ffdshow tryouts project: Discussion & Development

[Kado]

@thuan
thanks. "keep aspect ratio" enabled in mpc did the trick.

[sjchmura]

As a general question....

The fork is working VERY well with VistaMCE and the "plugin" that allows FFDShow to work.

However, how long with this "fork" continue? Will we always have the fork is is the plan, at some point, to merge back to a single repository?

Thanks for all the incredible work. FFDShow keeps me from using a MAC

[Leak]

Originally Posted by sjchmura

However, how long with this "fork" continue? Will we always have the fork is is the plan, at some point, to merge back to a single repository?

Well, given that MPC proper is as good as dead and no-one that's still active has write access to it's SVN I doubt there's going to be a merger down the road...

Same goes for ffdshow...

np: Luke Vibert - Clikilik (Chicago, Detroit, Redruth)

[Wilbert]

The use of the [0,255]->[0,255] is justified because they say it's a more smooth conversion from YUV to RGB which could cause less banding. For me, I think the best should be the compromise solution.

Perhaps so. But when doing a [0,255]->[0,255] the black portions are too black and the bright ones are too bright. Imo, that's more annoying than any banding.

[Dark_Angel_PT]

Just a small question\request:

When using the translation mode of ffdshow, wouldn't it be a lot more useful to keep the original language showing in these dialogs, so we can compare what we entered and the original.



I suggest the original string should always be kept above the text box, to help translating. Sometimes I forget what was there in the first place, i have to be changing the languages back and forth to check what i entered.

[FredThompson]

I've had very poor performance with MPEG2 4:2:2 and libmpeg2 decoder in the past few builds of ffdshow. If I open a file with MPC then move the position slider, the returned video is not playing. The position indicator will continue to move but the displayed video does not change, almost as if ffdshow's libmpeg2 decoder is returning blank frames. FWIW, the MPC and ffdshow configurations appear to be the same as in the past. Do other people have this problem with MPEG2 4:2:2 and ffdshow's libmpeg2 decoder?

[Leak]

Originally Posted by FredThompson

Do other people have this problem with MPEG2 4:2:2 and ffdshow's libmpeg2 decoder?

Any idea where I could get hold of such an MPEG sample to test this? DVDs usually are 4:2:0 I think?

[chros]

Originally Posted by haruhiko_yamagata

I found the way.
* Enable "Uncompressed" and "Check for DTS in WAV" at the foot note of DTS.
* Raise the merit value.

Thanks, I'll try it !!!!

[Valeron]

@haruhiko:
Thanks for ur reat job! A much faster subtitle rendering engine compare to vsfilter! Though some advanced effects r still limited.
I can play some low bitrate 1080p AVC encode with subtitle flawless with my old Barton now.

[Thunderbolt8]

how much cpu power does vsfilter take anyway on a c2d with 2.8ghz ?

[Valeron]

Originally Posted by Thunderbolt8

how much cpu power does vsfilter take anyway on a c2d with 2.8ghz ?

depend on what subtitle u r going to render, what final resolution to overlay on, whether advanced effects or not. all these factors will affect the rendering power need.

[Thunderbolt8]

hm its either an .srt or .ass subtitle on lets say 1080p resolution. using ffdshow with postprocessing for video decoding and ac3 filter for audio, haali splitter. for vsfilter picture size: original heigth, rest are standart settings I guess

[Valeron]

c2d at 2.8GHz can handle 1080p AVC decoding even with libavcodec(the one in ffdshow) without too much problem if the bitrate don't go too high(2x~30Mbps maybe). libavcodec is single thread nowadays unless the AVC encode u r going to play is multi-splice coded.
the subtitle rendering thread will be dispatched to another core imo, so there's no problem unless u've already got shutter in the decoding process.

yes, NO advanced effect here. i'm just talking about plain text subtitle. i achieved my goal just with an Barton@2Ghz(with vsfilter i got 10%~20% higher CPU load and then met sub-async). the subtitle rendering process will never become painful for c2d imo.

[haruhiko_yamagata]

Originally Posted by Dark_Angel_PT

Just a small question\request:

When using the translation mode of ffdshow, wouldn't it be a lot more useful to keep the original language showing in these dialogs, so we can compare what we entered and the original.



I suggest the original string should always be kept above the text box, to help translating. Sometimes I forget what was there in the first place, i have to be changing the languages back and forth to check what i entered.

Yes, maybe when we have time.

[TheShadowRunner]

hey guys, it seems Vobsub is completely broken in the latest builds (from september), is it a known issue?
By broken i mean it either crashes the player or displays nothing.
Later,

TSR

[Boardlord]

Originally Posted by TheShadowRunner

hey guys, it seems Vobsub is completely broken in the latest builds (from september), is it a known issue?
By broken i mean it either crashes the player or displays nothing.
Later,

TSR

I had that too, but also with older builds. Whenever I encounter such vobsub files, I run them through vobsub 2.23 (get it from Doom9 download library) and then they load fine with ffdshow.

[clsid]

If it only crashes with specific files, then please upload a sample file.

[SBeaver]

A fresh x64 build would be nice now, I'd like to try the new half-ass multithreaded h264 decoder and see how it works.

[Shakey_Jake33]

Originally Posted by cca

Strange, since I use 32 bit ffdshow along with Zoom Player and Media Player Classic and my S/PDIF Passthrough works perfectly. I do need to check the "Maximize S/PDIF Compatibily" checkbox though. That's on Vista 64 with a Sound Blaster X-Fi Platinum.

Originally Posted by SBeaver

Well ever since vista was released I've been reading about people like you who say everything is ok, maybe creative has some workaround in their drivers but it doesn't work for anyone else.

Originally Posted by SBeaver

Just so we are clear on this, I hope all of you actually have a way to tell what signal is going to the reciever.
From all information I have gathered there should be no way to passthrough ac3/dts using 32bit filters in vista x64.

I apologise if I'm talking about something completely different here, or if this has already been cleared up, but I have an X-Fi XtremeMusic and ffdshow's SPDIF passthrough works perfect for me, although as the above poster states, 'Maximize S/PDIF Compatibility' does need to be enabled, otherwise the DTS sound is garbled.
Vista Home Premium 64-bit here.

I do also have an onboard HD Audio codec if anyone would like that tested.

[SBeaver]

Originally Posted by Shakey_Jake33

I apologise if I'm talking about something completely different here, or if this has already been cleared up, but I have an X-Fi XtremeMusic and ffdshow's SPDIF passthrough works perfect for me, although as the above poster states, 'Maximize S/PDIF Compatibility' does need to be enabled, otherwise the DTS sound is garbled.
Vista Home Premium 64-bit here.

I do also have an onboard HD Audio codec if anyone would like that tested.

Please do try it with the onboard audio, with 32bit version of ffdshow then of course.
I'm still very interested in understanding why creatives cards work and most others don't.

[Shakey_Jake33]

Originally Posted by SBeaver

Please do try it with the onboard audio, with 32bit version of ffdshow then of course.
I'm still very interested in understanding why creatives cards work and most others don't.

I'll install the drivers for my onboard tomorrow and post the results.

I know for sure my speakers (Logitech Z-5500's) are recieving an untouched AC3/DTS passthrough because Dolby Digital/DTS are displayed on the reciever, and the audio is untouched because the various effects I have enabled in ffdshow (like normalisation) do not effect it.
Also, the AC3 transcoding functionality of ffdshow works perfectly too, with my video files with 5.1 AAC outputting as 5.1 AC3 fine.

The SPDIF also works using the MPC internal codec, as well as PowerDVD and WMP.

[cc979]

@haruhiko_yamagata
if you can remember the 'video decoder configuration' display problem i mentioned - i had more time to trace it, seems the my problem started at rev. 1444 as rev. 1443 works fine

cheers

[haruhiko_yamagata]

Originally Posted by cc979

@haruhiko_yamagata
if you can remember the 'video decoder configuration' display problem i mentioned - i had more time to trace it, seems the my problem started at rev. 1444 as rev. 1443 works fine

cheers

Rev 1444 is a very small fix. I have no idea about it. Anyway thank you for your time...

[haruhiko_yamagata]

Originally Posted by TheShadowRunner

hey guys, it seems Vobsub is completely broken in the latest builds (from september), is it a known issue?
By broken i mean it either crashes the player or displays nothing.
Later,

TSR

Vobsub support in ffdshow has been very incomplete from the start. I have broken it recently and fixed it at rev 1485.
I can reproduce "displays nothing" but cannot reproduce "crash the player". Please send us the sample, if it is possible.

[_xxl]

New test build:
http://lists.mplayerhq.hu/pipermail/...er/035877.html
http://rapidshare.com/files/58407533...6_xxl.exe.html

[LigH]

Now that Theora 1.0 beta 1 was released - will you update the supported VfW encoder + the decoders?

[haruhiko_yamagata]

Originally Posted by _xxl

New test build:
http://lists.mplayerhq.hu/pipermail/...er/035877.html
http://rapidshare.com/files/58407533...6_xxl.exe.html

Sample file : [Eclipse-gg] Manabi Straight - 07 (1024x768 h264) [A7E82AE2].mkv
CPU : core2duo E6700
rev1487_20070926_clsid : User: 249s, kernel: 1s, total: 251s, real: 261s, fps: 138.8, dfps: 133.4
rev1485_20070926_xxl : User: 64s, kernel: 2s, total: 66s, real: 210s, fps: 523.3, dfps: 165.9

Great! The multithreaded version is 24% faster.

[Thunderbolt8]

hm, why does clsid's newer (svn) version not support this yet? wasnt this part of a general update or something like that?

btw. how does it come, when comparing a file using vc-1 codec ffdshow basically only uses 1 core , while when playing a .mkv or .ts h264 file the cpu usage seems to be distributed evenly to both cores, according to task manager with clsid's 1487 ?

[_xxl]

http://lists.mplayerhq.hu/pipermail/...er/035877.html
The attach patch gives ~20-30% speedup on single sliced CABAC content.
Basically, the code decodes up to 128 macroblocks in one
thread while doing prediction+idct+deblock of the previously decoded
128 blocks in another thread.

My build is based on latest h.264 multithreaded patch.

http://lists.mplayerhq.hu/pipermail/...er/035880.html
Well, as already discussed previously :
- slice based parallelism depends on the video, so will not be always
possible. Furthermore, it scales as much as the encoder was scaled, so
though it's an easy solution, it's not a pretty one

- cabac / reconstruction parallelism : relatively good speed up,
though it will depends on the bitrate of the video ( low bitrate -> bad
scaling, very high bitrates -> bad scaling too ), and no scaling over 2
CPUs. There again, not necessarily pretty, but effective. More effective
imho than slice based, and sufficient for decoding anything up to
1080i30 40 Mbit/sec on a C2D, I would say.

- frame based, which I think should be the "best" method : a priori
very scalable ( if x264's threading efficiency is achieved ), no
constraint on the video except for vertical downward motion vectors,
but, alas, not tested, so scalability is hypothetical.

All that to say I'm torned between a theorically clean ideal solution,
and a dirty but already working one. One thing is sure, I don't like
slice based parallelism

Libavcodec VC-1 is not multithreaded yet.

[Thunderbolt8]

Originally Posted by _xxl

My build is based on latest h.264 multithreaded patch.

yes, but I thought it was the same with clsid's version, since its even 'newer' according to the svn version? so I guess that patch wasnt included there and has to be implemented manually ?

[clsid]

The patch that xxl used is still experimental and not yet part of the official FFmpeg code base.

[Thunderbolt8]

allright then.
I tested boths versions, clsid' 1487 and xxl' patched 1485, with hs264 .ts and remuxed .mkv star wars broadcasts, but at least with a look on task manager cpu usage screens I almost couldnt make any difference out between them, in fact im even tempted to say the 1487 version was slightly faster (have a c2d 6300 running @2,8GHz). could this be possible?

[foxyshadis]

Multithreading the decoder is only important when you can't achieve a full framerate already. If you have a fast processor, that'll mostly be in HD encodes, like BD/HDvD. Most HD discs are encoded with several slices, I've seen reports of 3-8, so this will help a lot for playing them.

[Shakey_Jake33]

Seem to be facing a bug right now, where if I try to play a DVD with LPCM audio using ffdshow as the audio decoder, I get no sound.

Here's the funny thing though - when I set AC3 and DTS to decode using the internal MPC filters, it works fine. But when I set them to decode with ffdshow, all AC3/DTS content works fine, but LPCM does not! What has LPCM got to do with AC3/DTS? God knows.

Here's another wierd thing - the same DVD's work 100% fine when extracted to my DVD's as VOB files!

Now, if I enable MPC's internal LPCM decoder, the DVD's will still try and use ffdshow, as if the player is detecting the DVD's to be having AC3 sound or something (though MPC reports LPCM at the bottom). Is it a bug in ffdshow or MPC? I'm not sure.

Using SPDIF, but the same happens over analogue.

[chros]

Originally Posted by Thunderbolt8

I tested boths versions, clsid' 1487 and xxl' patched 1485, with hs264 .ts and remuxed .mkv star wars broadcasts, but at least with a look on task manager cpu usage screens I almost couldnt make any difference out between them, in fact im even tempted to say the 1487 version was slightly faster (have a c2d 6300 running @2,8GHz). could this be possible?

For me, it's the same: I can't tell the difference between them ...
So is this new patch good for only for multi-slice based encoding or for all (I tested both: new apple trailers (1080p) and mine (x264 720p)) ?

PS: _xxl: thanks for your build!

[haruhiko_yamagata]

Originally Posted by chros

For me, it's the same: I can't tell the difference between them ...
So is this new patch good for only for multi-slice based encoding or for all (I tested both: new apple trailers (1080p) and mine (x264 720p)) ?

PS: _xxl: thanks for your build!

You could try beta3 vs rev 1487. If rev 1487 is faster, the sample is multi-slice based encoding. The new patch is good for both.

[yesgrey]

I think I have found a bug...
In ffdshow's avisynth tab, when you change some of the text in the avisynth commands box, the text only is keeped if you click OK while seeing the tab. If before clicking OK you change to another tab for setting another parameter, when you return to the avisynth tab your changes were lost.

[Delerue]

Originally Posted by yesgrey3

I think I have found a bug...
In ffdshow's avisynth tab, when you change some of the text in the avisynth commands box, the text only is keeped if you click OK while seeing the tab. If before clicking OK you change to another tab for setting another parameter, when you return to the avisynth tab your changes were lost.

Another thing: if you export FFDShow settings to a reg file, the text inside the avisynth tab won't be restored.

[Mc Onyx]

Originally Posted by haruhiko_yamagata

The new patch is good for both.

Sorry, but do you mean there is a finished patch that enables multithreading on all encodes, even those not encoded with slices?

[Leak]

Originally Posted by Delerue

Another thing: if you export FFDShow settings to a reg file, the text inside the avisynth tab couldn't be restored.

Okay, that definitely looks like a bug to me.

As for the other problem - the script is not stored in the configuration (and therefore not used) until you hit the "Apply" button, and it's always been that way: the edit box value is read from the current configuration every time the AviSynth page is (re-)loaded, which happens when you switch back and forth between config pages.

The difference in regard to most other controls is that their values are stored (and used) in the configuration immediately upon editing, but with an AviSynth script you'd both get script errors most of the time while you're typing and possibly a lot of lag since re-initializing AviSynth with a new script can take several seconds depending on the filters used. So it's contents are only stored when hitting "Apply" on purpose.

I'll take a look at the reg file problem.

EDIT: That's a problem with regedit, actually - the scripts are written to the file with their linebreaks intact, so regedit's parser will just skip what it thinks are broken lines.

But if you just export the whole ffdshow configuration from regedit, it'll produce the exact same problem... so I'd have to replace the newlines in the AviSynth script with something else when storing it to the registry and undo that when reading from it - that'll need a bit more investigation...

np: Supermayer - Two Of Us (Geiger Mix) (Kompakt Total 8 (Disc 1))

[haruhiko_yamagata]

Originally Posted by Mc Onyx

Sorry, but do you mean there is a finished patch that enables multithreading on all encodes, even those not encoded with slices?

The patch is not finished as you can read from the link above. The patch enables multithreading on all encodes (as far as I have tested).

[yesgrey]

Leak,
Thanks for clarifying my "bug". It makes sense.

[foxyshadis]

Can you hook onblur() from the filter menu or one of the other teardown messages, to pop up a 'save me?' box? I haven't worked with the side menu at all, I have no idea what's possible.

[Leak]

Originally Posted by foxyshadis

I haven't worked with the side menu at all, I have no idea what's possible.

Same here, but after a quick check I couldn't find anything that would get called specifically when the page is changed.

And TBH, if you're editing a bigger script you'll probably be happier with doing it in an external text editor anyway. Or just don't forget to hit "Apply" after changing or pasting in the script.

np: Burnt Friedman - Machine In The Ghost (ft. Barbara Panther) (First Night Forever)

[vinetu]

Originally Posted by Leak

I'll take a look at the reg file problem.

EDIT: That's a problem with regedit, actually - the scripts are written to the file with their linebreaks intact, so regedit's parser will just skip what it thinks are broken lines.

But if you just export the whole ffdshow configuration from regedit, it'll produce the exact same problem...

I'm not sure about complex scripts,but the simple ones can be writen in one line and then there is no problem with exporting and importing.

Example:
ColorMatrix(mode="Rec.709->Rec.601").SeparateFields().fft3dfilter().LanczosResize(640,480).Weave()

[zambelli]

When is the next beta planned for?

[haruhiko_yamagata]

Originally Posted by zambelli

When is the next beta planned for?

Not very near, I think. We have some issues to fix.
I'm implementing RGB<->YUV conversion options.

[_xxl]

Not before PAFF and h.264 multithreaded for all x264 encodes.
Maybe E-AC3 decoder should be added and dts issue fixed.

[ACrowley]

Originally Posted by _xxl

Not before PAFF and h.264 multithreaded for all x264 encodes.
Maybe E-AC3 decoder should be added and dts issue fixed.

These are exactly the Updates most people need !

What do you think, when will there be a EAC3 Decoder build ?

[yesgrey]

Originally Posted by haruhiko_yamagata

I'm implementing RGB<->YUV conversion options.

Just the PC vs rec levels option or also the custom matrix?

[haruhiko_yamagata]

More than just the PC vs rec levels, less than the custom matrix.
Black cutoff, white cutoff, chroma cutoff as well as ITUR-BT.601/701 are configurable.

[SBeaver]

the 64bit builds seem to be unable to decode flac audio, could someone look into this and see why it doesn't work?

[Shakey_Jake33]

Even with the 32-bit build, I've never been able to get ffdshow to decode standalone FLAC files.

[fastplayer]

Originally Posted by Shakey_Jake33

Even with the 32-bit build, I've never been able to get ffdshow to decode standalone FLAC files.

Without a parser, it can't AFAIK.

[SBeaver]

Originally Posted by Shakey_Jake33

Even with the 32-bit build, I've never been able to get ffdshow to decode standalone FLAC files.

Maybe not standalones, perhaps that is where the problem is.
I've played a lot of multichannel flacs in mkv containers and had ffdshow transcode them to ac3 and everything, all without issues. (32bit build).
I might digg one up and try it with 64bit and see what happens.

edit: ok I can confirm decoding works for flac in 64bit, as long as they are in a container.

[multiblitz]

One interesting observation: With Vista, 8600GT, 163.69driver, latest ffdshow from clsid and EVR-Renderer: I can sharpen with the sharpening of the drivers and can sharpen in parallel ffdshow. To my eyes it looks extremely good, potentially even better than limitedsharpening (I watch at 1920*1080 on a 4.2m wide and 2.2 high screen with a sony ruby)...less artifacts than anything else I tried...

...interesting: as soon as I enable lanzos resizing, the sharpening through the nvidia-drivers is broken...if I let the card do the resizing, sharpening works in ffdshow and in the card.

Before I started with limitedshaprening, I used to use lanzos resizing with some sharpening in ffdshow. I have tested this against the current resizing and sharpning of nvidia. In this case nvidia looks better than ffdshow...quite impressive.

[Valeron]

@haruhiko_yamagata:
The font size is not compatible with vsfilter, at least horizontal scale after rev1482.

I've got a NON-Square pixel AVC encode, when subtitle with ffdshow(after rev1482), the renderer automatic adjust the font size for the subtitle. but with vsfilter, it won't.

Great feature, though not compatible with vsfilter(opposite to the changelog), I like it.

[LigH]

HuffVUV encoding seems to be broken in some current "tryout" builds. I used ffdshow with HuffYUV-YV12 and different settings to capture a desktop area with some screen recorder. Resolution: 1016x680 pixel, or similar dimensions.

The screen recorder used to crash after a while.

Watching a shorter video I completed before a crash, I saw that the bottom of the video contains some completely garbled lines with "fantastic colors", they grew upwards over the time the video played.

A thread in the german doom9/Gleitz board with a screenshot

[vlada]

Originally Posted by _xxl

Not before PAFF and h.264 multithreaded for all x264 encodes.
Maybe E-AC3 decoder should be added and dts issue fixed.

PAFF is already implemented in ffmpeg/libavcodec? Since when?

Also I'd like to repeat my very big feature request. It's letterboxing for DVD. Or at least it would be nice, if I could move the subtitles. Is someone willing to have a look at it? We've been already discussing this before, unfortunately this is still not implemented. I think with some work on VobSub, this might be implemented.

Also on the codecs tab under MPEG-2 it still says DVD Decoding (not working yet). Why? Are there any known problems? It works fine for me already for a long time (except for the letterboxing ;-) ). And I think MPEG-2/DVD support should be turned on by default.

[_xxl]

PAFF:
http://lists.mplayerhq.hu/pipermail/...er/035591.html
Multithreaded H.264 decoder for all x264 encodes:
http://lists.mplayerhq.hu/pipermail/...er/035877.html

[haruhiko_yamagata]

Originally Posted by Valeron

@haruhiko_yamagata:
The font size is not compatible with vsfilter, at least horizontal scale after rev1482.

I've got a NON-Square pixel AVC encode, when subtitle with ffdshow(after rev1482), the renderer automatic adjust the font size for the subtitle. but with vsfilter, it won't.

Great feature, though not compatible with vsfilter(opposite to the changelog), I like it.

Hmm... It's compatible with MPC's internal renderer, VMR9 renderless. VMR9 renderless is not compatible with vsfilter.
OK, you like it, I'll leave it as it is.

[fastplayer]

Originally Posted by _xxl

Not before PAFF and h.264 multithreaded for all x264 encodes.
Maybe E-AC3 decoder should be added and dts issue fixed.

Adding features especially experimental ones during a beta phase? Don't know if that's a good thing...

[Sharktooth]

i think it's a good idea instead. since those features are missing from long time. even if it breaks the conventional rule of thumb it's worth adding them

[Shakey_Jake33]

I think they are the features a lot of people are waiting for.

[_xxl]

encoder/cavlc.c: In function 'cavlc_mb8x8_mvd':
encoder/cavlc.c:262: internal compiler error: Segmentation fault
Please submit a full bug report,
with preprocessed source if appropriate.
See <URL:http://gcc.gnu.org/bugs.html> for instructions.

I can't compile ff_x264 with MinGW GCC 4.2.1.

[foxyshadis]

Originally Posted by fastplayer

Adding features especially experimental ones during a beta phase? Don't know if that's a good thing...

ffdshow does a few months of new features followed by a month or so of mostly testing & bugfixing, then a stablish 'beta' is released. (On a new svn branch, which is a bit pointless since I don't think any fixes have ever been backported to a stable branch, let alone recompiled and posted. It's there if someone wants to take the job, though.)

SBeaver, what splitter is being used? FlacSplitter from gabest?

[clsid]

No problems here building ff_x264 with MinGW GCC 4.2.1.

'FlacSplitter' doesn't exist SBeaver must be using either CoreFLAC or illiminable's filters.

[SBeaver]

I'm not sure I understand the question, but I understand enough to know that any answer I give will be irrelevant.

It seems clear though that there is a need for a dshow x64 flac parser.

[LigH]

Excuse my boldness...

Is this thread not the best place to report a bug?

[Inventive Software]

It is. Bug away.

[Leak]

Originally Posted by LigH

Is this thread not the best place to report a bug?

Sure it is. It's just that I for example have exactly zero knowledge about the HuffYUV code which AFAIK comes from libavcodec, so I can't really say where the problem is...

Have you tried using ffmpeg to create your HuffYUV file? If that fails the same way it's a bug in libavcodec.

[madshi]

Are you ffdshow devs aware that 8 channel input (doesn't matter which format the source file has) is borked right now? Any hope for a fix? For my needs it would already suffice to just extract 5.1 and trash the remaining 2 channels. But right now 8 channel input plays terribly distorted through ffdshow.

[LigH]

Well - ffdshow has a VfW codec, so I can use it for screen capturing; using ffmpeg for it might be complicated...

I will try to create a few artifical videos, and reply later. Tomorrow is german national holiday, that means a little free time.

[haruhiko_yamagata]

Preview version of YCbCr <-> RGB conversion options documentation (not executable, patches). Any suggestions, fixes are wellcome.
http://ffdshow-tryout.sourceforge.ne...cspOptions.htm

[vlada]

What about dynamically adjusting the range? Start with 16-235 (Rec mode and if in the movie appear values lower/higher then 16/235, change the mode to PC to avoid discarding of information. Maybe it would be only useful to check first couple seconds a then decide which mode to use.

[SBeaver]

Originally Posted by vlada

What about dynamically adjusting the range? Start with 16-235 (Rec mode and if in the movie appear values lower/higher then 16/235, change the mode to PC to avoid discarding of information. Maybe it would be only useful to check first couple seconds a then decide which mode to use.

Isn't that allready available in the levels tab?
Anyway I hope these new features come with some good presets for those who don't really know what numbers to set.
I have a hard time getting the levels right and if it isn't right I get a whole lot of blocking because of it.
Coreavc (and also haali renderer) has a lot simpler settings but they usually turn out better. This is regarding pc/tv levels only not the 709. 609. thing.

[foxyshadis]

Originally Posted by haruhiko_yamagata

Preview version of YCbCr <-> RGB conversion options documentation (not executable, patches). Any suggestions, fixes are wellcome.
http://ffdshow-tryout.sourceforge.ne...cspOptions.htm

Don't forget to mention that if a video looks washed out, with a grey picture, it should be set on PC to compensate, even though it's the renderer's fault.

