ffdshow and ffdshow_tryout Builds Listing [Archive]

Inventive Software

9th January 2006, 15:18

Right. Since this thread (http://forum.doom9.org/showthread.php?t=98600) has more than 2000 posts to filter through, I figure we need a new thread that just has one post for each builder, and each builder can update this as necessary. If the builder could state what MMX and SSE support there is, and what compiler(s) has/have been used this will clarify for newbies. If you have a problem with a particular build, post here or in the other thread and we'll get back to you.

If you have a general ffdshow problem that you believe is not the cause of a specifically compiled build, start another thread and don't pollute this one please!

Inventive Software

Slave01

13th January 2006, 13:17

I support the request...It seems that aziendeassociate....link is stopped to 1103 build....the other builds are slower on my sempron 3000+...What is the best build organized by processor...

Thanks
John Slave

Inventive Software

13th January 2006, 13:51

Sempron 3000+, that speaks SSE2 to me. You have lots of choices there!
Milan's "official" daily build, not very recent, but supported. (http://cutka.szm.sk/files/ffdshow-20051129.exe) Dated 2005/11/29
Bob0r's build, (http://files.x264.nl/ffdshow/ffdshow-20051208-gcc4.0.2-sse2-x264.nl.exe) dated 2005/12/08.
Drevil_xxl's build, (http://www.myfilehut.com/userfiles/5120/CollectedBuilds/ffdshow-20051221.zip), dated 2005/12/21.

Lots of choices, take your pick.

Slave01

14th January 2006, 12:11

Sempron 3000+ xp core...only sse...not 64-bit :(. By the way thanks

breez

14th January 2006, 14:28

You can use Milan's and Drevil_xxl's builds, they don't require SSE. I'm using the latter with my Athlon XP and it is a good build (for me it is faster than any GCC compiled build).

Negi

16th January 2006, 06:40

For a "comprehensive" thread, you didn't include many versions of FFDSHOW.

What about SSE2/SSE/X64 build lists? Also, why hasn't there been a new build in a month? Also, put your updates in the first post. That way it's much more organized.

ExtraEye

16th January 2006, 13:20

but i don't think making new build has anything to do with nice Inventive Software person who's trying to help you choose the right builds.

Inventive Software

16th January 2006, 14:56

@Negi: I do the best I can. For the record this was a thread for the builders to post regular updates for their builds, not for me to do a catalogue of every build there has been, as that would take several hours, something I don't have at the moment. In future, I may, but don't hold your breath.

@Slave01: I apologise for getting the specs for your CPU wrong. I was going off the top of my head, and I got it wrong. :(

ICL9 and GCC combined builds are quite fast.

clsid

16th January 2006, 16:55

Build: 2005-12-22 (latest CVS)
Compiler: ICL9 (most files) + GCC 3.4.4 (for libavcodec.dll and mplayer.dll)
Minimum CPU requirement: MMX
Optimized for (used only if supported by CPU): MMX/MMXext/SSE/SSE2/SSE3
Download: click here (http://www.badongo.com/file.php?file=ffdshow+2005-12-22+ICL9-GCC402__2006-01-16_ffdshow-20051222.exe)

Inventive Software

19th January 2006, 13:34

@clsid: I tried that build, and it doesn't open the VFW configuration dialog. The other dialog boxes open fine, just that one.

It says: "ff_vfw.dll requires another module to load".

Thing is, I have an earlier build that works without hassles. Any suggestions?

clsid

19th January 2006, 17:12

It works here.

Dependency Walker (http://www.dependencywalker.com/) might be able to tell you what module is missing.

Inventive Software

22nd January 2006, 17:01

Thanks. I ran "Dependency Walker". It told me I needed "msvcr71.dll". I have about 5 copies of that scattered around my hard disk everywhere except "windows\system".

lithoc

23rd January 2006, 02:13

DigitalDivide

23rd January 2006, 02:25

Edited and deleted:

clsid

24th January 2006, 00:05

Build: 2006-01-23
Compiler: ICL9 (most files) + GCC 3.4.5 (for libavcodec.dll and mplayer.dll)
Minimum CPU requirement: MMX
Optimized for (used only if supported by CPU): MMX/MMXext/SSE/SSE2/SSE3
Download: click here (http://www.badongo.com/file.php?file=ffdshow+CVS+build+2006-01-23+ICL9-GCC3.4.5__2006-01-23_ffdshow-20060123.exe)

bob0r

25th January 2006, 00:07

Do anybody build ffdshow for old machines (eg Pentium-MMX, Pentium 2, older Celeron)?

I still have Pentium-MMX(overclocked) to watch movies sometimes(for sake of torturing it :devil: )
It plays perfectly on Linux + mplayer via VESA 2.0 output.

Try clsid's build, it should work.
Mirrored on http://files.x264.nl/ffdshow/ffdshow-20060123_clsid.exe

GCC version out once it can be compiled :o

LoRd_MuldeR

28th January 2006, 15:32

Try clsid's build, it should work.
Mirrored on http://files.x264.nl/ffdshow/ffdshow-20060123_clsid.exe

Some filters are broken in this build (e.g. Resize-Sharpen)

clsid

28th January 2006, 18:56

For old CPUs I recommend Milan's own builds (ICL8.1). Those are more stable than ICL9 builds. Hopefully he will put a new one online sometime soon, since his most current one dates from november 29th.

LoRd_MuldeR

28th January 2006, 19:15

Try clsid's build, it should work.
Mirrored on http://files.x264.nl/ffdshow/ffdshow-20060123_clsid.exe

GCC version out once it can be compiled :o

Can you make a new SSE build like "ffdshow-20051221-gcc4.0.2-sse-x264.nl.exe" from the current version?

cc979

6th February 2006, 23:17

Can you make a new SSE build like "ffdshow-20051221-gcc4.0.2-sse-x264.nl.exe" from the current version?

i think current cvs, is broke for gcc

Inventive Software

7th February 2006, 12:21

If it makes more sense, use a stable version of GCC for compiling. Most often, the CVS version is not worth it when compiling, as they don't give such noticable speed increases with program running time.

cc979

7th February 2006, 13:09

could someone upload the last compile-able version, as i've lost it .. thanks

Inventive Software

7th February 2006, 13:35

GCC or ffdshow?

cc979

7th February 2006, 16:45

GCC or ffdshow?

ffdshow i you can, thanks

zambelli

20th February 2006, 09:50

Does anyone know why the x264.nl hosted build of ffdshow hasn't been updated since 12-21-05? What are they waiting for?

bob0r

20th February 2006, 11:35

Does anyone know why the x264.nl hosted build of ffdshow hasn't been updated since 12-21-05? What are they waiting for?

ffdshow fails to compile with gcc 4.0.2, Milan updated ffdshow, we reported the bug, but he never went active again.

http://sourceforge.net/tracker/?group_id=53761&atid=471489
> http://sourceforge.net/tracker/index.php?func=detail&aid=1413410&group_id=53761&atid=471489

( also http://sourceforge.net/tracker/index.php?func=detail&aid=1425654&group_id=53761&atid=471489 )

Trying gcc-4.1.0-20060219 now
ftp://ftp.nluug.nl/mirror/languages/gcc/prerelease-4.1.0-20060219/
Its not a snapshot, but a true prerelease it seems.

It has compiled on mingw.

ffdshow fails:
gcc-4.0.2
http://files.x264.nl/ffdshow/ffdshow_output_20_feb_2006_gcc-4.0.2.txt
gcc-4.1.0.20060219
http://files.x264.nl/ffdshow/ffdshow_output_20_feb_2006_gcc-4.1.0.20060219.txt

cc979

20th February 2006, 17:34

ffdshow fails to compile with gcc 4.0.2, Milan updated ffdshow, we reported the bug, but he never went active again.

http://sourceforge.net/tracker/?group_id=53761&atid=471489
> http://sourceforge.net/tracker/index.php?func=detail&aid=1413410&group_id=53761&atid=471489

( also http://sourceforge.net/tracker/index.php?func=detail&aid=1425654&group_id=53761&atid=471489 )

Trying gcc-4.1.0-20060219 now
ftp://ftp.nluug.nl/mirror/languages/gcc/prerelease-4.1.0-20060219/
Its not a snapshot, but a true prerelease it seems.

It has compiled on mingw.

ffdshow fails:
gcc-4.0.2
http://files.x264.nl/ffdshow/ffdshow_output_20_feb_2006_gcc-4.0.2.txt
gcc-4.1.0.20060219
http://files.x264.nl/ffdshow/ffdshow_output_20_feb_2006_gcc-4.1.0.20060219.txt

have tried to re-compile gcc-4.1 with gcc-4.1

clsid

24th February 2006, 21:38

Build: 2006-01-27 (newest CVS)
Compiler: ICL 9.0.029 (most files) + GCC 3.4.5 (libavcodec.dll, mplayer.dll)
Minimum CPU requirement: MMX
Download: click here (http://rapidshare.de/files/14051213/ffdshow-20060127.exe.html)

cc979

2nd March 2006, 04:35

zambelli

2nd March 2006, 07:28

It'd be nice if somebody ran performance and functionality tests on these new builds as they become available. It wouldn't need to be anything elaborate, just a few basic functionality tests and some perf numbers on basic decoding and such. I feel like there are too many competing builds of ffdshow popping up and we have absolutely no bar and no way of measuring them up.

seehowyouare

2nd March 2006, 10:53

I feel like there are too many competing builds of ffdshow popping up and we have absolutely no bar and no way of measuring them up.

IDEA - If all the builds were distributed via Bit Torrent (the main audience anyway) it would
1) be "one" place to get your latest codec
2) natural attrition might just kill off "bad" builds
3) NO more searching forums and threads and ..
4) Reduces storage/bandwidth needs for builders.

Lots of disadvantages. For example, the super-beta-mega-quality-best-ever-XviD-codec.exe file will always be in circulation because of the filename, even if it is a "bad" build.

clsid

2nd March 2006, 18:11

what dx sdk do you use ?
The one included in VS.net. I also have the oct2005 one, but that should only be used in case something isn't found in the standard VS libs.

Egh

2nd March 2006, 20:35

IDEA - If all the builds were distributed via Bit Torrent (the main audience anyway)

I thought they were distributed via Bittorrent already, and several different builds in one package.

zambelli

2nd March 2006, 20:45

several different builds in one package.
I don't think that's helpful at all; in fact, I think that's confusing. Once again, I think there should be some sort of performance baseline for these builds. Just compiling Ffdshow and advertising its more recent build date shouldn't be any guarantee of a build's quality or performance value.

Egh

2nd March 2006, 21:22

I don't think that's helpful at all; in fact, I think that's confusing. Once again, I think there should be some sort of performance baseline for these builds. Just compiling Ffdshow and advertising its more recent build date shouldn't be any guarantee of a build's quality or performance value.

Of course. Trouble is that for performance issues, proper test should be done on those segments on CPU kingdom :p

Celeron and P3 (SSE1 only)
Athlon XP (SSE1 only)
P4 (SSE2)
Sempron and Athlon 64 (SSE2)
Athlons prior XP (no SSE at all).

It's safe to assume that for each group above the the ultimate proper build is different.

GCC believed to produce better results for AMD, ICL for Intel. Then it's the question of additional CPU instruction sets. Many builds won't work on casual Athlons at all. Also for AMD it should be separate builds for SSE1 and SSE2 capable processors. Don't forget there're different versions of gcc and ICL as well.

I tell all that to show that hardly anyone would volunteer to do proper testing. And w/o proper testing it all remains speculative and dependent on specific systems, as it is now.

clsid

3rd March 2006, 11:59

ICL can be patched so that it works good on AMD processors as well.

GCC compilations need at least SSE. ICL and MSVC compilations also work without SSE.

Most cpu hungry routines in ffdshow/libavcodec are already hand optimized in assembler code. So making a build that is optimized for a specific CPU instruction set won't give huge performance boosts.

cc979

3rd March 2006, 13:00

The one included in VS.net. I also have the oct2005 one, but that should only be used in case something isn't found in the standard VS libs.

cheers, i've been using dec2005 but had a look at feb2006 some interesting dx10 things in it

Inventive Software

3rd March 2006, 13:37

If people want to run tests and post results, may I suggest a new thread? This thread's primarily for ffdshow builders to report new builds. I would run some tests, but my system isn't great, and I have a lack of varied material.

ExtraEye

3rd March 2006, 16:28

how do i run tests? i have an athlon 64 3000+ if you need me to test it

zilexa

8th March 2006, 21:45

Why not just use the newest version on the ffdshow page?
Just search Google: ffdshow wiki
and you will find the page!

Getting FFDShow (http://ffdshow.sourceforge.net/tikiwiki/tiki-index.php?page=Getting+ffdshow)

Liisachan

9th March 2006, 00:56

milan's builds are not the newest, but actually I myself am using 2005-11-15 not -24 nor -29. Life would be too easy if the newest was always the best, but if you like to test the newest versions, here (http://ffdshow.faireal.net/)'s a random pack. It's torrented because bw is expensive and I'm not rich, only have 500GB/mo while need 2000.

Yong

9th March 2006, 11:09

Can someone pls tell me where to download "mingw compatible" GCC 4.xx binary? i wan to try to compile ffdshow too. ;)
i tried this one
http://oss.netfarm.it/mplayer/misc/gcc-mingw32-4.2.0.svn.20060122.tar.gz
but it seem lack some libs...

cc979

9th March 2006, 22:46

Liisachan

13th March 2006, 10:35

By the way, I'd really appreciate it if you would compile ff_tremor in high-accuracy mode, like celtic_druid did.

tremor is not good in default compile, I can easily abx at 15/15, meaning Ogg Vorbis decoded thru ffdshow/tremor is damaged in quality. (That's why I can't respect any codec packs that use ffdshow for Vorbis decoding. They don't have ears -_- )

Egh

13th March 2006, 21:23

clsid

3rd April 2006, 14:57

Liisachan

3rd April 2006, 15:44

Thanks very much clsid!!!

@Egh
libavcodec is ok, but tremor is the default for vorbis, isnt it? So you'd have to manually change the setting. Ordinary ffdshow users don't know that. I'd like ppl to enjoy the real quality of Ogg Vorbis.

LoRd_MuldeR

3rd April 2006, 20:59

videomixer9

3rd April 2006, 21:32

most important is that it has updated parts in libavcodec and updated libmpeg2. The h264 code was also updated but there seems to be no major speed improvement, and actually it seems libavcodec goes a different direction there, actually even VLC got recently more problems with 1080 content on my box, i'm still waiting for them to beat coreavcs performance.

LoRd_MuldeR

3rd April 2006, 21:49

most important is that it has updated parts in libavcodec and updated libmpeg2. The h264 code was also updated but there seems to be no major speed improvement, and actually it seems libavcodec goes a different direction there, actually even VLC got recently more problems with 1080 content on my box

Thanks for info :)

i'm still waiting for them to beat coreavcs performance.

Good luck :D

videomixer9

3rd April 2006, 22:07

Honestly I'm waiting for a powerful as vsfilter subtitle filter for *nix and some other things as a media addict so I can kiss goodbye to Windows finally, open source is mostly freeware and it's easy to maintain, usually frequently updated and easy to obtain with hazzles like registrations etc. So my windows system uses mostly FOSS too as it's just less hazzle than all this payware that install adware, trojans and intrusive copyright protection stuff.

But well sadly, i watch too much subtitled stuff to just ignore this issue of ugly subs that ppl sometimes styled for hours and of course the missing mass of games. I rarely use my pc for work , lol, and funstuff is still way more popular on windozzze and also most development for media related stuff is too.

