (already done but I think it's not very usefull, since it keeps the original width of the file, and then the resize have to be done by the videocard drivers)
- Expand to Next Multiply of (already done)
- Multiply by (already done)
- Screen Size (to be done)
Aspect Ratio (drop list: original; 4:3; 16:9; 2.35:1)

Or it could just be incorporated in the Specify Aspect Ratio.

I think this is a very important feature, because the resize by ffdshow is of a better quality than the resize of the videocard drivers.

Let me know what you think.

[Bathrone]

Not a bug, but why if flac not installed by default? And I know one installer atleast doesnt seem to offer it during installation as an option.

[bob0r]

x264.nl added clsid ffdshow revision 497.

I must say i am amazed, using only one core of my Intel D930, a BBC-HD H.264 1080MBAFF sample almost runs smooth (enabled the "no deblocking tweak" ofcourse, but same for coreavc)

If ffmpeg would use multi core/cpu code i am sure it could match, or even surpass all the other H.264 decoders.

[Isochroma]

Unicode characters in filenames played with MPC & ffdshow show in the ffdshow dialog as ?

[Px]

Reported but unconfirmed bugs:
4) Instant crash when trying to play this MSS2 WMV file (http://ftp.mplayerhq.hu/MPlayer/samples/V-codecs/MSS2/mss2_speech.wmv). Happens on Windows 2000. No crash on drevil_xxl's XP system. (reported by clsid)

I tested this file in my w2000 system, in WMP all plays fine (of course), in mpc video plays fine, but without sound.....

[Px]

Some speed comparsions with ffdshow (previous results was on other system with slower cpu)
Video - cornell_m1080p.mp4, H.264, 1920x1080, 8634 kbps

fps vmr9 overlay
ffdshow_rev135_20060904_icl91 33.4 32.8
ffdshow_rev420_20061020_clsid 33.4 32.8
ffdshow_rev482_20061101_clsid_icl9 33.3 32.8
ffdshow_rev494_20061103_clsid 33.4 33.0


Video - kane_reveal_1280x720_h264.mp4, H.264, 1280x720, 2255 kbps

fps vmr9
ffdshow_rev135_20060904_icl91 117.5
ffdshow_rev420_20061020_clsid 120.3
ffdshow_rev482_20061101_clsid_icl9 119.5
ffdshow_rev494_20061103_clsid 122.5

Only small speedup in second case, but it's better than nothing....

[Egh]

I think ffdshow misses one thing in the resize dialog: the option to resize to screen size.

I use two screens for seeing my stuff: a projector and a monitor. Since both screens have different resolutions, I have always to change the resolution by typing the new one. It would be easier to just set it to screen size and then ffdshow get it and resize to it. Even better if we could also set the aspect ratio, then ffdshow should only use the screen width as the reference.


You dont' have to specify AR, or to specify resolution in full.

Only thing required is to specify target width, and leave 0 for height. Then ffdshow automatically applies DAR from video.

An option to resize to current screen resolution would be nice, imo. But some questions arise, for instance, what to do if resolution is not constant during video playback. Should ffdshow use only resolution which it gets on initialisation or readjust it whenever it changes? Multimonitor configuration might also cause some problems.

[haruhiko_yamagata]

Unicode characters in filenames played with MPC & ffdshow show in the ffdshow dialog as ?
What do you mean? Please tell us the detail.
Whose build are you using? I mean, is it ANSI build or UNICODE build?

[Amour]

Unicode characters in filenames played with MPC & ffdshow show in the ffdshow dialog as ?
Works fine for me.

[yesgrey]

Only thing required is to specify target width, and leave 0 for height. Then ffdshow automatically applies DAR from video.


Yes I know this, but it does not work always. Sometimes I have to specify also the height because the video DAR is not correct. It also not solves the problem of having to change the width.