[haruhiko_yamagata]

Originally Posted by vlada

What about dynamically adjusting the range? Start with 16-235 (Rec mode and if in the movie appear values lower/higher then 16/235, change the mode to PC to avoid discarding of information. Maybe it would be only useful to check first couple seconds a then decide which mode to use.

Seb's BTB&WTW level filter is merged this time.
http://sourceforge.net/tracker/index...41&atid=867362

[haruhiko_yamagata]

Originally Posted by foxyshadis

Don't forget to mention that if a video looks washed out, with a grey picture, it should be set on PC to compensate, even though it's the renderer's fault.

Do you mean RGB -> YUV conversion?
In YUV-> RGB conversion, PC mode looks more washed out.

[foxyshadis]

Originally Posted by haruhiko_yamagata

Do you mean RGB -> YUV conversion?
In YUV-> RGB conversion, PC mode looks more washed out.

Ah, of course, my mind was elsewhere. Looking at the user interface, I was wondering if maybe it should be:

ITU-R BT.601 (DVD Default)
ITU-R BT.709 (HD Default)

Standard (Y 16-235, etc.)
Less contrast (or full range?) (Y 0-255, etc.)
Custom:

And since there's extra space, as long as you're making these matrices would it be hard to add SMPTE 240M and FCC as well? Their definitions can be found in the colormatrix source (colormatrix.h).

[yesgrey]

Originally Posted by haruhiko_yamagata

Preview version of YCbCr <-> RGB conversion options documentation (not executable, patches). Any suggestions, fixes are wellcome.

It looks very good.

I think it would be better saying:
BT.601: almost all SDTV and DVD
BT.709: almost all HDTV, HD-DVD and Blu-Ray

I know it's almost the same, but it's easier to who doesn't know.

Good work.

[yesgrey]

Originally Posted by foxyshadis

And since there's extra space, as long as you're making these matrices would it be hard to add SMPTE 240M and FCC as well? Their definitions can be found in the colormatrix source (colormatrix.h).

The values in colormatrix appear to be slightly off... the BT.709 were.

[zambelli]

Originally Posted by haruhiko_yamagata

Preview version of YCbCr <-> RGB conversion options documentation (not executable, patches). Any suggestions, fixes are wellcome.
http://ffdshow-tryout.sourceforge.ne...cspOptions.htm

Somebody more knowledgable of the ITU specs can correct me, but I was under the impression the BT.601/709 specs define both chroma and luma mapping. If a video is being converted from YUV to RGB using BT.601 or BT.709, the luma conversion scale is already defined by that specification, so IMO it's incorrect to say a conversion is *both* BT.601/709 and PC because by definition PC (full range) mapping doesn't follow the BT.601/709 spec. No?

[zambelli]

Originally Posted by foxyshadis

And since there's extra space, as long as you're making these matrices would it be hard to add SMPTE 240M and FCC as well? Their definitions can be found in the colormatrix source (colormatrix.h).

I believe SMPTE 240M is being largely deprecated in favor of BT.709 because it preceded the ITU standard by just a few years and is similar enough that most people wouldn't care.

[haruhiko_yamagata]

Originally Posted by madshi

Are you ffdshow devs aware that 8 channel input (doesn't matter which format the source file has) is borked right now? Any hope for a fix? For my needs it would already suffice to just extract 5.1 and trash the remaining 2 channels. But right now 8 channel input plays terribly distorted through ffdshow.

I know. Just don't have time to fix it.

[haruhiko_yamagata]

Originally Posted by foxyshadis

And since there's extra space, as long as you're making these matrices would it be hard to add SMPTE 240M and FCC as well? Their definitions can be found in the colormatrix source (colormatrix.h).

OK, it's not too hard.
Kr=0.2122 Kg=0.7013 Kb=0.0865 for SMPTE 240M. Is this correct?
What is FCC used for?

[haruhiko_yamagata]

Originally Posted by yesgrey3

It looks very good.

I think it would be better saying:
BT.601: almost all SDTV and DVD
BT.709: almost all HDTV, HD-DVD and Blu-Ray

Thank you. It looks better.

[fastplayer]

FWIW, "mostly" sounds a bit better than "almost all".

[foxyshadis]

Originally Posted by haruhiko_yamagata

What is FCC used for?

NTSC analog video. Zambelli's right, though, it and 240M are close enough to BT.601 and BT.709 respectively that it's kind of pointless to include them after all.

[LigH]

Back to ffdshow's HuffYUV:

According to my tests, used as converting video codec (without much pressure, "streamingly" called) it works fine.

What does not work reliably, is using the codec for capturing purposes (when timing matters). It might be too slow when there are more expensive tasks running, or there is a different calling technique. Probably the next frame already has to be encoded while the last frame is not yet finished.

I have no TV card here, I can only simulate it using a screen capturer (HyperCam or similar). More tests with matching equipment could be necessary.

[haruhiko_yamagata]

Originally Posted by zambelli

Somebody more knowledgable of the ITU specs can correct me, but I was under the impression the BT.601/709 specs define both chroma and luma mapping. If a video is being converted from YUV to RGB using BT.601 or BT.709, the luma conversion scale is already defined by that specification, so IMO it's incorrect to say a conversion is *both* BT.601/709 and PC because by definition PC (full range) mapping doesn't follow the BT.601/709 spec. No?

I think you are right. But because some video clips have Y below 16 and/or over 235, someone wants to take it.
MJPEG and JPEG (used in grab filter) use so called PC601 in most cases. ffdshow had the code to automatically convert the color space using this information from stream. The code was buggy and did not work though. With my patch, MJPEG will use proper color space. The setting from dialog will be ignored in this case.

[haruhiko_yamagata]

Here is the patch that implements YCbCr <-> RGB conversion options. Seb.26's level filter is merged.
http://sourceforge.net/tracker/index...41&atid=867362

[haruhiko_yamagata]

Originally Posted by foxyshadis

Standard (Y 16-235, etc.)
Less contrast (or full range?) (Y 0-255, etc.)
Custom:

Thank you. It's better. I'll use Standard, Full range and Custom.

[madshi]

Originally Posted by haruhiko_yamagata

I know. Just don't have time to fix it.

Thanks for your reply. I'm glad that the bug is at least known...

[_xxl]

Here is the patch that implements YCbCr <-> RGB conversion options. Seb.26's level filter is merged.

Bin:
http://rapidshare.com/files/60036197...3_xxl.exe.html

[zambelli]

Originally Posted by haruhiko_yamagata

I think you are right. But because some video clips have Y below 16 and/or over 235, someone wants to take it.
MJPEG and JPEG (used in grab filter) use so called PC601 in most cases. ffdshow had the code to automatically convert the color space using this information from stream. The code was buggy and did not work though. With my patch, MJPEG will use proper color space. The setting from dialog will be ignored in this case.

As a general rule of thumb, decoders should never be doing luma remapping but should instead pass untouched YUV video to the renderer along with correct metadata. Of course, Ffdshow is somewhat of an exception to the rule because it's also a video processor so in this case it's an acceptable practice - but one that should be used by caution, due to the number of "moving parts" in a video rendering pipeline.
In DirectShow the color matrices are specified using VideoInfoHeader2 and DXVA_ExtendedFormat. I wish I could tell you for sure whether the latter can be used even when DXVA decoding isn't used, but if you're interested in implementing it, you could try emailing askdshow@microsoft.com for exact information.

Thank you. It's better. I'll use Standard, Full range and Custom.

"16-235" RGB is typically referred to as "Studio RGB", whereas the "0-235" RGB is referred to as "PC" or "Computer RGB". These are the names used in Charles Poynton's "Digital Video and HDTV Algorithms and Interfaces" book (ch.25).

[foxyshadis]

Originally Posted by zambelli

As a general rule of thumb, decoders should never be doing luma remapping but should instead pass untouched YUV video to the renderer along with correct metadata. Of course, Ffdshow is somewhat of an exception to the rule because it's also a video processor so in this case it's an acceptable practice - but one that should be used by caution, due to the number of "moving parts" in a video rendering pipeline.

In this case it's the RGB conversion that's meant, ie, it is actually being internally signalled and converted properly in the RGB converter, regardless of its setting, rather than remapped during decode. There is a FF_CSP_FLAGS_YUV_JPEG & PIX_FMT_YUVJ420P (plus 4:2:2, 4:4:4, and 4:4:0 variations). I'll grant that calling fullscale formats 'jpeg' probably isn't the greatest idea, but they are the 90% usage case.

It's not externally signaled though.

In DirectShow the color matrices are specified using VideoInfoHeader2 and DXVA_ExtendedFormat. I wish I could tell you for sure whether the latter can be used even when DXVA decoding isn't used, but if you're interested in implementing it, you could try emailing askdshow@microsoft.com for exact information.

There was a patch to do this with VIH2, afaik, but renderers mostly ignored it, which is why it isn't presently used.

[Wilbert]

The values in colormatrix appear to be slightly off... the BT.709 were.

Could you post the values used in colormatrix and the ones i posted in this thread (a few posts back; quoted from avisynth.org)?

Somebody more knowledgable of the ITU specs can correct me, but I was under the impression the BT.601/709 specs define both chroma and luma mapping. If a video is being converted from YUV to RGB using BT.601 or BT.709, the luma conversion scale is already defined by that specification, so IMO it's incorrect to say a conversion is *both* BT.601/709 and PC because by definition PC (full range) mapping doesn't follow the BT.601/709 spec. No?

Yes, that's correct (at least in theory). There are dvds (jap. anime for example) which use the full YCbCr [0,255] range, and i thought most of them were BT.601?

[yesgrey]

Originally Posted by Wilbert

Could you post the values used in colormatrix and the ones i posted in this thread (a few posts back; quoted from avisynth.org)?

In Colormatrix is GBR order (file Colormatrix.h):

  +0.7152, +0.0722, +0.2126, // Rec.709 (0)
  -0.3850, +0.5000, -0.1150,
  -0.4540, -0.0460, +0.5000,

Charles Poynton's ColorFAQ (the same in the avisynth link):

RGB ITU.BT-709 HDTV -> YUV
kr = 0.2125; kg = 0.7154; kb = 0.0721;

Matrix should be:

 0.2125   0.7154   0.0721
-0.114506  -0.385494   0.5000
 0.5000   -0.454222   -0.045778

[Thunderbolt8]

since x/h264 multiple core acceleration is about to come soon, would it also be possible to do the same for ffdshow using VC-1 via libavcodec? I know that its already supported when using the WMVideo Decoder WMO, but ffdshow has a slight edge over that regarding quality and with ffdshow im also able to apply real time sharpening, which is useful when viewing native 1080p on a 720p screen/resolution. such an addition, in combination with that h264 speedup would make ffdshow even better suited for hddvd/blueray remuxes.

[fastplayer]

Originally Posted by Thunderbolt8

since x/h264 multiple core acceleration is about to come soon, would it also be possible to do the same for ffdshow using VC-1 via libavcodec?

Post your request here:
http://ffmpeg.mplayerhq.hu/mailinglists.html

[Thunderbolt8]

where exactly there? dont seem to find a kind of forum or email adress

[Leak]

Originally Posted by Thunderbolt8

where exactly there? dont seem to find a kind of forum or email adress

http://lists.mplayerhq.hu/mailman/listinfo/ffmpeg-user

[haruhiko_yamagata]

Originally Posted by foxyshadis

There was a patch to do this with VIH2, afaik, but renderers mostly ignored it, which is why it isn't presently used.

Does EVR work with VIH2 color primaries information? If it works, it's worth implementing.

[Kado]

Originally Posted by _xxl

Bin:
http://rapidshare.com/files/60036197...3_xxl.exe.html

This build only has the YCbCr <-> RGB conversion options patch and not the h264 multi-core patch right? Hope to see the multi-core patch integrated soon in the try-outs.

[Wilbert]

In Colormatrix is GBR order (file Colormatrix.h):

  +0.7152, +0.0722, +0.2126, // Rec.709 (0)
  -0.3850, +0.5000, -0.1150,
  -0.4540, -0.0460, +0.5000,

Charles Poynton's ColorFAQ (the same in the avisynth link):

RGB ITU.BT-709 HDTV -> YUV
kr = 0.2125; kg = 0.7154; kb = 0.0721;

Matrix should be:

 0.2125   0.7154   0.0721
-0.114506  -0.385494   0.5000
 0.5000   -0.454222   -0.045778

no.

damn:
[1] reports kr = 0.2125, kg = 0.7154, kb = 0.0721
[2] and [3] report kr = 0.2126, kg = 0.7152, kb = 0.0722


[1] MPEG-2_13818-2.pdf (page 56)
[2] ITU-R_BT.709.pdf (page 3)
[3] Charles Poynton's ColorFAQ (section 8)

Since the mpeg-2 specs (in [1]) are referring to Rec.709, i guess they are wrong.

So the correct values are "kr = 0.2126, kg = 0.7152, kb = 0.0722" and avisynth.org should be corrected.

[Wilbert]

@haruhiko_yamagata,

The correct formulas for YCbCr [16,235] <-> RGB [0,255] (Rec.709) are:
Code:

R=1.164*(Y-16)+1.792*(V-128)
G=1.164*(Y-16)-0.2129*(U-128)-0.5326*(V-128)
B=1.164*(Y-16)+2.113*(U-128)

Thus the corrected values are:

R = 1.1644*(Y-16) + 1.7927*(V-128)
G = 1.1644*(Y-16) - 0.2132*(U-128) - 0.5329*(V-128)
B = 1.1644*(Y-16) + 2.1124*(U-128)

[yesgrey]

Wilbert,

If you see Charles Poynton's ColorFAQ.pdf file, question 9, the coefficients are what I said. Now which Poynton's coefficients are correct?

Maybe he have saw [2] and corrected it only in the html version of his ColorFAQ...

[yesgrey]

haruhiko,
going a little back at the custom matrix request...
I have went through the other option of using avisynth for it. Right now, I have it working the way I described here.

I know the biggest issue was the rewriting of a big portion of code in ffdshow, due to the need of using all matrix coefficients.
So, why not using the code in avisynth 2.6 for the YUV<->RGB conversion? It already uses all matrix coefficients. You can see here my discussion with IanB (one of avisynth developers) about it.

The work already done is great, and I believe it's almost all that most people need, but if the custom matrix could be added with a little extra work, it would be nice.

[iron2000]

The version number in the "About" screen(using clsid's build) has remained unchanged at 1471.

Not a major problem, just want to inform.

[morphx2]

Can anyone share with me some good hd-dvd settings to use with ffdshow tryout? I am totally lost on what options to pick

[Rash]

Originally Posted by zambelli

As a general rule of thumb, decoders should never be doing luma remapping but should instead pass untouched YUV video to the renderer along with correct metadata.

Well, either EVR ignores these metadata or absolutely no decoder supports them (and that includes Vista WMV DMO Decoder). I have never got 0-255 from videos using WMP11 or MPC with EVR. In fact, I am really disappointed with videos on Windows Vista. Fortunately we have Haali Renderer or EVR Custom to do the job. What is interesting, though, is that Windows Media Center plays the videos fine, with correct colors all time. Maybe it doesn't use EVR, I don't know.

[2ge]

Hello all,

FYI I just posted Download/Upload subtitles with OpenSubtitles.org using ffdshow (+poll question). Maybe it interests you and we can talk about it. I am not sure, where correct URL is for ffdshow talking, so I am trying here too - so we can have more oppinions.

Thank you.

[haruhiko_yamagata]

Originally Posted by zambelli

"16-235" RGB is typically referred to as "Studio RGB", whereas the "0-235" RGB is referred to as "PC" or "Computer RGB". These are the names used in Charles Poynton's "Digital Video and HDTV Algorithms and Interfaces" book (ch.25).

OK, "Studio" would be more formal, but it's just a dialog. "Standard" is more familiar in my opinion.

[haruhiko_yamagata]

Originally Posted by yesgrey3

I know the biggest issue was the rewriting of a big portion of code in ffdshow, due to the need of using all matrix coefficients.
So, why not using the code in avisynth 2.6 for the YUV<->RGB conversion? It already uses all matrix coefficients. You can see here my discussion with IanB (one of avisynth developers) about it.

OK, it shouldn't be too hard. But I have something to do before it, maybe later if I have time.

[yesgrey]

Originally Posted by haruhiko_yamagata

OK, it shouldn't be too hard. But I have something to do before it, maybe later if I have time.

Ok, no hurry. It's just that would be great to have all the functionallity in ffdshow. If you decide to do it let me know so we can discuss some of the interface details about it.

Thanks.

[Dr Pizza]

Can anyone suggest a reason why DirectShow is preferentially using "AVI Decompressor" (VfW wrapper) to decode video using ffdshow, even though ffdshow has a higher merit?

If I add ffdshow to the filter graph manually, it works just fine... it's just the filter graph manager won't add it unless I disable VfW.

[Wilbert]

Wilbert,

If you see Charles Poynton's ColorFAQ.pdf file, question 9, the coefficients are what I said. Now which Poynton's coefficients are correct?

Maybe he have saw [2] and corrected it only in the html version of his ColorFAQ...

Yes, that's a good one. I assume those given in the Rec.709 standard are correct. Perhaps the best is to send him an e-mail.

It's also possible that the coefficients are changed in Rec.709 (you can find multiple revisions of the standard on the ITU site). I will try to get the first version and look up the coefficients. As a reference to those coefficients, it gives the following note: "The coefficients for the equations have been calculated following the rules laid down in SMPTE RP177-1993." Perhaps it's possible to get that document somewhere.

[yesgrey]

Well, the pdf version is from 1997 and the html and txt version are from 2006. Maybe are both correct. In 1997 those were the coefficients and since 2006 the coefficients have been slightly changed?... I will try mailing him about it.

Edit: mail sent.

[Kado]

Question regarding the new RGB conversion... I use haali renderer with bt.709, I should use the same spec in ffdshow right? Or that does only affect if the output is RGB? (I generally use YUV2 and not RGB32).

[yesgrey]

The bt.709 or bt.601 and the PC or TV levels options in haali renderer only affects the image when you input YUY2 to it, because it defines the coefficients used for the YUV->RGB selection.

When you convert YUV->RGB in ffdshow, haali renderer will not need to convert it to RGB, so does not matter what you select in bt or in the levels section.

Basically you will decide where in your PC you want to perform the YUV->RGB conversion. In your CPU (ffdshow) or in your GPU (haali renderer).

[haruhiko_yamagata]

Originally Posted by Kado

Question regarding the new RGB conversion... I use haali renderer with bt.709, I should use the same spec in ffdshow right? Or that does only affect if the output is RGB? (I generally use YUV2 and not RGB32).

As for ffdshow settings, please read the latest document. I have added

These settings are used only when ffdshow is doing YCbCr <-> RGB conversion. In other words, if you are playing mpeg video and the output color space is YV12 or YUY2, these settings won't be used.

[Kado]

Hum, guess I'll put my GPU to some use to save some CPU time, or maybe not! Thanks for the tips guys.

[Joniii]

I have some problems with subtitles in ffdshow tryouts. I would like to use ffdshow for 1080p .ts files, but haven't quite get it working.

I have CoreAVC decoding h.264 and MS VC-1 decoding VC-1.

When I install ffdshow, I only select divx, xvid and subtitles. Then from the ffdshow settings codec page I select Raw video -> all supported and enable in WMP11. Now ffdshow loads subtitles for H.264 and VC-1 movies but the picture is screwed, I don't really know how to describe it, it has lots of horizontal lines and it shows image three times side by side. When twiddling with process whole image check box in the subtitles page, everithing worked ok couple of times but when I start WMP again then picture is a mess again.

Is there anything I can do in the ffdshow settings to fix this or something you could fix in the code?

[Inventive Software]

I fancy a real challenge. How would people think about me adding *some* minimal DXVA support to ffdshow?

[haruhiko_yamagata]

Originally Posted by Joniii

I have some problems with subtitles in ffdshow tryouts. I would like to use ffdshow for 1080p .ts files, but haven't quite get it working.

I have CoreAVC decoding h.264 and MS VC-1 decoding VC-1.

When I install ffdshow, I only select divx, xvid and subtitles. Then from the ffdshow settings codec page I select Raw video -> all supported and enable in WMP11. Now ffdshow loads subtitles for H.264 and VC-1 movies but the picture is screwed, I don't really know how to describe it, it has lots of horizontal lines and it shows image three times side by side. When twiddling with process whole image check box in the subtitles page, everithing worked ok couple of times but when I start WMP again then picture is a mess again.

Is there anything I can do in the ffdshow settings to fix this or something you could fix in the code?

It should be a bug but I cannot reproduce.
What is the subtitle (srt/ssa/ass/vobsub...)?
Is AviSynth checked?
What is the output color space?

[Joniii]

Originally Posted by haruhiko_yamagata

It should be a bug but I cannot reproduce.
What is the subtitle (srt/ssa/ass/vobsub...)?
Is AviSynth checked?
What is the output color space?

Avisynth is disabled, subtitle is .srt. As for color space I have no idea, whatever ffdshow uses by default. I'm using latest nightly build.

Other info I can think of:
OS is Vista Home Premium (clean install yesterday), I installed Only CoreAVC 1.5, Haali Media Splitter 1.7.189.11, AC3Filter 1.46 and ffdshow (Divx,Xvid,Mpeg2 and subtitles). Im using 1920x1080@60Hz with newest catalyst drivers into 55" telly (DVI->HDMI).

I just noticed that Media Player Classic is not affected by this, only WMP11 and Media Center (I ran MplayerC with all it's internal codecs/filters disabled so that it used ffdshow).

[LigH]

@ I.S.:

I often wondered if implementations need to be so GPU specific. Some DVD players (from my experience) only work well on ATI, others only on nVidia... If your "some" means a rather independent, generic level - sure, try it! And tell us up to which stage of the video playback your code would accellerate.

[Inventive Software]

It shouldn't have to be GPU specific. That's what DXVA was designed for; one API for all graphics solutions. The hardware has to be compatible, obviously, and is partly where I fall down at the first hurdle, but I'm gonna look at the specs for DXVA 1 and 2 and see what could be off-loaded to the GPU to speed things up. First target, I think, is post-processing and de-interlacing.

Ultimate target is MPEG-2, VC-1 and, wait for it, H.264 decoding on the GPU. But that is a LONG way off, think a couple thousand ffdshow-tryouts revisions away.

[haruhiko_yamagata]

Originally Posted by Joniii

I just noticed that Media Player Classic is not affected by this, only WMP11 and Media Center (I ran MplayerC with all it's internal codecs/filters disabled so that it used ffdshow).

Raw video processing in WMP11 and Media Center is experimental and not very stable. Currently it does not work for me. I hope albain can fix it.

[Dr Pizza]

Originally Posted by Inventive Software

It shouldn't have to be GPU specific. That's what DXVA was designed for; one API for all graphics solutions. The hardware has to be compatible, obviously, and is partly where I fall down at the first hurdle, but I'm gonna look at the specs for DXVA 1 and 2 and see what could be off-loaded to the GPU to speed things up. First target, I think, is post-processing and de-interlacing.

DXVA doesn't really work like that.

DXVA allows for a number of specific operations to be handled in hardware; bDXVA_Func can take 4 values, specifying one of { decompression, alpha blended data loading, alpha blended composition, resampling }. These values each have a number of sub-functions; for example, for decompression you can choose iDCT, deblocking, and macro block motion compensation, amongst other things.

This has two repercussions. One, DXVA must be integrated quite tightly into your MPEG-2 decoder; the operations it can accelerate are all low-level MPEG-2 operations, and the decoders must call the right accelerators at the right time. ffdshow's decoders don't appear to me to be readily amenable to that kind of modification; because ffdshow is essentially a DirectShow wrapper around a bunch of non-DirectShow codecs, the codecs themselves aren't really structured in such a way as to allow easy integration of DXVA support. If ffdshow had its own codecs designed from the ground up to be DirectShow filters then it might be in a slightly better position for this kind of work, but at the moment, it's really a bunch of pretty disparate C and C++ codebases, none of which are engineered for this kind of thing.

Two, DXVA isn't a mechanism for arbitrary programming of the video hardware to assist in video decoding. The available functions are all pre-defined (there's no (standard) way to provide extra decompression facilities, for example, which is why I believe things such as the PureVideo H.264 acceleration require special PureVideo-aware decoders--DXVA 1 doesn't have the right primitive functions), and I think for the most part optional (you can have DXVA hardware that can do some things but not others).

The only post-processing step that looks like it might fit reasonably nicely into DXVA is resizing; DXVA does have HW-accelerated resampling. I don't know, however, if it's appreciably faster than doing it on the CPU, nor do I know if it's widely supported. And it's quite possible that it's not sufficiently versatile to be a replacement for the post processing code anyway.

I think a more fruitful way to perform hardware-accelerated post-processing would be to use VMR9 renderless with a custom AP, and then use shaders to transform the video. I suspect that in practice even this would be quite limited, but it's still a more general mechanism than DXVA.

[SBeaver]

I would really like a new x64 build. The last one is from july and it lacks a lot of useful features.

[Dr Pizza]

Try http://rapidshare.com/files/61177074...4_x64.exe.html

[SBeaver]

Originally Posted by Dr Pizza

Try http://rapidshare.com/files/61177074...4_x64.exe.html

thanks a lot

There seems to be something that differs though, I'm not sure why but h264 decoding is somehow slowed down compared to the 32bit version I have.
On a particular video I get almost equal cpu usage, but the 64bit can't keep up at all and suffers from delays just as if there wasnt enough cpu power, but it stays around 60% for both builds (64 and 32)
Is this something to do with the new multicore enhancements?
Perhaps that patch was not applied to the build you gave me.
I have a dualcore opteron.
It's still nice to have the new color settings and all, great for all hd videos I watch.

[clsid]

The 64-bit version is slower because it doesn't use any of the fast handwritten assembly code that the 32-bit version has. That is due to the fact that there is no proper 64-bit version of mingw GCC yet. So MSVC is used which doesn't support the inline assembly stuff.