Shapierian

4th April 2006, 02:19

SVN: 2006-04-03
Compilers used:
- MSVC71 (ffdshow.ax, etc)
- GCC 3.4.5 (libavcodec.dll, mplayer.dll, kerneldeint.dll)
Minimum CPU requirement: MMX
Download: click here (http://rapidshare.de/files/17103116/ffdshow-svn20060403.exe.html)

It may be better to identify builds by the changeset number and not the build date.

signatory

4th April 2006, 04:11

SVN: 2006-04-03
Compilers used:
- MSVC71 (ffdshow.ax, etc)
- GCC 3.4.5 (libavcodec.dll, mplayer.dll, kerneldeint.dll)
Minimum CPU requirement: MMX
Download: click here (http://rapidshare.de/files/17103116/ffdshow-svn20060403.exe.html)

Stupid me for trying something new.

http://img156.imageshack.us/img156/369/ffd0af.jpg
Yes, uninstalled the old version first.
Started this install. Ka-boom!

celtic_druid

4th April 2006, 04:53

MSVC71 means that ffdshow.ax requires msvcrt71.dll. If you don't have it and the installer doesn't include it, then the ax won't register.

Kador

4th April 2006, 08:33

Hi, I was frequently using http://m17n.cool.ne.jp/freeware/mpc/ to seek new ffdshow builds, is it dead ?

videomixer9

4th April 2006, 08:50

that's just one mirror of many others ... http://celticdruid.no-ip.com/xvid/

celtic_druid

4th April 2006, 09:34

I stopped compiling ffdshow when everyone else started. No need having lots of people compiling the same thing really.

Liisachan

4th April 2006, 10:20

@Kador
the last ffdshow compiled by celtic_druid is 2005-11-03
if you are in france, you can visit esby's mirror (http://esby.free.fr/CelticDruid/#ffdshow). It's hosting many other files by celtic_druid. The specific mirror you mentioned was canceled.

I myself once hoped that celtic_druid's "monthly ffdshow builds" would be restarted, then celtic_druid told me that his intention was to try to fill a gap, i.e. to compile 'rarewares' nobody else compiles, and now there was no gap to fill for ffdshow. I thought that made sense, too.

foxyshadis

4th April 2006, 13:52

bob0r has done the recent SSE builds, right? I don't know if recent changes warrant it, but it might be nice to have another of those.

LoRd_MuldeR

4th April 2006, 13:54

bob0r has done the recent SSE builds, right? I don't know if recent changes warrant it, but it might be nice to have another of those.
:goodpost:

Avish

4th April 2006, 18:10

SVN: 2006-04-03
Compilers used:
- MSVC71 (ffdshow.ax, etc)
- GCC 3.4.5 (libavcodec.dll, mplayer.dll, kerneldeint.dll)
Minimum CPU requirement: MMX
Download: click here (http://rapidshare.de/files/17103116/ffdshow-svn20060403.exe.html)
Guys, What are the longforms of these letters SVN, GCC, MSVC, ICL? I see these letters being used very often, so I'm curious to know what they mean?

videomixer9

4th April 2006, 18:38

noobalert ...

SVN = Subversion
GCC = Gnu Compiler Collection
MSVC = Microsoft Visual C/C++
ICL = Intel Compiler

next time use wikipedia or google duh ... details on all these things are there too

Liisachan

4th April 2006, 18:49

subversion
GNU Compiler Collection
Microsoft Visual C++
Intel Compiler (?)

LoRd_MuldeR

4th April 2006, 20:27

SVN (as well as CVS) is a tools for developing OpenSource software. It is needed if several developers want to work on the same source-codes. Each time the source-code is changed in any way, the latest version will be uploaded to SVN. Sometimes the SVN is updated several times a day. If you want to get the very latest version of a software, you will have to grap the latest source-codes form SVN and compile them yourself. The "official" software realeses are usually outdated (compared to the SVN), but they have been tested and approved before release. Usually builds from SVN are identified by the date they were made and not by an "official" version number.

GCC, ICL and MSVC are different compilers. You can use one of them to compile the source-codes and get a running program. I think only GCC is avaible for free. Sometimes different compilers produce a faster/slower program from the same source-codes. Which compiler is better for you, might also depend on the type of your CPU...

Lemonzest

4th April 2006, 20:45

god why do people still use rapid share? i can never user it :( how about a mega upload link?

videomixer9

4th April 2006, 20:55

megaupload is damn crap, slow as hell while rapidshare gives me full downspeeds, and i don't see a reason for rapidshare not working. rapidshare was awful for some time but since they upgraded their lines it's all fine.

dk75

4th April 2006, 23:01

Stupid me for trying something new.

http://img156.imageshack.us/img156/369/ffd0af.jpg
Yes, uninstalled the old version first.
Started this install. Ka-boom!
Had You rebooted before?

Lemonzest

5th April 2006, 01:35

well rapid share never works here because my ISP is a dumb ass and puts us all behind transparent web proxy's, so to rapid share we all look like the same person and there for not allowed to download, and videomixer9 no need to be so hostile its not like I'm personally asking you to add a new link for those of us who cant use rapid share, i asked for anyone.

Koti

5th April 2006, 03:10

SVN: 2006-04-03
Compilers used:
- MSVC71 (ffdshow.ax, etc)
- GCC 3.4.5 (libavcodec.dll, mplayer.dll, kerneldeint.dll)
Minimum CPU requirement: MMX
Download: click here (http://rapidshare.de/files/17103116/ffdshow-svn20060403.exe.html)
Thx for the build :)
how about a mega upload link , rapid share never works here because my ISP puts us all behind transparent web proxy's

http://www.megaupload.com/?d=AVJ7PJ8X
I "had" a isp like that

cyburn

5th April 2006, 03:56

Good day,

I am trying to find how to patch ffmpeg to enable avisynth input. I have found that ffmpeg builds at http://ffdshow.faireal.net/mirror/ffmpeg/ are properly patched. The site point to doom9 for technical questions about the ffmpeg builds... so here I am.

I tried to patch the latest cvs with 3GP_Converter051103 www.nurs.or.jp/~calcium but this did not work.

Anything special I should know?

celtic_druid

5th April 2006, 05:16

The builds are mine and yes I did use MobileHackerz patch. Don't really see what this has to do with ffdshow though.

Link to MobileHackerz source: http://www.nurs.or.jp/~calcium/3gpp/sources/ffmpeg-051130-0.33.tar.gz
Builds also require avsredirect to work.

Avish

5th April 2006, 08:31

videomixer9

5th April 2006, 09:53

Microsoft Compiler is also available for free, it's not open as GCC though and it's not really producing fast code either, however the resulting code is usually quite stable. First only the compiler was free and could be used via commandline or via make, now there is also a free edition of Visual C++ itself called Visual C++ Express Edition, you need to seperatly download the platform sdk though with this edition. Intel Compiler is only free for non-commercial use on Linux.

namchik

5th April 2006, 12:24

whose build is this one (http://www.free-codecs.com/FFDShow_download.htm) on freecodecs.net?

Avish

5th April 2006, 12:45

GCC, ICL and MSVC are different compilers. You can use one of them to compile the source-codes and get a running program. I think only GCC is avaible for free. Sometimes different compilers produce a faster/slower program from the same source-codes. Which compiler is better for you, might also depend on the type of your CPU...Another noob question...If i may... My CPU is Athlon XP 2600+, 1.91 GHz, 512 RAM, so which compiler build would be better n' faster for it? I searched before asking, but couldn't find any definitive answer.

Matrics

5th April 2006, 12:46

whose build is this one (http://www.free-codecs.com/FFDShow_download.htm) on freecodecs.net?
Quote from free-codecs.com page:
FFDShow 2006-04-03 was compiled by clsid using MSVC 7.1+GCC 3.4.5

Reino

5th April 2006, 13:41

Avish;
http://images.tomshardware.com/2005/11/21/the_mother_of_all_cpu_charts_2005/cpu_table_amd_big.gif

namchik

5th April 2006, 13:54

Quote from free-codecs.com page:
FFDShow 2006-04-03 was compiled by clsid using MSVC 7.1+GCC 3.4.5
but it's different size from this: http://rapidshare.de/files/17103116/ffdshow-svn20060403.exe.html

Liisachan

5th April 2006, 14:29

same file, free-codecs just zipped it
; 2726294 02:32.00 2006-04-04 ffdshow-svn20060403.exe
ffdshow-svn20060403.exe D6E74CC2
; 2726294 14:34.26 2006-04-03 ffdshow-svn20060403.exe
ffdshow-svn20060403.exe D6E74CC2

Lemonzest

5th April 2006, 15:06

no gcc sse2 builds? they run a little faster on this rig then the msvc builds.

Lemonzest

5th April 2006, 15:07

Thx for the build :)

http://www.megaupload.com/?d=AVJ7PJ8X
I "had" a isp like that

Thanks

Avish

5th April 2006, 15:40

Avish;
http://images.tomshardware.com/2005/11/21/the_mother_of_all_cpu_charts_2005/cpu_table_amd_big.gifHi CoRoNe, thanks for the chart, but how does it answers my question?
Edit: Never mind! I found it http://forum.doom9.org/showthread.php?p=809642#post809642

namchik

5th April 2006, 15:55

same file, free-codecs just zipped it
ah... my bad...

Matrics

5th April 2006, 16:51

no gcc sse2 builds? they run a little faster on this rig then the msvc builds.
Did you try Kuroshu build?
http://kurosu.free.fr/ffdshow-20060403-23H17-athlon.exe

LoRd_MuldeR

5th April 2006, 19:01

Another noob question...If i may... My CPU is Athlon XP 2600+, 1.91 GHz, 512 RAM, so which compiler build would be better n' faster for it? I searched before asking, but couldn't find any definitive answer.

I think GCC with optimizations for MMX, 3DNow! and SSE enabled.
SSE2 is not supported by AthlonXP...

Lemonzest

5th April 2006, 20:35

I'm using the k8 build from that site, but as the change log above the files indicates

Simple rebuild with no CVS updating.

the versions of the components like libavcodec.dll etc are not the up to date version released the other day , on the 3rd

videomixer9

5th April 2006, 20:47

gcc and ICL both failed on errors during compile here, though for the .ax file it doesn't matter much. MSVC is the only one that currently builds the ax file successfully.

btw. I tried and used -ftree-vectorize on GCC 4.1 to build libav ... feedback on any slowdowns would be nice (yes it seems it breaks the handoptimizations?)
http://rapidshare.de/files/17289764/libavcodec_athlonxp_gcc41_autovec.7z.html

Reino

5th April 2006, 22:49

Hi CoRoNe, thanks for the chart, but how does it answers my question?Well, in this chart you can look-up your CPU, to see whether your CPU supports SSE2 or not.

Egh

6th April 2006, 04:23

Did you try Kuroshu build?
http://kurosu.free.fr/ffdshow-20060403-23H17-athlon.exe

you took the URL i gave in the parallel thread but you're actually wrong ^^

The chap asked about sse2, and athlon-only build is 3DNow+SSE.

Besides, kuroshu already updated :approved:

So, gcc+sse2 build is

http://kurosu.free.fr/ffdshow-20060405-21H28H42-k8.exe

magnusr

6th April 2006, 12:42

Which version should i get for this dual core cpu: athlon 64 x2 4200+ ? The cpu supports mmx (+), 3dnow! (+), sse and sse2, see3.

And do you have a link to that version. I dont wanna compile anything myself. I just wanna install it.

What will be the fastest and best version for this cpu? sse2?

ffdshow-20060405-21H28H42-athlon.exe maybe?

And i dont wanna screw up anything purevideo does with h264 and mpeg2 (hardware support). Should i disable h264 after install then, and make sure mpeg2 is deselected during install?

Thx for any answers you might give.

Egh

6th April 2006, 15:54

Which version should i get for this dual core cpu: athlon 64 x2 4200+ ? The cpu supports mmx (+), 3dnow! (+), sse and sse2, see3.

What will be the fastest and best version for this cpu? sse2?

ffdshow-20060405-21H28H42-athlon.exe maybe?

yeah. If you want the newest changes in ffdshow, there's no better build for you (i.e. for your CPU you need gcc build with SSE2 enabled, and the link above is exactly for that).

magnusr

6th April 2006, 17:40

Should a codec also be dxva compatible? Or is that up to the player? I got 7800gtx with purevideo installed.

ExtraEye

6th April 2006, 20:29

you took the URL i gave in the parallel thread but you're actually wrong ^^

The chap asked about sse2, and athlon-only build is 3DNow+SSE.

Besides, kuroshu already updated :approved:

So, gcc+sse2 build is

http://kurosu.free.fr/ffdshow-20060405-21H28H42-k8.exe

i have problems playing mpg1 through this build with libmpeg2. i was able to play it through libavcodec. now im back to the 20060405 svcd build and it works fine

Shirokuu

7th April 2006, 06:37

ExtraEye's complaint is confirmed by my own experience. Reverting to 20060328 build solved the problem.

clsid

7th April 2006, 13:06

SVN: 2006-04-07 (rev. 2503)
Compilers used:
- MSVC71 (ffdshow.ax, etc)
- GCC 3.4.5 (libavcodec.dll, mplayer.dll, kerneldeint.dll)
Minimum CPU requirement: MMX
Download: click here (http://rapidshare.de/files/17407154/ffdshow-svn2503-20060407-unicode.exe.html)

It's an unicode build.

Liisachan

7th April 2006, 13:33

Thank you as always :D

For those who can't use rapidshare (http://www.faireal.org/tmp/ffdshow-svn2503-20060407-unicode.exe)

videomixer9

8th April 2006, 15:12

LotharZ

8th April 2006, 15:53

i have problems playing mpg1 through this build with libmpeg2. i was able to play it through libavcodec. now im back to the 20060405 svcd build and it works fine

Same problem here, I cant run mpg1 with libmpeg2 under ffdshow-20060405-21H28H42-k8
but perfectly with ffdshow-20060405-21H28H42-athlon

PC Config: AMD Sempron 2800+ with Win2003+SP1

Liisachan

8th April 2006, 16:08

Thank you very much videomixer9 :)

For those who can't use rapidshare (http://www.faireal.org/tmp/ffdshow-20060408_msvc8_gcc41_sse_videomixer9.exe)

Inventive Software

8th April 2006, 16:11

Wow, it's been ages since I last checked this thread. Been having problems with my internet at home.

If possible, can this thread be limited to just the builds and the updates for them? If you wanna post bugs, or report that it works, then fine. But maybe development-related stuff should be confined to another thread. I don't particularly want this one to end up like the previous one (see link in thread starter).

LotharZ

8th April 2006, 16:40

SVN: 2006-04-08 (rev. 2504)
Compilers used:
- MSVC8 (ffdshow.ax)
- GCC 4.1.1 (libavcodec.dll, mplayer.dll, ...)
Minimum CPU requirement: SSE
Download: click here (http://rapidshare.de/files/17494132/ffdshow-20060408_msvc8_gcc41_sse_videomixer9.exe.html)

Also unicode, not much changes, just other compilers and slighty other settings and SSE requirement.

Error during the instalation: "Error while registering ffdshow.ax"

Output folder: C:\Programs\ffdshow
Extract: ffdshow.ax... 100%
Extract: ffdshow.ax.manifest... 100%
Extract: libavcodec.dll... 100%
Extract: TomsMoComp_ff.dll... 100%
Extract: libmplayer.dll... 100%
Extract: libmpeg2_ff.dll... 100%
Extract: ff_liba52.dll... 100%
Extract: ff_wmv9.dll... 100%
Extract: ff_tremor.dll... 100%
Extract: ff_theora.dll... 100%
Extract: ff_libmad.dll... 100%
Extract: ff_libdts.dll... 100%
Extract: ff_libfaad2.dll... 100%
Extract: ff_realaac.dll... 100%
Extract: ff_samplerate.dll... 100%
Extract: ff_unrar.dll... 100%
Extract: ff_x264.dll... 100%
Extract: ff_kernelDeint.dll... 100%
Could not load: C:\Programs\ffdshow\ffdshow.ax

videomixer9

8th April 2006, 17:25

Missing runtime libs for Microsoft VC8?