[yesgrey]

But some questions arise, for instance, what to do if resolution is not constant during video playback. Should ffdshow use only resolution which it gets on initialisation or readjust it whenever it changes? Multimonitor configuration might also cause some problems.

I think we can clear up these situations. It could use the resolution it gets on initialization, since I don't know why (yes, I know someone could) anyone would change the screen resolution in the middle of a file viewing. The multimonitor it's not also a big question, because it could simply get the resolution of the screen in which the media application is opened during the initialization process.

I will try to do it and see how it works...

[Egh]

Yes I know this, but it does not work always. Sometimes I have to specify also the height because the video DAR is not correct.

Which files are those? Anamorphic resolution AVI? You can try to remux them into mkvs and assign proper DAR that way.

Not that I see much anamorphic avis anyway :P

[Inventive Software]

I think ffdshow misses one thing in the resize dialog: the option to resize to screen size.
Media Player Classic? It has a "Stretch to Window" option which is appropriate for your needs I think.

[Egh]

Media Player Classic? It has a "Stretch to Window" option which is appropriate for your needs I think.

The problem is that it's not using ffdshow resizer in that case.
And if you haev high-speed CPU but slow GFX card in terms of shaders (for instance my present system configuration) you can't actually use MPC shader-based bicubic or Haali renderer. Such cards could be, for instance, old FX5xxx series, or moden built-in in the motherboard (like 6100/150 in NForce). Not all computers are built keeping gaming in mind (office computers are one such particular case :D)

So if you are not using ffdshow resizer you're able to use only standard bilinear and for me that's not enough.

[Liisachan]

Ok, I now know the "problem"--i was mislead by this report:

1) Wavpack decoder doesn't work with hybrid wavpack files. Sample file (http://rapidshare.com/files/1609252/SoC__XviD_WavPack_MP3_-007.mkv.html). (reported by clsid)

The audio in that sample file is Wavpack Lossy, not Wavpack Hybrid. So we can simply say "Wavpack decoder doesn't work," which is much bigger problem than not just supporting Hybrid.

[_xxl]

Ok, I now know the "problem"--i was mislead by this report:
The audio in that sample file is Wavpack Lossy, not Wavpack Hybrid. So we can simply say "Wavpack decoder doesn't work," which is much bigger problem than not just supporting Hybrid.
This file is working ok.
http://ftp.mplayerhq.hu/MPlayer/samples/A-codecs/lossless/luckynight.wv

[MatMaul]

Yes it's a lossless 2ch sample and it is the only configuration which works.

[Liisachan]

That file is Wavpack Lossless, not Wavpack Lossy nor Hybrid.
Wavpack decoder doesn't work generally, not only for Hybrid:

Current ffdshow
Wv 4.3 4.4
Lossy x x
Hybrid x x
Lossless o #

# works, if --optimize-mono is not used

[MacAddict]

(snip)

An option to resize to current screen resolution would be nice, imo.
(/snip)


That would be great and about the only feature I can think of that I'm missing on my HTPC!

[Px]

Reported but unconfirmed bugs:
4) Instant crash when trying to play this MSS2 WMV file (http://ftp.mplayerhq.hu/MPlayer/samples/V-codecs/MSS2/mss2_speech.wmv). Happens on Windows 2000. No crash on drevil_xxl's XP system. (reported by clsid)

I tested this file in my w2000 system, in WMP all plays fine (of course), in mpc video plays fine, but without sound.....
I cleared my system from codecs garbage, reinstalled last DX and wmp9, now this file plays normally (sound and video) in mpc and bsplayer, so I think that is problen with system, not ffdshow....

[Jeremy Duncan]

There's a bug in the Levels Tab.

Didée said that the ylevels, ylevelsG, ylevelsS, ylevelsC.
That Milan Cutka put into FFdshow, wasn't done properly.

He said that what Milan put into FFdshow was exactly backwards and he doesn't recommend using the ylevels, ylevelsG, ylevelsS, ylevelsC in FFdshow.