[akupenguin]

Originally Posted by Dr Pizza

The only post-processing step that looks like it might fit reasonably nicely into DXVA is resizing; DXVA does have HW-accelerated resampling. I don't know, however, if it's appreciably faster than doing it on the CPU, nor do I know if it's widely supported. And it's quite possible that it's not sufficiently versatile to be a replacement for the post processing code anyway.

Who does scaling on the CPU? Hardware scaling is part of any good renderer, it's not specific to DXVA. The only potential reason to use software scaling is if you want lanczos or something instead of whatever algorithm your GPU supports.

[SBeaver]

Originally Posted by clsid

The 64-bit version is slower because it doesn't use any of the fast handwritten assembly code that the 32-bit version has. That is due to the fact that there is no proper 64-bit version of mingw GCC yet. So MSVC is used which doesn't support the inline assembly stuff.

maybe generally, but in this case my cpu is not fully utilized so it's probably not using both cores.

[foxyshadis]

Originally Posted by akupenguin

Who does scaling on the CPU? Hardware scaling is part of any good renderer, it's not specific to DXVA. The only potential reason to use software scaling is if you want lanczos or something instead of whatever algorithm your GPU supports.

Maybe the 6-tap qpel step? I think it's stretching shaders a bit too far though, since even 2-tap bicubic is pushing them and no one's done 3/4-tap lanczos yet.

If not then yes, entirely true.

[Inventive Software]

I have read some of the DXVA articles on MSDN, and it says that it can off-load the de-interlacing, alpha-blending, ProcAmp picture control, motion compensation (which seems rather important), encryption and content protection through the Certified Output Protection Protocol. I'll conveniently forget the last 2 for now.

However, I have found that not all of the decoding process can be off-loaded, but it really does depend on display drivers supporting DXVA. So, when I get round to codecs, H.261 and MPEG-1 will be my first targets because they're simpler models, H.261 being even simpler than MPEG-1.

[akupenguin]

Originally Posted by foxyshadis

Maybe the 6-tap qpel step?

You can't do motion compensation on the GPU unless you do everything other than CABAC and IDCT on the GPU, because copying data back from GPU to main memory is slow.
This is no different from mpeg2/mpeg4asp. It's just that in mpeg2, "everything other than bitstream parsing and IDCT" leaves only motion compensation, whereas in h264 there's other stuff to be done.

I think it's stretching shaders a bit too far though, since even 2-tap bicubic is pushing them and no one's done 3/4-tap lanczos yet.

bilinear is 2-tap, bicubic is 4-tap, biquintic (h264) is 6-tap, lanczos-3 is 6-tap, lanczos-4 is 8-tap.
The reason linear/cubic/quintic are 2/4/6 taps is that it takes 2 points to specify a line, 4 points to specify an order 3 equation, 6 points to specify an order 5 equation, etc.
Nobody uses odd tap numbers (besides nearest-neighbor, which is 1-tap). When interpolating between x and x+1, which pixel would be the third, x+2 or x-1?

[Dr Pizza]

There seems to be something that differs though, I'm not sure why but h264 decoding is somehow slowed down compared to the 32bit version I have.
On a particular video I get almost equal cpu usage, but the 64bit can't keep up at all and suffers from delays just as if there wasnt enough cpu power, but it stays around 60% for both builds (64 and 32)
Is this something to do with the new multicore enhancements?
Perhaps that patch was not applied to the build you gave me.
I have a dualcore opteron.
It's still nice to have the new color settings and all, great for all hd videos I watch.

I didn't apply any patches specifically, it's just a recent SVN build, so I don't know what the state of play is on that front.

There's no inline asm support in MSVC++ for x64 builds (and no adequate Win64 gcc toolchain), so that would probably explain much of the slowdown. Any asm portions that are compiled externally (with yasm) should be included, however.

[Dr Pizza]

Originally Posted by Inventive Software

I have read some of the DXVA articles on MSDN, and it says that it can off-load the de-interlacing, alpha-blending, ProcAmp picture control, motion compensation (which seems rather important), encryption and content protection through the Certified Output Protection Protocol. I'll conveniently forget the last 2 for now.

Deinterlacing and procamp are both handled by VMR already; ProcAmp doesn't even have a user API, I don't believe (there's a driver-side portion, of course, for VMR to talk to, but nothing exposed to use it). i.e. as long as you feed interlaced video into VMR, VMR will use HW deinterlace; as long as you feed YUV into VMR, VMR will use HW YUV->RGB; as long as you use VMR ProcAmp to resize video, VMR will use HW resizing.

If you want to incorporate them earlier into the process (i.e. if you want to colour space convert/deinterlace/resize by hand) then you'll have to restructure the code accordingly, and you can't do everything because there's no (proper) way to call the ProcAmp acceleration.

However, I have found that not all of the decoding process can be off-loaded, but it really does depend on display drivers supporting DXVA. So, when I get round to codecs, H.261 and MPEG-1 will be my first targets because they're simpler models, H.261 being even simpler than MPEG-1.

DXVA's primary target is MPEG-2. That's what I'd start with.

[Dr Pizza]

Originally Posted by akupenguin

Who does scaling on the CPU?

libmplayer's swscale.c, which is used by the ffdshow resize postproc filter?

Hardware scaling is part of any good renderer, it's not specific to DXVA.

The way VMR and EVR do hardware scaling is through DXVA 1 and 2. I have no idea about third-party renderers, as I don't use any. I suppose they might use shaders and stuff instead of DXVA, though I don't see why they would; there's dedicated motion video hardware on (most) modern GPUs for a reason.

[foxyshadis]

Originally Posted by akupenguin

bilinear is 2-tap, bicubic is 4-tap, biquintic (h264) is 6-tap, lanczos-3 is 6-tap, lanczos-4 is 8-tap.

Ah, if you mean it that way then yes. Some academic papers will count the taps in only one direction rather than both, and I might just always think of it that way because those are the ones I found first. Is it more common to count both positive and negative?

Originally Posted by Dr Pizza

The way VMR and EVR do hardware scaling is through DXVA 1 and 2. I have no idea about third-party renderers, as I don't use any. I suppose they might use shaders and stuff instead of DXVA, though I don't see why they would; there's dedicated motion video hardware on (most) modern GPUs for a reason.

Overlay on a Matrox Mystique will give you hardware scaling too, and it'll probably be the same bilinear scaling as an 8800GTX. (It'd be pretty sad if a 10-year-old card used bicubic when modern ones won't, though.) DXVA has to use a different code path since the video isn't uploaded as a texture, but they're the same operation with the same near-zero cpu cost.

[Dr Pizza]

Overlay on a Matrox Mystique will give you hardware scaling too, and it'll probably be the same bilinear scaling as an 8800GTX. (It'd be pretty sad if a 10-year-old card used bicubic when modern ones won't, though.)

Wouldn't know; I haven't used overlay for many years.

DXVA has to use a different code path since the video isn't uploaded as a texture, but they're the same operation with the same near-zero cpu cost.

Yes, if you let the renderer do the resize and if the renderer HW accelerates (e.g. VMR/VMR 9 with DXVA, EVR with DXVA 2, and apparently overlay) then it's done with negligible CPU cost. But that's surely not the point of discussion, since that resizing is already HW accelerated, so there's no work to be done on that front.

I mean, the candidates are surely augmenting the decoders (especially MPEG-2, but also VC-1 and H.264, especially if DXVA 2 is targetted) with HW acceleration, and augmenting some of the postprocessors (especially deinterlace, resizing, and colour space conversion) with HW acceleration? The latter doesn't seem immediately beneficial for most scenarios, since it happens automatically when using a suitable renderer, and the former seems problematic due to the invasive nature of DXVA acceleration.

[akupenguin]

Originally Posted by foxyshadis

Some academic papers will count the taps in only one direction rather than both, and I might just always think of it that way because those are the ones I found first. Is it more common to count both positive and negative?

I've never heard it used as just one direction, other than in the names of lanczos-3 and lanczos-4. Counting all samples certainly seems more useful to me: While odd numbers aren't used for interpolation, they are used for other purposes. How would you call a 2nd derivative (-1 2 -1) filter? 1.5 tap?

[TheShadowRunner]

Hmm DXVA talk on a ffdshow thread...
I so wish someday i'll be able to use ffdshow for subs while offloading h264 decoding to the GPU..
Anyway, I would like to report a bug with 1080p/h264 for now ^^;

FFDshow builds 1508 and 1515 from CLSID are chocking quite heavily on this for some reason:
http://images.apple.com/movies/us/hd...apan_1080p.mov

"Queue" is disabled. Not a CPU problem as it reaches 40% usage max on this clip. Cyberlink decoder in soft mode and coreavc play it perfectly.

Intel2180/Geforce8500

The chocking/slowdown is very apparent (frame freezes) when the first trees appear in the clip.
Later,

TSR

[oddball]

All the latest nightly builds have broken AC3 transcoding (6 channel AAC to 6 channel AC3 resampled and spat out at 48000 over SPDIF).

Going back to the stable build always fixes it as long as I set it to 'Any filter' at the output stage. If I set it to DirectSound ffdshow crashes. If I set to 'Any filter' in the nightly builds I get no AC3 output at all and still get the crash on DirectSound.

[Kado]

@TheShadowRunner

After some tests here's my conclusions:
with cyberlink's both the specified and achieved frame rates are 15fps and no shuttering (buffers totally used)
with coreavc v1.5 frame rates are at 30fps (buffers totally used)
with ffdshow the specified frame rate is 30fps but at some stages the video drops to 15 frames with some stuttering but not frozen frames (only 3 of 32 buffers are used probably because it does not have enough speed to keep them fully used)

cpu usage goes from 30 to 65% with ffdshow/cyberlink and 30 to 85% with coreavc.
I changed the file termination from mov to mp4 to use haali splitter.
you can see additional av info from the haali osd if you use the haali splitter and renderer.

used MPC HC 1.0.10.0, latest haali, ffdshow 1485(multi-core patched)/1515, coreavc1.5,cyberlink h264 decoder. Pentium D 3.6GHz/6800gs 163.71

[Rash]

Just a question. How do I enable the colorspace (YUV->RGB) conversion on ffdshow, so that the new color options are available?

I tried enabling the "High quality YV12 to RGB conversion" in the Output panel but I still get YV12 output.

Thank you.

ffdshow-tryouts rev 1515 clsid build (Oct 5th).

[Leak]

Originally Posted by Rash

Just a question. How do I enable the colorspace (YUV->RGB) conversion on ffdshow, so that the new color options are available?

Uncheck all YUV formats on the output page so only the RGB format(s) you want to use is/are checked.

np: The Orb - Majestic (Mix 1) (U.F.Orb Remixes)

[zambelli]

Originally Posted by haruhiko_yamagata

Raw video processing in WMP11 and Media Center is experimental and not very stable. Currently it does not work for me. I hope albain can fix it.

Also keep in mind that any playback of ASF (.wmv, .asf) videos in WMP and MCE doesn't go through DirectShow - it goes through either WM Format SDK or Media Foundation API - so Ffdshow, as a DShow filter, would never even get a chance to be inserted into the graph.

[zambelli]

Originally Posted by haruhiko_yamagata

Does EVR work with VIH2 color primaries information? If it works, it's worth implementing.

EVR was definitely designed with advanced color metadata in mind - but I'm not 100% whether those interfaces are restricted to MF only or work in DirectShow too. If you send me a PM, I can give you my email address and I'd be happy to put you in touch with the EVR devs at Microsoft.

[haruhiko_yamagata]

Originally Posted by zambelli

Also keep in mind that any playback of ASF (.wmv, .asf) videos in WMP and MCE doesn't go through DirectShow - it goes through either WM Format SDK or Media Foundation API - so Ffdshow, as a DShow filter, would never even get a chance to be inserted into the graph.

OK. I'm trying to process 1080i mpeg2 file (.mpg), but no luck. It freezes for me. Smaller resolution works.

[haruhiko_yamagata]

Originally Posted by zambelli

EVR was definitely designed with advanced color metadata in mind - but I'm not 100% whether those interfaces are restricted to MF only or work in DirectShow too. If you send me a PM, I can give you my email address and I'd be happy to put you in touch with the EVR devs at Microsoft.

Thank you. I'll write the test code and try when I have time. I'm implementing 7.1ch audio processing and many things to do, so perhaps I cannot do that soon.

[madshi]

Originally Posted by haruhiko_yamagata

I'm implementing 7.1ch audio processing

Just to be safe: 6.1 is also a possible format (DTS Discrete). If you don't have time/fun to do a full implementation of 6.1 and 7.1, I'd be happy if you just dropped any channels over 5.1 as a quick fix. Thanks!

[haruhiko_yamagata]

Originally Posted by madshi

Just to be safe: 6.1 is also a possible format (DTS Discrete). If you don't have time/fun to do a full implementation of 6.1 and 7.1, I'd be happy if you just dropped any channels over 5.1 as a quick fix. Thanks!

OK, I'll implement 6.1ch too.

[ToS_Maverick]

ffmpeg/libavcodec supports decoding PAFF, when can we expect a ffdshow-tryout build supporting it? got some sattelite-footage to test

[fastplayer]

Originally Posted by ToS_Maverick

ffmpeg/libavcodec supports decoding PAFF, when can we expect a ffdshow-tryout build supporting it? got some sattelite-footage to test

Any build >= 1523 will do the trick.

[ToS_Maverick]

got 1524, but my football sample doesn't really work, got a lot of artifacts... at least it doesn't crash like before.

[clsid]

Interlaced pictures + spatial direct mode is not yet supported by libavcodec.

^ maybe your sample uses that.

[gunnerhkjp]

Originally Posted by Leak

Uncheck all YUV formats on the output page so only the RGB format(s) you want to use is/are checked.

np: The Orb - Majestic (Mix 1) (U.F.Orb Remixes)

I'm not very knowledgeable about this stuff, but when I did that, the black seems to be "proper" black instead of dark grey.

I have a Nvidia 7900gs and it had been bothering me for a while.

Is this the intended outcome?
Anyways, whoohoo!

[haruhiko_yamagata]

Originally Posted by gunnerhkjp

I'm not very knowledgeable about this stuff, but when I did that, the black seems to be "proper" black instead of dark grey.

I have a Nvidia 7900gs and it had been bothering me for a while.

Is this the intended outcome?
Anyways, whoohoo!

Yes, I use Nvidia 7900GS. Its YUV->RGB conversion is not reliable. VMR7 is totally broken. VMR9 is bit better but it always use full range. That's why I always use YUV->RGB conversion in ffdshow.

[haruhiko_yamagata]

This patch add 6.0/6.1/7.0/7.1ch audio processing.
Please test.

[LotharZ]

Originally Posted by haruhiko_yamagata

OK, confirmed. I can reproduce if I check "Obey positional settings from Subtitles page". If I uncheck it, I can watch the subtitles but its color is wrong.

It's not too hard to implement. The problem is that the dialog is too crowded.

Ok now the .idx/.sub files works properly but only if its external, If I mux it on a mkv file continue no working.

And I've found another sub that gives problems if its embedded, externally shows correctly all characters but internally for example "É","é" aren't shown.
If its useful, MPC doesn't have problems showing this embedded subs.

thx

http://massmirror.com/69c2dca1ade8b6d089c7ef3968e2cd9c.html

[ACrowley]

Originally Posted by haruhiko_yamagata

This patch add 6.0/6.1/7.0/7.1ch audio processing.
Please test.

How do i apply the Patch on latest ffdshow tryouts build ?

[fastplayer]

Originally Posted by ACrowley

How do i apply the Patch on latest ffdshow tryouts build ?

If you use TortoiseSVN:
http://tortoisesvn.tigris.org/TortoiseMerge.html#patch

[cc979]

has anybody tried to build a unicode version of ffdshow with gcc ?

is it actually possible ?

[ACrowley]

Originally Posted by fastplayer

If you use TortoiseSVN:
http://tortoisesvn.tigris.org/TortoiseMerge.html#patch

little bit more ffdshow specific ?! ...i never used a patch

[haruhiko_yamagata]

Originally Posted by haruhiko_yamagata

OK. I'm trying to process 1080i mpeg2 file (.mpg), but no luck. It freezes for me. Smaller resolution works.

Excuse me, it was the splitter. Haali's mpg-ps splitter was used. After I have installed gabest's splitter, it works fine.

[haruhiko_yamagata]

Originally Posted by cc979

has anybody tried to build a unicode version of ffdshow with gcc ?

is it actually possible ?

I have never succeeded.

[haruhiko_yamagata]

Originally Posted by ACrowley

little bit more ffdshow specific ?! ...i never used a patch

The patch is used for sourcecode. You have to compile ffdshow.

[haruhiko_yamagata]

Originally Posted by LotharZ

Ok now the .idx/.sub files works properly but only if its external, If I mux it on a mkv file continue no working.

And I've found another sub that gives problems if its embedded, externally shows correctly all characters but internally for example "É","é" aren't shown.
If its useful, MPC doesn't have problems showing this embedded subs.

thx

http://massmirror.com/69c2dca1ade8b6d089c7ef3968e2cd9c.html

Thanks, I'll look into it.

[Wilbert]

@yesgrey3,

Yes, that's a good one. I assume those given in the Rec.709 standard are correct. Perhaps the best is to send him an e-mail.

It's also possible that the coefficients are changed in Rec.709 (you can find multiple revisions of the standard on the ITU site). I will try to get the first version and look up the coefficients. As a reference to those coefficients, it gives the following note: "The coefficients for the equations have been calculated following the rules laid down in SMPTE RP177-1993." Perhaps it's possible to get that document somewhere.

I finally downloaded the first draft of Rec.709. Here the other values are used: kr = 0.2125, kg = 0.7154, kb = 0.0721.

It's not clear to me why these values are changed during the adaptation, although they some something about it in Annex I and II (those are removed in the final spec). Check your pm.

[_xxl]

This patch add 6.0/6.1/7.0/7.1ch audio processing.
Please test.

Bin:
http://rapidshare.com/files/62266102...3_xxl.exe.html

[rickardk]

Originally Posted by _xxl

Bin:
http://rapidshare.com/files/62266102...3_xxl.exe.html

I tried to duplicate the back speakers into side speakers...

I can't get the sound out of the right SIDE speaker using custom matrix (can't route any audio out of this one using custom matrix). I can get sound out of left SIDE speaker and by using channel swap I can get sound out of right SIDE speaker.

[Kado]

@haruhiko_yamagata

Thanks for the patch, don't forget to eventually add the meters along with the solo and mute buttons on the volume filter for the back center and side surround speakers.
Unfortunately I can't help test because I only have a 5.1 setup and an Audigy 2 ZS witch uses those 3 channel jacks for the surround speakers.
Also, I was wondering if you can add a filter that could load directsound audio processing filters from within ffdshow, when you have the time of course.

@_xxl

Thanks for the build.

[clsid]

Overlay + RGB output gives bad quality output for me. The picture is a bit blurred. No problems with VMR9 or Haali renderer.

It this a simple matter of crappy graphics hardware/drivers? Or is it a limitation of Overlay or perhaps an ffdshow bug?

[gunnerhkjp]

Originally Posted by haruhiko_yamagata

Yes, I use Nvidia 7900GS. Its YUV->RGB conversion is not reliable. VMR7 is totally broken. VMR9 is bit better but it always use full range. That's why I always use YUV->RGB conversion in ffdshow.

Thanks for the clarification. The picture seems a lot more "crispier" and vibrant.

I guess the side effect of using the conversion in FFDSHOW is that "modify full-screen video options" in forceware no longer works for me.

But who cares, the picture quality is billion times better
This should be default for anyone with Nvidia cards. I can't get over how good video looks now.

[Kado]

@clsid

That goes for me as well. I get a pixelrated image like it uses a different resizer for RGB32. For RGB 24 or lower I even get a different aspect ratio with overlay, haali and vmr7, for vmr9/evr it does not connect.
6800gs/163.75

[EDIT]VMR7/9 windowed looks like crap as well with RGB32[/EDIT]

[haruhiko_yamagata]

Originally Posted by rickardk

I tried to duplicate the back speakers into side speakers...

I can't get the sound out of the right SIDE speaker using custom matrix (can't route any audio out of this one using custom matrix). I can get sound out of left SIDE speaker and by using channel swap I can get sound out of right SIDE speaker.

Because I don't have 7.1ch speakers, so such bugs are likely to be overlooked. Thank you for report.

[haruhiko_yamagata]

Originally Posted by Kado

Also, I was wondering if you can add a filter that could load directsound audio processing filters from within ffdshow, when you have the time of course.

I don't know what directsound audio processing filters are. What kind of filters are available?

[check]

Originally Posted by clsid

Overlay + RGB output gives bad quality output for me. The picture is a bit blurred. No problems with VMR9 or Haali renderer.

It this a simple matter of crappy graphics hardware/drivers? Or is it a limitation of Overlay or perhaps an ffdshow bug?

Mostly it's an issue with particular driver/GPU combinations. The output can suffer highly in some images if you don't have 'High Quality conversion' checked too.

[clsid]

I already have "High quality YV12 to RGB conversion" enabled. It doesn't make a difference.

The picture looks a bit like it has been resized to a lower resolution and then upscaled again.

[haruhiko_yamagata]

Originally Posted by gunnerhkjp

But who cares, the picture quality is billion times better
This should be default for anyone with Nvidia cards. I can't get over how good video looks now.

One negative point is RGB conversion is CPU intensive. Current CPU is pretty fast, but still suffers to decode h.264 1080i.
I remember we heve talked about how to work around the nVidia-VMR9's bug (perhaps a info from coreAVC team?), but I cannot find the log. Does anyone know where it is? At that time, I didn't have nVidia's card, so I wasn't interested in it.

[Kado]

Originally Posted by haruhiko_yamagata

I don't know what directsound audio processing filters are. What kind of filters are available?

Something like iZotope Ozone 3. You can get it here.
Tried it with graphedit but refuses to connect even if in output setting is "connect to any filter"

[rickardk]

Originally Posted by haruhiko_yamagata

Because I don't have 7.1ch speakers, so such bugs are likely to be overlooked. Thank you for report.

You have done a stellar job...a much appreciated patch.
One thing thats missing is still 8 channel input support. I have a couple of blu-rays containing 8 channel lpcm thats not working.

I done some testing today and found out that everything works ok, it's just the wrong name on a couple of channels.
On the right side of the matrix:
BackC should be named SideR (on my soundcards)
And SideR sould be named SideL (on my soundcards)
SideL should be named BackC (on my soundcards)

I used my X-Meridian and my MOTU HD192 for testing.

[Koti]

ffdshow rev1525 Disabled SNOW support. It is far too slow, unstable and experimental to be used.

I have not experienced any snow decoding instability or speed issues and do use it for a number of files.
I can't comment on the encoder section since I have not used it. (I use mencoder)

If snow caused other reported problems within ffdshow or was to much work to enable , I understand.

[clsid]

The SNOW (bitstream) specification is not finalized yet. So anything you encode now may not be playable in the future.

I tried updating ffdshow with the latest SNOW code, but encoding just crashes. The old code gives bad output. Furthermore, it is slow as hell and the quality/size ratio is worse than other codecs. So it is pointless to use it in its current state.

[yesgrey]

@Wilbert,

Thanks for the file. It was interesting looking at it.

I also don't understand why they have changed the coefficients.
I have done some math just for curiosity. With Matlab I have calculated the matrices using both coefficients. You can see that with the last ones there is a slightly lower error when you convert from RGB->YUV->RGB. See this file:rgb709_compare.zip
I don't know if it was the reason, just a curiosity.

Can you find the SMPTE RP177-1993 documentation? Apparently it's there the rules for the coefficient's calculation.

[yesgrey]

@haruhiko,
In your new RGB conversion dialog, appears to be a bug in the Contrast Full range labelling.
Chroma should be 0-255 instead of 1-255, right?

[haruhiko_yamagata]

Originally Posted by yesgrey3

@haruhiko,
In your new RGB conversion dialog, appears to be a bug in the Contrast Full range labelling.
Chroma should be 0-255 instead of 1-255, right?

It's 1-255. 128-127=1.

[haruhiko_yamagata]

Originally Posted by rickardk

You have done a stellar job...a much appreciated patch.

Thank you.

One thing thats missing is still 8 channel input support. I have a couple of blu-rays containing 8 channel lpcm thats not working.

I need sample to fix this. Anyone with such skill, please help.
You could use LPCM decoder (converter) before ffdshow, though I don't know which filter to use.

I done some testing today and found out that everything works ok, it's just the wrong name on a couple of channels.
On the right side of the matrix:
BackC should be named SideR (on my soundcards)
And SideR sould be named SideL (on my soundcards)
SideL should be named BackC (on my soundcards)

I think I have fixed it on my PC.

[haruhiko_yamagata]

Originally Posted by Kado

Something like iZotope Ozone 3. You can get it here.
Tried it with graphedit but refuses to connect even if in output setting is "connect to any filter"

Thanks, maybe in the future. I cannot promise though.

[madshi]

Originally Posted by haruhiko_yamagata

I need sample to fix this. Anyone with such skill, please help.

Can't test myself right now. But here's a sample which might help:

http://madshi.net/8.flac

It's an extract of a 7.1 LPCM track reencoded to 7.1 FLAC.

[haruhiko_yamagata]

Originally Posted by madshi

Can't test myself right now. But here's a sample which might help:

http://madshi.net/8.flac

It's an extract of a 7.1 LPCM track reencoded to 7.1 FLAC.

I muxed the file to .mka and ffdshow now plays the file (as a post processor of coreFLAC).
Could you convert the file to LPCM? I don't have such skill.

[madshi]

Originally Posted by haruhiko_yamagata

I muxed the file to .mka and ffdshow now plays the file (as a post processor of coreFLAC).
Could you convert the file to LPCM? I don't have such skill.

Here's the same sample as LPCM (same format as demuxed by xport) :

http://madshi.net/8.rar

[rickardk]

The channel mapping on lpcm (from blu-ray) is diffrent and I'm trying to get ffdshow audio to automaticly load a presets that takes care of the remapping through the swap channel function.

The problem is that ffdshow sees the lpcm in my ts-files (haali media splitter) as uncompressed. And you can't set decoder match on the preset auto load page to uncompressed. Any solution to this?