MSVCR80.dll and MSVCP80.dll iirc ...
http://rapidshare.de/files/17503992/msvcr8runtime.7z.html
extract and put in windows or windows system32 dir

LotharZ

8th April 2006, 19:10

No, even with your dlls the problem isnt solved :/

videomixer9

8th April 2006, 19:58

edit: you don't need .NET Framework ... you just need vc8_redist ... a 2.5 MB download. I assumed it needed .NET as VC8 files also come with it and I tried on my fresh installed notebook XP via installing .NET Framework.

you may try this here:
vcredist_x86.exe (http://faux.warwickcompsoc.co.uk/vcredist/vcredist_x86.exe)

LotharZ

8th April 2006, 20:03

Then thats the problem :)
thx

videomixer9

9th April 2006, 16:24

SVN: rev. 2504
Minimum CPU requirement: x64 compatible CPU
OS: Win64
Download: click here (http://rapidshare.de/files/17577748/ffdshow64-20060409.exe.html) or here (http://www.megaupload.com/?d=GN7EQAR5) or get a better ISP

x64 build, beware, may not work at all or perform horrible, I don't have a x64 CPU so I couldn't test it.

SeeMoreDigital

9th April 2006, 18:04

Could you use a different "files host" please!

It's practically impossible for NTL (and probably TeleWest) broadband users in the UK to access RapidShare's free service.

Cheers

videomixer9

9th April 2006, 18:40

seriously go and get a better ISP ... can you call an ISP an ISP if they don't provide proper internet access?

Lemonzest

9th April 2006, 20:06

well NTL has just bought out Telewest in the UK, and there the only cable company's around most of the UK, and all the other ISP, are still ADSL on British Telecom lines running at about 2Mb/s tops, UK is home to monopolies, High prices and crap service no point changing ISP's as there all as bad as each other.

GmorG McRoth

9th April 2006, 23:06

It is legal to host custom bulids? If yes anyone could use pages.google.com for it, 100MB of free hosting space. You just need gmail account. uplaoding and page editing is thru browser, easy and fun.

namchik

9th April 2006, 23:51

videomixer9
i also have error registering ffdshow.ax in Windows x64

will this download help?
vcredist_x64.exe from http://www.microsoft.com/downloads/details.aspx?FamilyID=25ae0cd6-783b-4968-a841-38a2743307d9&DisplayLang=en#filelist

videomixer9

10th April 2006, 00:02

I guess so, it is MSVC8 again. So you need the runtime libs on Win64 too of course. It should complain about the missing libs when you try to register the filter by hand. Also I'm not to sure about the nsis installer and the installer script, I used /DX64 but i guess the nsis installer is still 32bit and prolly runs under wow32.

namchik

10th April 2006, 16:05

ok... but 10 megs seems too much for my dial-up :(

by the way about the broken libmpeg2 in the latest k8 build on athlon64 cpus... Disabling SSE2 in ffdshow settings solves the problem

bob0r

10th April 2006, 16:16

I will try some new tests, when:

1: svn co https://svn.sourceforge.net/svnroot/ffdshow/trunk ffdshow
might work, cus for me it does not work, any tips?

svn co https://svn.sourceforge.net/svnroot/ffdshow/trunk ffdshow
Error validating server certificate for 'https://svn.sourceforge.net:443':
- The certificate is not issued by a trusted authority. Use the
fingerprint to validate the certificate manually!
Certificate information:
- Hostname: *.sourceforge.net
- Valid: from Dec 8 13:40:07 2005 GMT until Feb 7 13:40:07 2007 GMT
- Issuer: Equifax Secure Certificate Authority, Equifax, US
- Fingerprint: 49:b8:cb:87:04:8c:49:39:45:83:dd:4c:cf:c7:54:57:b0:9e:84:5d
(R)eject, accept (t)emporarily or accept (p)ermanently? p

Then nothing happens.....

2: i will test gcc 3.4.5 4.0.3 4.1.0, bug report on all if compiling fails and ask Milan which version(s) he will support

If i make a new build it will be gcc only, so it depends on SF and Milan to make this possible.

Liisachan

10th April 2006, 17:24

I don't think this will be of any help, but just in case... Since I'd like to see/test your new builds again :)
http://ffdshow.faireal.net/tmp/svn.png
I permanently accepted that thing, thinking to myself like 'SSL certificates are too expensive. no way sourceforge could be hacked'

Egh

10th April 2006, 18:29

I will try some new tests, when:

1: svn co https://svn.sourceforge.net/svnroot/ffdshow/trunk ffdshow
might work, cus for me it does not work, any tips?

best thing is to ask Kuroshu. He uses gcc for his AthlonXP and Athlon/w/SSE2 builds.

But iirc he used his own builds of gcc for that. He used 2501 svn revision for his last (atm) builds.

From his website:

April 3
* Not sure which gcc version is used, as gcc 4.1.0 autovectorization hardly works and can lead to faulty binaries
* Using SVN branch (oh I am such a l33t dud3 n0w!)

April 6
* Files are first built with CC as defined in update.cfg, then the hardcoded 4.0.2.

videomixer9

10th April 2006, 19:22

GCC 4.1.0 prolly won't work, GCC 3.4.5 will and 4.0.3 i'm not sure about ... I only tried a 4.1.1 and that one failed, mingw gcc 3.4.4 works fine.

LoRd_MuldeR

11th April 2006, 01:52

kurosu's "experimental" gcc 4.2 build (AthlonXP) seems to work fine here.

Egh

11th April 2006, 02:17

kurosu's "experimental" gcc 4.2 build (AthlonXP) seems to work fine here.

Yeah, since 20060127 kuroshu's builds were practically best.

http://kurosu.free.fr/ffdshow-20060407-19H04-athlon.exe (sse1 only build, sauce -- svn 2504)

@Bob0r:
Now what kuroshu wrote:

April 10
* Experimental builds: gcc-4.2 is used to compile (by hand), then gcc 4.1.0 when 4.2 fails (mostly simd.h asm stuff), then 4.0.2 (baseclasses and related stuff).

Lemonzest

11th April 2006, 14:59

tryied the latest gcc k8 build from kurosu's site and libmpeg still crashes MPC when playing back MP1 files, works fine when libavcodec is decoding mp1/2 dont know about other builds.

namchik

11th April 2006, 16:36

disable SSE2 in settings and it will not crash

Did anyone try prescott builds (http://kurosu.free.fr/ffdshow-20060407-19H04-sse3.exe) on AMD 64 CPUs? Coz afaik latest processors since Venice core (if i'm not mistaken) support sse1, sse2 and sse3.
How fast are they compared to k8 buidls?

Carpo

11th April 2006, 18:02

Could you use a different "files host" please!

It's practically impossible for NTL (and probably TeleWest) broadband users in the UK to access RapidShare's free service.

Cheers

im on telewest and have never had a prob getting onto there - except when dns when tits up :)

cristox

11th April 2006, 23:42

Since a couple of months the gradual denoise is broken (at least on my system).
It is blurring, colouring and adding a kind of smear...slo-mo effekt to the picture.
The last build from the developers homepage is working.

Has anyone the same experience?

DrNein

12th April 2006, 04:33

I have an odd and likely very rare problem here I hope someone can help with. Perhaps consider it a challenge ;)

I am using mpc2kxp6490 to play everything with all internal source filters enabled and all internal transform filters disabled, with ffdshow-20060202-icc-sse2 handling decoding of most common video types and a just a few audio types.

Audio is output solely via S/PDIF to a HT receiver and sometimes headphones are plugged into it for isolated listening of movies, in which case Cyberlink's Audio Decoder (PowerDVD6) Speaker Environment is toggled within MPC from S/PDIF to Headphone in order to down-mix AC3 and DTS into PCM 2ch surround in one of various mock surround modes (Dolby, SRS, Cyberlink).

Naturally, the receiver can only decode AC3, DTS and PCM so for AAC 6ch I use ffdshow's output to AC3 feature to maintain true surround to 5.1 speakers rather than down-mixing to PCM.

Everything is working fine up to there. However, I started wondering how I could get WMA3 6ch to work as well (for WMV HD). ffdshow does not seem to offer WMA3 decoding (proprietary?) and so Microsoft's WMAudio Decoder DMO must be used. But I was able to figure out that enabling ffdshow's Uncompressed format decoder does the trick.

However, and here's the problem bit... when enabled it for some reason sets Cyberlink Audio Decoder's Speaker Environment to 2-Speaker and prevents it from being changed in MPC until Uncompressed is disabled again. From within PowerDVD it remains able to be changed since it of course does not use ffdshow so there is no conflict.

Any idea what is going on here and how to fix it or alternatively another way to get WMA3 output as AC3 without having to change any settings (same as AAC)?

-
Hmm, maybe the WMAudio Decoder DMO is panicking at S/PDIF and somehow setting the Cyberlink Audio Decoder to 2-Speaker in an attempt to deny 6ch digital and only allow 2ch PCM or up to 6ch analog? But these do not appear to be protected files (if that matters) and in any case CAD is not even being used as proven with GraphEdit.

videomixer9

12th April 2006, 07:37

just don't waste cpu cycles and more by using the AC3 out function? The soundcard drivers should bother with getting the multichannel audio properly to your receiver. The AC3 out in ffdshow decodes to PCM and then reencodes to AC3 ... my receiver gets 6ch audio fine without this from my soundcard though too, so I don't bother.

DrNein

12th April 2006, 10:04

CPU usage is negligible. Most PC audio has no trouble passing AC3/DTS/PCM via S/PDIF to a receiver or decoding AC3/DTS/AAC/WMA3/&c. and sending analog to a receiver however as said I only use digital out because it is not near the PC and I otherwise have no use for analog. I am not aware of any sound card driver or decoders available to pass a 6ch AAC or WMA3 stream via S/PDIF as with AC3 and DTS, so fddshow's output seems the only way around that. How do you manage it with yours? Or do you mean it has DTS Connect and/or Dolby Live to re-encode AAC and WMA3 streams which I guess is pretty much the same thing but done by a C-Media DSP instead of CPU?

celtic_druid

12th April 2006, 10:17

There are Dolby Live and DTS connect sound cards that do hardware AC3 and DTS encoding.

videomixer9

12th April 2006, 13:08

celtic_druid

12th April 2006, 13:44

http://www.hidiaudio.com/

bob0r

12th April 2006, 14:39

Does
svn co https://svn.sourceforge.net/svnroot/ffdshow/trunk ffdshow
work for anyone? Or is SF having some problems, if so, how do you guys get the sources?

videomixer9

12th April 2006, 15:11

works fine, check your routing to sf svn server or get a better ISP(tm)

bob0r

12th April 2006, 21:50

works fine, check your routing to sf svn server or get a better ISP(tm)

isp: amx-is best and biggest provider in the world
routing: see isp

What software do you guys use? svn.exe for mingw is what i use.

Edit:
Updated to svn, version 1.3.1 (r19032)
compiled Mar 30 2006, 02:41:28
Same result:
svn co https://svn.sourceforge.net/svnroot/ffdshow/trunk ffdshow
Error validating server certificate for 'https://svn.sourceforge.net:443':
- The certificate is not issued by a trusted authority. Use the
fingerprint to validate the certificate manually!
Certificate information:
- Hostname: *.sourceforge.net
- Valid: from Dec 8 13:40:07 2005 GMT until Feb 7 13:40:07 2007 GMT
- Issuer: Equifax Secure Certificate Authority, Equifax, US
- Fingerprint: 49:b8:cb:87:04:8c:49:39:45:83:dd:4c:cf:c7:54:57:b0:9e:84:5d
(R)eject, accept (t)emporarily or accept (p)ermanently?

I press "p", but nothing happens..... Any tips?

videomixer9

12th April 2006, 22:06

using this http://subversion.tigris.org/servlets/ProjectDocumentList?folderID=91 which also works fine with no mingw installed at all and never saw a request because of the certificate. Prolly you're missing the root certificate and it cannot secure the connection or establish it properly or your openssl libs are broken. Equifax root certificate usually comes with Windows. You prolly removed it or fucked it up somehow.

Get the Equifax root certificate and install it: http://www.geotrust.com/resources/root_certificates/index.asp

bob0r

12th April 2006, 22:09

Hmm i did checkout via CMD, the cert and checkout worked.
Tried again via mingw and now it works! Oh well.... Thnx for the tips *looks at stphn (irc)

bob0r

12th April 2006, 22:54

I tried compiling rev 2504 with 3 different gcc versions.
Unlike in the past, all version are failing at some point, gcc 4.0.3 should be working, but is not, so i reported this as bug (my version(s) will be 1 gcc version only)

If the bug(s) are fixed, ill host a new version again.

The bug report:
http://sourceforge.net/tracker/index.php?func=detail&aid=1469519&group_id=53761&atid=471489

dragongodz

13th April 2006, 02:50

What software do you guys use?
its pretty hard to go wrong with tortoisesvn aswell.
http://tortoisesvn.sourceforge.net/

Lys

13th April 2006, 07:06

These are all GCC compiles, any possibility of having an ICL instead?

videomixer9

13th April 2006, 08:55

ICL compiles most parts of this revision correct for me, except for one funny error in Tdll::loadFunction in the subtitle part. However, for me it always fails on linking the .ax file. I guess I'll try again later today, currently I'm at university. Milan seems to have planned ICL only for 64bit compiling as the makefiles have icl parameters set to win64 and amd64 target machine, so that needed fixing. Maybe something else is still wrong and that's why linking fails, didn't have a closer look yet.

DrNein

17th April 2006, 00:03

The cheaper solution usually is to use Creative's propiertary multichannel cables, I dunno how exactly it works but I used it a long time with my DTT 3500 soundsystem from creative. In the digital out of your soundcard you put an adapter for a special cable that can be connected to a special multichannel connector on the soundsystem. It's one cable and outputs anything to the soundsystem, I still wonder how this really works. After all it's the usual SPDIF output plug, no matter what kind of multichannel audio, it works through this singe cable. Question is if it is multichannel PCM or if they just send some kind of analog signal that's split up into 6 channels, after all, the decoding of things is done before. Negative thing, only Creative uses this. Also, I dunno if a second adapter in front of the receiver might magically convert it back to sth. usual receivers accept.

Yeah, that's a proprietary DIN for 6ch PCM. Too bad something like this did not become standard on HT receivers. There have been half-arsed/obscur efforts like the proprietary DenonLink and also a FireWire audio spec. If there were combiners available for 3 S/PDIF to HDMI then maybe any SoundBlaster that has the 4-pole digital jack could work with HDMI receivers now.

In any case, it seems we are finally on the verge of everthing coming together with HDMI. With the receivers available it is only a matter of waiting for the PC audio. Sapphire has announced an HDMI video card with combined audio output via SiL 1390 chip.

It will be interesting to see how this schtuff develops. Real-time encoders and analog will soon be passe and we won't have to compromise. Indeed, DSP soundcards should become cheaper -or at least digital-only models could be without the necessity of analog convertors or connectors etc. Then again, it will be one less reason to own a good soundcard if on-board software audio already has HDMI output (but that may be awhile).

DrNein

17th April 2006, 00:56

http://www.hidiaudio.com/

Yeah, as implied in post #126 I am aware of those cards but they don't do anything the software ain't doing for me now (plus, I get OpenAl and full EAX support from reliable drivers). Indeed, I suspect one of those cards would prolly necessitate toggling settings frequently as most sources are preferably streamed as-is. Right now, I just find that I have to toggle one setting in ffdshow if I want to encode WMA3 to AC3 and only so that it does not conflict with the Cyberlink decoder setting -and admittedly WMA3 will remain a seldom used novelty.

But I am still curious about this. Isn't the AC3 output supposed to work with any audio playback? I do not understand the difference between "any filter, DirectSound and WaveOut" as none seem to allow connecting to WMAudio Decoder DMO, or indeed any codec not activated from within ffdshow. Also, how is LPCM used with the Cyberlink decoder? Activating LPCM just makes noise.