Here's the thread Didée started that shows how ylevels, ylevelsG, ylevelsS, ylevelsC works and the code it uses.
Link (http://forum.doom9.org/showthread.php?s=&threadid=79898)

May I ask that this bug is fixed and Didée's ylevels, ylevelsG, ylevelsS, ylevelsC is properly implemented into the FFdshow Levels tab. :)

[_xxl]

/**
* @file wavpack.c
* WavPack lossless audio decoder
*/
Wavpack Lossy & Hybrid aren't supported.

[Liisachan]

maybe they didn't know Wavpack had lossy mode. They might want to fix this limitation after 4.4 is officially out, which would make ffdshow's Wavpack decoder even better than CoreWavpack.

[Inventive Software]

Small request: Vista only ffdshow_tryout builds that have the CPU detection code disabled, and for XP/2000 CPU detection on...

[clsid]

Or perhaps detect the OS version at runtime. Or let the installer set a flag in the registry (HKLM) to disable the CPU usage monitor.

[Jeremy Duncan]

There's a bug in the Levels Tab.

Didée said that the ylevels, ylevelsG, ylevelsS, ylevelsC.
That Milan Cutka put into FFdshow, wasn't done properly.

He said that what Milan put into FFdshow was exactly backwards and he doesn't recommend using the ylevels, ylevelsG, ylevelsS, ylevelsC in FFdshow.

Here's the thread Didée started that shows how ylevels, ylevelsG, ylevelsS, ylevelsC works and the code it uses.
Link (http://forum.doom9.org/showthread.php?s=&threadid=79898)

May I ask that this bug is fixed and Didée's ylevels, ylevelsG, ylevelsS, ylevelsC is properly implemented into the FFdshow Levels tab. :)

I wonder if this will be fixed.
I sure hope it will.
haruhiko_yamagata, clsid, drevil_xxl. Will you please fix this bug.

:)

[frodeste]

What are your thoughts about starting building 64 bit editions of ffdshow?

Also, any thoughts about building an autoupdate in to the product, now that the builds and patches are beeing produced at such a high level?

This is something I would be very thankful for.

Keep up the good work.

[Egh]

Also, any thoughts about building an autoupdate in to the product, now that the builds and patches are beeing produced at such a high level?


Don't think there's any need to use an autoupdater, not before a "stable" version is released in any case.

[yesgrey]

Which files are those? Anamorphic resolution AVI? You can try to remux them into mkvs and assign proper DAR that way.

Not that I see much anamorphic avis anyway :P

Maybe badly encoded files, with the wrong DAR.

[yesgrey]

The problem is that it's not using ffdshow resizer in that case.
And if you haev high-speed CPU but slow GFX card in terms of shaders (for instance my present system configuration) you can't actually use MPC shader-based bicubic or Haali renderer. Such cards could be, for instance, old FX5xxx series, or moden built-in in the motherboard (like 6100/150 in NForce). Not all computers are built keeping gaming in mind (office computers are one such particular case :D)

So if you are not using ffdshow resizer you're able to use only standard bilinear and for me that's not enough.

But even bicubic has not enough quality. For me the best is Lanczos with 2 taps, which gives the higher detail possible without ringing. Since I use a projector with a big screen, the resize algorithm is very important!

I have made several test with all ffdshow algorithms and several resolutions and the best option IMO is resizing with Lanczos2 for the screen resolution.

We could also resize for a higher resolution and then let the videocards drivers downsize it for the screen resolution, but besides being lower quality is also more CPU intensive.

I will try to get some time to do it. I will probably need some help from some of the developers here but when the time arise I'll ask for it.

[cc979]

Maybe badly encoded files, with the wrong DAR.

i remember i had problems with anamorphic avis xvid ones anyway, will play fine in mpc with vmr9 not overlay

[Egh]

But even bicubic has not enough quality. For me the best is Lanczos with 2 taps, which gives the higher detail possible without ringing. Since I use a projector with a big screen, the resize algorithm is very important!