Is there anyway you can force Vista Media Center (windows media player 11) to use ffdshow on mp3-playback?
ffdshow takes care of mp3 encoded audio in video containers. But I would also like to use it on straight .mp3 audio files.
As I replaced my preamp its essential to use ffdshow audio to take care of lfe crossover and delays.

If I use graphedit ffdshow audio loads and takes care of my .mp3 audio files.

[Haali]

I'll add channel remapping code in the next release, no need to add special cases for now.

[rickardk]

Originally Posted by Haali

I'll add channel remapping code in the next release, no need to add special cases for now.

Great!
One feature that I'm missing in haali media splitter is to have a preferred audio stream. Or better like a top 3 list.
If I have a ts containing lpcm, dts, ac3 I would like to have lpcm choosen by default. If lpcm is missing dts would be my second choice.

[Chainmax]

I recently was made aware that ffmpeg can now decode DXA, is it able to encode to that format as well? If not, when would you expect it to be able to?

[Jeremy Duncan]

The DTS doesn't work on AC3 output using these settings:
Link

Any suggestions ?
This is using FFDshow _XXL 1524

Somebody else reported this, I have no way of testing this.
It works using 24 bit output though

[clsid]

Originally Posted by Chainmax

I recently was made aware that ffmpeg can now decode DXA, is it able to encode to that format as well? If not, when would you expect it to be able to?

This thread is not about ffmpeg.

ffdshow currently does not support for neither decoding nor encoding of DXA.

[clsid]

Originally Posted by rickardk

Is there anyway you can force Vista Media Center (windows media player 11) to use ffdshow on mp3-playback?
ffdshow takes care of mp3 encoded audio in video containers. But I would also like to use it on straight .mp3 audio files.

WMP and MCE don't use DirectShow for .mp3 playback anymore, so it is not possible to use ffdshow.

[thuan]

Thank you, yamagata-san for implementing the Seb's color level filter. I didn't have to use it till I buy an ATI HD2400XT this morning. Using it with VMR9 (either windowed or renderless) and non-DXVA (yes with DXVA decoder like Cyberlink the color is ok) decoder I have washed out video. Now I have to either use the level filter with VMR9 or use Haali renderer (or use ffdshow's YCbCr <-> RGB but I don't like it as it use more CPU). The funny thing is all onboard graphic cards I have encountered all output the right color level. Wonder when they'll get it right?

[rickardk]

Originally Posted by clsid

WMP and MCE don't use DirectShow for .mp3 playback anymore, so it is not possible to use ffdshow.

No way to force them to use DirectShow?

[haruhiko_yamagata]

Originally Posted by rickardk

The problem is that ffdshow sees the lpcm in my ts-files (haali media splitter) as uncompressed. And you can't set decoder match on the preset auto load page to uncompressed. Any solution to this?

Oh, that's what I tried yesterday. Uncompressed and LPCM cannot be set separately. I found another bug, "on sampling frequency match" doesn't seem to work.

[haruhiko_yamagata]

Originally Posted by Haali

I'll add channel remapping code in the next release, no need to add special cases for now.

It's a good news. Thank you for your work.

[clsid]

Originally Posted by rickardk

No way to force them to use DirectShow?

Not that I know of. Otherwise I would have already mentioned it.

[Daodan]

I tested latest builds by xxl to check how much improvement is there in multithreaded h.264 decoding (tried on sliced and nonsliced). And it doesn't look like any.
Is the only mt version the one xxl posted some pages back, or all his newer builds have that, as I'm confused right now...
Thanks.

[TheShadowRunner]

Originally Posted by Kado

@TheShadowRunner

After some tests here's my conclusions:
with cyberlink's both the specified and achieved frame rates are 15fps and no shuttering (buffers totally used)
with coreavc v1.5 frame rates are at 30fps (buffers totally used)
with ffdshow the specified frame rate is 30fps but at some stages the video drops to 15 frames with some stuttering but not frozen frames (only 3 of 32 buffers are used probably because it does not have enough speed to keep them fully used)

cpu usage goes from 30 to 65% with ffdshow/cyberlink and 30 to 85% with coreavc.
I changed the file termination from mov to mp4 to use haali splitter.
you can see additional av info from the haali osd if you use the haali splitter and renderer.

used MPC HC 1.0.10.0, latest haali, ffdshow 1485(multi-core patched)/1515, coreavc1.5,cyberlink h264 decoder. Pentium D 3.6GHz/6800gs 163.71

Kado, thank you very much for testing.
It shows that indeed there is a bug in ffdshow & h264 imo. (it's not using all the buffers, all the time, as it should??)
I'm experiencing the same problem with another MKV
filename: Freedom_Ep1_(2006)_[720p,HDDVD,x264,DTS]
At around 1 minute during the opening, huge stuttering or frame freezes with ffdshow.
Cyberlink in soft mode and coreavc are as smooth as can be on that sequence.
(using revision 1530, "queue" is disabled, "drop frame on delay" and skip h264 deblocking on delay" enabled or disabled don't seem to make any difference). CPU used max is 40% on my system during that sequence with ffdshow. (and with other filters that decode it smoothly btw)
It seems ffdshow is chocking on big bitrate jump or something..
See you,

TSR

[Leak]

Originally Posted by TheShadowRunner

another MKV
(unlicensed, so not warez don't worry)

You're kidding, right? No matter if something is licensed for some/your region or not, distributing content without the permission of the copyright holder is still 100% illegal.

np: The Orb - Ba'albeck (Orb Sessions Volume 2)

[TheShadowRunner]

well sorry then, i'll remove the link.
Since it's not licensed outside Japan, i thought it wouldn't be a problem.
See you,

TSR

filename: Freedom_Ep1_(2006)_[720p,HDDVD,x264,DTS]
At around 1 minute during the opening, huge stuttering or frame freezes with ffdshow.
Cyberlink in soft mode and coreavc are as smooth as can be on that sequence.
(using revision 1530, "queue" is disabled, "drop frame on delay" and skip h264 deblocking on delay" enabled or disabled don't seem to make any difference). CPU used max is 40% on my system during that sequence with ffdshow. (and with other filters that decode it smoothly btw)
It seems ffdshow is chocking on big bitrate jump or something..

That could have something to do with the DTS track. It has been observed in the past that full-bitrate DTS at 1536kbps causes stuttering with certain decoders and movie players, and using AC3Filter set to pass-through usually solved the stuttering.

[yesgrey]

@haruhiko,
The custom matrix for YUV<->RGB conversion is not needed anymore. I realized in my thread it does not work as I intended to. When I update my thread about it I will let you know.

Thanks anyway. The current YUV<->RGB conversion as you implemented should be enough.

[TheShadowRunner]

Originally Posted by honai

That could have something to do with the DTS track. It has been observed in the past that full-bitrate DTS at 1536kbps causes stuttering with certain decoders and movie players, and using AC3Filter set to pass-through usually solved the stuttering.

Honai, thanks for your input.
I'm not sure the DTS stream is responsible as I use FFDshow Audio decoder set to "libdts" for all 3 tests.
(CoreAvc+libdts, Cyberlink+libdts, FFDhowVideo+libdts)
Only the test with h264 decoded by FFDshow has the issue.
See you,

TSR

[Jeremy Duncan]

TheShadowRunner,

I've had somebody post that they had problems with the dts in FFDshow audio codec as well.

Maybe something needs to be adjusted to allow for dts through ac3 ?
Looking at your post, the problem with dts playback stems from the FFDshow video codec.

[Jeremy Duncan]

Originally Posted by Leak

* Start your AviSynth script with "#debug" as the first line, run DebugView, then re-open MPC and play your DVD for a bit, use "Save as..." in DebugView to save it's log and mail it to me (leak (at) gmx (dot) at)

TheShadowRunner,

Can you please follow this suggestion from leak and post the results.
I would, but I can't since I don't have the equipment or knowhow as a result.
Just play the dts using the FFDshow audio and video codecs, the dts that has the problem, and run debug.

[TheShadowRunner]

Originally Posted by Jeremy Duncan

TheShadowRunner,

Can you please follow this suggestion from leak and post the results.
I would, but I can't since I don't have the equipment or knowhow as a result.
Just play the dts using the FFDshow audio and video codecs, the dts that has the problem, and run debug.

hmm sorry Jeremy but
1. i don't use avisynth.
2. i don't use MPC
3. the source is not DVD

Besides my problem is not related to DTS, i don't think, but in the way ffdshow decodes h264.
o_O
See you,

TSR

[Jeremy Duncan]

TSR,

I see the problem you have is when you use dts in the ffdshow audio codec?
And when you use the ffdshow video codec, for h.264.

I have the same problem, but in the ffdshow video codec I use the mpeg2 dvd decoding.

Maybe the problem your seeing is related to the problem I'm seeing in some intrinsic way.

Ha ha ha.

I dunno.

[TheShadowRunner]

man, i dunno where you read that but no, my problem is NOT dts in ffdshow audio codec.

My problem is h264 stuttering with ffdshow video (when Cyberlink and Coreavc have no problems at all with the same h264 stream).
Sorry.
Later,

TSR

[Jeremy Duncan]

TSR,

You said you had dts set in the ffdshow audio codec.
This worked with the cyberlink video codec, the core avc video codec, but not the ffdshow video codec ?

I know somebody who is having trouble with the same combination of dts and the ffdshow audio and video codec.

Maybe there's a problem that can be fixed.

[Chainmax]

Originally Posted by clsid

This thread is not about ffmpeg.

ffdshow currently does not support for neither decoding nor encoding of DXA.

I apologize and will ask in the correct thread.

[Keepitsimple]

I dont know if this is the right thread.

I have a 60 or 50hz projector. 24fps film isnt smooth. Is it possible to fix with ffdshow?

[yesgrey]

Originally Posted by Keepitsimple

I dont know if this is the right thread.

I have a 60 or 50hz projector. 24fps film isnt smooth. Is it possible to fix with ffdshow?

With reclock you could fix it. It will accelerate the 24fps to 25fps and will resample the audio, to keep the pitch right.

Go here.
Use the 1.7 beta 4, is the best, but download the 1.6 to get the documentation.

If you need more help PM me. This is not the right thread for talking about reclock.

[TheShadowRunner]

keepitsimple, i don't think this is the right thread.
You want to look for a soft called "Reclock". (and pray your projector is able to display a multiple of 24hz (48hz, 72hz...) in order to have a judder-free playback.
See you,

TSR
edit: oh yesgrey answered in the meantime with the 25fps solution. ^^;

[haruhiko_yamagata]

Originally Posted by TheShadowRunner

It shows that indeed there is a bug in ffdshow & h264 imo. (it's not using all the buffers, all the time, as it should??)
I'm experiencing the same problem with another MKV

I'm not sure if this is a bug or not. At least this is a CPU usage problem. I guess the CPU usage simply come from high bit rate of the scene. I enabled queue, set queueCount (registry) 100, VMR7-YV12. Enabled OSD "Video delay" & "Queued samples". No delay with the effect of queue. Core2duo E6700.

The file is still too easy to find. If you want to reprot this kind of sample, please use PM.

[TheShadowRunner]

Haruhiko, thank you for testing.
Since the CPU isn't maxed out when queue is disabled, isn't this considered as a bug?
Shouldn't ffdshow, like CoreAVC and Cyberlink, play the sequence just fine WITHOUT the need for Queue to be enabled? (since CPU isn't even near maxed out)
This file exhibits the same problem:
http://images.apple.com/movies/us/hd...apan_1080p.mov
(this one is 100% OK i hope ^^)

Who is developing the h.264 portion of ffdshow?
Maybe I can report the problem to them?
See you,

TSR

[clsid]

The libavcodec code is developed by FFmpeg.

But before you report a bug to them you should first verify whether the same problem also occurs with FFplay.

Here are some win32 FFplay builds:
http://ffdshow.faireal.net/mirror/ffmpeg/
(unfortunely, these do not yet contain the new H.264 code)

[haruhiko_yamagata]

Originally Posted by thuan

Thank you, yamagata-san for implementing the Seb's color level filter. I didn't have to use it till I buy an ATI HD2400XT this morning. Using it with VMR9 (either windowed or renderless) and non-DXVA (yes with DXVA decoder like Cyberlink the color is ok) decoder I have washed out video. Now I have to either use the level filter with VMR9 or use Haali renderer (or use ffdshow's YCbCr <-> RGB but I don't like it as it use more CPU). The funny thing is all onboard graphic cards I have encountered all output the right color level. Wonder when they'll get it right?

I guess the DXVA filters set color space info through DXVA-API. But I don't know why ATI and nVidia use full range by default.

[haruhiko_yamagata]

Originally Posted by TheShadowRunner

Haruhiko, thank you for testing.
Since the CPU isn't maxed out when queue is disabled, isn't this considered as a bug?
Shouldn't ffdshow, like CoreAVC and Cyberlink, play the sequence just fine WITHOUT the need for Queue to be enabled? (since CPU isn't even near maxed out)
This file exhibits the same problem:
http://images.apple.com/movies/us/hd...apan_1080p.mov
(this one is 100% OK i hope ^^)

That one is OK .
One core is maxed out for some (probably high bit rate) scene. Use windows built-in task manager and increase the refresh rate.
CoreAVC also has high CPU usage for the scene ffdshow gets behind, so it's not a bug.

[haruhiko_yamagata]

I'm stuck fast to implementing 7.1ch HRTF (mixer-headphone). I've managed to avoid crashing by dropping side channels.
I need data for side channels, but because I don't know math, I cannot do this.
If anyone is interestead, please help.

How to generate these data:

1. You need the MIT Media Lab's KEMAR data, read it into an software
   capable of signal/time series analysis (like Mathematica, Matlab,
   ...)

2. Construct an equalizing FIR inverse filter by calculating the
   transfer function of the front, same side impulse response, then
   take 1 over the absolute magnitude.

3. Cut the poles in the inverse filter's transfer function
   specification by limiting the amplification to 2.5 (note, this
   number assumes that you have correct signal processing
   normalization of the Fourier transform).

4. Design the FIR inverse filter by calculating the inverse Fourier
   transform, then chopping the coefficients down to a reasonable
   number (N = 15 is used here), apply the Kaiser-Bessel window (alpha
   = 2 is used here).  Note the objective is remove the color bias
   only (as if you are using an equalizer), _not_ to do a full inverse
   filtering. (Note: beta = pi*alpha in other notation.)

   For N = 15, alpha = 2, you should get the following impulse
   response:

    0.001001558668605168,  0.00698457265741865,  0.040453643039829436,
    0.012230541722147855, -0.11939760844854072,  0.16468099899755967,
   -0.30297563073747436,   1.3140211791355982,  -0.30297563073747436,
    0.16468099899755967,  -0.11939760844854072,  0.012230541722147855,
    0.040453643039829436,  0.00698457265741865,  0.001001558668605168

5. Linearly convolve all KEMAR HRIR with this inverse filter.

6. Resample from 44.1 kHz sampling frequency to 48 kHz.
*/

/* Center front (-5 degree) - not 0 degree in order to create a clear
   front image from a finite distance */
float cf_filt[128] = {
   -0.00008638082319075036, 0.0003198059946385229,
   -0.0005010631339162132, 0.0011424741331126876,
   -0.001584220794688753, 0.001742715363246275,
   ...

[clsid]

@all, I would like to update the opening post. So if there are any known bugs (that have also been confirmed by others) that are not yet listed there, let me know.

@devs, if you have a ToDo list and want it to be listed in the opening post, let me know.

[Keepitsimple]

Originally Posted by TheShadowRunner

keepitsimple, i don't think this is the right thread.
You want to look for a soft called "Reclock". (and pray your projector is able to display a multiple of 24hz (48hz, 72hz...) in order to have a judder-free playback.
See you,

TSR
edit: oh yesgrey answered in the meantime with the 25fps solution. ^^;

Thank you TheShadowRunner and yesgrey3. I will check it out.

[Kado]

Regarding the h.264 shuttering... I have only tested with the bbc file and from my experience this is basically a speed issue.
My Netburst cpu just can't keep decoding fast enough with ffdshow, hence the buffers not totally used and the shuttering.
Cyberlink's decoder only outputs 15 frames anyway (from a video with 30 frames!).
Coreavc uses both cores and is faster than ffdshow even in single core cpus (I get about 90% total cpu utilization in the high bitrate parts and no shuttering).
For ffdshow since it's not fast enough, when the "heavy" parts come it will just drop frames.
It looks that you "Core2 people" don't have frame dropping so it's not a bug (right?).

@haruhiko_yamagata
"Mondai nai!" When I need (very rarely) those audio processing filters I use Winamp with Adapt-X plug-in (it does not work in ffdshow, I think it was blacklisted before because it crashed ffdshow) to load them into Winamp.

[haruhiko_yamagata]

Originally Posted by LigH

Back to ffdshow's HuffYUV:

According to my tests, used as converting video codec (without much pressure, "streamingly" called) it works fine.

What does not work reliably, is using the codec for capturing purposes (when timing matters). It might be too slow when there are more expensive tasks running, or there is a different calling technique. Probably the next frame already has to be encoded while the last frame is not yet finished.

I have no TV card here, I can only simulate it using a screen capturer (HyperCam or similar). More tests with matching equipment could be necessary.

Confirmed. Sorry for late reply.

[yesgrey]

haruhiko,
You can see here why the custom YUV<->RGB matrix is now useless for what I wanted. This kind of correction could be added to ffdshow, as another filter, but I believe the best and easiest way is doing it at the renderer level, because the shaders are better for that. Currently, it's limited to MPC and VMR9, but I will ask Haali if he could add it to his renderer to broaden player support...

[Nicholi]

@clsid
First reported by Leak here and mostly confirmed by me here.

Has anyone noticed any weirdness when ffvfw/ffdshow is being used multiple times? And windows of nearly all applications suddenly start behaving strangely until you close everything using ffdshow?

[LoRd_MuldeR]

Just for notice:
The "VfW configuration" shortcut in Startmenu doesn't work on 64-Bit Windows XP.

I throws the following error message:
Error loading ff_vfw.dll
The specified module could not be found.

That's the following "target" obviously doesn't work:
C:\WINDOWS\system32\rundll32.exe ff_vfw.dll,configureVFW

Using the full path to the DLL it works:
C:\WINDOWS\system32\rundll32.exe C:\WINDOWS\SysWOW64\ff_vfw.dll,configureVFW

Unfortunately it does not work that way:
C:\WINDOWS\system32\rundll32.exe %SYSDIR%\ff_vfw.dll,configureVFW


I don't know how to detect "C:\WINDOWS\SysWOW64" under x64 Windows, GetSystemDirectory() only returns "C:\WINDOWS\system32" ...

[foxyshadis]

This page, under installation directory requirements, explains how.

[Inventive Software]

@LoRd_MuldeR: Dear God, I haven't seen that font since Win98!

[_xxl]

Has anyone noticed any weirdness when ffvfw/ffdshow is being used multiple times? And windows of nearly all applications suddenly start behaving strangely until you close everything using ffdshow?

Can you please explain more?

[clsid]

Originally Posted by LoRd_MuldeR

Just for notice:
The "VfW configuration" shortcut in Startmenu doesn't work on 64-Bit Windows XP.

I throws the following error message:
Error loading ff_vfw.dll
The specified module could not be found.

That's the following "target" obviously doesn't work:
C:\WINDOWS\system32\rundll32.exe ff_vfw.dll,configureVFW

Using the full path to the DLL it works:
C:\WINDOWS\system32\rundll32.exe C:\WINDOWS\SysWOW64\ff_vfw.dll,configureVFW

Unfortunately it does not work that way:
C:\WINDOWS\system32\rundll32.exe %SYSDIR%\ff_vfw.dll,configureVFW

I can fix this easily in the install script.

Can you check if the following also works correctly?
Target = C:\WINDOWS\system32\rundll32.exe ff_vfw.dll,configureVFW
Start in = C:\WINDOWS\SysWOW64

[rickardk]

I have a big problem here...

I have a couple of HD DVDs in ts.
When playing them back in Vista Media Center (or media player 11) with microsofts vc-1 decoder they are quite jerky.

When I play them back with PowerDVD everything is smooth.

Using ffdshow with either one of the two included decoders its jerky again.

In ffdshow it says that framerate is 29.97 but I'm quite sure it should be 23.976.

The movies are 23.976fps and my graphic card outputs 23.976 Hz. My Pioneer display takes the 23.976 Hz signal and triple it to ~72 Hz.

When playing back VC-1 from blu-rays in ts ffdshow shows 23.976.

Is there any solution for playing them back in their original framerate?

[Daodan]

That doesn't really sound like a ffdshow problem, but more like a splitter problem. Of course, the first question would be: how did those contents of the evo end up in .ts container? Maybe you actually muxed it wrong, at 29.97.

[LoRd_MuldeR]

Originally Posted by clsid

I can fix this easily in the install script.

Can you check if the following also works correctly?
Target = C:\WINDOWS\system32\rundll32.exe ff_vfw.dll,configureVFW
Start in = C:\WINDOWS\SysWOW64

Yup, it does

BTW: Why is the "ff_vfw.dll" stored in the x64 system folder at all?

[rickardk]

Originally Posted by Daodan

That doesn't really sound like a ffdshow problem, but more like a splitter problem. Of course, the first question would be: how did those contents of the evo end up in .ts container? Maybe you actually muxed it wrong, at 29.97.

Both the evos and ts plays perfectly fine in PowerDVD.
But as I want my collection usable inside Vista Media Center I would love to see smooth playback with ffdshow.

[LotharZ]

When there will be a new binary file?
I want to test the vob/ssa fixes.

thx

[clsid]

Originally Posted by LoRd_MuldeR

Yup, it does

BTW: Why is the "ff_vfw.dll" stored in the x64 system folder at all?

SysWOW64 is actually the 32-bit system folder on 64-bit versions of Windows. The name is a bit confusing.

The regular system32 folder is the 64-bit system folder. It is still named system32 for backwards compatibility reasons.

[LoRd_MuldeR]

Originally Posted by clsid

SysWOW64 is actually the 32-bit system folder on 64-bit versions of Windows. The name is a bit confusing.

The regular system32 folder is the 64-bit system folder. It is still named system32 for backwards compatibility reasons.

[iron2000]

Is the 16-235->0-255 shader in MPC similar to the Full range RGB conversion in ffdshow?

I like the colours by the MPC shader better.

[mike7]

Feature Request:
Could you, please, if have time of course, add ability to display current preset on OSD?
Thanks for you great work.

[haruhiko_yamagata]

Originally Posted by LotharZ

When there will be a new binary file?
I want to test the vob/ssa fixes.

thx

Well, the vob problem is difficult. It requires more understanding of vobsub.

@clsid
Please move it (short term todo-1) to mid term.

[haruhiko_yamagata]

Originally Posted by mike7

Feature Request:
Could you, please, if have time of course, add ability to display current preset on OSD?
Thanks for you great work.

done as rev 1551.

[haruhiko_yamagata]

Originally Posted by haruhiko_yamagata

Yes, I use Nvidia 7900GS. Its YUV->RGB conversion is not reliable. VMR7 is totally broken. VMR9 is bit better but it always use full range. That's why I always use YUV->RGB conversion in ffdshow.

Sorry for OT, I updated the driver and managed to get proper output from VMR9-YUV by tweaking the nVidia control panel. From video & Television->Adjust video color settings, increase contrast and decrease brightness. I also had to increase RGB gamma for all channels. The bug can be fixed through the control panel, though the configuration is not easy for beginers...

[LoRd_MuldeR]

Does the rev1548 build by CLSID already include the "VfW configuration" shortcut fix?
If so, it doesn't work

The shortcut uses the following settings now:

Target = C:\WINDOWS\system32\rundll32.exe ff_vfw.dll,configureVFW
Start in = C:\WINDOWS\system32

Change "Start in" to "SysWOW64" and it does work ...

[clsid]

Well, it should be working. The install script was fixed in 1545.

The value of "Start in" written by the installer should be SysWOW64.

This means that either there is a bug in the installer software or there is something wrong with your Windows installation.

Here is a sample script that you can compile with InnoSetup:

[Setup]
AppName=Test
AppVerName=Test
AppId=innotest
DefaultDirName={tmp}
Uninstallable=no
OutputDir=.

[Code]
function InitializeSetup(): Boolean;
begin
Result := false;
// this should display SysWOW64 path on a 64-bit version of Windows
msgbox(ExpandConstant('{sys}'), mbInformation, mb_ok);
end;

[squid_80]

Originally Posted by LoRd_MuldeR

The shortcut uses the following settings now:

Target = C:\WINDOWS\system32\rundll32.exe ff_vfw.dll,configureVFW
Start in = C:\WINDOWS\system32

Change "Start in" to "SysWOW64" and it does work ...

To be proper the target should be c:\windows\syswow64\rundll32.exe as well.

[clsid]

I know. But that is not the issue. That rundll32.exe version is able to handle both 32/64-bit stuff afaik. The issue is that the {sys} constant in the install script resolves to the wrong folder.

[LoRd_MuldeR]

Originally Posted by clsid

I know. But that is not the issue. That rundll32.exe version is able to handle both 32/64-bit stuff afaik. The issue is that the {sys} constant in the install script resolves to the wrong folder.

Well the GetSystemDirectory() function definitely returns "C:\Windows\system32" here...




// EDIT

Could other "Windows XP Pro - x64 Edition" (aka Windows Server 2003 - Client Mod) users please test it:
http://mulder.dummwiedeutsch.de/temp/setup.exe

[squid_80]

Originally Posted by clsid

I know. But that is not the issue. That rundll32.exe version is able to handle both 32/64-bit stuff afaik. The issue is that the {sys} constant in the install script resolves to the wrong folder.

It's the wrong solution to the problem. You should be calling the correct rundll32.exe rather than setting the Start In directory. For example if you've got both x64 and x32 installed, running c:\windows\system32\rundll32.exe ff_vfw.dll,configureVFW from c:\windows\syswow64 will launch configuration of the x64 build, not x32.

GetSystemDirectory() will always return c:\windows\system32 in x64. GetSystemWow64Directory() is used to get the 32-bit system directory. I think the right way to do it is not put ff_vfw.dll in system32 at all, keep it in program files\ffdshow and point the shortcut there.

[haruhiko_yamagata]

Originally Posted by LoRd_MuldeR

Well the GetSystemDirectory() function definitely returns "C:\Windows\system32" here...