Due to above limitations and as said previous I can only get WMA3 output as AC3 when the codec "Uncompressed" is activated which I understand to be akin to "Raw" on the video side (a kind of catch-all for any source being played rather than connecting to a specific decoder). So, I even tried creating a ffdshow Preset with "Uncompressed" enabled but only auto load upon a decoder match or DirectShow filter presence (using variously WMAudio Decoder DMO, WMA3, WMA*, wmadmod.dll, etc.) or 6ch match. However the preset never loaded when WMA3 material was played.

How can I either get WMA3 to be encoded as AC3 without enabling "Uncompressed" codec or otherwise only have that done by a Preset Auto Load condition which detects the WMA3 decoder?

clsid

21st April 2006, 18:10

SVN: 2006-04-20 (rev. 2523)
Compilers used:
- MSVC71 (ffdshow.ax, etc)
- GCC 3.4.5 (libavcodec.dll, mplayer.dll, kerneldeint.dll)
Minimum CPU requirement: MMX
Download: click here (http://rapidshare.de/files/18585145/ffdshow-svn2523-20060420.exe.html)

JarrettH

23rd April 2006, 05:17

When I updated to the SSE2 build my image settings would crash the player when I try to use them. Like for Live TV, DVD, DivX/XviD, etc. Crashed Media Player Classic and Got All Media (for Live TV). I installed the unoptimized 2006-04-20 build instead and that seemed to fix things.

XmSurfer

23rd April 2006, 06:14

videomixer9

25th April 2006, 17:20

ffdshow rev. 2527 (25th April 2006)

requires: a CPU, with SSE preferably
compilers: GCC 4.1.1 + MSVC 8
specials: hopefully high accuracy tremor, maybe libavcodec default vorbis decoder
links: here (http://rapidshare.de/files/18909341/ffdshow-rv2527.exe.html) and here (http://www.megaupload.com/de/?d=B69NO8JR)
dependencies: if registering fails get and install this (http://faux.warwickcompsoc.co.uk/vcredist/vcredist_x86.exe) first

videomixer9

26th April 2006, 07:32

ffdshow rev. 2529

requires: a CPU, with SSE preferably
compilers: GCC 4.1.1 + MSVC 8 (ffdshow.ax)
specials: high accuracy tremor
tested: yes, working
links: here (http://rapidshare.de/files/18954480/ffdshow-rv2529.exe.html) or here (http://www.megaupload.com/de/?d=ZVPYYON9) (identical files just other hoster)

needs MSVC8 runtime libs, otherwise fails registering!!!

videomixer9

27th April 2006, 16:05

ffdshow rev. 2532

requires: SSE
compilers: GCC 4.1.1 + MSVC 8
specials: high accuracy tremor
download: here (http://rapidshare.de/files/19057679/ffdshow-rev2532.exe.html) or here (http://www.megaupload.com/de/?d=E6EGBBNQ) or here (http://www.filepoint.de/download/2841-dl-ffdshow-rev2532_exe) (no nagging) (2.93 MB)

namchik

27th April 2006, 18:27

i guess celtic_druid resumed his builds
http://ffdshow.faireal.net/mirror/ffdshow/

btw i can't understand why ffdshow topic is splitted into 2 parts???

Shirokuu

27th April 2006, 18:49

This thread is just for posting new builds accompanied with links to where the binaries can be downloaded. The idea is to keep this thread clean from bugreports, questions and discussion. The other FFDSHOW thread is meant for the latter.

i guess celtic_druid resumed his builds
http://ffdshow.faireal.net/mirror/ffdshow/

btw i can't understand why ffdshow topic is splitted into 2 parts???

videomixer9

27th April 2006, 21:15

ffdshow rev. 2533

requires: SSE
compilers: GCC 4.1.1 + MSVC 8
specials: high accuracy tremor
download: here (http://rapidshare.de/files/19084161/ffdshow-rev2533.exe.html) or here (http://www.megaupload.com/de/?d=IZD51S29)

changes: VMR9 color controls

videomixer9

28th April 2006, 07:28

ffdshow rev. 2534

requires: SSE
compilers: GCC 4.1.1 + MSVC 8
specials: high accuracy tremor, more experimental gcc switches
download: here (http://www.filepoint.de/download/2870-dl-ffdshow-rev2534_exe) or here (http://rapidshare.de/files/19110555/ffdshow-rev2534.exe.html) or here (http://www.megaupload.com/de/?d=MKML8A61)

netwolf

28th April 2006, 08:12

I have a general question regarding SSE/SSE2 etc.:

when you say "requires SSE", does that mean that only SSE is supported (so no SSE2), or does it rather mean that SSE is a minimum requirement, but if your CPU supports more, also SSE2 etc. are used ?

If the latter is true, what's the point in providing SSE2 builds anyway (if you get the same results with SSE builds)?

Or do SSE2 builds still work better on SSE2 capable CPUs than SSE-only builds would do?

thx.

videomixer9

28th April 2006, 09:10

requires means that it has to at least have SSE, of course SSE2 intrinsics and other do also work. The only project that doesn't come with specific optimization is the ffdshow.ax that may or may not profit much from automatic optimization done by the compiler. My test show that I had no real performance gain on my SSE2 P4 with specials SSE2 builds, the only differences are there where automatic vectorization kicks in, and for libmplayer and libavcodec which are both the most used workhorses and are usually compiled with GCC this doesn't make much sense as GCC doesn't autovectorize good enough yet. Generally there are not much optimizations for SSE2 or even SSE3 in ffdshow as I see it.

libavcodec always makes use of anything available automatically, even with generic builds that only require MMX.

Take VLC as an example, VLC also uses libavcodec and is compiled usually for generic CPU and still fully uses all optimizations. mplayer is the same, there's only a slight gain cause not every passage of code is fully hand optimized and e.g. memory copy actions can be handled different.

Actually some generic builds seem to be better performing on my P4 just for the decoding part than some optimized builds, so I don't think all these versions are really neccessary.

e.g. I doubt that clsid's builds that only require MMX are worse performing than all other builds. The only thing with the SSE2 thing is that some libs will assume that SSE2 is available and automatically use the optimizations always without any checks, thus on non-SSE2 CPUs they will randomly crash at some point, usually this only happens with libdts or liba52 most often.

videomixer9

28th April 2006, 12:42

ffdshow rev. 2535

requires: SSE
compilers: GCC 4.1.1 + MSVC 8
specials: high accuracy tremor, more experimental gcc switches
download: here (http://www.filepoint.de/download/2883-dl-ffdshow-20060428-rev2535_exe) or here (http://rapidshare.de/files/19125295/ffdshow-20060428-rev2535.exe.html) or here (http://www.megaupload.com/de/?d=H9P10Z1M)

changes: vertical and horizontal image flipping

videomixer9

28th April 2006, 14:29

ffdshow rev. 2536 (unicode as always)

requires: SSE
compilers: GCC 4.1.1 + MSVC 8
specials: high accuracy tremor, even more experimental gcc switches
download: here (http://www.filepoint.de/download/2895-dl-ffdshow-20060428-rev2536_exe) or here (http://rapidshare.de/files/19132438/ffdshow-20060428-rev2536.exe.html) or here (http://www.megaupload.com/de/?d=GH3BR3ES) (2.90 MB)

changes: DeBand filter

videomixer9

29th April 2006, 23:32

ffdshow rev. 2538 MULTITHREADED VERSION

requires: SSE
compilers: GCC 4.1.1 + MSVC 8
specials: high accuracy tremor
download: here (http://bittekeinspam.googlepages.com/ffdshow-20060429-mt-rev2538.exe)

the patch is note exactly made for this revision so it's a unknown to me if it works.

netwolf

30th April 2006, 14:28

@videomixer9: thank you for the detailed explanation !
So in the futureI don't have to eagerly wait for a special SSE2 build if there are important changes in SVN and there is "only" an SSE2 version available yet :)

videomixer9

4th May 2006, 15:42

ffdshow rev. 2544

requires: SSE
compilers: GCC 4.1.1 + MSVC 8
specials: high accuracy tremor, high accuracy libmad, runtime cpu detection
download: here (http://bittekeinspam.googlepages.com/ffdshow-20060504-rev2544.exe)

Multithreading patch by Haruhiko Yamagata is now officially part of ffdshow. Also new is Aud-X support, audxlib.dll comes with installer, get sure to read the license here (http://bittekeinspam.googlepages.com/AudX_licence.pdf).

Liisachan

4th May 2006, 16:48

Multithreading patch by Haruhiko Yamagata is now officially part of ffdshow. Congrats, haruhiko! :cool:

odyn1982

4th May 2006, 16:51

Thank you videomixer9 for your great ffdshow builds!:thanks:

LoRd_MuldeR

4th May 2006, 20:58

ffdshow rev. 2544
Also new is Aud-X support, audxlib.dll comes with installer

I never used Aud-X before, but I can't see "Aud-X" on the Audio Codecs page of ffdshow. How to enable Aud-X support? Enabled by default?

videomixer9

4th May 2006, 21:29

no clue, I missed it there too but I guess it is enabled per default via mp3 decoding as it is an extension to mp3. Supposedly the dll should appear in the list in the about dialog, whatever ... nothing here, no clue what works.

LoRd_MuldeR

4th May 2006, 21:34

no clue, I missed it there too but I guess it is enabled per default via mp3 decoding as it is an extension to mp3.

Found it :D

It is available as a decoder for MP3. You can select it instead of "mp3lib" or "libmad". Don't know if it's good to use Aud-X decoder, when your MP3 audio is not Aud-X though...

videomixer9

4th May 2006, 21:56

maybe it upsamples to surround ?_?

ffdshow rev. 2545

requires: SSE
compilers: GCC 4.1.1 + MSVC 8
specials: high accuracy tremor, high accuracy libmad, runtime cpu detection
download: here (http://bittekeinspam.googlepages.com/ffdshow-20060504-rev2545.exe)

software scaler cosmetics? whatever, this also comes with audxlib.def as I noticed that there are prolly load errors without it, gotta hate this external stuff. Aud-X may still crash if it's used as regular mp3 audio decoder all the time, at least it crashed for me after a while.

Avish

5th May 2006, 09:45

videomixer9

5th May 2006, 18:50

ffdshow rev. 2546

requires: SSE
compilers: GCC 4.1.1 + MSVC 8
specials: high accuracy tremor, high accuracy libmad, runtime cpu detection, aud-x libs incl.
download: here (http://bittekeinspam.googlepages.com/ffdshow-20060505-rev2546.exe)

LPCM seems to be broken for some unclear reason, it seems to be the case in other builds too so I don't think it's an compiler error. LPCM is not hard on the CPU and doesn't need advanced decoding. The decoder is embedded into the ffdshow.ax. If you use e.g. MPC just use the internal LPCM decoder, it's uncompressed anyways thus quality is equal on any decoder too.

IgorC

7th May 2006, 04:04

Maybe it will be better to suscribe ffdshow new revision as sticky here http://forum.doom9.org/showthread.php?t=89979

and don't make a new post each time.

Just advice.

Liisachan

8th May 2006, 15:06

As many of you already knew, recently celtic_druid restarted to build ffdshow too.
http://esby.free.fr/CelticDruid/
http://ffdshow.faireal.net/

clsid

8th May 2006, 21:25

SVN: 2006-05-05 (rev. 2546)
Compilers used:
- MSVC71 (ffdshow.ax, etc)
- GCC 3.4.5 (libavcodec.dll, mplayer.dll, kerneldeint.dll)
Patches:
- high accuracy tremor
- high accuracy libmad
Minimum CPU requirement: MMX
Download: click here (http://rapidshare.de/files/19968328/ffdshow-20060505-rev2546.exe.html)

igor1st

9th May 2006, 13:03

SVN: 2006-05-05 (rev. 2546)

Great build, thank you.
But for me any ffdshow build since rev2544 crashes on seeking. And this occurs without enabling multithreaded video decoding in options.
So I proceed with rev2541 and wait until Milan will fix this.

MatMaul

9th May 2006, 13:14

Great build, thank you.
But for me any ffdshow build since rev2544 crashes on seeking. And this occurs without enabling multithreaded video decoding in options.
So I proceed with rev2541 and wait until Milan will fix this.

yeah !
I have the same problem, but a lot of person don't have this problem.
You have this problem with MPC?
Can you post your cpu type please ?

Gord

9th May 2006, 14:00

Mine did the same thing until I changed the MP3 decoder.

libmad and aud-x don't crash it... mp3lib does.

igor1st

9th May 2006, 15:18

yeah !
I have the same problem, but a lot of person don't have this problem.
You have this problem with MPC?
Can you post your cpu type please ?
Athlon XP 3200+, MPC rev609.
Seeking crashes reproduces easily by seeking with hotkeys on few steps (one step == 10s for me). Crashes with seeking by 'seek bar' are much more rare.

BTW, not ffdshow himself crashes - problem with libavcodec. And this confirms by switching the decoder (for example from lavc to xvid for xvid video).

libmad and aud-x don't crash it... mp3lib does.
I try default ffdshow configuration and my tweaked one (btw, I always use libmad). Problem exist since rev2544.

Gord

9th May 2006, 15:28

I try default ffdshow configuration and my tweaked one (btw, I always use libmad). Problem exist since rev2544.

Changing the mp3 decoder lib was what fixed it for me anyway... might be another issue though.

MatMaul

9th May 2006, 16:41

it's not mp3lib for me too : I use libmad

3dsnar

9th May 2006, 20:24

Hi,
@ Videomixer:
Is it possible to distribute the audxlib.dll
with the installer? (I think it is not available by default).
Please let me know, 3d

videomixer9

9th May 2006, 20:46

it's in the installer already, aud-x should work per default with my builds. Check About -> Version details, it should say audxlib.dll present there.

Liisachan

9th May 2006, 23:17

seek->crash ocasionaly occurs to me, even with ffdshow MP3 decoder disabled.

MacAddict

10th May 2006, 01:01

Seeking in MPC has been crashing the past few days for me as well, just randomly. I'm only playing VOB files _without_ any sort of MP3 decoding though. MPC build is 609.

3dsnar

10th May 2006, 06:43

it's in the installer already, aud-x should work per default with my builds. Check About -> Version details, it should say audxlib.dll present there.
OK. I have downloaded FFDshow (rev2546-SSE2) from www.free-codecs.com and the DLL is not in the installer. Why?
------
The mp3 Aud-X stream contains stream indicator (once in a while string 'Aud-X1.0' is present in the stream. Based on this it is possible to distinguish regular mp3 from aud-x mp3. Would it be possible to provide for the user some 'activate mp3 5.1 decoding' switch (instead of the library choice, as an alternative mp3 decoding engine). So if the switch would be ON, and after detecting the 'Aud-X1.0' string, Aud-X DLL would be automatically used. I hope you know what I mean.
Please let me know what you think.
:thanks:

videomixer9

10th May 2006, 07:06

cause that is not my build. ask celtic_druid for that.

3dsnar

10th May 2006, 07:34

OK, how about the option that I proposed?

videomixer9

10th May 2006, 07:35

ask milan for that, he's still the author of ffdshow, use sourceforge feature requests page for that or submit a patch.

http://sourceforge.net/tracker/?group_id=53761&atid=471492

kurt

10th May 2006, 07:41

I also have problems wit celtic_druid's latest Build (ffdshow-rev2546-SSE2.exe). I can't play xvid files anymore, libvacodec doesn't seem to be registered...

with rev2529 (sse2) there are no such problems...

celtic_druid

10th May 2006, 07:44

Yeah, I didn't include it.

libavcodec doesn't need registering. I'll have a look when I get home. I know at least that AVC decoding works and that uses libavcodec.

3dsnar

10th May 2006, 07:49

Yeah, I didn't include it.

libavcodec doesn't need registering. I'll have a look when I get home. I know at least that AVC decoding works and that uses libavcodec.
How about adding audxlib.dll to the installer (it is only less than 200 kB, when compressed)?

kurt

10th May 2006, 14:43

Yeah, I didn't include it.

libavcodec doesn't need registering. I'll have a look when I get home. I know at least that AVC decoding works and that uses libavcodec.
after reinstalling it again, it works now. It might be because of my strange system. Sorry for the inconvenience :p

chipzoller

10th May 2006, 20:34

I hope this is the right place to post this, if not please move it.