I have made several test with all ffdshow algorithms and several resolutions and the best option IMO is resizing with Lanczos2 for the screen resolution.

We could also resize for a higher resolution and then let the videocards drivers downsize it for the screen resolution, but besides being lower quality is also more CPU intensive.


Well some love Lanczos with 4 taps, some favor Spline36resize. Depends also if it's anime or RL content and if the source is sharp in the first place.

Regarding bicubic -- did you try to adjust it's parameters first? :D Default bicubic and something like a=0, b=0.5 are two different things in terms of sharpness :P

But generally I support this kind of patch, if you are able to provide it, thanks alot.

[baka]

I use crystal player and always newest build of ffdshow, and when i want to seek, it freezes... :(

that queue thing doesn't help, what to do?

[Anima123]

I use crystal player and always newest build of ffdshow, and when i want to seek, it freezes... :(

that queue thing doesn't help, what to do?

Same here, even with queue disabled. I have to stop using ffdshow. The problem might happened with buffer mechanism of crystal player. There's no problem while using XviD as decoder.

[Jeremy Duncan]

Please read the link:
Link (http://forum.doom9.org/showthread.php?t=79898)
The code in the threads first post was put into FFdshow by Milan Cutka. Only Milan put it in wrong, according to Didée. Read reply # 21 to see Didée's comment on this.

In a thread I erased. Didée commented that when Milan put the code in the thread I linked to in FFdshow. He did it so it was exactly wrong.

Please ask Didée about it in the thread I linked too.

[Didée]

Regarding "Ylevels" in ffdshow -

forget any rumbling that ever has been done. It was *me* who implemented my own idea *wrongly*. :o
ffdshow just reproduces my faulty implementation, quite exactly. ;)

"Right" and "wrong" are relative, since it's an arbitrary operation ... it just does not what I wanted it to do.

[baka]

and now we just have to wait to get that right or? :)

[haruhiko_yamagata]

Please read the link:
Link (http://forum.doom9.org/showthread.php?t=79898)
The code in the threads first post was put into FFdshow by Milan Cutka. Only Milan put it in wrong, according to Didée. Read reply # 21 to see Didée's comment on this.

In a thread I erased. Didée commented that when Milan put the code in the thread I linked to in FFdshow. He did it so it was exactly wrong.

Please ask Didée about it in the thread I linked too.
OK, just wait untill we have time to fix it, please.

Didée, thank for your code and coming here.

[Eragon4ever]

@developers: You really should put yourselves into authors.txt!
You are doing a great job and this should be documented.

[Didée]

Well, in that forgotten PM that Alain2 once had sent me, he included changes that seem to do things the correct way.
Thanks, Alan!

I've put up that message in the test forum (http://forum.doom9.org/showthread.php?p=897854#post897854), to not spoil this thread (message takes much space).

(Surprisingly, the linked images are still online on Imageshack.)

[haruhiko_yamagata]

@developers: You really should put yourselves into authors.txt!
You are doing a great job and this should be documented.
Thank you.

[Romario]

I am sorry, but I must to ask this. What's going on with Milan Cutka? Is he abandon ffdshow developing?

[LoRd_MuldeR]

I am sorry, but I must to ask this. What's going on with Milan Cutka? Is he abandon ffdshow developing?

yes. afaik he just disappeared.
that's why official ffdshow website was not updated for a LONG time...

[yesgrey]

Well some love Lanczos with 4 taps, some favor Spline36resize. Depends also if it's anime or RL content and if the source is sharp in the first place.

Regarding bicubic -- did you try to adjust it's parameters first? :D Default bicubic and something like a=0, b=0.5 are two different things in terms of sharpness :P

But generally I support this kind of patch, if you are able to provide it, thanks alot.