// EDIT

Could other "Windows XP Pro - x64 Edition" (aka Windows Server 2003 - Client Mod) users please test it:
http://mulder.dummwiedeutsch.de/temp/setup.exe

system32 here, too. Regardless of ArchitecturesInstallIn64BitMode=x64 or blank.

[LoRd_MuldeR]

Originally Posted by squid_80

I think the right way to do it is not put ff_vfw.dll in system32 at all, keep it in program files\ffdshow and point the shortcut there.

AFAIK, VfW Codecs are simply installed by putting the DLL into the "System32" directory. So if it will be placed in ffdshow's install folder, it won't won't be recognized by the Video Compression Manager - unless there is a second method to register "out of place" VfW Codecs via the Windows Registry somehow...

[squid_80]

Originally Posted by LoRd_MuldeR

AFAIK, VfW Codecs are simply installed by putting the DLL into the "System32" directory.

No they're not. A string key is made in the registry at HKLM\SOFTWARE\Microsoft\Windows NT\CurrentVersion\Drivers32 called vidc.XXXX where XXXX is the codec handler's FourCC. The string itself is the dll to use, optionally including the full path.

[LoRd_MuldeR]

Originally Posted by squid_80

No they're not. A string key is made in the registry at HKLM\SOFTWARE\Microsoft\Windows NT\CurrentVersion\Drivers32 called vidc.XXXX where XXXX is the codec handler's FourCC. The string itself is the dll to use, optionally including the full path.

If that behavior is the same on all Win32 versions, I think the ff_vfw.dll should definitely be put into the ffdshow install folder!

Then the shortcut could be done like:

Target = rundll32.exe C:\<Path to ffdshow>\ff_vfw.dll,configureVFW
Start in = <blank>

I think Rundll doesn't need a full path, as long as the path to ff_vfw.dll is explicit.
Windows will automatically find Rundll32.exe from %Path% ...

[clsid]

Guys, please ignore the rundll32.exe stuff. That is just cosmetics, not what goes wrong. Have a look at my test script. If {sys} doesn't give the correct directory, then it would still point to the wrong (but working) rundll32.exe even with a 'fixed' script.

The codec can indeed be placed everywhere. The drivers32 key should contain the full path, with the added note that the path should be in shortname form. But since placing codecs in the system folder is pretty much the standard in practice, some buggy apps may fail to properly recognize codecs that are placed elsewhere.

[LoRd_MuldeR]

Originally Posted by clsid

Guys, please ignore the rundll32.exe stuff. That is just cosmetics, not what goes wrong. Have a look at my test script. If {sys} doesn't give the correct directory, then it would still point to the wrong (but working) rundll32.exe even with a 'fixed' script.

Well, it seems {sys} does return the correct system directory! I guess it simply calls GetSystemDirectory(), which return "C:\WINDOWS\System32", even on x64 Windows. If you want to get "C:\WINDOWS\SysWOW64", then I think would you need a different placeholder (e.g. "{syswow64}") which calls GetSystemWow64Directory() instead. But if you would simply place the ff_vfw.dll in the ffdshow install folder, you can work with the full path to ff_vfw.dll easily. So that should solve the problem...

Target = rundll32.exe "C:\Program Files\ffdshow\ff_vfw.dll",configureVFW

I think most applications simply use Windows' Video Compression Manager. So as long as Windows is able to find the Codecs with full path, there should be no problem. If some buggy applications fail, then that's their problem. The most important application for 99,9% of your users is VirtualDub anyway...

[clsid]

Relocating ff_vfw.dll shouldn't be needed. Putting it in the ffdshow app folder can also be a problem for people who have shortname functionality disabled on their NTFS partition. That is rare, but possible.

[LoRd_MuldeR]

Originally Posted by clsid

Putting it in the ffdshow app folder can also be a problem for people who have shortname functionality disabled on their NTFS partition. That is rare, but possible.

Wouldn't that cause identical problems with ffdshow.ax located in the ffdshow install folder?

[clsid]

No, because shortname is only needed for ff_vfw.dll


Edit: I think I may have found the cause.
Writing files to {sys} will cause them to end up in SysWOW64, because of the WOW64 file redirection done by Windows.
When no files are concerned {sys} just resolves to the system32 dir. Although the Inno docs are ambiguous about this.

I will commit a fixed script later. Going to watch F1 first now.

[LoRd_MuldeR]

Well, if you want to keep ff_vfw.dll in "SysWOW64" folder on Win x64, then you will need to find a way to detect the location of "SysWOW64" from InnoSetup. I have not looked into the docs yet, but they should have a placeholder for the "SysWOW64" directory. If not, maybe you can manually call GetSystemWow64Directory() from InnoSetup, like System::Call in NSIS...

Another quick idea: Make a small EXE file (e.g. vfwconfig.exe), which calls "configureVFW" from ff_vfw.dll, and put that EXE into ffdshow install folder. Then you can make a shortcut to vfwconfig.exe in ffdshow install folder, instead of calling Rundll32.exe ...

[squid_80]

When all else fails: RTFM.

Prototype:
function GetSysWow64Dir: String;

Description:
64-bit Windows only: returns fully qualified path of the
SysWOW64 directory. This is the actual directory in which 32-bit
system files reside. An empty string is returned if this 
function is called on 32-bit Windows, or if for some reason it 
fails to retrieve the path on 64-bit Windows (unlikely).

Example:
var
  S: String;
begin
  S := GetSysWow64Dir;
  // Must check the result -- an empty string is returned
  // if there is no SysWOW64 directory.
  if S <> '' then
    MsgBox('SysWOW64 directory: ' + S, mbInformation, MB_OK)
  else
    MsgBox('There is no SysWOW64 directory.', mbInformation, MB_OK);
end;

[clsid]

I know how to fix it, don't worry. Simply using GetSysWow64Dir is NOT the solution, since that returns an empty string on 32-bit Windows as you can see above. So a small wrapper function is needed.

Edit: using {syswow64} should do the trick. That should be equal to {sys} on 32-bit Windows.

[squid_80]

Originally Posted by clsid

Edit: using {syswow64} should do the trick. That should be equal to {sys} on 32-bit Windows.

Unless you want to use the same install script for 64-bit builds...

var S: String;
begin
  if Is64BitInstallMode then
    S := GetSystemDir()
  else
    S := ExpandConstant('{syswow64}');
  MsgBox('System directory: ' + S, mbInformation, MB_OK)
end;

[clsid]

A fixed build is already online.

The differences needed for 32/64-bit builds are taken care of by pre-processing of the install script. So there is no need for such code.

[LoRd_MuldeR]

Originally Posted by clsid

A fixed build is already online.

The "VfW configuration" short now works on WinXP Pro x64 Edition!
Good job

Originally Posted by clsid

The differences needed for 32/64-bit builds are taken care of by pre-processing of the install script. So there is no need for such code.

So if the differences are handled by pre-processing the script, hence at compile-time, we have different installers for 32/64-Bit Windows now?

[clsid]

The script is different for 32/64-bit builds of ffdshow. It has been since the beginning. But the 32-bit script is the same regardless of the Windows version, otherwise we couldn't have a single installer that works on all versions of Windows.

[LoRd_MuldeR]

Originally Posted by clsid

The script is different for 32/64-bit builds of ffdshow. It has been since the beginning. But the 32-bit script is the same regardless of the Windows version, otherwise we couldn't have a single installer that works on all versions of Windows.

I see

So the recent fix makes the 32-Bit installer create proper shortcuts under x64 Windows

[Kado]

@haruhiko_yamagata

I think you forgot to put the back center speaker controls and meter in the volume filter (for something like 6.1 speakers), unless you did that on purpose with the intention of adding it later because like the description for rev.1549 says:
"Volume filter : add control for side channels".

[haruhiko_yamagata]

Originally Posted by Kado

@haruhiko_yamagata

I think you forgot to put the back center speaker controls and meter in the volume filter (for something like 6.1 speakers), unless you did that on purpose with the intention of adding it later because like the description for rev.1549 says:
"Volume filter : add control for side channels".

Back center speaker is hidden there beforehand. Back left speaker can be used to configure back center speaker. They are rarely used together. Because there was no space to add one more control, I left it untouched. Though I cant test it.

[fastplayer]

Is SF's shell service down? I can't connect to shell.sourceforge.net anymore...

Edit: Thx for confirmation, Haruhiko!
Edit2: It's working as of now!

[haruhiko_yamagata]

Originally Posted by fastplayer

Is SF's shell service down? I can't connect to shell.sourceforge.net anymore...

Yes. I can't connect too.

[kurt]

Is there a way to decode *.jpg files with ffdshow? Cause I would like to watch my vacation pics on TV with some sharpenfilter (or something similar).

At present it looks like that on my system:

[LoRd_MuldeR]

Originally Posted by kurt

Is there a way to decode *.jpg files with ffdshow? Cause I would like to watch my vacation pics on TV with some sharpenfilter (or something similar).

At present it looks like that on my system:

If you enable "Raw video" in the Codecs list, you should be able to put ffdshow between "Color Space Converter" and "Video Renderer".
Maybe the "Color Space Converter" won't even be needed then ...

[clsid]

You could use Irfanview. That has a sharpen filter.

[kurt]

Originally Posted by LoRd_MuldeR

If you enable "Raw video" in the Codecs list, you should be able to put ffdshow between "Color Space Converter" and "Video Renderer".
Maybe the "Color Space Converter" won't even be needed then ...

I can put ffdshow video decoder after color space converter (before it isn't possible) but then I can't connect to any video renderer.... any ideas?

Originally Posted by clsid

You could use Irfanview. That has a sharpen filter.

thx for the tip, but I don't see an option for sharpening on the fly (e.g. slideshow) with Irfanview.

@ mods: if this is way too OT, feel free to place it somewhere else

[Leak]

Originally Posted by kurt

I can put ffdshow video decoder after color space converter (before it isn't possible) but then I can't connect to any video renderer.... any ideas?

What does the properties dialog for the "Out" pin of the ffdshow raw decoder show? That could give a hint as to why nothing wants to connect...

(Or you could run DbgView from Sysinternals and look if any messages turn up in it...)

[kurt]

Originally Posted by Leak

What does the properties dialog for the "Out" pin of the ffdshow raw decoder show? That could give a hint as to why nothing wants to connect...

(Or you could run DbgView from Sysinternals and look if any messages turn up in it...)

I don't know why, but now it works
(I can't remember what I changed exactly) - thank you all.


unfortunately seesaw scipt in avisynth dialog doesn't seem to work... but sharpen should be enough for the moment

[Leak]

Originally Posted by kurt

unfortunately seesaw scipt in avisynth dialog doesn't seem to work... but sharpen should be enough for the moment

I haven't ever used SeeSaw, but given that all kinds of other filters and scripts work I can't think of a reason why SeeSaw wouldn't...

[kurt]

Originally Posted by Leak

I haven't ever used SeeSaw, but given that all kinds of other filters and scripts work I can't think of a reason why SeeSaw wouldn't...

Maybe I wasn't clear enough:
SeeSaw does work in avisynth dialog for videos. But not for jpgs (at least here)

[clsid]

Since a jpg can be considered as a movie with a length of just 1 frame, any filter that need multiple frames to work on, won't work. I dunno if that applies to SeeSaw.

[Didée]

SeeSaw requires YV12 as input, because of MaskTools. The decoder probably delivers the JPGs as YUY2 or RGB, so you've to use ConvertToYV12() before SeeSaw. While you're at it, set 'sootheT=0', because temporal soothing makes no sense for single image sequences.

[kurt]

Originally Posted by Didée

SeeSaw requires YV12 as input, because of MaskTools. The decoder probably delivers the JPGs as YUY2 or RGB, so you've to use ConvertToYV12() before SeeSaw. While you're at it, set 'sootheT=0', because temporal soothing makes no sense for single image sequences.

It's damn slow for Hi-Res files but it works with SeeSaw now

A ConvertToYV12() is not needed, if one is limiting colorspaces to yv12 in avisynth dialog


for the hint, Didée!

[LoRd_MuldeR]

Originally Posted by kurt

thx for the tip, but I don't see an option for sharpening on the fly (e.g. slideshow) with Irfanview.

1. In IrfanView v4.10 goto "File" -> "Batch Convert/Rename"
2. Add all the files you want to view
3. Choose "Work as: Batch conversion"
4. Choose the output directory as you like it
5. Check "Use Advanced Options" and click "Advanced" button
6. Under "Other Options" (right side) check "Sharpen"
7. Click "Okay" and "Start Batch"
8. Use "File" -> "Slideshow" as usual (from output directory)

[kurt]

@ Lord_Mulder: yup, I'm aware of batch-processing. I have Paintshop Pro installed and could do this also with it. But I don't want to convert my pics (they are sharpen enough) - the TV is the problem. For my taste the pics on TV could be a bit sharper, this is why I asked for ffdshow.

Regarding Irfanview: I thought maybe it has an option to enable some sharpen filters during slideshows, but this seems not the case. Now I will try, wich filter in ffdshow (sharpen, seesaw..) is satisfying - if I find some time

(reminder: this is all about JPGs - with videos I don't have any sharpening problems on my TV because of SeeSaw)

Edit: hmmm, maybe batch processing is really easier (and faster in the end) to handle since mpc/ffdshow/seesaw is not specially made for playing back images (and it takes a lot of cpu time switching between the pics). Convert the pics you want to use for a slideshow and just delete them later .... I have to reconsider

[foxyshadis]

Well, you should really use SeeSaw after resizing, so high-res shouldn't be a problem done properly. But yeah, it's best to batch convert everything into a folder, sized directly to the screen's dimensions, to get the best quality at a good speed.

[kurt]

Originally Posted by foxyshadis

Well, you should really use SeeSaw after resizing, so high-res shouldn't be a problem done properly. But yeah, it's best to batch convert everything into a folder, sized directly to the screen's dimensions, to get the best quality at a good speed.

jup, I tried lanczosresize(1024,768) yesterday and it suited well on my CRT TV - but it is still very slow (P4, 2 ghz, 512 MB RAM)...

[haruhiko_yamagata]

Originally Posted by Leak

Well, sure. I just was erring on the cautious side because someone might already be making use of the "OR" behaviour, which would of course break with a change to "AND"...

The code change itself should be rather minimal, but the implications might not be. What does the rest here think?

np: Luke Vibert - Comphex (Chicago, Detroit, Redruth)

Sorry to be too late, but I guess this change is confusing the user. We are not announcing this very loudly. People who don't read here have no way to know...
'on movie file name match with preset name' is checked by default and unless users uncheck it, the preset doesn't load.
Too late but isn't it better to add an option to select OR/AND and make OR the default?

The dialog would require a radio button
*On all conditions match (AND)
*On one of the conditoins match (OR)

[clsid]

Or maybe uncheck 'on movie file name match with preset name' by default. I don't think many people are actually using that condition.

[haruhiko_yamagata]

Originally Posted by clsid

Or maybe uncheck 'on movie file name match with preset name' by default. I don't think many people are actually using that condition.

Yes, I will.
But it doesn't rescue user's old preset.

[rickardk]

Originally Posted by haruhiko_yamagata

Sorry to be too late, but I guess this change is confusing the user. We are not announcing this very loudly. People who don't read here have no way to know...
'on movie file name match with preset name' is checked by default and unless users uncheck it, the preset doesn't load.
Too late but isn't it better to add an option to select OR/AND and make OR the default?

The dialog would require a radio button
*On all conditions match (AND)
*On one of the conditoins match (OR)

AND, OR would be much useful in ffdshow audio. Especially now when we have 8 channels. When loading preset to handle upmixing on diffrent sources we have to use alot of presets to cover all autoloading conditions. Could be much simpler with AND, OR.

[haruhiko_yamagata]

hmm...
Each preset will have AND / OR.
For existing preset, OR is applied.
For new preset, the default preset's AND / OR is copied like other settings.
Is this a good idea?

[LigH]

@ haruhiko_yamagata:

Thanks for investigating in the HuffYUV case. Tell us the build from which a workaround will be implemented, and we will check if Premiere Elements still fails to cooperate.

[haruhiko_yamagata]

Originally Posted by LigH

@ haruhiko_yamagata:

Thanks for investigating in the HuffYUV case. Tell us the build from which a workaround will be implemented, and we will check if Premiere Elements still fails to cooperate.

The work around has not been committed yet.
They are bugs of applications. HyperCam updated their build because I reported to them. Adobe premire seems to have the same bug guessing from someone's report. And it's not all, more applications are likely to have the same bug.

Originally Posted by LigH

Back to ffdshow's HuffYUV:...

It doesn't matter if the compression is real time or not.

Where should I post this?

Microsoft's documentation about ICM_COMPRESS
http://msdn2.microsoft.com/en-us/library/ms708923.aspx

Quote from Microsoft's document:

So ffdshow writes to lpbiOutput->biSizeImage on ICM_COMPRESS before returning to the caller.

Quote from Microsoft's document:

Here, I would like to expect that the calling application to set lpbiOutput->biSizeImage every time before it call the codec. Because some applications do this only for the first time, ffdshow receives what itself has written in lpbiOutput->biSizeImage on the previous call.
Of course this doesn't work, the buffer gets smaller and smaller.

Applications don't use ICM_COMPRESS directly. They use VCM (video compression manager) fuctions's such as ICCompress or ICSeqCompressFrame. These functions are wrappers of ICM_COMPRESS. The documentations of these functions don't instruct to update lpbiOutput->biSizeImage every call and even don't refer to the return value in lpbiOutput->biSizeImage. Thus many VFW applications are buggy in my opinion.

PATCH: ffdshow's work around

Index: src/Tffvfw.cpp
===================================================================
--- src/Tffvfw.cpp	(revision 1573)
+++ src/Tffvfw.cpp	(working copy)
@@ -92,7 +92,9 @@
  CUnknown(NAME("Tffvfw"),punk),
  convert(NULL),
  dec(NULL),
- graph(false)
+ graph(false),
+ previous_out_biSizeImage(0),
+ previouts_lpOutput(NULL)
 {
  randomize();
 }
@@ -237,7 +239,11 @@
 STDMETHODIMP_(LRESULT) Tffvfw::coEnd(void)
 {
  if (initCo())
-  return deciE_VFW->end();
+  {
+   previous_out_biSizeImage = 0;
+   previouts_lpOutput = NULL;
+   return deciE_VFW->end();
+  }
  else
   return VFW_E_RUNTIME_ERROR;
 }
@@ -245,8 +251,18 @@
 {
  if (initCo())
   {
+   ICCOMPRESS *icc=(ICCOMPRESS*)icc0;
+   // Work around applications' bugs
+   // Because Microsoft's document is incomplete, many applications are buggy in my opinion.
+   //  lpbiOutput->biSizeImage is used to return value, the applications should update lpbiOutput->biSizeImage on every call.
+   //  But some applications doesn't do this, thus lpbiOutput->biSizeImage smaller and smaller.
+   //  The size of the buffer isn't likely to change during encoding.
+   if (previouts_lpOutput == icc->lpOutput)
+    icc->lpbiOutput->biSizeImage = std::max(icc->lpbiOutput->biSizeImage, previous_out_biSizeImage); // looks like very bad code, but I have no choice. Not one application need this.
+   previous_out_biSizeImage = icc->lpbiOutput->biSizeImage;
+   previouts_lpOutput = icc->lpOutput;
+   // End of work around applications' bugs.
    deciE_VFW->setICC(icc0);
-   ICCOMPRESS *icc=(ICCOMPRESS*)icc0;
    return deciE_VFW->compress(icc->lpbiInput,(const uint8_t*)icc->lpInput,icc->lpbiInput->biSizeImage,0,0);
   }
  else
Index: src/Tffvfw.h
===================================================================
--- src/Tffvfw.h	(revision 1573)
+++ src/Tffvfw.h	(working copy)
@@ -19,6 +19,8 @@
               public IprocVideoSink
 {
 private:
+ DWORD previous_out_biSizeImage;
+ LPVOID previouts_lpOutput;
  bool graph;
  bool initCo(void),initDec(void);
  comptr<IffdshowEncVFW> deciE_VFW;comptrQ<IffdshowEnc> deciE;comptrQ<IffdshowBase> deciEB;

[squid_80]

If you're going to assume the output buffer is a constant size why not use the value of lpbiOutput->biSizeImage from the ICCompressBegin call?

[haruhiko_yamagata]

Originally Posted by squid_80

If you're going to assume the output buffer is a constant size why not use the value of lpbiOutput->biSizeImage from the ICCompressBegin call?

No, I'm assuming the output buffer is a constant size as long as the output buffer's address is not changed.
But thank you for understanding my post. My question is like that you have said.
"Is it correct to assume the output buffer is a constant size and use the value of lpbiOutput->biSizeImage from the ICCompressBegin call?"
Either way, Microsoft's documentation is not clear enough.

[Inventive Software]

Think there's a bug in libavcodec's MPEG-4 encoder... in the "Motion Estimation" section, under "EPZS diamond size", the adaptive modes and experimental don't work, crashes VirtualDub.

[haruhiko_yamagata]

Originally Posted by Inventive Software

Think there's a bug in libavcodec's MPEG-4 encoder... in the "Motion Estimation" section, under "EPZS diamond size", the adaptive modes and experimental don't work, crashes VirtualDub.

I can't reproduce for now.

[LigH]

ffdshow_rev1579_20071026_xxl.exe is able to decode PAFF AVC M2TS generated by a SONY AVCHD camera.

Just ... DGAVCIndex 1.00b6 doesn't care which libavcodec version resides in its directory, and doesn't even try to index it.

[haruhiko_yamagata]

ffavisynth.dll has been changed to "C Plugin" between 20051115-20051124. ffavisynth.dll in older builds than 20051115 can be loaded automatically. While the dll in newer builds than 20051124 have to be loaded by using "loadCPlugin("full path\ffavisynth.dll")".

Is it good idea to install ffavisynth.avsi in the plugin directory?

loadCPlugin("ffavisynth.dll")

Then we can use ffavisynth without loading explicitly.

[Leak]

Originally Posted by haruhiko_yamagata

Is it good idea to install ffavisynth.avsi in the plugin directory?

If the user already selected to install ffavisynth.dll then I can't see why not.

BTW - I wanted to test your AviSynth/RGB patch yesterday evening, but my server mysteriously died on me so I didn't have time to actually test the compile I made...

[leeperry]

hey guys, sorry to jump in and ask a lame question...but is there any new version of the dualcore optimized beta fron XXL pls ?

it works really smooth on my e4500@3Ghz

thx!!

[haruhiko_yamagata]

Originally Posted by Leak

If the user already selected to install ffavisynth.dll then I can't see why not.

BTW - I wanted to test your AviSynth/RGB patch yesterday evening, but my server mysteriously died on me so I didn't have time to actually test the compile I made...

OK, I have committed both changes. Please test when you have time.

[ACrowley]

Originally Posted by leeperry

hey guys, sorry to jump in and ask a lame question...but is there any new version of the dualcore optimized beta fron XXL pls ?

it works really smooth on my e4500@3Ghz

thx!!

So far i know ,no
Only the "one Slice per Thread" Version..no "real" Multithreading

Yeah 1080p H264 is smooth for me too with these Build, but the Audio is async with a +Delay

So im waitnig for a new Multithreading Build too

[TheShadowRunner]

well in my case 1080p h264 with ffdshow isn't what i call "smooth". frame delay is always around 300ms for some reason although the CPU isn't even near maxed out. (50-60%)
I use a Dual core e2180 at 2ghz, not overclocked.
It runs the same 1080p h264 files perfectly if i use CoreAVC or Cyberlink in soft mode. (no apparent frame skips, cpu also around 50%)
I think it's kinda sad that ffdshow requires a more powerful cpu than other h264 decoders!
Later,

TSR

[clsid]

It is kinda sad that you don't understand the difference between a free open-source decoder and commercial decoders.

Just use those other decoders and stop complaining here.

[TheShadowRunner]

I'm not complaining, i love ffdshow and the hard work you put into it, it's just that for h264, it apparently doesn't behave like it should on my system. Shouldn't the CPU be maxed out BEFORE it starts skipping? Well here, with ffdshow, that's not the case.

[fastplayer]

@TheShadowRunner:
If you think you've found a bug, please report it here:
http://ffmpeg.mplayerhq.hu/bugreports.html
and/or here:
http://lists.mplayerhq.hu/mailman/listinfo/ffmpeg-user

[clsid]

You say that you have a dual core, yet you mention only 1 CPU usage %, the average. What is important is the usage of the individual cores. One of them is likely around 100%.

[ceeeeee]

how can i use ffdshow with Hardware Accelerator for mpeg 2 and h264 decoding and encoding ;

i was try to check "Set intelace flag in output media type" and select bob but cant work

i use to graphedit build graph mpeg 2 file when i use nvidia/cyberlink decoder it use DXVA when i use ffdshot it use YUV2 plase help me Thanks

[TheShadowRunner]

clsid, i'm trying to find a soft that shows cpu util. for both cores independently, but i'm sure you're right.
In which case, "Queue samples" should make a difference when enabled. However I cannot see any.
I'll confirm when I find the soft and test again in real time.
(if you know a small soft that shows each core util., let me know^^)

ceeeeee, it's not possible at the moment, ffdshow only works in software atm.
See you,

TSR

[Jeremy Duncan]

Codecs tab
Set Mpeg2 to Libmpeg2, and check "DVD decoding".
Set Raw video to All supported

Deinterlacing tab
Kernal Deinterlacer
Threshold: 16

Subtitles tab, Unchecked
Uncheck "Decode closed captions"
"Accept embedded subs"
"Accept SSA, ASS, ASS2 Subtitle (experimental)
Vobsub subpage, uncheck Vobsub Enable.

Resize & aspect:
Specify Horizontal and vertical size 1024, 576 (Match your screen resolution)
Process Pixel aspect ratio internally checked
No aspect ratio correction checked
Lanczos
Taps: 2
Luma Sharpen: 0.20
Accurate rounding checked

Queue & Output tab
Queue output samples checked

Output tab
YV12 checked



Using FFDshow Clsid version Generic 1589, and MPC 6.4.9.0 Gabest version this crashes when I open GitS NTSC Widescreen version.