I wanted to test out vobsub support in ffdshow and so unregistered vsfilter.dll from my system and shifted all the vobsub decoding to ffdshow but found it doesn't work 100% of the time. BTW, I searched this thread and found no mention of this problem. Apparently, it really doesn't like it (in my MKV with vobsubs) if you seek ahead then seek back...the ahead seek will usually display subs but seeking back won't. Is there something wrong here in ffdshow, or is its support experimental? I'd like to see if I can get this working so I can limit my software packages down to just 2 in order to redistro. my encodes.

Liisachan

10th May 2006, 20:44

Generally sub support of ffdshow is experimental, at least for SSA/ASS/ASS2. Not very sure about Vobsub.

foxyshadis

10th May 2006, 21:47

Does FFDShow use code from vsfilter? If it does, it must be quite old, since many things don't render correctly with it. (It'd be nice if it was as simple as updating the code, but I suppose it probably isn't.)

videomixer9

10th May 2006, 21:51

it shouldn't use even one line of code from vsfilter

chipzoller

11th May 2006, 15:05

Vobsub support in ffdshow seems to also be experimental at best...I sure hope this area gets some attention soon.

Inventive Software

11th May 2006, 17:15

Because NSIS is quite configurable, it should be quite easy to have a page for all the free codecs, then all the patent-risky ones, both for audio and video. I'll have a look tonight, been ages since I dabbled in NSIS!

Oh, in case this gets too off-topic, can we keep it focused pleeeaase? First post in this thread has the link for the unofficial thread (now at more than 1000) for bug reports and the like. Here it is again in case you've forgotten / can't be bothered to find it (;)): ffdshow bug report / feature request thread (http://forum.doom9.org/showthread.php?t=98600). This thread is for the builders to post what they have enabled, and to keep a log of all the builds made. Videomixer9 has done an incredible job with the regularity of his builds, and his work is VERY MUCH APPRECIATED!!

3dsnar

11th May 2006, 17:28

videomixer9

12th May 2006, 16:39

ffdshow64 rev. 2546

requires: SSE3
compilers: ICL 9.1.022
specials: high accuracy tremor, high accuracy libmad, libavcodec 6ch vorbis fix
download: here (http://bittekeinspam.googlepages.com/ffdshow64-20060512-rev2546-icl91.exe)

pure ICL9.1 build incl. libavcodec and libmplayer

LotharZ

13th May 2006, 01:00

Im geting a strange error with your new compilation.

http://img125.imageshack.us/img125/6272/ffdshowerror1cl.th.jpg (http://img125.imageshack.us/my.php?image=ffdshowerror1cl.jpg)

http://img89.imageshack.us/img89/7746/aviproperties5tn.th.jpg (http://img89.imageshack.us/my.php?image=aviproperties5tn.jpg)

Edit: Problem, seems that comes from the AC3 audio stream.

videomixer9

13th May 2006, 01:07

hm I don't have an 64bit CPU and thus no win64 so I really cannot test it. Wonder what this is about, I checked the projects and there was no duplicate linking to libguide ... wonder how that happens ...

libguide is used for automatic parallelization, liba52 uses this amongst others, this error may occur only randomly I was told somewhere else so it shouldn't make any problems all the time, in fact chances to get it are pretty low and I bet most people will never see it. there is nothing to change with the way of using it, everything is done automatic and I just added libguide.lib to other dependencies.

LotharZ

13th May 2006, 09:24

Oops my bad, isnt the 64bits build, I was speaking about the "ffdshow-20060512-rev2546-icl91.exe (3.96 MB)" compiled with ICL9.1.022 + GCC 4.1.1.

And isnt a random problem, always happens with AC3 files. Its the first time that happens with your builds.

videomixer9

13th May 2006, 11:20

That's odd, I didn't see any problems and so far noone else seems to had this one, I usually test an AC3, DTS, OGG and AAC movie before releasing a build. I tried multiple demos I had and had no problem, I also got that same file you got there and it works fine. I found that tremor can be crashed this way using parallelization, need to investigate some more on it and maybe turn it off, though it's odd if it's just happens on one system. Wonder if it only happens on specific cpu, e.g. your sempron if you still got that an win2003 iirc.

However I silently replace files sometimes and I did with this one too, check if your build date of liba52 is 20:01.

LotharZ

13th May 2006, 14:05

Yeah, its May 12 2006 20:01.32

Don't worry too much, probably its only a conjuntion of ffdshow options and my system, I'll try with your next build.

videomixer9

13th May 2006, 14:24

please try this (http://bittekeinspam.googlepages.com/ffdshow-20060513-rev2546-icl91.exe) build, auto parallelization is off for liba52, libdts and tremor, not that it parallelized anything that worthy anyways. However auto-parallelization manages to fuck up the code.

SeeMoreDigital

13th May 2006, 14:33

Is the ff_x264.dll decoder the most up-to date libavcodec version?

Cheers

LotharZ

13th May 2006, 14:39

@videomixer9

Perfect, problem solved.

thx

videomixer9

13th May 2006, 14:42

ff_x264 is an encoding lib and was last updated long ago.

LoRd_MuldeR

13th May 2006, 15:21

ff_x264 is an encoding lib and was last updated long ago.

Why not using x264 from SVN for compiling that lib?

videomixer9

13th May 2006, 16:11

you cannot just use the official lib, needs changes and without knowing them it's fiddling around for something prolly very few people even use and I guess milan picks the candy for updates and not too much borked code.

clsid

13th May 2006, 16:50

SVN: 2006-05-13 (rev. 2546)
Compilers used:
- MSVC71 (ffdshow.ax, etc)
- GCC 3.4.5 (libavcodec.dll, mplayer.dll, kerneldeint.dll)
Patches:
- high accuracy tremor
- high accuracy libmad
- MT fixes
- libavcodec 6ch vorbis fix
Minimum CPU requirement: MMX
Download: click here (http://rapidshare.de/files/20360880/ffdshow-20060513-rev2546.exe.html)

I'll make a ICL9.1 build later this weekend.

Liisachan

17th May 2006, 15:10

mirrored ffdshow64-rev2546.exe (http://ffdshow.faireal.net/mirror/ffdshow/ffdshow64-rev2546.exe)

BY: celtic_druid
FOR: x64

celtic_druid

17th May 2006, 15:23

Not the default nsis installer either. Used InnoSetup since it supports x64 better.

clsid

17th May 2006, 22:33

Here is another ICL9.1 build:

ffdshow.ax (rev2546) (http://rapidshare.de/files/20717549/ffdshow_ax_rev2546_icl91.rar.html)

Kostarum Rex Persia

18th May 2006, 01:08

Please use megaupload or mytempdir servers, Rapidshare is bad.

SeeMoreDigital

18th May 2006, 08:16

Please use megaupload or mytempdir servers, Rapidshare is bad.Agreed ;)

Inventive Software

18th May 2006, 11:23

Never had any problems with rapidshare... then again I have a good ISP!! ;)

I've seen a lot of people using ICL 9 to compile things that GCC can't handle. Would I be right in assuming that ICL 9 costs mega-bucks, and that there isn't a free version, except for Linux?

videomixer9

18th May 2006, 11:58

You can get discounted versions, regularly it costs quite too much just for compiling some free software.

Things GCC cannot handle, well, Intel Compiler is probably the best compiler available, it generates very fast code and it can automatically vectorize quite good, also it can now automatically produce code that makes use of dualcore features. That's more of a reason that people use it.

GCC is actually not a that good compiler, it has many quite annoying bugs and glitches. Consistency is very bad, what worked in a previous version can break easily in another, Linux distributions often have to carefully chose a version. Many of the optimizations can easily lead to broken code.

Microsoft's Compiler is a good compromise between compatibility and speed. Microsoft was blamed often for MS specific extensions, however this is the case for GCC, ICL or other compilers too. Microsoft Compiler however misses some of those neat automatic optimizations, however the code is usually quite good considering it's very fast compile times. Besides it hardly ever produces failures just through compilation.

Intel Compiler is a clear winner on speed of compiled code usually, the drawback is that it can easily takes hours to compile some things as extended code optimizations need long analyzing runs that eat quite much CPU power and memory.

There's also another thing, different compilers like to produce different quality of floating point code. ICL seems to produce fast operating code that is still very precise, you can notice this with things like tremor, libfaad and others, especially faad likes to break and make AAC sound worse just due to compiler glitches, especially GCC, using ICL or MSVC dramatically decreases this risk. milan's default makefiles are very basic probably for this very reason.

And of course Intel let's people pay for this, not too bad, considering you can easily gain some speed without doing anything it's an good deal.

Of course there is the ideal of compiling free software with free software, but my experience so far shows that ICL/MSVC had way less problems and errors due to the compiler. I dunno how often I redid a GCC compile sometimes cause some switch totally broke random code.

Shapierian

19th May 2006, 17:52

There's also another thing, different compilers like to produce different quality of floating point code. ICL seems to produce fast operating code that is still very precise, you can notice this with things like tremor, libfaad and others, especially faad likes to break and make AAC sound worse just due to compiler glitches, especially GCC, using ICL or MSVC dramatically decreases this risk. milan's default makefiles are very basic probably for this very reason.

Tremor is integer only. See: http://en.wikipedia.org/wiki/Tremor_%28software%29

clsid

26th May 2006, 14:46

SVN: 2006-05-26 (rev. 2546)
Compilers used:
- MSVC71 (ffdshow.ax, etc)
- GCC 3.4.5 (libavcodec.dll, mplayer.dll, kerneldeint.dll)
Patches:
- high accuracy tremor
- high accuracy libmad
- latest MT patch
- libavcodec 6ch vorbis fix
- LPCM fix
- DTS fix
Minimum CPU requirement: MMX
Download: click here (http://rapidshare.de/files/21433224/ffdshow-20060526-rev2546.exe.html) (rapidshare)

videomixer9

26th May 2006, 17:32

ffdshow rev. 2546 2006-05-26

requires: SSE or better
compilers: ICL 9.1.022, GCC 4.1.1
specials: high accuracy tremor, high accuracy libmad, libavcodec 6ch vorbis fix, newest mt patch, lpcm fixes, dts fixes, some changes from ffmpeg merged, more gcc optimizations, includes aud-x libs
download: here (http://www.fileking.to/download.php?n=fCT2Fcg5IB) or here (http://rapidshare.de/files/21595503/ffdshow-20060526-rev2546.exe.html) or here (http://www.filepoint.de/download/5378-dl-ffdshow-20060526-rev2546_exe) or here (http://s6.simpleupload.de/f5c9c52b1/ffdshow-20060526-rev2546.exe.html)

If no other errors will appear my hopefully last 2546 build.

clsid

26th May 2006, 17:57

Could you post a patch/diff for the ffmpeg and libfaad2 changes that you merged? Thanks.

videomixer9

26th May 2006, 18:19

libfaad2 you can just check out from sf.net and replace the files with the one in the ffdshow directory, any random compile of libfaad2 2.1 beta would even work with ffdshow if it uses the same def file, it'd only throw an error trying to show version info. I merged ffmpeg stuff per hand as the patchfiles I got of ffmpeg SVN server didn't wanna apply, wonder if I remember which files I changed hm, nothing major, let's see if i get them right, though prolly just run patch over the whole dir will work. Though it's just some files where I removed forced asserts and some minor changes to h264 code.

http://bittekeinspam.googlepages.com/ratherfunnyffmpeg.patch

some unneccessary things as I removed some lines accidently.

SeeMoreDigital

26th May 2006, 18:21

The "Installer" says "DivX4" but the "Video Decoder Config" says "Generic MPEG4": -

http://img157.imageshack.us/img157/1317/installer0ts.png

By-the-way, is there any reason why the "Other MPEG4" FourCC's aren't bundled together with the "Generic MPEG4" FourCC's?

Cheers

Liisachan

26th May 2006, 18:41

@clsid Thanks very much!!!

@all
Please update libfaad carefully. Celtic_druid and I have good reason to belive sbr_qmf.c revision 1.29 is buggy (can't decode some HE-AAC properly) and 1.28 is a good one.

The situation is still unclear and hypothetical but the best guess so far is: menno tried to make it faster by changing 32-point DCT-IV + DST-IV to DCT-IV 2 times (simpler - more optimizable), but he forgot to change the positive/negative. Just changing k -> 31-k is not enough. Signs should have been properly modified. This explanation may or may not help. Honestly I dont know what I'm talking about; I'm just repeating what I was told by someone who knows better. What I know for sure is, the newest libfaad doesn't decode this sample (http://www.minkymomo.info/~meroko_deathnote/tmp/he_aac_problem/pocket_1.aac).

videomixer9

26th May 2006, 18:53

bleh. fixed the issue with the sbr stuff and just changed to new links.

Liisachan

27th May 2006, 00:58

libfaad thing _is_ a bug, confrimed by menno.
already updated.
sbr_qmf.c 1.30 is basically the same with 1.28
in sbr_dct.c 1.17 the DST4_32 macro restored too
So you can now just use the newest CVS, at least that's what is supposed to be.

videomixer9

30th May 2006, 22:17

ffdshow rev. 2546 2006-05-30

requires: SSE
compilers: MSVC 2005, GCC 3.4.5
specials: high accuracy tremor, high accuracy libmad, libavcodec 6ch vorbis fix, newest mt patch, lpcm fixes, dts fixes, includes aud-x libs
download: here (http://www.filepoint.de/download/5859-dl-ffdshow-20060530-rev2546_exe) or here (http://rapidshare.de/files/21875771/ffdshow-20060530-rev2546.exe.html)

traditional style with maybe less side-effects and not much really worse decoding performance, down to 2.99 MB with possible more saving when removing aud-x and msvcrt, like 2.59 MB :o

LoRd_MuldeR

31st May 2006, 13:34

ffdshow rev. 2546 2006-05-30

requires: SSE
compilers: MSVC 2005, GCC 3.4.5
specials: high accuracy tremor, high accuracy libmad, libavcodec 6ch vorbis fix, newest mt patch, lpcm fixes, dts fixes, includes aud-x libs
download: here (http://www.fileking.to/download.php?n=X54L8NdnWd) or here (http://rapidshare.de/files/21796474/ffdshow-20060530-rev2546.exe.html) or here (http://www.filepoint.de/download/5713-dl-ffdshow-20060530-rev2546_exe)

traditional style with maybe less side-effects and not much really worse decoding performance, down to 2.99 MB with possible more saving when removing aud-x and msvcrt, like 2.59 MB :o

This one fixed the "libfaad2" problem with MPC.

Inventive Software

31st May 2006, 14:06

Is there ANY possibility that when building, people remove dependancies on DLLs and libs that are NOT compatible with 98? It kills me that I wanna try a new build SO BADLY and yet, it's not compatible with 98. Please don't tell me to upgrade, cause that's not an option at the moment!

@SeeMoreDigital: I'll take a look at the NSIS script tonight and see if it can be updated.

Liisachan

31st May 2006, 16:23

This one fixed the "libfaad2" problem with MPC.
What's that problem? Is it a MPC-side problem, and is it the same pb that was in CoreAAC+HE-AAC?

Btw, I'm using videomixer9's now too, as it's now ok with P4. Thanks a lot!

SeeMoreDigital

31st May 2006, 16:44

@SeeMoreDigital: I'll take a look at the NSIS script tonight and see if it can be updated.That would be cool.... many thanks ;)

EDIT: I'm sure somebody recently released a build that said "Generic MPEG-4" within the install page!

videomixer9

31st May 2006, 17:13

I did ... and then immediatly forgot about it an used the unchanged script for the next one again as I reverted the sources :p

LoRd_MuldeR

31st May 2006, 17:13

What's that problem? Is it a MPC-side problem, and is it the same pb that was in CoreAAC+HE-AAC?

Btw, I'm using videomixer9's now too, as it's now ok with P4. Thanks a lot!