I once also prefered the Lanczos4, but recently changed to 2 because of the ringing added by L4. I will try the bicubic as you describe. I have once read that Lanczos resize is implemented in ffdshow as bicubic but with different a and b. So, maybe you have just 're-invented' Lanczos... :-)

[haruhiko_yamagata]

I use crystal player and always newest build of ffdshow, and when i want to seek, it freezes... :(

that queue thing doesn't help, what to do?
I tryed crystal player. Seek works unless it freezes up when opening a file. Crystal player freezes up when opening some 720p mov file(Apple Quick time trailer).

What is the file that freezes? MOV, AVI ,... and h.264, divx,...

[SmallwooD13]

What is the file that freezes? MOV, AVI ,... and h.264, divx,...

avi (DivX, XviD) and mp4 (AVC1)

avi H264 (one film tried), wmv, mpg, are working normaly

FFdshow-Tryouts-20061025-rev446-sse2, CrystalPlayer-1.96
other players are working normaly (LA, media player classic)

[_xxl]

Is a ffdshow-tryout that works ok?
CrystalPlayer-1.97 freezes up for me when I try to exit or to seek video.I have tested some avi files.(XVID,AVC1 or H.264).
For audio is used ffdshow's libmad & libfaad2.
What is "ActiveMovie Filter Class Manager"?
http://img14.imagevenue.com/loc415/th_82320_Capture_122_415lo.jpg (http://img14.imagevenue.com/img.php?image=82320_Capture_122_415lo.jpg)

[haruhiko_yamagata]

What is the audio codec?

[clsid]

"ActiveMovie Filter Class Manager" manages DirectShow filters.

[BlindWanderer]

Hey, i got ffdshow to crash on picture properties with drevil_xxl's sse2 build when using "hue". I have an Athlon 64 X2, so it shouldn't be an issue of cpu not supporting it. The generic build by clsid doesn't crash under the same conditions (using rev 514 in both cases). New bug or an old one?

[_xxl]

Hey, i got ffdshow to crash on picture properties with drevil_xxl's sse2 build when using "hue". I have an Athlon 64 X2, so it shouldn't be an issue of cpu not supporting it. The generic build by clsid doesn't crash under the same conditions (using rev 514 in both cases). New bug or an old one?
No crash on AMD XP.(SSE)

[Bathrone]

Thanks again for FFdshow

Drevils build has FLAC as an option. CLSIDs build does not list flac. Why?

[FredThompson]

ffdshow_rev514_20061109_clsid and libmpeg2 as the MPEG2 decoder for MPC will allow proper playback of MPEG2 4:2:2 sources.

However, MPC seems to be more fault tolerant WRT muxing in some streams. Here is an example which will play properly with MPC's MPEG2 decoder but ffdshow will have problems using either libavcodec or libmpeg2.

MPC set to VMR7 Renderless (to prevent hardware acceleration messing with luma) and ffdshow libmpeg2 works great, except the mux on these DVR streams really screws it up. Is this a function of demuxing in ffdshow? Is there a way configure demuxing so this doesn't happen?

Here's the file (6.5 M): http://www.corrugatedmachines.com/FruitBeer.vob

[_xxl]

Is this a ffdshow bug?Did you try mplayer?
EDIT
http://tirnanog.fate.jp/mirror/mplayer/mplayer2006.11.09.7z
http://mulder.brhack.net/public/downloads/MPUI.2006-11-09.Full-Package.exe

[FredThompson]

I haven't tried another player. Will get mplayer and test. In any event, that sample stream does show a video skip and freeze. At 6.5M and pure DVR rip, it's a decent test case.

[haruhiko_yamagata]

ffdshow_rev514_20061109_clsid and libmpeg2 as the MPEG2 decoder for MPC will allow proper playback of MPEG2 4:2:2 sources.

However, MPC seems to be more fault tolerant WRT muxing in some streams. Here is an example which will play properly with MPC's MPEG2 decoder but ffdshow will have problems using either libavcodec or libmpeg2.