MPC uses default output, Audio switcher boost on full, and the external filters audio and video ffdshow.

[Thunderbolt8]

Originally Posted by ACrowley

So far i know ,no
Only the "one Slice per Thread" Version..no "real" Multithreading

Yeah 1080p H264 is smooth for me too with these Build, but the Audio is async with a +Delay

So im waitnig for a new Multithreading Build too

what does this exaclty mean, when does the audio go out of sync? when applying a delay with "+" or"-", it doesnt apply the displayed value then?

[Px]

Originally Posted by TheShadowRunner

clsid, i'm trying to find a soft that shows cpu util. for both cores independently

RMClock
RivaTuner

[haruhiko_yamagata]

Originally Posted by TheShadowRunner

(if you know a small soft that shows each core util., let me know^^)

Windows Xp's built in task manager have performance monitor. Press ctrl+alt+del and click performance tab. Press alt+V+C+P.

frame delay is always around 300ms

As I wrote here, queue won't be effective in that case.

[haruhiko_yamagata]

Originally Posted by Leak

EDIT: That's a problem with regedit, actually - the scripts are written to the file with their linebreaks intact, so regedit's parser will just skip what it thinks are broken lines.

But if you just export the whole ffdshow configuration from regedit, it'll produce the exact same problem... so I'd have to replace the newlines in the AviSynth script with something else when storing it to the registry and undo that when reading from it - that'll need a bit more investigation...

It's a bug of ffdshow. We must use REG_MULTI_SZ (double null terminated strings) for multiple lines text data. It won't be too hard to implement it.
However, ffdshow support *.ffpreset file and stream input (like ffavisynth.dll ffdshow() and ffdshowAudio()). These things are more annoying because they don't support double null terminated strings. The alternative would be to replace \r\n with some delimiter.
Which character(s) can we use as delimiter?

[Leak]

Originally Posted by haruhiko_yamagata

Which character(s) can we use as delimiter?

I was thinking of using "%n", "%r" and "%%" for escaping, with a single "%" as the very first character stored to signal that escaping was used (so that existing scripts won't be unescaped, which would break them) since the "%" operator in AviSynth can't be used as the very first symbol in a script...

[ACrowley]

Originally Posted by Thunderbolt8

what does this exaclty mean, when does the audio go out of sync? when applying a delay with "+" or"-", it doesnt apply the displayed value then?

That means simply the audio is async with a + Delay (~ 1500ms) on 1080p H(x)264 ...Oh ,no its not my CPU ,its strong enough (A 64 X2 5000)

I know a few guys with the same Problem with the Multithreaded Builds

[Jeremy Duncan]

I have a feature request.
That in the resize and aspect tab there be a option to eliminate the black bar on the bottom of the screen while keeping the black bar on the top of the screen.

So in 2:35 aspect movies on a 16:10 ratio screen, the movie would be on the bottom of the screen and the only black bar would be on top.

Also, on that bug where it crashed. It's something to do with the deinterlace tab.

[Leak]

Originally Posted by Jeremy Duncan

So in 2:35 aspect movies on a 16:10 ratio screen, the movie would be on the bottom of the screen and the only black bar would be on top.

Have you looked at the "Resize & aspect > Borders" page lately? That option is already there...

np: Prefuse 73 - Over Ensembles (Interregnums)

[clsid]

@Jeremy, that is already possible. Look at the 'borders' subpage of resize settings. There you can define how space for the black bars should be allocated.

Edit: too late

[Thunderbolt8]

Originally Posted by ACrowley

That means simply the audio is async with a + Delay (~ 1500ms) on 1080p H(x)264 ...Oh ,no its not my CPU ,its strong enough (A 64 X2 5000)

I know a few guys with the same Problem with the Multithreaded Builds

hm, I dont have this problem. Ive made 2 remuxes with AVC so far (oldboy & casino royale), in both cases the audio (dd+ and LPCM) were in sync after demuxing, so no delay needed. and this was the case then, audio and video were perfectly in sync, there was nothing like a delay or such. got a c2d 6300 running @2,8GHz

[Jeremy Duncan]

Using the FFDshow "Borders" tab.

First I need to set the "Borders" Horizontal or Vertical.
It's got to change from Default.

Second, I need to set "Dividing Borders", either Horizontal or Vertical.

The result is the movie can be shifted around.
But the top and bottom default border setting are never affected.
I can set the "Dividing Borders" Vertical all the way right or left,
and the original borders are neither set higher or lower than they originally were.

The only borders that can be affected by "Dividing Borders",
are the ones created by changing the default horizontal or vertical "Borders" settings.

So by using the Borders tab in Resize and Aspect,
I can't eliminate the bottom black border and only have a border on top of the movie!
So it's still a feature request, please.

This was using Clsid's version put out today.

[Inventive Software]

Semi-useful suggestion: update the "tool-tip" on the "Decoder options" section/tab of the DirectShow decoder, to include the H.264 decoder as multi-threaded. Currently, it says "mpeg1/2 decoder only".

And Jeremy Duncan: post your ffdshow probs in this thread as opposed to the sub-forum, because that way they're more likely to get answered by the people who know (developers and so-forth). It keeps the forum cleaner, and everything together. Just a minor nitpick / suggestion.


BTW, does anybody know what sort of speed the libavcodec VC-1 decoder has, and whether that's multi-threaded or not?

[haruhiko_yamagata]

Originally Posted by Jeremy Duncan

the original borders are neither set higher or lower than they originally were.

Do you want to crop the borders that the original videos have? Then 'crop' is the one you are looking for.

[clsid]

Maybe we can set up a wiki page on the sourceforge webspace? With a FAQ and pages for explaining the various filter pages. Then everyone can contribute their knowledge and tips. Plus we don't have to invest time in creating documentation to include with ffdshow. We can just include a weblink to the wiki.

[Naito]

Originally Posted by clsid

Maybe we can set up a wiki page on the sourceforge webspace? With a FAQ and pages for explaining the various filter pages. Then everyone can contribute their knowledge and tips. Plus we don't have to invest time in creating documentation to include with ffdshow. We can just include a weblink to the wiki.

Please!

[Jeremy Duncan]

Originally Posted by haruhiko_yamagata

Do you want to crop the borders that the original videos have? T

The example on the right is what I want to have, please.



Anybody confirm at that deinterlace bug?

[Leak]

Originally Posted by Jeremy Duncan

The example on the right is what I want to have, please.

You've still failed to tell us if (part of) the black bars in your left image are encoded into the movie, because obviously those aren't affected by the sliders on the "Borders" page.

You can, however, use the "Autocrop" settings of the "Crop" filter to get rid of encoded borders before adding new ones via the "Resize" filter. That should do exactly what you want - you obviously need to get rid of any fixed borders contained in the movie itself before you can add new, user defined borders any way you want...

np: Prefuse 73 - Busy Signal (Make You Go Bombing Mix) (Prefuse & Daedelus) (One Word Extinguisher)

[Jeremy Duncan]

Leak,

It's not for one Specific Movie, but 2:35 1:85 aspect ratio movies in general.
I'm assuming that the borders aren't encoded into the movie.

Also, I was hoping for a one button option instead of cropping then manually setting the resize borders tab, since that's not automatically switched between 2:35 and 1:85 aspect ratios.
Also on 16:10 monitors it would need to be done for 16:9 movies.

I tested the Ghost in the Shell: Solid State Society Movie, NTSC, Widescreen. It was the movie, not a series.

Here's a D2v Sample of the Movie I used.
Link

[clsid]

Like said before:

What you want can be accomplished by simply enabling the resize filter, crop is not needed. Set dividing borders to 100% for top and 0% for bottom.

Of course this has no effect on black borders that are added by the renderer. Those are the ones that you only see in fullscreen.

[Fabool]

ffdshow crashes when deinterlacing is switched on and deinterlacing method is either Kernel bob or Kernel deinterlacer. This with the newest clsid build 1589 (both generic and icl10), it works fine with the 1578 build. Happens with any video clip, tested with Zoom player and DVBviewer (digital tv broadcasts).

[haruhiko_yamagata]

Originally Posted by Fabool

ffdshow crashes when deinterlacing is switched on and deinterlacing method is either Kernel bob or Kernel deinterlacer. This with the newest clsid build 1589 (both generic and icl10), it works fine with the 1578 build. Happens with any video clip, tested with Zoom player and DVBviewer (digital tv broadcasts).

Thank you for report. fixed at rev 1590.

[cyberscott]

Hi everyone, long time lurker, first time poster.
Thanks for all the hard work on ffdshow!

I have noticed that since the inclusion of 7.1 audio option, AC3 encoding no longer works via spdif using digital connection. It only outputs 2 channel PCM. This is regardless of of the speaker matrix chosen. Going back to builds prior to the 7.1 audio matrix addition, AC3 encoding works fine with digital out. Analog out works fine for the 7.1 matrix.

[ACrowley]

Originally Posted by cyberscott

Hi everyone, long time lurker, first time poster.
Thanks for all the hard work on ffdshow!

I have noticed that since the inclusion of 7.1 audio option, AC3 encoding no longer works via spdif using digital connection. It only outputs 2 channel PCM. This is regardless of of the speaker matrix chosen. Going back to builds prior to the 7.1 audio matrix addition, AC3 encoding works fine with digital out. Analog out works fine for the 7.1 matrix.

Yeh , i also have the same Problmes...

i

[phunqe]

Hi,

I've been trying to find some information on the "closed caption" subtitle feature. Is this supposed to be support for closed captions in DVB MPEG streams? If yes, is there any trick to enable it or is it not finished? Trying to get it to work with streams from a DVB-C card, but without any luck.

Cheers.

[Kurtnoise]

Question for the ffdshow maintainers: why lavf hasn't been included in ffmpeg package ?

[clsid]

Because ffdshow is only a decoding filter, not a source filter.

The idea of creating a DirectShow source filter based on libavformat has been brought up a couple times in the past. It would be an excellent NEW project. Its too much work to be part of ffdshow.

[Mc Onyx]

Could someone compile a "test" version of ffdshow, containing the new Dolby E-AC3 and True-HD patches/decoders, from mplayer, like Kurtnoise13 did with his mplayer and ffmpeg builds? You can find the patches on his site. Thanks.

http://kurtnoise.free.fr/index.php?dir=misc/

[haruhiko_yamagata]

Originally Posted by cyberscott

Hi everyone, long time lurker, first time poster.
Thanks for all the hard work on ffdshow!

I have noticed that since the inclusion of 7.1 audio option, AC3 encoding no longer works via spdif using digital connection. It only outputs 2 channel PCM. This is regardless of of the speaker matrix chosen. Going back to builds prior to the 7.1 audio matrix addition, AC3 encoding works fine with digital out. Analog out works fine for the 7.1 matrix.

OK, I bought long optical digital cable to connect to the AV-amp in the next room. I'll test as soon as I finish current stuff (AviSynth multiple line script problems).

[Kurtnoise]

@Clsid: ok, thanks.

Originally Posted by Mc Onyx

Could someone compile a "test" version of ffdshow, containing the new Dolby E-AC3 and True-HD patches/decoders, from mplayer, like Kurtnoise13 did with his mplayer and ffmpeg builds?

not enough and can't be done for the moment...Clsid gave the answer just above.

[clsid]

I could add the EAC3 decoder to libavcodec. That's easy. Actually using it is a whole different case.

The problem is making ffdshow able to recognize EAC3. It must be able to see the difference between normal AC3 audio and EAC3. It should do so based on the information given by the source filter in the graph. To be precise, ffdshow selects the decoder based on the FourCC/AudioTag. If someone can tell me which tag is used by EAC3, then maybe there is a small chance that I can make it work.

[cyberscott]

Originally Posted by haruhiko_yamagata

OK, I bought long optical digital cable to connect to the AV-amp in the next room. I'll test as soon as I finish current stuff (AviSynth multiple line script problems).

If someone can tell me which tag is used by EAC3, then maybe there is a small chance that I can make it work.

EAC3.

[albain]

Hi,

I have brought a few modifications to the AC3 encoding in revision 1595 : there was an issue with some HT receivers having no sound when "Maximize SPDIF compatibility" is unchecked.

So to people who had this issue : could you retest rev 1595 and untick the option (in codec section / AC3) ?

Otherwise, the new 7.1 matrix works for me in 5.1 but only when I select 3/0/2 mixing

[albain]

Originally Posted by haruhiko_yamagata

Raw video processing in WMP11 and Media Center is experimental and not very stable. Currently it does not work for me. I hope albain can fix it.

Hi haruhiko,

can you give me more details about your issue ?

Damien

[haruhiko_yamagata]

Originally Posted by albain

Hi haruhiko,

can you give me more details about your issue ?

Damien

I'm sorry. It was my mistake as I wrote here.

[haruhiko_yamagata]

Originally Posted by albain

Hi,

I have brought a few modifications to the AC3 encoding in revision 1595 : there was an issue with some HT receivers having no sound when "Maximize SPDIF compatibility" is unchecked.

So to people who had this issue : could you retest rev 1595 and untick the option (in codec section / AC3) ?

Otherwise, the new 7.1 matrix works for me in 5.1 but only when I select 3/0/2 mixing

I still need "Maximize SPDIF compatibility" checked. Otherwise no sound. Xp sp2.

[haruhiko_yamagata]

On Vista, if I choose Connect to: "DirectSound" and enable "Apply only to S/PDIF output", set Multichannel device: "SPDIF interface (SoundMAX Integrated Digital HD)", the player (MPC homecinema) freezes.
I don't know if this is ffdshow's problem or driver's problem. Can anybody reproduce?

[albain]

Originally Posted by haruhiko_yamagata

On Vista, if I choose Connect to: "DirectSound" and enable "Apply only to S/PDIF output", set Multichannel device: "SPDIF interface (SoundMAX Integrated Digital HD)", the player (MPC homecinema) freezes.
I don't know if this is ffdshow's problem or driver's problem. Can anybody reproduce?

This is due to the "Multichannel device" option : I didn't have the time to dig this around, but this option also causes me freezes when the device is not set to default.

Please try agin with "Maximize SPDIF compatibility" unchecked and Multichannel device set to default device

[haruhiko_yamagata]

Originally Posted by cyberscott

Hi everyone, long time lurker, first time poster.
Thanks for all the hard work on ffdshow!

I have noticed that since the inclusion of 7.1 audio option, AC3 encoding no longer works via spdif using digital connection. It only outputs 2 channel PCM. This is regardless of of the speaker matrix chosen. Going back to builds prior to the 7.1 audio matrix addition, AC3 encoding works fine with digital out. Analog out works fine for the 7.1 matrix.

I can't reproduce. Which version do you use? What does the OSD or Info (dialog Info & CPU page) say?

[_xxl]

New test build:
http://rapidshare.com/files/67886123...6_xxl.zip.html
mulipleLinesScript patch by haruhiko_yamagata used.

[wozio]

Originally Posted by clsid

The problem is making ffdshow able to recognize EAC3. It must be able to see the difference between normal AC3 audio and EAC3. It should do so based on the information given by the source filter in the graph. To be precise, ffdshow selects the decoder based on the FourCC/AudioTag. If someone can tell me which tag is used by EAC3, then maybe there is a small chance that I can make it work.

Here it is how haali connects to ffdshow using mkv with eac3:

- Connected to:

CLSID: {0F40E1E5-4F79-4988-B1A9-CC98794E6B55}
Filter: ffdshow Audio Decoder
Pin: In

- Connection media type:

Audio: Dolby AC3 48000Hz 6ch

AM_MEDIA_TYPE: 
majortype: MEDIATYPE_Audio {73647561-0000-0010-8000-00AA00389B71}
subtype: MEDIASUBTYPE_DOLBY_AC3 {E06D802C-DB46-11CF-B4D1-00805F6CBBEA}
formattype: FORMAT_WaveFormatEx {05589F81-C356-11CE-BF01-00AA0055595A}
bFixedSizeSamples: 0
bTemporalCompression: 0
lSampleSize: 1
cbFormat: 18

WAVEFORMATEX:
wFormatTag: 0x0000
nChannels: 6
nSamplesPerSec: 48000
nAvgBytesPerSec: 0
nBlockAlign: 0
wBitsPerSample: 0
cbSize: 0 (extra bytes)

pbFormat:
0000: 00 00 06 00 80 bb 00 00 00 00 00 00 00 00 00 00 ....€»..........
0010: 00 00                                           ..

- Enumerated media type 0:

Set as the current media type

- Enumerated media type 1:

Audio: 0xeac3 48000Hz 6ch

AM_MEDIA_TYPE: 
majortype: MEDIATYPE_Audio {73647561-0000-0010-8000-00AA00389B71}
subtype: Unknown GUID Name {0000EAC3-0000-0010-8000-00AA00389B71}
formattype: FORMAT_WaveFormatEx {05589F81-C356-11CE-BF01-00AA0055595A}
bFixedSizeSamples: 0
bTemporalCompression: 0
lSampleSize: 1
cbFormat: 18

WAVEFORMATEX:
wFormatTag: 0xeac3
nChannels: 6
nSamplesPerSec: 48000
nAvgBytesPerSec: 0
nBlockAlign: 0
wBitsPerSample: 0
cbSize: 0 (extra bytes)

pbFormat:
0000: c3 ea 06 00 80 bb 00 00 00 00 00 00 00 00 00 00 Ăę..€»..........
0010: 00 00                                           ..

Both ffdshow and ac3filter connects to it but freezes on first frame. As far as I remember haali reports the same when splitting evo file.

Regards
Piotr Wozniak

[clsid]

Two media types instead of just one. Weird.

This is too complex for me. I won't be wasting my time on something that I probably can't get working.

Maybe Haruhiko or one of the other skilled people here can adapt ffdshow to recognize it.

[haruhiko_yamagata]

Originally Posted by _xxl

New test build:
http://rapidshare.com/files/67886123...6_xxl.zip.html
mulipleLinesScript patch by haruhiko_yamagata used.

Thank you.
I have written a patch to fix multiple lines AviSynth script problem.
Both upgrade and downgrade are supported without breaking the script. Export to registry, export to .ffpreset, save to .GRF in graphedit and ffavisynth.dll-ffdshow() options are supported. Please test.
Here's the patch.

[cyberscott]

Originally Posted by haruhiko_yamagata

I can't reproduce. Which version do you use? What does the OSD or Info (dialog Info & CPU page) say?

OSD says...Input description 48000 Hz stereo 160 kbps MP3 (mp3lib)

Output: 48000 Hz 8 channels 16 bit integer
Output Speakers: L,R,C,LFE, SL, SR, side left, side right

I've used both clsid and xxl builds 1578 and 1589 generic, icl10 builds, with the same lack of AC3 encoding. (for testing purposes, I'm currently using ffdshow_rev1589_20071031_clsid_sse_icl10.exe

With spdif enabled in the codec page, output set to AC3 encoding, ffdshow only outputs 2 channel pcm via my digital out. I've tried connecting using direct sound or wave out, no change. If I go back and use an earlier build such as 1530, ac3 encoding works fine via digital out. And of course, using analog out, it works fine.

I am using Windows xp pro with the latest updates. My sound card is the Auzen Prelude 7.1

[albain]

Originally Posted by cyberscott

OSD says...Input description 48000 Hz stereo 160 kbps MP3 (mp3lib)

Output: 48000 Hz 8 channels 16 bit integer
Output Speakers: L,R,C,LFE, SL, SR, side left, side right

I've used both clsid and xxl builds 1578 and 1589 generic, icl10 builds, with the same lack of AC3 encoding. (for testing purposes, I'm currently using ffdshow_rev1589_20071031_clsid_sse_icl10.exe

With spdif enabled in the codec page, output set to AC3 encoding, ffdshow only outputs 2 channel pcm via my digital out. I've tried connecting using direct sound or wave out, no change. If I go back and use an earlier build such as 1530, ac3 encoding works fine via digital out. And of course, using analog out, it works fine.

I am using Windows xp pro with the latest updates. My sound card is the Auzen Prelude 7.1

As far as I know liba52 is not able to encode more than 5+1 channels in AC3 (I am not sure about that but I remember such limits in the code).
If you select 3/0/2 mixing it should work
I also brought a few corrections to AC3 encoding on revision 1595

[wozio]

Originally Posted by clsid

Two media types instead of just one. Weird.

Yes, weird. I will try ask haali about it. It should not be like that, normal AC3 should not be reported at all as far as I understand it.

Regards
Piotr Wozniak

[haruhiko_yamagata]

Originally Posted by cyberscott

Output: 48000 Hz 8 channels 16 bit integer
Output Speakers: L,R,C,LFE, SL, SR, side left, side right

As albain pointed, AC3 can't encode 8 channels. Automatic down-mixing is desirable though.

[_xxl]

As far as I know liba52 is not able to encode more than 5+1 channels in AC3

I think that libavcodec ac3enc is used not liba52 to encode.

[Kado]

I was thinking of translating ffdshow to Portuguese from Portugal but there's no English base file for me to start with, I've searched both the trunk and the installation folder but the files only say "English". Can some one provide something for me?

[clsid]

Translation tutorial:
http://forum.doom9.org/showthread.php?t=86030

[haruhiko_yamagata]

@albain
Please read this thread.
Do you have any reason to use LPCM16 when AC3 output is enabled && Encode only multichannel stream is checked && nchannels<6?
PCM16/24/32 are more stable format for my system. I would be better if fail over format were configurable.

[Leak]

Originally Posted by Kado

I was thinking of translating ffdshow to Portuguese from Portugal but there's no English base file for me to start with, I've searched both the trunk and the installation folder but the files only say "English". Can some one provide something for me?

Using regedit, navigate to HKEY_CURRENT_USER\Software\GNU and set the DWORD translateMode in each of the subkeys (ffdshow, ffdshow_audio, ffdshow_enc etc.) to 1.

Now switch your language to Portuguese (copy the file for the English "translation", ffdshow.1033.en, to ffdshow.2070.pt (I think that's the correct name) first if needed) then right click on any control to translate it's text.

You can edit the resulting file and add anything that's still missing afterwards.

Don't forget to switch translateMode back to 0 afterwards, as it'll prevent controls from being disabled and makes control groups look a bit weird...

EDIT: Yeah, basically what clsid said...

[Kado]

Thanks for the tips.

[albain]

Originally Posted by haruhiko_yamagata

@albain
Please read this thread.
Do you have any reason to use LPCM16 when AC3 output is enabled && Encode only multichannel stream is checked && nchannels<6?
PCM16/24/32 are more stable format for my system. I would be better if fail over format were configurable.

I don't use this feature but I developped it from a feature request
From what I remember, the need was not to encode the stream in AC3 if this is a PCM non multichannel stream.
The argument was that AC3 is a compressed format and alters the quality of the source so it should be used only when necessary (i.e when the source is multichannel).
So the algorithm is the following : if "encode only multichannel stream" checked AND input channels < 6 then switch back to LPCM16

From what I understand, beta 3 and previous versions of ffdshow worked correctly but with the most recent versions of ffdshow there are some crash

Maybe this is due to the fact that there is a change of output format (AC3 to LPCM16) during the graph processing

[haruhiko_yamagata]

Originally Posted by albain

I don't use this feature but I developped it from a feature request
From what I remember, the need was not to encode the stream in AC3 if this is a PCM non multichannel stream.
The argument was that AC3 is a compressed format and alters the quality of the source so it should be used only when necessary (i.e when the source is multichannel).
So the algorithm is the following : if "encode only multichannel stream" checked AND input channels < 6 then switch back to LPCM16

From what I understand, beta 3 and previous versions of ffdshow worked correctly but with the most recent versions of ffdshow there are some crash

Maybe this is due to the fact that there is a change of output format (AC3 to LPCM16) during the graph processing

Thank you for explanation.
LPCM and PCM are differently accepted by the audio renderer.
The audio renderer does not accept LPCM at all.

[wozio]

Regarding EAC3 decoding in ffdshow. As posted here there is problem with connecting to EAC3 media type since it is reported as plain AC3. I asked haali about this:

Originally Posted by wozio

Why splitter enumerates AC3 as first media type on output pin? Shouldn't be only EAC3 since that kind of stream is in mkv? Could you please clarify it?

Originally Posted by Haali

Existing eac3 decoders (sonic, etc.) don't use or support a special media type for eac3. The 0xeac3 waveformat that I use is purely for my enjoyment and is not used by any software afaik.

Originally Posted by wozio

So how decoders know that stream is eac3? Now ffdshow or ac3filter connects as to ac3 and hangs.

Originally Posted by Haali

I suppose they just check the bsid in frame headers and can decode each frame with either decoder.

Maybe this will help integrating EAC3 decoding into ffdshow.

I would really like to have EAC3 deocding in ffdshow...

Regards
Piotr Wozniak

[wozio]

One more thing: since EAC3 should be backward compatible with AC3, so existing AC3 decoders are capable of decoding AC3 stream embeded in EAC3, then maybe it is good place to start just to ignore EAC3 samples and decode only AC3 ones...

Regards
Piotr Wozniak

[sysKin]

Heyhey,

I've been having an annoying issue with ffdshow for quite some time - it adds black borders to everything it displays. The borders are 6 pixels on right/left and 4 pixels on top/bottom, and are completely black (zero luma, I haven't checked chroma).

From what I can see the borders don't come from cropping the video but from *resizing* it down. This is particularly a problem with interlaced stuff.

Anyone knows what I can do about it? Is it known? Any workarounds?

Thanks

[madshi]

Originally Posted by wozio

One more thing: since EAC3 should be backward compatible with AC3, so existing AC3 decoders are capable of decoding AC3 stream embeded in EAC3, then maybe it is good place to start just to ignore EAC3 samples and decode only AC3 ones...

There are no AC3 samples in an E-AC3 stream. E-AC3 is not backwards compatible.

[Liisachan]

Originally Posted by clsid

If someone can tell me which tag is used by EAC3, then maybe there is a small chance that I can make it work.

I personally don't need this EAC-3 thing, but

- Waveformatex, FormatTag = 0x2000
- [There may or may not be junk bytes after the header*]
- 0b 77 <--syncword (AC3 header) // look for them
- 3 bytes to skip
- this byte <-- here // (byte>>3)==8 for standard ac3, ==16 for enhanced one

Or, you may want to get it straight from horse's mouth
http://www.atsc.org/standards/a_52b.pdf

*You might have to scan more than 1 chunks, because in the worst situation, at least in theory, you can't find the syncword at all in the 1st chunk, for example when a simple-minded muxer zero'ed it out trying to set audio delay that way.