Dont know. FAAD2 was broken for some videos with AAC audio (for example Superman trailer). With RealAAC there was no problem. But the FAAD2 problem was only there in MPC! In Graph Edit or WMP it worked fine with ffdshow & faad2.

SeeMoreDigital

31st May 2006, 17:36

Well here's my GUI proposal. Sorry I could not stop myself ;) : -

http://img312.imageshack.us/img312/3339/ffdshowgui1ia.png

foxyshadis

31st May 2006, 18:14

MSMPEG4v3 should cascade to the other two, which are hardly seen anyway, no reason to have all three listed. Maybe even all cascaded from Divx3 since they're the same codec. And no one would have a clue what Generic MPEG-4 Part 10 meant :p "AVC, H.264" would work.

If you're going to change it around, you might as well add Flash Video (FLV), maybe ZMBV, and "Legacy Codecs" (Sorensen, Cinepak, Indeo3, old MS stuff).

SeeMoreDigital

31st May 2006, 18:34

MSMPEG4v3 should cascade to the other two, which are hardly seen anyway, no reason to have all three listed. Maybe even all cascaded from Divx3 since they're the same codec.

And no one would have a clue what Generic MPEG-4 Part 10 meant :p "AVC, H.264" would work. Well they "bloody well should do" ;)

Here's another effort: -

http://img200.imageshack.us/img200/3885/ffdshowgui7vf.png

videomixer9

31st May 2006, 19:52

Hm, okay I did some small changes to the installer and added a funny WMV3 box too as way too many people seem to be interested in that "hack". Compressing one option into three isn't that easy with the current installerscript. Needs multi-4cc assigning to things and registry writing stuff and so on ... considering the importance of this screen rather useless. Also someone might want to enable only MPEG3v2 by MS, oddly some open movie people love this 4cc even though they obviously don't use MS MPEG4. Btw. this isn't the discussion thread :p

Avish

31st May 2006, 20:43

Dont know. FAAD2 was broken for some videos with AAC audio (for example Superman trailer). With RealAAC there was no problem. But the FAAD2 problem was only there in MPC! In Graph Edit or WMP it worked fine with ffdshow & faad2.Yes. It's fixed allright. What did u fixed reagrding this problem VM9? Can u shed some light? Can u tell us what was the problem?

videomixer9

31st May 2006, 20:54

It's just the old sources. Was there actually anything that didn't properly decode with those for anyone? besides new source didn't work with GCC somehow.

Negi

31st May 2006, 21:28

Perhaps someone could write some sort of script that adds new builds to an RSS feed. That would be awesome.

LoRd_MuldeR

31st May 2006, 22:16

@VM9: Why gone back to old GCC 3.4.5 in latest builds?

videomixer9

31st May 2006, 22:37

Actually I was told using 2.95 might even be a better choice. Also one mplayer dev told me gcc is shit. Also GCC 4.x had this weird random problems, so for a stable build I though I'd choose 3.4.5. Maybe I'll try 2.95 next time if it works :P For the hardcore optimization I might go GCC 4 again, but I dunno if I'm upto more fishes appearing in the video for several people and all this other stuff :p Also GCC is kinda contraproductive with optimization, the more you use the crappier the code, even the default -O3 is shitz ...

I encourage everyone to use an whatever built, no need to upgrade to the one with the newest date if your current one is fine, it's identical revisions after all. Also we got lots of people now doing builds in tons of flavours! :p

The newest ones this time might be a bit slower than the other ones as it's a stability build. Not much fancy.

sillKotscha

31st May 2006, 22:56

This one fixed the "libfaad2" problem with MPC.

yeahh, exactly!!!

now my problem described here (http://forum.doom9.org/showpost.php?p=833842&postcount=1431) is gone :)

thank you

LoRd_MuldeR

31st May 2006, 23:50

thuan

1st June 2006, 05:28

Because there's crash with the sample in this post (http://forum.doom9.org/showthread.php?p=833602#post833602), msvc ffdshow.ax and gcc one works fine (but gcc has stupid bug in some filter like gradual denoise).

BTW vm9, according to my testing and available ffdshow compiles then libavcodec compile with gcc 3.4.x is the fastest (don't know about 2.95 no one compiled with it) about libmplayer then your gcc4.1.1 compile is the fastest, libmad fastest with gcc 3.4.5 (this one I just look at Task manager so not so sure). The others haven't tested yet.

Inventive Software

1st June 2006, 13:28

Does nobody care about supporting Windows 98?????? (Except me.:()

GCC 2.95.1 would be the most stable 2.9 series of GCC, though for the 3 series, I'd recommend the latest in the 3.3 series, as 3.4 seems to induce a lot of unwanted hassle!

videomixer9

1st June 2006, 15:14

Why support Win98? I totally see not a single reason to support a part of the DOS based Windows Versions, use Windows NT, that one was released 1996 and is still way better than 98, or use Win2k which runs on older PCs fine too, it just requires a bit more RAM to install. Some NT Servers even made it to over 7 years uptime.

foxyshadis

1st June 2006, 21:35

Inventive, your best chance of getting a win9x install is getting mingw and making one. (Pretty sure you can do a pure gcc build, right?) Otherwise an old copy of VC6 or VC7.1 from someone. No one who builds it now can even test it on 9x before release, knows where the pitfalls are, or which libs may be non-9x compatible, or wants to deal with the frustration of patching working code to resucicate a dead OS with 5% marketshare and so many driver problems. At the least, it would probably mean installing old platform and directx SDKs, an old compiler, and excising any nt-specific code.

celtic_druid

1st June 2006, 22:25

Biggest problem would be unicode. At least I always did unicode builds. I would imagine that others do to.

clsid

1st June 2006, 22:58

My build will work on Win9x. It will need msvcr71.dll though.

videomixer9

1st June 2006, 23:03

Newer MSVC though still has a special switch that enables Win98 compatbility. But seriously, Win9x was so bad ... though i hopped onto NT 1996 already ... so no clue. There should be also an unicode addon for win98 ... *shudder*

Inventive Software

2nd June 2006, 20:10

@foxyshadis: I tried, but there are elements of the .ax filter that aren't compilable with GCC. I do have MinGW, haven't used it for ages cause I restored my Ghosted partition a few months ago, but the last time I tried to compile GCC 3.4.5, it choked. Compiling the rest of ffdshow wasn't a problem, getting the .ax to compile was (and that's the critical part!). I don't have access to MSVC6 or 7. I'd use the VC Toolkit, which is free, but that's only for 2000/XP.

@clsid: It's your build I'm using now, cause I can't get VM9's builds to work in 98. Time to go looking for your latest I think...

videomixer9

2nd June 2006, 20:12

Try installing this http://www.microsoft.com/downloads/details.aspx?FamilyId=73BA7BD7-ED06-4F0D-80A4-2A7EEAEE17E2&displaylang=en

Inventive Software

2nd June 2006, 20:26

Wow, thanks! I'll try that later, and see if that helps...

I pinpointed the problem to do with 2 DLLs that aren't included in the standard 98 SE install, userenv.dll and apphelp.dll. If they're not used, I wonder if dependancies on them can be removed...

clsid

2nd June 2006, 21:43

Try installing this http://www.microsoft.com/downloads/details.aspx?FamilyId=73BA7BD7-ED06-4F0D-80A4-2A7EEAEE17E2&displaylang=en
ffdshow needs to be linked against unicows.lib, otherwise it won't use unicows.dll on 9x. However, last time I tried that, ffdshow would just crash.

videomixer9

2nd June 2006, 21:48

Well then nothing that bothers me more than that, no sense fixing up ffdshow for a lonesome being keeping win98 and I sure won't turn off unicode just because of that or do other stunts ... and if your build works it's fine, plenty of builders plenty of choices!

Inventive Software

2nd June 2006, 23:02

Meh, fair cop! Just gotta wait for XP I suppose...

issa

3rd June 2006, 12:45

Meh, fair cop! Just gotta wait for XP I suppose...

Here is the sse gcc build without unicode, [MegaUpload] (http://www.megaupload.com/?d=BTXJIP8J)

videomixer9

3rd June 2006, 17:57

ffdshow always the latest

requires: SSE
compilers: may vary
patches: various
download: here (http://ffdshow.drition.net/current.php)

automatically always redirects to a random mirror with my latest ffdshow build. :D

Inventive Software

3rd June 2006, 21:55

I don't know what worked, but clsid's build seemed to do the business, but vm9's didn't.

issa

4th June 2006, 06:11

I don't know what worked, but clsid's build seemed to do the business, but vm9's didn't.

Is my build work on your system?

JarrettH

4th June 2006, 07:35

Can we close this thread. It makes no sense and I have to keep checking both for development. The first post in this thread points to the other thread. PLEASE PLEASE PLEASE KILL THIS THREAD!!:scared:

Inventive Software

4th June 2006, 21:59

@issa: No, it didn't. Sorry. :(
@JarrettH: This thread is for builders. The other thread is for developers. I wanted to keep the two separate, and it seems to be working. Live with it for a while. ;)

SeeMoreDigital

4th June 2006, 22:35

@JarrettH: This thread is for builders. The other thread is for developers. I wanted to keep the two separate, and it seems to be working. Live with it for a while. ;)Indeed....

That said it might make more sense to newbies if the thread title said something like: "FFDshow Beta Test/Development Builds: Where to get them and who does what!"

Cheers

Episode

4th June 2006, 23:02

ffdshow should get it's own forum section ;)

SeeMoreDigital

5th June 2006, 10:06

ffdshow should get it's own forum section ;)A "Decoder Setting-up and Help" section might not be a bad idea ;)

Inventive Software

5th June 2006, 10:12

Indeed....

That said it might make more sense to newbies if the thread title said something like: "FFDshow Beta Test/Development Builds: Where to get them and who does what!"

Cheers
Bearing that one in mind, I may change the thread title, but it would be more appropriate to change the other thread's title.

ffdshow should get it's own forum section

A "Decoder Setting-up and Help" section might not be a bad idea

Stay tuned. By the end of this week, I hope to have a simple guide to setting up ffdshow PROPERLY. This'll be in it's own thread, and ideally, stickied so newbies know where to turn.

Liisachan

5th June 2006, 23:48

After I updated AVG Antivirus' Virus definitions, it started bitching Downloader.Zlob.ANX was found in ffdshow-rev2546-SSE2.exe (http://ffdshow.faireal.net/mirror/ffdshow/ffdshow-rev2546-SSE2.exe) .

I already msg'ed to celtic_druid, and am 99.9% sure this is just a false positive (i.e. AVG Free is sillily bitching when there's nothing wrong there), but I'd like to re-confirm what I'm assuming.
Does anyone has any better, professional virii scanner to test this thing? Or any comment? TYIA

I already did http://virusscan.jotti.org/ and AVG is the only one who bitches.

http://ffdshow.faireal.net/tmp/avg.png

wyrd

6th June 2006, 02:16

@liisachan.
"am 99.9% sure this is just a false positive" me too.
http://tirnanog.fate.jp/tmp/scan/scan.png

JarrettH

6th June 2006, 07:59

That's very cool...VirusTotal:eek:

issa

7th June 2006, 02:26

@issa: No, it didn't. Sorry. :(
@JarrettH: This thread is for builders. The other thread is for developers. I wanted to keep the two separate, and it seems to be working. Live with it for a while. ;)

Could not install or the ffdshow.ax crash?

Inventive Software

7th June 2006, 12:36

When I try to configure the decoders the dialog box doesn't appear. Happens in the Start Menu and in the "configure" tab in VirtualDub. It installed fine!

videomixer9

7th June 2006, 12:47

I dunno if Win9x uses manifests but maybe deleting the external manifest may help, it defines the use of Common Controls 6.0 which may be not available on Win9x or need to be updated seperately. Though if e.g. milan version works this would be odd. But if it doesn't crash this might cause that the dialogs cannot be created. Though if I think about it you said clsid worked ... hm.

Inventive Software

7th June 2006, 14:34

Forgive my n00bness, but what does the manifest actually do?

Liisachan

7th June 2006, 14:37

AVG, that annoying antivirus that made a false accusation against celtic_druid's ffdshow binary, now updated its virus definitions, and it's now a good boy again.

This was nuts!
Virus base: 268.8.2/356
Release date: 2006-06-05 14:30:00

This is ok :)
Virus base: 268.8.2/357
Release date: 2006-06-06 12:20:00

therealjoeblow

21st June 2006, 22:07

ffdshow should get it's own forum section ;)

I agree, it's difficult to determine where to post comments/requests because the info is scattered all over.

Case in point, I have a request for the Keys & remote activation keys, and posted it at the following location. I'm just posting a pointer here in case that was not the correct location:

http://forum.doom9.org/showthread.php?p=843482#post843482

videomixer9

22nd June 2006, 22:19

I was bored a bit (in fact I needed one for my old Thunderbird Athlon :P) and compiled a generic working ICL 9.1 build, found on the website (http://ffdshow.da.cx) as usual. Only requires MMX as basic.

Other than that, some people may have noticed I did some random builds, those usually only vary in compilers and don't have any real changes, there is no real gain always getting the latest one if I put on up, if your current build is working fine, please just stick to it if it's the same revision anyways and you see no new patches listed, one build e.g. has broken HuffYUV and other codecs as I tried for fun to use GCCs tree vectorizer again which is much more stable by now, as it seems it wasn't stable enough to not break at least some things, unlike before it didn't break everything though so it seemed fine to me during regular tests.

Latest build that should be completly stable should be 20060604. I know there's some people that like fiddling with new versions but don't stick with builds if they are broken for you, no extra features you can expect, only varying speeds :P If you wait for new toys and features get new versions once there's new code added by milan. That's currently not the case, new builds are just there to keep people interested in ffdshow and have sth. they can play with or randomly benchmark :P

Liisachan

23rd June 2006, 06:39

Hm your "applied patches" list says nothing about lavc Vorbis multi-chan fix?

videomixer9

23rd June 2006, 07:09

it is applied though ... just forgot about it after I had it listed together with other ffmpeg stuff, though I doubt anyone would regularly even notice if it were missing :P

clsid

5th July 2006, 18:38

SVN revision 2543
Compilers used:
- MSVC71 (ffdshow.ax, etc)
- GCC 3.4.5 (libavcodec.dll, mplayer.dll, kerneldeint.dll)
Patches:
- accuracy.diff (high accuracy tremor and libmad)
- ogg-5.1ch.diff (libavcodec 6ch vorbis fix)
- TsampleFormat.patch
- inttypes.diff
Minimum CPU requirement: MMX
Download: click here (http://www.mytempdir.com/785868)

This build is based on the latest revision before the MT patches by haruhiko.

Can someone verify that DTS works correctly? I haven't applied the DTS patch since DTS decoding works ok for me. But I only tried one sample file. So some feedback is appreciated.

NULUSIOS

5th July 2006, 18:44

2543?

what about 2546?

I am confused...

Liisachan

5th July 2006, 19:51

2546 is the newest revision as of writing this.

Yamagata Haruhiko's multithreading (MT) patch was officially applied in Rev. 2544. So 2543 is the last revision that does not include the above patch. Simply put, if you suspect "tricky" multithreading makes ffdshow less stable, you might want to try 2543 instead of 2546.

NULUSIOS

5th July 2006, 22:15

ah ok

asasadad_1

6th July 2006, 07:19

clsid

6th July 2006, 13:05

Liisachan

8th July 2006, 06:34

kurosu's new builds:
http://kurosu.free.fr/ffdshow.htm

Simple rebuild based on source from 2006-07-06 20:49
No SVN/CVS updating, hence no log.