MPC set to VMR7 Renderless (to prevent hardware acceleration messing with luma) and ffdshow libmpeg2 works great, except the mux on these DVR streams really screws it up. Is this a function of demuxing in ffdshow? Is there a way configure demuxing so this doesn't happen?

Here's the file (6.5 M): http://www.corrugatedmachines.com/FruitBeer.vob
It works for me, except for screwing up on seek.
If you are talking about this seek problem, it should be ffdshow's issue.

[baka]

avi (DivX, XviD) and mp4 (AVC1)

avi H264 (one film tried), wmv, mpg, are working normaly

FFdshow-Tryouts-20061025-rev446-sse2, CrystalPlayer-1.96
other players are working normaly (LA, media player classic)

ye, freezes when playing avi, mp4, and i think some other, no freeze in wmv

mpc is working ok same as smallwood said

[FredThompson]

It works for me, except for screwing up on seek.
If you are talking about this seek problem, it should be ffdshow's issue.Are you SURE? When I play the vob with ffdshow as the MPEG2 decoder, a portion at the beginning is missed then the video freezes. The portion of a few words about putting a lime in beer doesn't get played.

[Jeremy Duncan]

- In Media Player Classic External Filters Tab.
Double Click the FFdshow Audio decoder to access the configuration.
- Check OSD,and Check Jitter.
- Close FFdshow Audio Decoder, and play a movie.
- While the movie is playing. You'll the the word "Jitter" in the top left of the screen.
Beside the word Jitter is a number. Write this number down, or remember what it was.
- Open the FFdshow Audio decoder again.
- Check the Delay Tab. Set each box to the number you wrote down in the previous step.
- Check the Volume Tab. Leave it at defaut.
- Check the Mixer Tab. Check "Normalize Matrix", Check the "Output Speakers Configuration" tab. If you are using only 2 speakers, or headphones select "2/0 - stereo". This step is very important, do it properly or the sound will be all messed up.
- Close the FFdshow audio decoder by pressing Apply and OK.

Following these setup instructions.
I find that the number beside Jitter is always 32
Regardless of the pc used.
It's happened to both my pc's and somebody else.

The number by jitter is different than the number beside "time one FFdshow". Are both jitter and time on ffdshow showing the ms number that is jitter ?
If so, the audio decoder isn't working, because it stays at a solid 32 ms and never changes.

I'm using the November 5, 2006 21:20:40 ansi version

[Jeremy Duncan]

Also, I can't get FFdshow to work with Gomplayer.
Please look at getting FFdshow to work with Gomplayer.

Link to Gomplayer (http://www.gomplayer.com/main.html)

[cc979]

Also, I can't get FFdshow to work with Gomplayer.
Please look at getting FFdshow to work with Gomplayer.

Link to Gomplayer (http://www.gomplayer.com/main.html)

ffdshow works with gom player if you disable built-in codecs
preferences > other > codec page

[haruhiko_yamagata]

Are you SURE? When I play the vob with ffdshow as the MPEG2 decoder, a portion at the beginning is missed then the video freezes. The portion of a few words about putting a lime in beer doesn't get played.
Yes, it's played. I tested three computers. Seek does not work properly.
Please make a detailed report about your environment including Video card, output color space, OS, CPU, etc.

[FredThompson]

Pfff...there's no way to fully document everything which could theoretically be causing a choke during playback. Summary info is Intel Core Duo 1.6 XP Pro SP2 Gateway laptop. I'll test on other systems to see if I have one which will play properly then look at precedence with GSpot.

[_xxl]

The number by jitter is different than the number beside "time one FFdshow". Are both jitter and time on ffdshow showing the ms number that is jitter ?
If so, the audio decoder isn't working, because it stays at a solid 32 ms and never changes.
I'm using the November 5, 2006 21:20:40 ansi version
http://i15.tinypic.com/2v92m85.jpg

[KoD]

That FruitBeer.vob file freezes for me even when using Nvidia's mpeg2 video and audio decoder. Using the built-in decoders from MPC will cause playback freezing as well. Are you sure the file is not broken itself ? (or maybe the vob splitter is at fault ?)