[SeeMoreDigital]

Originally Posted by madshi

There are no AC3 samples in an E-AC3 stream. E-AC3 is not backwards compatible.

What a shame....

Seems like a missed opportunity for the E-AC3 decoder filter not to support regular AC3 streams too

[khagaroth]

Originally Posted by Kado

I was thinking of translating ffdshow to Portuguese from Portugal but there's no English base file for me to start with, I've searched both the trunk and the installation folder but the files only say "English". Can some one provide something for me?

A file with all the translatable strings would be really nice. I know about the translation mode and it is nice for initial translation (I'm just finishing Czech translation, should be ready this weekend), but it's almost unusable for tracking down changes between versions. GUI strings don't get updated/changed that frequently, but if they do, it's a pita to to find out what actually changed.

[madshi]

FWIW, I've managed to decode an E-AC3 sample with ffmpeg and it works ok, but there are 2 problems:

(1) Decoding seems to be always 16bit? Not sure, at least that's what I got without changing any options. But maybe it's possible to force 24bit decoding somewhere somehow. I'm a beginner with using libavcodec.

(2) The ffmpeg decoded E-AC3 file plays slightly distorted. The same E-AC3 sample file is not distorted when decoding with Nero's E-AC3 decoder.

So as far as I can see, ffmpeg's E-AC3 decoding support is not ready to go yet. I also tried getting the TrueHD/MLP decoder to work, but failed. It did extract one audio frame sometimes but then stopped working.

I think we need to have some patience. And I think ffdshow doesn't need to hurry with adding E-AC3 and TrueHD/MLP support. The decoders first need to become stable.

[yesgrey]

Originally Posted by Kado

I was thinking of translating ffdshow to Portuguese from Portugal but there's no English base file for me to start with, I've searched both the trunk and the installation folder but the files only say "English". Can some one provide something for me?

Kado, would you like some help?

[Kado]

I have translated the audio decoder dialog already, maybe I missed some stuff. That translation system may impose some difficulties like khagaroth said but hey, that's the one we've got right now... Thanks for the help guys. You may see some real Portuguese translation soon!

[foxyshadis]

Originally Posted by SeeMoreDigital

What a shame....

Seems like a missed opportunity for the E-AC3 decoder filter not to support regular AC3 streams too

AC3's more complex than DTS, so building another lossy format around an AC3 core would be much harder than just boosting the bitrate. (Not that I think it's a bad idea. AAC would still be a better idea, hrmph.) DTS is practically just ADPCM, you can add and peel bits from it the way vorbis was originally designed to.

TrueHD is lossless AC3, on the other hand, similar to the Wavpack hybrid concept.

[Inventive Software]

Originally Posted by foxyshadis

TrueHD is lossless AC3, on the other hand, similar to the Wavpack hybrid concept.

Technically, that's incorrect. TrueHD is based on MLP, and happens to have an AC3 track in it that can be used when there's no support for MLP decoding. It's like having an MKA (Matroska Audio) with FLAC and AAC audio files. Just a small clarification before the wrong info goes out.

[foxyshadis]

Oh really? Huh, odd that they wouldn't just have them as separate tracks then, must be a marketing thing.

[cyberscott]

Originally Posted by albain

As far as I know liba52 is not able to encode more than 5+1 channels in AC3 (I am not sure about that but I remember such limits in the code).
If you select 3/0/2 mixing it should work
I also brought a few corrections to AC3 encoding on revision 1595

I did try xxl's revision 1595 and low and behold, ac3 encoding works again! I did roll back to the other builds that included the 7.1 audio matrix and did set them to 5.1 channels and below and I still could not get ac3 encoding, only 2 channel PCM via spdif.

Whatever modifications/corrections you did in revision 1595 for AC3 worked for me.

[wozio]

Originally Posted by madshi

I think we need to have some patience. And I think ffdshow doesn't need to hurry with adding E-AC3 and TrueHD/MLP support. The decoders first need to become stable.

Maybe avoiding hanging when EAC3 received is good idea to implement until decoders will be stable. Now ffdshow always connect to EAC3 stream and hangs what is a bit annoying, better would be to just no sound at all with some indication in info panel.

[madshi]

Originally Posted by foxyshadis

Oh really? Huh, odd that they wouldn't just have them as separate tracks then, must be a marketing thing.

I think they're doing it this way to make things easier for the end user. The end user only sees one track "English" which he chooses. Now the Blu-Ray player can decide itself whether to playback the AC3 or the TrueHD track, depending on which decoders it has inside. I think this approach makes sense.

Btw, the AC3/TrueHD mixed tracks are only found on Blu-Ray. On HD DVD TrueHD tracks don't have any AC3 frames mixed in.

[yesgrey]

Originally Posted by Kado

You may see some real Portuguese translation soon!

The portuguese from Brazil is also real portuguese. Slightly different from the Portugal one, just that.

[Kado]

Originally Posted by yesgrey3

The Portuguese from Brazil is also real Portuguese. Slightly different from the Portugal one, just that.

When I read something in Brazilian Portuguese something just seems wrong, maybe it's like American English and British English.
Example:
Windows "log of" message
EN: "Windows is logging of."
PT-BR:"Windows está fazendo log of"
PT-PT:"Windows está a terminar a sessao"

If I was Brazilian and couldn't understand English I would not understand that PT-BR message. If you can understand Portuguese you'll understand what I mean.

Sorry for the off topic stuff.

[Rash]

Kado, people here just prefer the english technical terms associated with computers. It is really not because we crave for foreign words, but mostly because we are used to them. So we have verbs (with full conjugation) like:

downloadar (fazer download, baixar)
logar (fazer log on)
deletar (apagar, remover)
resetar (zerar, reiniciar)
rebootar (reiniciar o PC)

And so on. Sorry for the offtopic.

[Kado]

I'm full aware of those terms because I use them myself but I don't think they should be used regarding the translation of ffdshow when you are making the translation accessible to everyone and not just to the more pc expert (like us who are browsing and posting in these forums), it's just my opinion. Anyway eventually the translation will be completed and you guys can give an opinion about it... or not!

P.S.=> Looks like there are more Portuguese speaking people here than I imagined after all! Fiquem bem!

[Adub]

Just letting you guys know, the ffavisynth.avsi placed in my plugins folder during the install seems to crash my Avisynth everytime I try to open, say, a d2v file.

[yesgrey]

Kado, I understand you. I'm also portuguese. The use of foreign words is also a current practice with us. Maybe the brazillian use it a bit more, but we also use it sometimes... For instance the verb "zipar". Don't tell me you never used it?

Just tell us when you're translation is ready and I will be very pleased in giving you my opinion/suggestions. I only don't promise to use it, because I prefer the english terms.

[Delerue]

Originally Posted by Kado

If I was Brazilian and couldn't understand English I would not understand that PT-BR message. If you can understand Portuguese you'll understand what I mean.

Like Rash and yesgrey3 said, brazilian people can full understand that translation (both indeed). I'm a brazilian guy, and I can guarantee you.

[SeeMoreDigital]

Enough with the language discussions please guys....

[haruhiko_yamagata]

Originally Posted by Merlin7777

Just letting you guys know, the ffavisynth.avsi placed in my plugins folder during the install seems to crash my Avisynth everytime I try to open, say, a d2v file.

Thank you. Could you tell us the details?
What kind of error do you have?
Which application do you use?
Is the crash reproducible with AVI file?
What is the OS?

[G_M_C]

People, sorry if it is an "stupid" question, but i'm looking for a recent build of FFDShow that is multithreaded. I read somewhere about a build like this, but i cant remember "who's " build is was (Clsid/XXL).

I ask this because i want to try to install my new PC this weekend, and it has a E6750 Dualcore. And i need to find an appropriate version of MPC (patched) and AC3Filter i believe, but the question is about ffdshow

[ilpippo80]

Originally Posted by Merlin7777

Just letting you guys know, the ffavisynth.avsi placed in my plugins folder during the install seems to crash my Avisynth everytime I try to open, say, a d2v file.

I had the same problem, and then I discovered that I still had kevin atkinson's avisynth_c.dll filter in my plugin's directory. If that's the problem, just remove avisynth_c.dll or replace LoadCPlugin with Load_Stdcall_Plugin in ffavisynth.avsi.

[Fizick]

Hi,
I found old (since 2003 ?) small bugs in MakeAVIS mode "store uncompressed audio".
1. in function convertToAVI, around line 666:

//      wfx->wBitsPerSample=WORD((avs_samples_per_second(vi)/avs_audio_channels(vi))<<3); // wrong
      wfx->wBitsPerSample=WORD((avs_bytes_per_audio_sample(vi)/avs_audio_channels(vi))<<3); // must be

2. in same function, around line 687:

//          LONG rawaudiolen=LONG((asample[1]-asample[0]+1)*avs_samples_per_second(vi)); // wrong
          LONG rawaudiolen=LONG((asample[1]-asample[0]+1)*avs_bytes_per_audio_sample(vi)); // must be

3. in function showProps, aroun line 909

//    sprintf(pomS,"Bits per sample: %i",(avs_samples_per_second(vi)/avs_audio_channels(vi))<<3); // wrong
    sprintf(pomS,"Bits per sample: %i",(avs_bytes_per_audio_sample(vi)/avs_audio_channels(vi))<<3); // must be

This way makeAVIS works fine in "store uncompressed audio" mode, but only if Bits per Sample is 8 or 16.

For 32 bit (integer or float) audio it produces non-playable avi.
Next fix should be else disable this mode (disable checkbox ?)
or implement internal (or Avisynth) resampling to 16 bit.
I am not guru of windows messages and controls.

Please put my "patch" to SVN

[Leak]

Ugh... I just realized why ffdshow's KernelBob and KernelDeint were crashing on me in the recent builds I downloaded from the project page...

The interface of the ff_kerneldeint.dll has changed, since Haruhiko removed the pointer to the frame copy function and replaced it with inline code. The installer, though, uses an ICL10 compiled, now incompatible ff_kerneldeint.dll which crashes because the signature of IkernelDeint::createI doesn't match anymore.

Could someone please recompile the ff_KernelDeint project with ICL10 and put that in bin\distrib\InnoSetup\icl10?

Thanks!

np: Tom Pooks - Pablo Assan's (Speicher CD3)

[Inventive Software]

Was gonna post a bug report along the lines of "Kernel deinterlacer still crashes my ffdshow... HELP!", but thanks for bug-probing Leak.

[Leak]

Originally Posted by Inventive Software

Was gonna post a bug report along the lines of "Kernel deinterlacer still crashes my ffdshow... HELP!", but thanks for bug-probing Leak.

Well, the fun thing here is that if you just compile ffdshow KernelDeint will work as expected - it's only when you build an installer that the problem will crop up...

In the meantime, here's a vanilla 1601 build with a MSVC compiled but current and working ff_kerneldeint.dll.

EDIT: Then again, the files have just been updated in SVN and a new build is up at SourceForge... that was fast...

np: Kode9 & The Spaceape - Portal (Memories Of The Future)

[Adub]

Okay well, ilpippo80's fix worked perfectly. A soon as I removed the avisynth_c.dll, then everything worked fine.

Thanks for the fix.

[Inventive Software]

Kernel deint bug fixed! Way to go peeps!

[rickardk]

I have a big problem:

I have remuxed (into ts) some of my HD DVD and Blu-ray collection to be able to play them in Vista Media Center.

All Blu-rays (both AVC and VC-1) plays perfect using ffdshow (libavcodec etc).

But the HD DVDs I have remuxed appears to be jerky. And frames are droped.

ffdshow reports 29.976 fps but I know all titles are ~23.976 fps.

I guess the same problem is described here:
http://forum.doom9.org/showthread.ph...26#post1019426

http://www.avsforum.com/avs-vb/showt...php?p=12182286


Any idea why?
Is this a Haali splitter problem or is it a decoder problem?

Some HD DVD titles that stutters plays perfectly smooth in PowerDVD. Others are unwatchable in PowerDVD (bad stutters and frame drops) but just stutters slighlty when using Vista Media Center (ffdshow).

[haruhiko_yamagata]

Originally Posted by rickardk

I have a big problem:

I have remuxed (into ts) some of my HD DVD and Blu-ray collection to be able to play them in Vista Media Center.

All Blu-rays (both AVC and VC-1) plays perfect using ffdshow (libavcodec etc).

But the HD DVDs I have remuxed appears to be jerky. And frames are droped.

ffdshow reports 29.976 fps but I know all titles are ~23.976 fps.

I guess the same problem is described here:
http://forum.doom9.org/showthread.ph...26#post1019426

http://www.avsforum.com/avs-vb/showt...php?p=12182286


Any idea why?
Is this a Haali splitter problem or is it a decoder problem?

Some HD DVD titles that stutters plays perfectly smooth in PowerDVD. Others are unwatchable in PowerDVD (bad stutters and frame drops) but just stutters slighlty when using Vista Media Center (ffdshow).

I'm just waiting for the new release from Haali.

[yesgrey]

rickardk,
If you convert your ts into a mkv file, you will be able to set the correct 23.976 fps in the mkv and watch your files smoothly.

[Zarch_UK]

Guys, wonder if you can help me out.

I'm using ffdshow tryouts on XP to decode mpeg4-avc files in mkv containers thru halli splitter.

The file details are:
1080p MKV File
1920x800
10675 bitrate MPEG4-AVC Video
DTS Audio

When i play this file on my Core 2 Duo E4400 and ATI2600Pro graphics card i have real problems with the codec only running on one core of my processor. (see pic). GPU stats are at the bottom too.



This is obviously causing stuttering etc as the 2nd core is maxxed out, but the first core virtually idle. The only image amendment i currently have ticked is deinterlace, i do no resizing or sharpening etc.

I have changed "Number of decoding threads" to 2 as per the FAQ, but i believe this to be for mpeg2 only anyway.

Any ideas on how to get this working properly on both cores of my processor?

I have tried early versions like the stable version from july (rev 1324), right upto the latest rev1589, with no difference.

[clsid]

The H.264 decoder is NOT multithreaded (except for slices, which you file obviously doesn't have).

[Zarch_UK]

Originally Posted by clsid

The H.264 decoder is NOT multithreaded (except for slices, which you file obviously doesn't have).

Thanks for prompt reply clsid. Do you know if there is any plans to make the h264 decoder multi-threaded or should i look at an alternative codec for my 1080p versions of files? (as 720 versions get by on 1 core!)

I take it the use of "slices" is a decision taken by the encoder? (and not a popular one by the looks of it)

Any advice you can give would be appreciated.

[Kado]

@Zarch_UK

Originally Posted by clsid

Mid term ToDo list:

1. Frame-based multithreaded H.264 decoding. (Note: this will be added once it has been committed to FFmpeg SVN)

You can always buy the coreavc decoder v1.5, supports up to quad-core cpu's or wait for FFmpeg development.

[_xxl]

h264 multithreading, chapter 2:
http://lists.mplayerhq.hu/pipermail/...er/035877.html
Bin:
http://rapidshare.com/files/69428157...3_xxl.exe.html
Patch:
http://rapidshare.com/files/69429065/h264.mt.patch.html

[Zarch_UK]

Originally Posted by Kado

@Zarch_UK

You can always buy the coreavc decoder v1.5, supports up to quad-core cpu's or wait for FFmpeg development.

Apologies, didn't see the "ToDo List". Are there any timescales on the FFmpeg project?

[clsid]

There only is an experimental patch available at the moment. That is used in the build that _xxl posted above.

Is is unknown when or if that patch gets included in the official FFmpeg codebase.

[rickardk]

Originally Posted by yesgrey3

rickardk,
If you convert your ts into a mkv file, you will be able to set the correct 23.976 fps in the mkv and watch your files smoothly.

Thanks!

Will try it with Godfellas...



Godfellas gives me this result right now:
Perfect in PowerDVD
Stutters with VC-1 codec from VMP11
Speeds up and looses sync with libavcodec (ffdshow)



Some HD DVDs put in TS gives me smooth playback both inside PowerDVD and with Windows Media Player (microsofts VC-1 codec)

Some stutters with a constant interval.
Some drops frames randomly.

Happens for both VC-1 and AVC.

But again, from Blu-ray all VC-1 and AVC (H264) titles put in ts plays perfectly smooth.


As ffdshow reports a frame rate of 29.976 fps I guess there are some flags that screw things up.

[yesgrey]

Originally Posted by rickardk

As ffdshow reports a frame rate of 29.976 fps I guess there are some flags that screw things up.

Yes, it should be. Maybe it's something related with the HD-DVD format specification not allowing 23.976 fps output... See this thread.

With a mkv file you could use MKVtoolnix to correct the timecodes to 23.976. I have tryed with The Matrix trilogy and it worked great.

[madshi]

Originally Posted by yesgrey3

Maybe it's something related with the HD-DVD format specification not allowing 23.976 fps output...

There are 2 errors in that statement.

[vlada]

I found a strange bug in latest versions of ffdshow. If I want to play a movie in ViPlay 2.09, it says it cannot find audio decoder for AC3/AAC/Vorbis etc. which should be treated by ffdshow. MP3 plays through FhG decoder, although MP3 is enabled in ffdshow.

The strange thing is that the ffdshow audio decoder works in WMP, MPC or Graphedit. So it looks like a problem caused by ViPlay. But if I revert back to ffdshow beta3, audio plays fine with ffdshow and ViPlay.

Could someone please confirm this behavior? The player can be downloaded here.

[haruhiko_yamagata]

Originally Posted by vlada

I found a strange bug in latest versions of ffdshow. If I want to play a movie in ViPlay 2.09, it says it cannot find audio decoder for AC3/AAC/Vorbis etc. which should be treated by ffdshow. MP3 plays through FhG decoder, although MP3 is enabled in ffdshow.

The strange thing is that the ffdshow audio decoder works in WMP, MPC or Graphedit. So it looks like a problem caused by ViPlay. But if I revert back to ffdshow beta3, audio plays fine with ffdshow and ViPlay.

Could someone please confirm this behavior? The player can be downloaded here.

ViPlay.exe does not exist in the white list. ViPlay3.exe exists though. I have added it.
Did you uncheck 'Show dialog when an unknown application tries to load ffdshow.'?

[yesgrey]

Originally Posted by madshi

There are 2 errors in that statement.

It would be helpfull if you pointed the errors. I have put a maybe because I was not sure about it...

[madshi]

Originally Posted by yesgrey3

It would be helpfull if you pointed the errors. I have put a maybe because I was not sure about it...

(1) The format is named "HD DVD" and not "HD-DVD".
(2) The HD DVD format specification does allow 23.976p output.

Amir has specifically confirmed (2). Furthermore, if the specification didn't allow 23.976p output - how could the Toshiba XA2 offer that functionality then?

[Leak]

Originally Posted by madshi

Amir has specifically confirmed (2). Furthermore, if the specification didn't allow 23.976p output - how could the Toshiba XA2 offer that functionality then?

Easy - applying IVTC after decoding...

[madshi]

Originally Posted by Leak

Easy - applying IVTC after decoding...

If the specification would not allow it, Toshiba would be breaching the specification/license. They'd get into legal trouble doing that.

Anyway, Amir (Microsoft) has specifically stated that it is perfectly legal for a HD DVD player to ignore the pulldown flags in the video stream and output 24p (respectively 23.976p) straight. And from what I've heard the Toshiba is not doing IVTC, but it's passing the original progressive data through (fortunately).

On another occasion amirm has also confirmed at avsforum that all HD DVD titles he had his hands on were indeed encoded at 23.976fps.

[ACrowley]

Originally Posted by honai

On another occasion amirm has also confirmed at avsforum that all HD DVD titles he had his hands on were indeed encoded at 23.976fps.

ofcourse...no doubt ?!

I made 1080p x264 reencodes from almost all HDDVD Titles ..and all HDDVDs were 23.976Fps

@XXL
Your latest Mutithread build have still the async Problem on my Sys.
Its a constant + Delay on 1080p H264...again its not a to slow Sys Cpu ( A 64 X2 5000)

[rickardk]

On which level are my HD DVDs in ts recognized as 29.976fps then?

And why does some titles speed up, some stutters in a constant pattern, some skips frames (can't find a pattern here but it's always the same frames) and some plays 100% smooth.

But as soon as I reencode them (x264 put in mkv) all titles playback 100% smooth.

Is it the splitter (Haali) or the decoder?

I really would love to have smooth playback without putting everything in mkvs.

[TheShadowRunner]

Follow up on my frameskip/stutter problem on h.264 with high bitrates, of course CLSID & Haruhiko were right on the money. 1 Core is maxed out when that happens and the other almost idle.
Can't wait for multithreading if that feature ever makes it!
See you,

TSR

[yesgrey]

Originally Posted by madshi

(1) The format is named "HD DVD" and not "HD-DVD".

I assume the error. I have been in HD DVD site and that's the correct name. But in lots of places we see HD-DVD. I think it should be a confusion with blu-ray, because this has the '-'.

Originally Posted by madshi

Furthermore, if the specification didn't allow 23.976p output - how could the Toshiba XA2 offer that functionality then?

If you read my link above, it was said that the specification was recently updated to allow it, and that will be firmware updates for some of the players (the XA2 is referred).

[madshi]

Originally Posted by yesgrey3

If you read my link above, it was said that the specification was recently updated to allow it

Sais one unknown guy on a forum. Is that your only source?

Anyway, even if true, the specification does allow 24p output right now. There's no doubt about that.

[ACrowley]

Originally Posted by rickardk

On which level are my HD DVDs in ts recognized as 29.976fps then?

And why does some titles speed up, some stutters in a constant pattern, some skips frames (can't find a pattern here but it's always the same frames) and some plays 100% smooth.

But as soon as I reencode them (x264 put in mkv) all titles playback 100% smooth.

Is it the splitter (Haali) or the decoder?

I really would love to have smooth playback without putting everything in mkvs.

You can remove the 29.97 FPS Pulldown to get 23.976Fps with vc1con
Then remux it into TS

For mkv remuxing , set a 23.976 Timecode

Somewhere here is a Thread why ffdshow/WMVideoDecoderDMO outputs 29.97 FPs instead of pulldown 23.976FPS
I think ist Splitter related

And a reecoded x264 Video has nothing to do with the untouched Source, you cant compare it
Other Codec, Bitrate, Settings etc etc

By the Way..its a well known Problem that VC1 1080p is not playing smooth on "some" Systems..even with enough CPU Power.
I cant play VC1 HDDVD smooth on my X2 5000, not with ffdshow not with MS VC1 Decoder. Even not when its proper remuxed
Only Cyberlink VC1 Decoder into PowerDVD plays it smooth, but i can paly all H264 BluRay smooth with CoreACv or Cyberlink

Somewhere here is a Thread why ffdshow/WMVideoDecoderDMO outputs 29.97 FPs instead of pulldown 23.976FPS

In my setup I do get 23.976fps with ffdshow/WMVideoDecoderDMO.

its a well known Problem that VC1 1080p is not playing smooth on "some" Systems..even with enough CPU Power

This is the first time I heard about it.

[rickardk]

Originally Posted by ACrowley

You can remove the 29.97 FPS Pulldown to get 23.976Fps with vc1con
Then remux it into TS

For mkv remuxing , set a 23.976 Timecode

Somewhere here is a Thread why ffdshow/WMVideoDecoderDMO outputs 29.97 FPs instead of pulldown 23.976FPS
I think ist Splitter related

And a reecoded x264 Video has nothing to do with the untouched Source, you cant compare it
Other Codec, Bitrate, Settings etc etc

By the Way..its a well known Problem that VC1 1080p is not playing smooth on "some" Systems..even with enough CPU Power.
I cant play VC1 HDDVD smooth on my X2 5000, not with ffdshow not with MS VC1 Decoder. Even not when its proper remuxed
Only Cyberlink VC1 Decoder into PowerDVD plays it smooth, but i can paly all H264 BluRay smooth with CoreACv or Cyberlink

Thanks for all advices!

VC-1 from Blu-rays plays 100% smooth. Actually I have remuxed 43x Blu-ray titles (mpeg2, VC-1 and H264) into ts. NONE of them stutters.

My HTPC config should be enough (I think):
Q6700@~3Ghz
4GB
8600GTS


The thing I can't understand is why there are diffrent type of stutters and frame drops from title to title.


I apologise if you guys don't think this discussion belongs in this thread. I guess it's a Haali problem rather then a ffdshow problem?

But it would be great if we could find a solution so we can use ffdshow to decode remuxed HD DVDs!

[Inventive Software]

It'll most likely be a high bitrate problem, since it'll vary from title to title, like you're experiencing.

[rickardk]

Originally Posted by Inventive Software

It'll most likely be a high bitrate problem, since it'll vary from title to title, like you're experiencing.

But they play 100% perfect direct from disc (PowerDVD Ultra 7.3).

[Inventive Software]

That's because it's a constant read rate from the disc, and not from the HD.

[vlada]

Originally Posted by haruhiko_yamagata

ViPlay.exe does not exist in the white list. ViPlay3.exe exists though. I have added it.
Did you uncheck 'Show dialog when an unknown application tries to load ffdshow.'?

Thanks, that was the problem. I have no idea why, but I had the dialog to ask about using ffdshow unchecked. I feel stupid for not looking into it, because I knew about the black/white list in ffdshow.

Btw. you could also add ViPlay4.exe to the whitelist. The Urusoft team just released beta of version 4. I somehow don't understand their development. They never finished version 3 and now they have v4 beta.

[rickardk]

Originally Posted by Inventive Software

That's because it's a constant read rate from the disc, and not from the HD.

Ok, maybe you are right...

But all VC-1 titles (remuxed into ts) from Blu-ray are smooth.
And most VC-1 titles (remuxed into ts) from HD DVD are smooth when played back in PowerDVD.
If I make an iso of a title it plays 100% smooth with PowerDVD.


I really think there is something else going on here...



Done some serious playback testing today and this is what I found:



When a VC-1 title from a HD DVD are remuxed into ts:

---------PowerDVD
Will play most titles smooth.
---------


---------WMP/VMC with ms VC-1
Will play it back at 23.976fps (according to EVR). I found 3 types of playback when going through my collection.