* Build Athlon XP (has 3DNow, SSE1 but no SSE2), 20060706-20H07 (3207 KB)
* Build Pentium4 (has SSE1 and SSE2), 20060706-20H07 (3078 KB)
* Build Pentium 3 (only has SSE1), 20060706-20H07 (3337 KB)
* Build Prescott (has from SSE1 up to SSE3), 20060706-20H07 (3088 KB)
* Build Athlon 64 (has 3DNow, SSE1 and SSE2), 20060706-20H07 (3219 KB)

Jubijub

9th July 2006, 13:16

SVN revision: 2543
Build date: 2006-07-06
Compilers used:
- MSVC71 (ffdshow.ax, etc)
- GCC 3.4.5 (libavcodec.dll, mplayer.dll, kerneldeint.dll)
Patches:
- accuracy.diff (high accuracy tremor and libmad)
- ogg-5.1ch.diff (libavcodec 6ch vorbis fix)
- TsampleFormat.patch
- inttypes.diff
- dts.patch
Minimum CPU requirement: MMX
Download: click here (http://www.mytempdir.com/787331)

This build is based on the latest revision before the MT patches by haruhiko.

Absolutely wonderful : this is the first ffdshow I test where I don't have random hang when I scroll the video timeline, I can also manage to play h264 matroska without problems...

I just tried before a build I believe was build against SVN epository and I still had the above mentionned bugs...

Adopted :)

midiboy

12th July 2006, 10:00

Hi everyone !

Does anyone have an idea why AC3 passthrough (decoding AAC and encoding to AC3, then passing this via SPDIF) does not work anymore with some of the latest 6 channel AAC encoded Apple quicktime 7 trailers for instance this (http://images.apple.com/movies/lionsgate/theusvsjohnlennon/theusvsjohnlennon_h720p.mov) one or this (http://movies.apple.com/movies/wb/ladyinthewater/ladyinthewater-tlr3_h720p.mov) one ?

It still works fine with older clips like this (http://images.apple.com/movies/lionsgate/the_descent/the_descent_h720p.mov) one.

I tried all the latest ffdshow builds. The only way I can decode this properly is via the 3ivx Audio decoder, however encoding and passing via SPDIF then does not work ...

Thanks for your help !
Alex

SeeMoreDigital

12th July 2006, 16:44

Most probably because Apple elected to associate extra data for those particular 6Ch AAC-LC streams within the MOV container.

The MOV container is quite different to the MP4 container - despite what some people might like to think.

videomixer9

12th July 2006, 20:15

Well, the clips themselves decode fine for me, the AC3 encoding produces some stuttering more high pitched sounds though.

Taurus

12th July 2006, 21:13

SVN revision: 2543
Build date: 2006-07-06
Compilers used:
- MSVC71 (ffdshow.ax, etc)
- GCC 3.4.5 (libavcodec.dll, mplayer.dll, kerneldeint.dll)
Patches:
- accuracy.diff (high accuracy tremor and libmad)
- ogg-5.1ch.diff (libavcodec 6ch vorbis fix)
- TsampleFormat.patch
- inttypes.diff
- dts.patch
Minimum CPU requirement: MMX
Download: click here (http://www.mytempdir.com/787331)

This build is based on the latest revision before the MT patches by haruhiko.

My virus scanner says your build is the "TR/Drop.Zlob.LH.3" trojan.
Can you please investigate this?

Cheers

Taurus

clsid

12th July 2006, 21:21

Conclusion: crappy virus scanner

Taurus

12th July 2006, 21:28

Conclusion: crappy virus scanner
I will investigate on my own expense ;)

sillKotscha

12th July 2006, 21:33

Hi everyone !

Does anyone have an idea why AC3 passthrough (decoding AAC and encoding to AC3, then passing this via SPDIF) does not work anymore with some of the latest 6 channel AAC encoded Apple quicktime 7 trailers

Personally I gave up trying to "fix" it - I've tried a lot of settings I can think of - but the only version that does really work for me is still milan's latest release... and I stick with it.

remember one thing, for other members/ users realtime aac-> ac3 transcoding still works with videomixer's builds, so in our case maybe it is a hardware/ technical issue.

I guess videomixer didn't touch the spdif part, once it worked why should he change it... so, maybe it is like SMD said and apple changed the way they store 6ch aac in mov - who knows?

sill

videomixer9

12th July 2006, 21:45

My virus scanner says your build is the "TR/Drop.Zlob.LH.3" trojan.
Can you please investigate this?

Cheers

Taurus

Finally stop with this f*cking Zlob already, it's totally annoying seeing lame companies blacklist the NSIS installers. If this goes on with stupid false claims of viruses in open source products I'll seriously think about sueing those companies for badmouthing and ask them to pay billions of compensation to the authors of all projects that are affected. If that'd happen maybe milan and others wouldn't have to disappear for months randomly :P

Taurus

12th July 2006, 21:49

Conclusion: crappy virus scanner
So OK!
You're right.
The binary is clean.
Sorry

Cheers

Taurus

videomixer9

12th July 2006, 22:05

Always happy to see people trust clueless companies that much, I bet it was that crap called AntiVir, maybe you should scan that for some unwanted behaviour first, it phones home more than ffdshow or most trojans do. First give out free copies everywhere and then start using it to make open source look like it's infected. Great plot.

breez

12th July 2006, 23:11

First give out free copies everywhere and then start using it to make open source look like it's infected. Great plot.

Indeed :rolleyes:

celtic_druid

13th July 2006, 03:20

Must be quite a few closed source commercial apps using NSIS though yeah? At least I think I can recall installing non OSS that used NSIS.

videomixer9

13th July 2006, 07:14

Well NSIS is open source, I bet the company producing InstallShield bought some virus detections to come out with a new slogan "don't get your products identified as virii" before people have a chance to install them ;P

Inventive Software

13th July 2006, 12:17

@celtic_druid: DivX for one.

videomixer9

13th July 2006, 12:46

Taurus

13th July 2006, 14:50

I sure hope anyone got that this is a joke on the paranoia and wild assumptions people create when they got too much time on their hands, so with people that honestly think they have to investigate shit like viruses in the builds available here.
I don't like your words.
Better get back on topic.
Wanted to be helpful and show you guys something from the outside world....
Using ffdshow right from the start years ago, but for the first time ever I got a virus alarm on it (even so it was a false alarm).
So I thought that it might be interesting to see for the devs, compilers and users.
I duck my head now and dissapear in the vanish.

Cheers

Taurua

clsid

13th July 2006, 16:24

VirusTotal (http://www.virustotal.com) is a good site to get a second-opinion on virus alerts.

Taurus

13th July 2006, 19:44

VirusTotal (http://www.virustotal.com) is a good site to get a second-opinion on virus alerts.
That's what I did @ second thought:D
So, c'mon let's get back to business.

Thanks clsid.

Taurus

Liisachan

14th July 2006, 00:03

Taurus, I agree with you for some parts but you could just use "Search this Thread" with the key word Virus. This false positive thing had been discussed more than once already before your post.

Also, the same was repeated here
http://forum.doom9.org/showthread.php?p=847605#post847605

Taurus

14th July 2006, 08:36

clsid

22nd July 2006, 15:17

SVN revision: 2543
Build date: 2006-07-22
Compilers used:
- MSVC71 (ffdshow.ax, etc)
- GCC 3.4.5 (libavcodec.dll, kerneldeint.dll, TomsMoComp_ff.dll)
- GCC 4.0.3 (mplayer.dll)
Patches:
- accuracy.diff (high accuracy tremor and libmad)
- ogg-5.1ch.diff (libavcodec 6ch vorbis fix)
- TsampleFormat.patch
- inttypes.diff
- dts.patch
- custom installer
Minimum CPU requirement: MMX
Download: rapidshare (http://rapidshare.de/files/26625385/ffdshow_rev2543_20060722.exe.html) mytempdir (http://www.mytempdir.com/821145)

SVN revision: 2546
Build date: 2006-07-22
Compilers used:
- MSVC71 (ffdshow.ax, etc)
- GCC 3.4.5 (libavcodec.dll, kerneldeint.dll, TomsMoComp_ff.dll)
- GCC 4.0.3 (mplayer.dll)
Patches:
- accuracy.diff (high accuracy tremor and libmad)
- ogg-5.1ch.diff (libavcodec 6ch vorbis fix)
- TsampleFormat.patch
- inttypes.diff
- dts.patch
- patches by haruhiko up to and including 2006-07-21
- custom installer
Minimum CPU requirement: MMX
Download: rapidshare (http://rapidshare.de/files/26625534/ffdshow_rev2546_20060722.exe.html) mytempdir (http://www.mytempdir.com/821151)

Edit: mirrors added for SMD ;)

SeeMoreDigital

22nd July 2006, 16:28

RapidShare file hosting strikes again eh!

Tzim

23rd July 2006, 09:26

Well, the clips themselves decode fine for me, the AC3 encoding produces some stuttering more high pitched sounds though.

Late reaction... I just came to see what's news in the ffdshow world since the last official build.

For the AC3 encoding problem, it seem's that the "Connect To", with "Apply only to SPDIF output" doesn't apply (anymore ?) to AC3 encoded output.

Some sound hardware or drivers (like my XPlosion Soundcard) doesnt want SPDIF from DirectSound Output so you have to use WaveOut. If I force WaveOut in Media Player Classic, it works.

midiboy

26th July 2006, 07:59

Well, the clips themselves decode fine for me, the AC3 encoding produces some stuttering more high pitched sounds though.

Hi videomixer and others,

as for the "AC3 encoding and passing via SPDIF" problem with latest quicktime HD trailers ..

they do not work for me at all. I tried your latest ffdshow versions, I tried Milans latest version and the lastest by CLSID, I tried it on 3 different audio cards (M-audio audiophile, Terratec Firewire X24, RME HDSP9632). The receiver does not switch to AC3 mode, except for maybe a few milliseconds every few minutes. With older trailers no problem ...

I also tried with Waveout or DS but only Waveout works generally for me.

Is this something only Milan can fix or can you guys help ? I could write a bug report on the CVS site if you think it might help ...

Thanks,
Alex

Antaeogo

27th July 2006, 08:06

SVN revision: 2546
Build date: 2006-07-22
Compilers used:
- MSVC71 (ffdshow.ax, etc)
- GCC 3.4.5 (libavcodec.dll, kerneldeint.dll, TomsMoComp_ff.dll)
- GCC 4.0.3 (mplayer.dll)
Patches:
- accuracy.diff (high accuracy tremor and libmad)
- ogg-5.1ch.diff (libavcodec 6ch vorbis fix)
- TsampleFormat.patch
- inttypes.diff
- dts.patch
- patches by haruhiko up to and including 2006-07-21
- custom installer
Minimum CPU requirement: MMX
Download: rapidshare (http://rapidshare.de/files/26625534/ffdshow_rev2546_20060722.exe.html) mytempdir (http://www.mytempdir.com/821151)

Just a very tiny issue, with the custom installer, the 'VFW interface' option can Not be unchecked during installation.

-Out

spinstate

28th July 2006, 12:26

check this

http://www.free-codecs.com/download/FFDShow.htm

Are those the builds by drevil_xxl & clsid?

Inventive Software

28th July 2006, 12:52

Well, the names clsid and drevil_xxl sort of give it away! ;)

clsid

29th July 2006, 13:19

ffdshow patches: download (http://rapidshare.de/files/27434584/ffdshow_patches.rar.html)
custom InnoSetup install script (with CPU detection): download (http://rapidshare.de/files/27538681/ffdshow_innosetup_script.rar.html)

SVN revision: 2543
Build date: 2006-07-30
Compilers used:
- MSVC71 (ffdshow.ax, etc)
- GCC 3.4.5 (libavcodec.dll, kerneldeint.dll, TomsMoComp_ff.dll)
- GCC 4.0.3 (mplayer.dll)
Patches:
- accuracy.diff (high accuracy tremor and libmad)
- ogg-5.1ch.diff (libavcodec 6ch vorbis fix)
- TsampleFormat.patch
- inttypes.diff
- dts.patch
- custom installer
Minimum CPU requirement: MMX
Download: RapidShare (http://rapidshare.de/files/27540717/ffdshow_rev2543_20060730.exe.html)

SVN revision: 2546
Build date: 2006-07-30
Compilers used:
- MSVC71 (ffdshow.ax, etc)
- GCC 3.4.5 (libavcodec.dll, kerneldeint.dll, TomsMoComp_ff.dll)
- GCC 4.0.3 (mplayer.dll)
Patches:
- accuracy.diff (high accuracy tremor and libmad)
- ogg-5.1ch.diff (libavcodec 6ch vorbis fix)
- TsampleFormat.patch
- inttypes.diff
- dts.patch
- patches by haruhiko up to and including 2006-07-30
- custom installer
Minimum CPU requirement: MMX
Download: RapidShare (http://rapidshare.de/files/27540840/ffdshow_rev2546_20060730.exe.html)

NULUSIOS

29th July 2006, 16:05

can you please remind (again) the difference between the two builds? (why we care)

foxyshadis

30th July 2006, 00:30

One does resize multithreading and queues to ensure hard-to-decode scenes cause less stutter (as well as a few other updates by haruhiko), one does no multithreading at all.

iron2000

30th July 2006, 02:07

can the files be on some other hosting?
Rapidshare don't work for me.

foxyshadis

30th July 2006, 02:26

conando

30th July 2006, 09:28

hi! i experienced the drivers32 deletion myself yesterday :-/ thankfully had a registry backup... another question i have: i use (and love) automatic preset loading with ffdshow quite a bit.. however i tried a myriad of builds now because i've got the problem that ff_vfw (the app i use only has vfw interface *sigh*) doesn't care about it's different presets at all - it always uses the one named "default".. regardles of the autoload conditions and even which one you select as default :-( .. is this simply not implement or a bug? could it be done?

clsid

30th July 2006, 12:24

clsid, there's a report that your uninstaller deleted the drivers32 key (where the acm/vfw codecs are) in general. Maybe you could try backing it up and testing. Probably the same issue that koepi's builds had for a while.
I really doubt that. My initial script didn't even contain a reference to the drivers32 key. I forgot to add the keys needed for ff_vfw.dll. I have added them now and I am uploading new builds. Tested them and they work ok.

conando

30th July 2006, 12:52

hi again... so i guess i found the point in the source where always the default profile is loaded for ff_vfw:

decVFW=new TffdshowDecVideoVFW(&hr,this);
decVFW->initPresets();
decVFW->setActivePreset(FFPRESET_DEFAULT,0);
return !!decVFW;

in Tffvfw.cpp ... would it perhaps work if i'd just add the correct preset autoload code there? can i simply access globalSettings from there? sorry i know this is not the best place for that kind of discussion i guess :-) :-(

foxyshadis

30th July 2006, 15:09

Hmm, he said it was the latest build on free-codecs caused it, but now that I think about it he also mentioned crashing - that might mean sse2 on a non-sse2 cpu, which means it wouldn't be your build. Oh well.

conando, the other ffdshow thread is usually used for development. :p See TffdshowDec::initPreset in TffdshowDec.cpp for how the DS side does it, it looks like you need to call presets->getAutoPreset, but I'm not sure if fvfw inherits from ffdshowdec or not...

conando

30th July 2006, 15:47

hm it seems as if it was pretty easy.. or it does have strang side effects that i don't know of yet because i don't fully know the source :)
i just changed:

decVFW->initPresets();
decVFW->setActivePreset(FFPRESET_DEFAULT,0);

to:

decVFW->initPreset();

in Tffvfw.cpp and changed some other files to make TffdshowDec::initPreset a "publicly" available interface function.. it seems to handle all the "automatics" involved pretty much by itself ;)
now i just have to get an optimized build running here because it's not up on par yet with the others ;) ... how could my very little change be submitted? mail to milan or something? guess it's not only useful to myself to finally be able to use automatic preset loading in ffvfw too...

foxyshadis

30th July 2006, 18:47

Here's the diff, though I didn't test yet; you can post it on the ffdshow tracker or send it to Haruhiko (he's become somewhat the unofficial manager lately).