[haruhiko_yamagata]

That FruitBeer.vob file freezes for me even when using Nvidia's mpeg2 video and audio decoder. Using the built-in decoders from MPC will cause playback freezing as well. Are you sure the file is not broken itself ? (or maybe the vob splitter is at fault ?)
It works only when MPC's internal source filter "MPEG PS/TS/PVA" is checked. Why?

[Rgeek]

I got a file(Container AVI - Codec WMV3) but some reason ffdshow doesn't decode It.

Someone can help me?


Thanks in advance.


Sample: http://rapidshare.com/files/2933552/SS.avi.html

[Inventive Software]

WMV3 isn't fully supported by ffdshow. :search:

[Rgeek]

WMV3 isn't fully supported by ffdshow. :search:


I know but all WMV3 files I got work with ffdshow, except one.


I'm curious why. What's different about this file?

[foxyshadis]

Probably one of the more advanced Main Profile features, like deblocking, new frame types, and so on. Kind of hard to tell without a sample, we aren't psychic. Anyway, always test with mplayer before reporting a codec bug, because it's most likely actually an ffmpeg bug/limitation. (Definitely is in this case.)

[FredThompson]

It works only when MPC's internal source filter "MPEG PS/TS/PVA" is checked. Why?

Enabling that setting means MPC will decode, not ffdshow. That's why I questioned your report of proper playback.

That file is a satellite DVR stream with mux to accomodate dropouts, most often those happen in audio. It might also use the field duplication flag. It's MPEG2 valid but not DVD spec valid. It's a nice small real-world test sample. Is it choking ffdshow's demux or one of the video decoders or a combination of both? I don't know. My hunch is this is a demux issue because both MPEG2 decoders choke in the same manner.

It's not an odd profile, btw. Burn it to a disc and it will play in almost any hardware DVD player although some of the Sony units are touchy. It will also play in "proper" DVD players like PowerDVD.

[Rgeek]

Probably one of the more advanced Main Profile features, like deblocking, new frame types, and so on. Kind of hard to tell without a sample, we aren't psychic.

I already send a sample to Rapidshare. Link in my first post in this Thread.


Anyway, always test with mplayer before reporting a codec bug, because it's most likely actually an ffmpeg bug/limitation. (Definitely is in this case.)

In Media Player Classic, adding WMVideo Decoder DMO in external filters, ffdshow decode the file. Is recognized as YV12 - uncompressed.

[FredThompson]

FruitBeer.vob is screwing up with mplayer. The audio stream is properly decoded, something MPC with ffdshow will choke on, but the video still skips and sync is lost. Grrr...

[_xxl]

Hey, i got ffdshow to crash on picture properties with drevil_xxl's sse2 build when using "hue". I have an Athlon 64 X2, so it shouldn't be an issue of cpu not supporting it. The generic build by clsid doesn't crash under the same conditions (using rev 514 in both cases). New bug or an old one?
Confirmed.

[foxyshadis]

Enabling that setting means MPC will decode, not ffdshow. That's why I questioned your report of proper playback.
No, it's the MPC internal splitter, not decoder. So whichever splitter you're using is likely screwing something up.

In Media Player Classic, adding WMVideo Decoder DMO in external filters, ffdshow decode the file. Is recognized as YV12 - uncompressed.
Of course it does, because ffdshow isn't decoding anything! WMV Decoder DMO is microsoft's wmv decoder (obviously), ffdshow just piggybacks on the uncompressed output.

As I said, test in mplayer/ffmpeg (not MPC) before reporting against ffdshow. Anyway, mplayer doesn't play it either, it just opens the same DMO decoder to pass it through first.

It looks like for whatever reason, ffmpeg won't accept this file, so you'll have to take it up with the ffmpeg mailing list.