ffdshow-vfwpresets.diff

Index: src/Tffvfw.cpp
===================================================================
--- src/Tffvfw.cpp (revision 2546)
+++ src/Tffvfw.cpp (working copy)
@@ -121,8 +121,7 @@
if (decVFW) return true;
HRESULT hr=S_OK;
decVFW=new TffdshowDecVideoVFW(&hr,this);
- decVFW->initPresets();
- decVFW->setActivePreset(FFPRESET_DEFAULT,0);
+ decVFW->initPreset();
return !!decVFW;
}

conando

30th July 2006, 22:15

ok I'll submit it to him and/or the patch "corner" @ ffdshows sourceforge page... it's not only the Tffvfw.cpp change but a few lines in some of the ffdshowDecVideo interface declarations to make initPreset a public interface function (until now it was private).. hopefully without strange side effects ;) but that's up to people who know the code much better than me

foxyshadis

31st July 2006, 00:52

It should probably be protected, in that case. (nm, I see what you mean now, it already was protected.)

conando

31st July 2006, 12:01

*sigh* didn't fully work.. it now respects which preset you set as default and i added a workaround so that the fourcc can be checked as well buuuut: i just realized that there's no way to get the source filename from a vfw decompressor/codec .. craaaaap :-( .. or does anyone have an idea for some wicked workaround? :D

haruhiko_yamagata

31st July 2006, 13:07

another question i have: i use (and love) automatic preset loading with ffdshow quite a bit.. however i tried a myriad of builds now because i've got the problem that ff_vfw (the app i use only has vfw interface *sigh*) doesn't care about it's different presets at all - it always uses the one named "default".. regardles of the autoload conditions and even which one you select as default :-( .. is this simply not implement or a bug? could it be done?
Hello, conando. I would like to reproduce the problem. What is the application you use?

conando

31st July 2006, 13:25

well it's simply not implemented... i have patched the source a little so that the selected default preset is at least loaded/"respected" and that automatic loading depending on the process name and fourcc works.. but i don't see any way on how to get the filename of the original avi if we're in ff_vfw "mode".. :-(

i use vegas video...
or you can simply use virtualdub and open an avi using a ff_vfw codec.. now try to switch image filtering presets.. it always loads the "default" one...
i just looked around (google, virtualdub sources, msdn) but i don't have any idea on how you could get the filename - as with vfw you're only a decompressor getting some data.. there simply isn't any connection to the original avi file.. :(

foxyshadis

31st July 2006, 14:09

What you could do, is get the process ID of your host, scan the process's heap, and search for anything that looks like a filename that ends in .avi, .avs, or whatever. ;D

(I jest.)

conando

31st July 2006, 14:20

hehe :D yes i thought about that too.. unfortunately in my current case (sony vegas) there are a dozen open AVIs and i want different presets for some of them.. *sigh* .. why can't they just use directshow for christs sake :-(

conando

1st August 2006, 14:39

ok... i'm currently trying something.. guess that wouldn't make it into a public patch *g* .. with vegas for example it makes a decompressor_begin call for every single avi.. and i'm hooking createfile to see which avi it opened before calling the dec_begin... :)) seems to work... now i'll clean that up and make a setting for it (because doing the hooking always isn't a good idea i guess ;o)

Inventive Software

8th August 2006, 13:21

Is there any chance of this getting stickied, to keep it falling off the topic page? ;)

conando

8th August 2006, 15:34

if anybody is interested: i really got this working! :-D
i added hooking of CreateFileA/W to ffdshow and actually it works quite good by saving the last opened avi filename to an internal buffer and looking at it when opening the decoder!
i'll clean it up and make a patch/diff for anybody who's interested... although this needs quite some more testing i guess. ;)
but so far it looks fine.. finally automatic preset loading by filename with ffvfw - yeah! *g*

Inventive Software

8th August 2006, 17:33

Let Milan know, and he may include that in the SVN trunk if he's feeling nice!

MatMaul

8th August 2006, 18:04

igor1st

11th August 2006, 09:47

ffdshow SVN revision: 2543
Build date: 2006-08-10
Compilers used:
- GCC 3.4.5 (libavcodec.dll)
- GCC 4.0.3 (libmplayer.dll, ff_kernelDeint.dll, TomsMoComp_ff.dll)
- MSVC71 (ffdshow.ax, etc)
Changes:
- high accuracy libmad & tremor;
- fixed PCM audio playback;
- fixed vorbis 5.1 decoding in libavcodec;
- fixed volume normalization;
- updated libdts to SVN rev 28;
- updated libfaad2 to CVS 2006-08-08;
- custom installer with my tweaked settings.
Minimum CPU requirement: MMX
Download: ffdshow_rev2543_20060810.exe

ffdshow-tryout SVN revision: 61
Build date: 2006-08-11
Compilers used:
- GCC 3.4.5 (libavcodec.dll)
- GCC 4.0.3 (libmplayer.dll, ff_kernelDeint.dll, TomsMoComp_ff.dll)
- MSVC71 (ffdshow.ax, etc)
Changes:
- updated libfaad2 to CVS 2006-08-08 (hcr.c, mp4.c, syntax.c, cfft.c, common.h, neaacdec.h);
- custom installer with my tweaked settings.
Minimum CPU requirement: MMX
Download: ffdshow-tryout_rev61_20060811.exe

igor1st

16th August 2006, 09:03

Both builds was updated with latest libfaad2 (CVS 2006-08-15):

ffdshow_rev2543_20060816.exe

ffdshow-tryout_rev69_20060816.exe

igor1st

19th August 2006, 07:58

libsamplerate updated to 0.1.2 (diff (http://d.turboupload.com/d/893624/libsamplerate.diff.zip.html))

ffdshow_rev2543_20060822 (http://d.turboupload.com/d/905286/ffdshow_rev2543_20060822.exe.html)

ffdshow-tryout_rev76_20060819
ffdshow-tryout_rev83_20060822

Inventive Software

22nd August 2006, 18:50

@igor1st: What's with the ffdshow_tryout builds?

Isochroma

22nd August 2006, 19:03

The recent rush of ffdshow builds has gained at most a few percent in speed, while throwing away what's really important - reliability and features. Speedup can be purchased with a new CPU, but broken features cannot be used. All I can say is perhaps now is a good time for a radical rethink of build philosophy. The CoreAVC team (despite their troubles with automated purchasing and website issues) made an excellent filter, which in one step bested ffmpeg's avc decoding by a significant margin, without compromising stability.

drevil_xxl's builds have working MakeAVIS, just checked. clsid's don't have it, and haruhiko_yamagata's don't work. However, we seem to have lost WMV7/8 encoding in the vfw interface, and WMV3/9 has duplicate entries in the decoder filter ui.

clsid

22nd August 2006, 21:38

So what is buggy and/or unstable in the tryouts? If you don't report bugs, then there is a chance they won't get fixed.

CoreAVC has plenty of bugs too. They could and should IMHO have released an interim version before v1.2. And don't forget that it is a commercial product, while ffdshow is a free open-source product.

My builds don't have MakeAVIS (nor any of the other extras) yet. I will update my custom script to include them when I have some spare time.

Isochroma

22nd August 2006, 21:54

Excellent! In my previous post I forgot to thank the big three for their hard work producing all these nice builds, so I'll do it here. Thanks guys. Apologies for the harsh tone of previous post. Sometimes I wonder if life in a dictatorship with only one choice of bread, etc. would be better... people like choice, but it brings so many problems.

Maybe it is possible to copy makeavis.exe from drevil_xxl's build's ffdshow folder, then uninstall and install some other build, then copy back?

igor1st

23rd August 2006, 15:26

@igor1st: What's with the ffdshow_tryout builds?
If you asking about changes - this is simple build from SVN. Only my tweaked default settings.

drevil_xxl's builds have working MakeAVIS, just checked. clsid's don't have it, and haruhiko_yamagata's don't work.
Try this fresh build:
ffdshow-tryout_rev85_20060823
It has makeavis and avisynth plugin (auto-copy in plugin folder if avisynth installed).

Isochroma

24th August 2006, 02:35

Thanks! Quite busy now, but I will try it ASAP.

MatMaul

26th August 2006, 19:38

we seem to have lost WMV7/8 encoding in the vfw interface, and WMV3/9 has duplicate entries in the decoder filter ui.

I think it's my fault, I try to correct that when I go back to home.

EDIT : oups, corrected by drevil_xxl, sorry.

igor1st

26th August 2006, 20:17

ffdshow-tryout_rev94_20060826 (http://d.turboupload.com/d/921271/ffdshow-tryout_rev94_20060826.exe.html)

ffdshow-tryout_mix_20060826 (http://d.turboupload.com/d/921263/ffdshow-tryout_mix_20060826.exe.html)
Based on rev13 with patches from revs 15, 37, 38, 64, 70, 71, 76.

Also I deleted my old builds from turboupload except this one - ffdshow_rev2543_20060822 (http://d.turboupload.com/d/905286/ffdshow_rev2543_20060822.exe.html).

_xxl

26th August 2006, 22:26

Based on rev13 with patches from revs 15, 37, 38, 64, 70, 71, 76.
Can you post all patches?

haruhiko_yamagata

27th August 2006, 00:55

ffdshow-tryout_mix_20060826 (http://d.turboupload.com/d/921263/ffdshow-tryout_mix_20060826.exe.html)
Based on rev13 with patches from revs 15, 37, 38, 64, 70, 71, 76.
Interesting build, but the installer seems to wipe almost all of "HKEY_CURRENT_USER\Software\GNU\ffdshow\default".

Another thing, your build is fast, but once your build is installed, all other builds become turtles as if MMX is turned off. I tryed to overwrite by mine, clsid's, and milan's. Unistall did not resolve the problem.

haruhiko_yamagata

27th August 2006, 06:25

Another thing, your build is fast, but once your build is installed, all other builds become turtles as if MMX is turned off. I tryed to overwrite by mine, clsid's, and milan's. Unistall did not resolve the problem.
I found why now. Soon after the registry was wiped, I installed my ffdshow-20060730-Q.exe. It enables postprocessing if the registry is empty. Anyway current setting should not be deleted.

Boulder

27th August 2006, 09:26

Could someone please create a sticky which would *only* have the latest build information from everyone who releases ffdshow builds?

igor1st

27th August 2006, 10:33

Can you post all patches?
ffdshow-tryout_mix_diff.zip (http://www.mytempdir.com/890169)

Anyway current setting should not be deleted.
Ok, fixed:
ffdshow_rev2543_20060827.exe (http://www.mytempdir.com/890159)
ffdshow-tryout_mix_20060827.exe (http://www.mytempdir.com/890162)
ffdshow-tryout_rev94_20060827.exe (http://www.mytempdir.com/890164)

haruhiko_yamagata

27th August 2006, 10:51

Thank you, igor1st. This time it's very good.

DSP8000

27th August 2006, 12:44

Could someone please create a sticky which would *only* have the latest build information from everyone who releases ffdshow builds?

That is a very good idea, with all of the revisions & updates it is hard to keep up.It is good however that so many people are involved in patching ffdshow.

Also, changelog or what's fixed with each build would be great.

Tnx. guys.
DSP8000

vega22

27th August 2006, 13:08

haruhiko_yamagata

27th August 2006, 13:34

Best build today for Pentium D dual core, ss3?
Latest builds from Clsid, Drevill and Igor (2546) run perfectly with divx vmr7-9 and jerky (frames dropped) in overlay. No problem with Clsid, Drevill and Igor 2543. Latest build ss3 from Kiroshu run OK (does not have the MT patch?)
It's a bug.
http://forum.doom9.org/showthread.php?p=868509#post868509
igor1st's "ffdshow-tryout_mix_20060827.exe" decode divx fast.

clsid

27th August 2006, 14:55

I haven't noticed any slowdowns with MPEG-4 ASP on my AMD Athlon. So perhaps an SSE issue?

vega22

27th August 2006, 15:04

Thanks, Haruhiko. Because i not have speed problems with 920@940 cpu (in all builds <9% CPU in standard divx/xvid, sharpen on), i go to install Clsid 2543 or latest ss3 Kiroshu, while the overlay mixer/MT bug is solved. The ffdshow-tryout_mix_20060827.exe have this bug too.

PS: All tests in Zoom Player. WMP in Windows xp is vmr7 default, so never there are this problem.

igor1st

27th August 2006, 17:59

Best build today for Pentium D dual core, ss3?
Latest builds from Clsid, Drevill and Igor (2546) run perfectly with divx vmr7-9 and jerky (frames dropped) in overlay.

PS: All tests in Zoom Player. WMP in Windows xp is vmr7 default, so never there are this problem.

Trying now overlay mixer with MPC on my Athlon XP. All ok if not try to enable 'queue output samples'.

vega22

28th August 2006, 14:01

MPC same problem on my Pentium D: overlay mixer jerky, vmr7-9 ok. 'queue output samples' on and off.

haruhiko_yamagata

28th August 2006, 14:41

vega22

28th August 2006, 23:46

XFX Nvidia 6600GT, driver 84.43.
No problem in 2543.

haruhiko_yamagata

28th August 2006, 23:58

XFX Nvidia 6600GT, driver 84.43.
No problem in 2543.
What happens with queue and overlay?

vega22

29th August 2006, 00:52

2546 + overlay mixer + queue = jerky (frames dropped)
2546 + overlay mixer + queue off = jerky (frames dropped)
2546 + vmr7 or vmr9 + queue = all good
2546 + vmr7 or vmr9 + queue off = all good

zoom player, same in MPC

haruhiko_yamagata

29th August 2006, 12:42

vega22

30th August 2006, 09:30

News:

Build 20060604-rev2546 by videomixer9 + overlay mixer + queue = All good

Build 20060604-rev2546 by videomixer9 + overlay mixer + queue off = All good

igor1st

30th August 2006, 12:12

ffdshow-tryout_rev107_20060830 (http://d.turboupload.com/d/934088/ffdshow-tryout_rev107_20060830.exe.html)

haruhiko_yamagata

30th August 2006, 15:35

News:

Build 20060604-rev2546 by videomixer9 + overlay mixer + queue = All good

Build 20060604-rev2546 by videomixer9 + overlay mixer + queue off = All good
Thank you. Your imformation will help me to debug.

vega22

1st September 2006, 10:54

Thanks to you and all the developers of this wonderful project

haruhiko_yamagata

1st September 2006, 13:22

Bug fix : frames are dropped on overlay mixer since rev 37.

FredThompson

7th September 2006, 04:48

I'm looking for a build which supports MPEG2 4:2:2 on SSE. Does such a beast exist? I've found discussions about 4:2:2 in the decoder libraries but can't tell if those have made their way into ffdshow.

belay that, I'm an idiot. I forgot to turn off MPC's internal MPEG2 decoder...

SeeMoreDigital

7th September 2006, 08:45

Hi Fred,

Could you upload a "snippet" of MPEG-2 4:2:2, by any chance?

Cheers

Amour

7th September 2006, 10:09

Don't know, but maybe igor1st you should join the http://ffdshow-tryout.sourceforge.net/ project.

By the way, where to find the source code of the current releases of ffdshow?

igor1st

7th September 2006, 10:31

By the way, where to find the source code of the current releases of ffdshow?
http://sourceforge.net/svn/?group_id=173941

FredThompson

7th September 2006, 11:43

ASCII and ye shall receive (after the attachment is approved.)

NTSC DV -> VirtualDub w/411Helper set for interpolation and Donald Graft's Reverse Field Dominance
TMPGEnc 2.5 Plus - 300/8000/15000 2-pass 4:2:2 High Profile and YUV set to 4:2:2

Virtually indestinguishable from the source and 1/4 the storage space with none of the loss you'd get from 4:1:1 -> 4:2:0

Yes, that's a corrugator...God bless industrialization...but damn those industrial lights!

igor1st

7th September 2006, 12:20

ffdshow-tryout_rev149_20060907 (http://www.mytempdir.com/915125)

Also I made a very little tweak of volume nolmalization (for my own use). If someone use normalization with ffdshow - please test it and reports results (the goal was to eliminate easy audible volume jumps - in my tests it almost always better now, but could be worse with complicated dynamic streams, so it needs testing).
ffdshow.ax for rev149 (http://www.mytempdir.com/915127)

MatMaul

7th September 2006, 12:43