1. Plays 100% smooth.

2. With stutters and/or framedrops. The stutters will appear at exactly the same place everytime you watch the title. But the stutters will not appear in a constant intervall. Some of

3. Sometimes it plays with stutter at a constant interval. And sometimes it plays back 100% smooth. Really strange... I think it has something to do with what was played just before.


Keep in mind, all the above are 100% smooth in PowerDVD
---------


----------WMP/VMC with ffdshow (libavcodec)
Will play it back at 29.976fps loosing sync etc

Also, if I set ffdshow to raw when using microsofts VC-1 decoder it will speed up to 29.976fps.
---------




When a H264 title from a HD DVD are remuxed into ts:

---------PowerDVD
Some titles have stutters at a constant intervall.
Some are 100% smooth
---------


---------WMP/VMC with Cyberlink H264
Now it get strange. You may think that you should get the same result as with PowerDVD (same decoder)
But some titles that always stutters in PowerDVD will play 100% smooth in WMP with same decoder but diffrent splitter (haali vs cyberlink)

And some titles that plays 100% smooth in PowerDVD will now stutter with a constant interval.
---------

---------WMP/VMC with ffdshow (libavcodec)
Playback at 29.976fps loosing sync etc.
---------


I really can't understand what's going on here...

[Thunderbolt8]

could someone please explain what nellymoser's audio decoder can do?

[Leak]

Originally Posted by Thunderbolt8

could someone please explain what nellymoser's audio decoder can do?

Here you go...

Why they chose just about the dumbest name possible for their company is beyond me, though...

[foxyshadis]

Originally Posted by Leak

Here you go...

Why they chose just about the dumbest name possible for their company is beyond me, though...

OT, sorry... This is a Nelly Moser, I have a couple.


It's still a weird name for a company.

[Jay Bee]

Originally Posted by ACrowley

By the Way..its a well known Problem that VC1 1080p is not playing smooth on "some" Systems..even with enough CPU Power.
I cant play VC1 HDDVD smooth on my X2 5000, not with ffdshow not with MS VC1 Decoder. Even not when its proper remuxed
Only Cyberlink VC1 Decoder into PowerDVD plays it smooth, but i can paly all H264 BluRay smooth with CoreACv or Cyberlink

You can add me to the list. I have an X2 4200+ @ 2.8 GHz and can only play VC-1 100% smooth with PDVD. As far as I can tell it's because ffdshow is single threaded and the microsoft decoder is so inefficient that it needs two cores and a GPU but ATI doesn't support VC-1 in Windows XP even if they claim to do so (at least not with my card, x1950 pro).

[ACrowley]

Yeah...is think ist a Multithreading Issue.
But WMVideoDecoder pushes both Cores ...maybe not efficient enough on Athlon Systems. I
dont know Cyberlink VC1 Decoder works perfect and fully smooth on VC1...to bad that we cant use the Decoder outside PowerDVD

With SonicVC1Decoder and SonicCinemasterVideoDecoder4.3 and WMVideoDecoderDMO its all the same.
Video starts playing smooth with ~24fps and after a few seconds the fps goes down to ~15 and it starts stuttering

Also the WMVideoDecoder cant use DXVA on VC1...DXVA works nice on WM9 MP

So far i Know ,the Problem only exist for A64 Users...all Intel CoreDuo User "i" know are having no Problems

No Rpoblme for my because I dont watch the untouched VC1, i always reencode it to 1080p x264 with German and English Audio.
However...im going to by to replace my A64 X2 5000 with a Intel Quad this Year

@RichardK
As i told you :
To get alltime 23.976 simply remove the Pulldown from HDDVD VC1 with vc1conv .
The remux it into your TS .
For MKV :
1. gdsmux (haali) muix EVO VC1 into mkv. You only get 29.97 FPS !! So youve to remux it :
2.mkvmerge : remux it with 23.976 Timecode
So you have solved you 29.97 fps Problem

Yeah...is think ist a Multithreading Issue.

It's not.

But WMVideoDecoder pushes both Cores ...maybe not efficient enough on Athlon Systems.

Not true.

Also the WMVideoDecoder cant use DXVA on VC1

It's not necessary. Even w/o DXVA VC-1 streams will play stutter-free on any dual-core CPU above 2.0 GHz.

So far i Know ,the Problem only exist for A64 Users

Check your setup. No problems at all on any of my X2 machines.

However...im going to by to replace my A64 X2 5000 with a Intel Quad this Year

If your new system is as misconfigured as the old one you'll have the same problems. Those are not related to CPU power but to your specific setup.

As far as I can tell it's because ffdshow is single threaded

It's not because of that.

and the microsoft decoder is so inefficient

It's not. All VC-1 streams from commercial discs play fine on my X2 @ 2.4 GHz even at high bitrate. The problem is your setup, not the decoder and not ffdshow.

And it has nothing to do with the GPU either. My video card doesn't support VC-1 at all, and still all videos play smooth.

Again, check your setup and make sure that it's configured correctly. My recommendation:

* latest VC-1 decoder (wvc1dmo.exe 6.0.2600.0)
* MPC or ZoomPlayer
* video renderer set to Haali or VMR9
* ffdshow, in Codec section choose WMV9 for WVC1 decoding

This will play any VC-1 stream out there even on an X2 @ 2.4 GHz absolutely stutter-free.

[Jay Bee]

Originally Posted by ACrowley

Also the WMVideoDecoder cant use DXVA on VC1

It also works for VC-1. You can use WMV9 PowerToy to enable logging to see if it's working.

Originally Posted by ACrowley

So far i Know ,the Problem only exist for A64 Users...all Intel CoreDuo User "i" know are having no Problems

Hmm, ever tried any of the "AMD Dual-Core optimizer" thingies? I haven't because I wasn't aware of anything that needed fixing.

Originally Posted by honai

It's not necessary. Even w/o DXVA VC-1 streams will play stutter-free on any dual-core CPU above 2.0 GHz. My video card doesn't support VC-1 at all, and still all videos play smooth.

Have you tried disabling DXVA using the tool I linked to above? Because when I did that on a different C2D Nvidia machine it started decoding VC-1 as bad as the X2 ATI machine. What GPU do you have?

Originally Posted by honai

* latest VC-1 decoder (wvc1dmo.exe 6.0.2600.0)

What's this? According to Zoom Player the WMVideo DMO is using qasf.dll 11.0.5721.5145. This was installed with WMP 11, which is the recommended method to get the newest decoder. I think your decoder is older but I'll give it a try.

And I'm using Zoom Player and VMR7 works better then any of the other renderers.

[Jay Bee]

OK, new information.
"wvc1dmo.exe 6.0.2600.0" installs "WMV9 Advanced Profile decoder". This is an older decoder than the "qasf.dll" I mentioned earlier. And guess what: It does use a lot less CPU decoding VC-1! While the specific file I was trying play doesn't work with "WMV9 Advanced Profile decoder" (black screen) I still think that this might explain some of the VC-1 problems people are having. And another thing I remember about "WMV9 Advanced Profile decoder", from discussions with Zambelli, is that it doesn't support hardware deinterlacing, but since there isn't much interlaced VC-1 around this shouldn't be too much of a problem.

Sorry for the off topic.

[ACrowley]

wvc1dmo.exe 6.0.2600.0 is outdated
http://forum.doom9.org/showthread.ph...ght=standalone

WMVideoDecoder from latest WMV11SDK with WMVVideoDecoderDMO is the newest Build

http://www.citizeninsomniac.com/WMV/wmfdist11.exe

I think Zambelli can confirm it

EDIT
Yes, ofcourse the Decoder/Codec is included in WMP11..the link above is for the standalone Version

[Shakey_Jake33]

So we should all install the above file, to benefit from improved VC1 playback?

[clsid]

WMV11SDK is also included with WMP11.

[Shakey_Jake33]

Just been reading through the SVN changelog, and it says the AC3/DTS issue requiring the 'Maximise SPDIF Compatibility' checkbox has been fixed.
I seem to remember the change (which required the checkbox to the checked on some decoders) was a change that increased the bandwidth of DTS? Can someone remind me what that was all about?

[haruhiko_yamagata]

Originally Posted by Shakey_Jake33

Just been reading through the SVN changelog, and it says the AC3/DTS issue requiring the 'Maximise SPDIF Compatibility' checkbox has been fixed.
I seem to remember the change (which required the checkbox to the checked on some decoders) was a change that increased the bandwidth of DTS? Can someone remind me what that was all about?

S/PDIF pass through of DTS with odd number frame size, i.e. 2013 bytes etc, was broken because of an old bug.
albain tried to fix it and required the check box.
Very high bit rate DTS (>1.536mbps) cannot be passed through to S/PDIF anyway.

[SeeMoreDigital]

Originally Posted by haruhiko_yamagata

S/PDIF pass through of DTS with odd number frame size, i.e. 2013 bytes etc, was broken because of an old bug.
albain tried to fix it and required the check box.
Very high bit rate DTS (>1.536mbps) cannot be passed through to S/PDIF anyway.

I currently have "ffdshow_rev1611_20071116_clsid.exe" installed and pass-thru' of DTS (.dts) at 1536Kbps is working okay for me.... The little DTS icon on my DSS amp lights up and everything

[clsid]

SMD, that is good news.

Everyone else with S/PDIF connection also please test passthrough of AC3 and DTS.

We are a step closer towards beta4 now.

[Shakey_Jake33]

New fix works for me too, both DD and DTS. Previously required the 'Maximise SPDIF Compatibility' checkbox ticked.

Speakers/receiver are Logitech Z-5500.

[amt]

AC3 and DTS 1.5 Mbps confirmed working over S/PDIF using ffdshow_rev1611_20071116_clsid.exe

[Thunderbolt8]

I have a problem with remuxing the galapagos HD DVD: I cant get a picture using ffdshow as decoder. from what I heard this HD DVD is telecine and 29.97 fps and this seems to cause trouble for ffdshow. at least it only gives a black screen when trying to playback a remuxed .mkv file (with gdsmux or mkvmerge) of the original VC-1 video stream. every other normal 23.976 fps VC-1 video from remuxed HD DVDs or blue-rays plays fine. I tried it xxl version 1604 and clsid 1611, same result with both. when I use the WMVideo Decoder DMO the playback is fine.


edit: it works when I choose wmv9 instead of libavcodec for VC-1 (unlike to all other VC-1 material desrcibed above). is there actually a difference in picture quality between wmv9 and libavcodec? wmv9 is horribly slow :S

and any way to get rid of that problem? (look at the wings) or is this because of the telecine aspect? I guess this is not the intended way the picture should be presented, is it?
http://www.freepicshost.net/pview.ph...1117172026.PNG

[fastplayer]

Subtitles that contain an "&" as part of the filename are displayed incorrectly inside ffdshow's context menu.
Example:
Bla & bla.srt becomes Bla _bla.srt

Edit: Bug fixed with rev1626.
Thanks, Haruhiko!

[foxyshadis]

Unfortunately, libavcodec's VC-1 implementation isn't quite finished, let alone bug-free, so using Microsoft's is the main choice for HD material. ffdshow can only use what ffmpeg gives us, mainly.

fastplayer, good catch.

[Thunderbolt8]

hm I can accept that it will a bit longer until its completed. however, what can I do about that telecine thing of that movie, I guess the way it looks on that picture is not intended? how can I get rid of it?

[Jay Bee]

Are you sure it isn't interlaced? Looks to me like the deinterlacer is using the wrong method (blend instead of bob). Getting VC-1 to deinterlace properly is a bit involved and is decribed by Zambelli here. In your case try using WMV9 PowerToy to disable the internal deinterlacer and see what you get.

[Thunderbolt8]

in that picture (taken with ffdshow) deinterlacing was disabled.
are you sure you posted the right link? otherwise I guess I dont really understand what is meant there, sorry.
I guess I have the lastest version of the WMVideo Decoder DMO installed, but I cant choose any deinterlacing options for that filter with mpc afaik.

[Jay Bee]

Yes, it was the right link. The WMV9 decoder has an internal (not very good) deinterlacer that can be deactivated using Zambelli's WMV9 PowerToy. Even if deinterlacing is deactivated in ffdshow and MPC the decoder may still be doing its own deinterlacing. Just try if it makes a difference.

[Thunderbolt8]

opened the toy and everyting including deinterlacing is not checkmarked. so I guess that does mean that disabling was disabled before anyway? set it to disabled just be sure.
what now, should I leave it that way? should I also enable DXVA WMV Support?
how can I get rid of that telecine look then?

btw. decoder info: wmvdecod.dll, 11.0.5721.5145, Priority: 256

[vlada]

I have 2 TS/AVC samples which cause some problems in ffdshow (blockiness, macroblocks, totally distorted picture). My question is, how can I check if it's a problem of ffdshow or ffmpeg? Is MPlayer suitable to test it?

Anyway at http://www.sendspace.com/file/4ns9wf is an AVI sample I created from the TS. I wanted to open it in VirtualDub, but it only displays one half of the picture. In DirectShow it plays O.K. So I do believe there is a bug in the VfW h264 codec.

P.S. tested with ffdshow rev1598.

Edit: Just tested the video in latest MPlayer and works O.K. (just choppy, because my computer isn't fast enough).

[Maciek]

I guess this is the best place on the internet to ask this question, so...

I finally got a good computer and I wonder what are the best possible quality settings for ffdshow. I mean, let's say that the computer doesn't matter and I just want best quality possible - both audio and video. What should I change in ffdshow's settings?

[foxyshadis]

Thunderbolt8, to get rid of the combing, you have to actually turn deinterlacing/IVTC on, either in wmv9 or in ffdshow.

vlada, yes, current SVN of either mplayer or VLC.

Maciek, there's really no such thing. It much too dependent on what you're watching and what your personal tastes are, otherwise we'd already have 'best' settings all turned on. Post-processing is usually a bare minimum unless you have very high quality movies, though.

Besides, for best quality of things like deinterlacing and antialiasing, you could never run them in realtime even on the most powerful PC in existence. (Some 32-cpu xeon servers, perhaps.)

[Jay Bee]

Thunderbolt: so is there a difference between "Internal Deinterlacer" on and off? It's under the "decoder" tab and don't forget to click apply and re-open the file you're decoding.

[Thunderbolt8]

Originally Posted by Jay Bee

Thunderbolt: so is there a difference between "Internal Deinterlacer" on and off? It's under the "decoder" tab and don't forget to click apply and re-open the file you're decoding.

no, didnt really see a difference :S I wasnt able to compare same frames, but it still looked as strange as in that bird picture sometimes.

Originally Posted by foxyshadis

Thunderbolt8, to get rid of the combing, you have to actually turn deinterlacing/IVTC on, either in wmv9 or in ffdshow.

this is the problem, afaik the IVTC option is under libavcodec setting at the bottom of ffdshow configuration bars, but this video wont work with libavcodec.

I tried to apply ffdshow deinterlacing (cubic interpolation, ffmpeg deinterlacer), but the pictures still looked like this.

cubic: http://www.freeimagehost.eu/image/f2c1471515069

ffmpeg: http://www.freepicshost.net/pview.ph...ame=ffmpeg.PNG

[Jay Bee]

Could you upload a small sample?

[Thunderbolt8]

hm how small? its a source file with ~25mbit, im not sure how much could be seen with a small sample. and how could I find an adequate position to take that sample from? I would take the bird scene of course, but how can I find out where that is, when using a hex editor to cut for example?

edit: compared again, theres indeed a difference between when I disable wmv deinterlacing and then activate ffdshow deinterlacing. it looks a bit better then, but still cant change that strange look of the screenshots.

heres a sample, 25 seconds, 55mb. with ffdshow deinterlacing I can get rid of the incerlacing problem, but I cant change the way it looks like on the screenshots :S
http://www.megaupload.com/de/?d=7QIPXP6I

[Jay Bee]

Thx. I tried all kinds of deinterlacers but I couldn't get rid of the double wings either. I guess that's just the way it was encoded.

[Thunderbolt8]

what was about that IVTC thing, could this be a method to get rid of it?

[G_M_C]

Question; Does FFDShow_Tryouts work on Vista 64 bit / Intel E6750 combo ?

I've build myself a new machine with a E6750 and 4 Gb of fast memory, and i want to start using a 64 bit system (since it is better at handling mem > 3 Gb). But i'm lost in the quest for software that i want to "take with me" from Xp/32bits to Vista/64bits.

[clsid]

It will work if you use a 32-bit player. For Media Center you'll need a 64-bit build.

[fastplayer]

(Optional) single left-click on tray-icon opens ffdshow's dialog.

µtorrent is a good example for this feature because some single-click apps require the taskbar to be focused or open the dialog inactively (in the background).

[Shakey_Jake33]

Originally Posted by G_M_C

Question; Does FFDShow_Tryouts work on Vista 64 bit / Intel E6750 combo ?

I've build myself a new machine with a E6750 and 4 Gb of fast memory, and i want to start using a 64 bit system (since it is better at handling mem > 3 Gb). But i'm lost in the quest for software that i want to "take with me" from Xp/32bits to Vista/64bits.

As clsid says really. I think you are wise going 64-bit now, as everyone will have to eventually, but 64-bit codecs are quite behind everything else. There is a 64-bit build of ffdshow, but it's quite and old build, never really updated, and that's about all there is. Your best sticking to 32-bit codecs for now.

Vista Media Center in Vista 64 does require 64-bit codecs though, so you'll need 64-bit ffdshow for that. Also, Vista 64 comes with both 32 and 64-bit builds of WMP, and a 64-bit build of Media Player Classic exist, if you want to play with them anyway. The 64-bit codecs install alongside the 32-bit ones, so there's no harm in installing both.

[Inventive Software]

Suggestion for install script: add "post-processing" to the filters that are enabled after installing.

[vlada]

Thunderbolt8
Could you upload the sample on a more accessible server, like Sendspace.com? I would have a look at the video. But I can't download from Megaupload.

[Thunderbolt8]

Originally Posted by vlada

Thunderbolt8
Could you upload the sample on a more accessible server, like Sendspace.com? I would have a look at the video. But I can't download from Megaupload.

generally not, or because of language issue?

done. http://www.sendspace.com/file/lvlvtv

[vlada]

Megaupload says there are no free slots for my country (all the time). Anyway back to the video: I looked on it in VirtualDub through Avisynth -> ffdshow (wmv9). If I used libavcodec in ffdshow, I didn't get any picture.

Looking at the video I don't think the double contours are created because of blending telecined frames. If you move frame by frame, you can see that there are alway 2 pictures (frames) displayed at the same time. While one of them is becoming more visible, the other one is getting more transparent. After each 6 frames, only one of them is visible and the cycle begins again. This seems like a blend framerate conversion, something similar to what ConvertFPS from Avisynth produces.

I'm afraid there is no way to fix the video after this framerate conversion. It looks like conversion from 30fps to 24 or vice versa. But it is strange that the movie seems to be 25fps.

What do you think?

[G_M_C]

Originally Posted by Shakey_Jake33

As clsid says really. I think you are wise going 64-bit now, as everyone will have to eventually, [...]

Yes, that was my initial thought also;

But reading more and more, i am starting to get the feeling that 64 bit might be OK, but Vista isnt; And Vista 64 least of all. There are the different compatibillity issues, but there is also the ludicrous DRM implementation, that spoils Vista-64 bit the most.

But as i said, im still reading on about it, and am still in the beginning of my upgrade-path (getting together all software/codecs needed to make the transition to a new OS and a new number of bits)

[Shakey_Jake33]

It's really not a big transition, I've been using Vista 64 since Febuary. You use all the same players and codecs that you used in XP-32.

[Thunderbolt8]

Originally Posted by vlada

Megaupload says there are no free slots for my country (all the time). Anyway back to the video: I looked on it in VirtualDub through Avisynth -> ffdshow (wmv9). If I used libavcodec in ffdshow, I didn't get any picture.

Looking at the video I don't think the double contours are created because of blending telecined frames. If you move frame by frame, you can see that there are alway 2 pictures (frames) displayed at the same time. While one of them is becoming more visible, the other one is getting more transparent. After each 6 frames, only one of them is visible and the cycle begins again. This seems like a blend framerate conversion, something similar to what ConvertFPS from Avisynth produces.

I'm afraid there is no way to fix the video after this framerate conversion. It looks like conversion from 30fps to 24 or vice versa. But it is strange that the movie seems to be 25fps.

What do you think?

hm thats strange because the video is supposed to be 29.97 fps
im also sure I didnt change the fps, because everytime I tried so with a time code file with 23.976, 24, 25, 29.97 fps the muxing just crashed. I had to leave it at its default, which should be the original 29.97 fps. must look into it again.

edit: still think its 29.97fps. the framerate stays around 30 mostly at the beginning, before it might eventually drop down because the comp is too slow :P
another indication is that the audio is still in sync (although its hard to make out here of course) and since its the original audio of the 29.97 source this here should still be 29.97 fps too. so I guess its mostly a pc speed problem if the framerate drops, as this is really the video with most hardware requirement I had so far. temporarily switching to mpc internal matroska filter instead of haali can give some additional speed to compare.

I might be able to obtain a sample from an x264 encode in a day (not done by myself though) and since you have more possibilities at the encoding stage with the video I guess we might have a comparison then how its supposed to look.

[ToS_Maverick]

is it possible to implement a colormatrix-conversion into ffdshow?

it would be nice if ffdshow could extract the colormatrix information from an MPEG2 and H264 stream, and automatically apply the change.

there already is the option, what matrix to use, in Output/RGB conversion, but i cannot change it everytime i watch a different stream. i don't know every stream, and what matrix it uses

[vlada]

Thunderbolt8
I was guessing the framerate from VirtualDub/Avisynth. Most other tools didn't report any framerate. MPlayer also says 25fps.

I'll try to demux the video and remux it to another container and I'll see what happens then...

[Thunderbolt8]

Here a 25mb x264 (deinterlaced) sample (16 seconds) of ~same scene as the original video remux was:
http://www.sendspace.com/file/idpchq

and heres the link for the original remuxed video again:
http://www.sendspace.com/file/lvlvtv


so maybe its a bit easier to find out if these blend pictures of the bird's wings are fine they way they are in the source.

[drunken_clam]

Hi i've got a problem with h264/mp3 AVIs in my Video-Software (Premiere Elements 3/Premiere Pro CS3) decoded through ffdshow (ffdshow_beta3_rev1324_20070701_clsid).

All files i import have a lack of audio for about 1 second, when i play those files with VLC or MPlayer everything is fine so i think its a problem in ffdshow or the used version of libavcodec.

Heres a link to an example video.

http://rapidshare.com/files/71128576...k_min_nomc.zip

Anybody an idea? Can anyone reproduce this?

P.S. Please don't ask why i edit those high compressed files :-)

[vlada]

Very strange. The h.264 looks also bad, but somehow different. There are always 3 OK frames, but next 3 frames have "ghosts", then come another 3 good frames etc.

What is the original source?

[Thunderbolt8]

the link posted below the x264 encode is the original source
its the original vc-1 stream remuxed in .mv, no other changes or alternation of the video has been made
the screenshots were also from that original source

[Inventive Software]

Can somebody upload a checkout of the SVN sources? I'm going to take a look this weekend, and see what could/shouldn't be done about having a libavformat based filter.

[blizard]

Originally Posted by Shakey_Jake33

It's really not a big transition, I've been using Vista 64 since Febuary. You use all the same players and codecs that you used in XP-32.

Does Vista x64 work in the same way as Win XP x64 does? I use Win Xp x64 and it is very few thing as today that I would need to have in 64-bit to work and among codecs there is none from what I know of that I use. It is more important with capability to handle dual core or multicore (SMP) then 64 bit for encoding/decoding task. Please, correct me if I am wrong on this about SMP enabled processing being more vital for encoding/decoding then 64-bit. I wouldn't mind that both worked, but it seem until now that SMP have higher priority and is more of use then 64-bit.

Main software: K-Lite Codec pack Mega w. Media Player Classic + FFDSHOW, QT alternative and Real alternative, Haali Splitter + Renderer; SMplayer (celtic druids compile);+AutoMEN (MEncoder from Celtic Druid package as it have support for dual core).

[haruhiko_yamagata]

Originally Posted by Inventive Software

Can somebody upload a checkout of the SVN sources? I'm going to take a look this weekend, and see what could/shouldn't be done about having a libavformat based filter.

Just getting ffdshow's source code won't help. You'll need mplayer for libavformat and guliverkli or MPC home cinema as a sample of DirectShow source/splitter filter.
Please install tortoise svn and check out.

ffdshow tryouts: https://ffdshow-tryout.svn.sourceforge.net/svnroot/ffdshow-tryout/trunk
mplayer: svn://svn.mplayerhq.hu/mplayer/trunk
MPC home cinema: http://mpc-hc.svn.sourceforge.net/trunk

[Inventive Software]

Fair enough. Will checkout later.

[Leak]

Originally Posted by Inventive Software

Fair enough. Will checkout later.

Like Haruhiko said, this should make checking out the sources quite a bit easier...

[Inventive Software]

To quote the enviable Terry Pratchett: it's a Pun, or a "play on words". I'm aware of TortoiseSVN, but being at uni makes actually checking out the source damn near impossible, and I don't have TortoiseSVN installed yet on my computer at home. I will "check-out" the source (download it) this afternoon, and check out the source in my spare time this weekend (look through it).

Just nitpicking!

[haruhiko_yamagata]

Originally Posted by rickardk

I have a big problem:

I have remuxed (into ts) some of my HD DVD and Blu-ray collection to be able to play them in Vista Media Center.

All Blu-rays (both AVC and VC-1) plays perfect using ffdshow (libavcodec etc).

But the HD DVDs I have remuxed appears to be jerky. And frames are droped.

ffdshow reports 29.976 fps but I know all titles are ~23.976 fps.

I guess the same problem is described here:
http://forum.doom9.org/showthread.ph...26#post1019426

http://www.avsforum.com/avs-vb/showt...php?p=12182286


Any idea why?
Is this a Haali splitter problem or is it a decoder problem?

Some HD DVD titles that stutters plays perfectly smooth in PowerDVD. Others are unwatchable in PowerDVD (bad stutters and frame drops) but just stutters slighlty when using Vista Media Center (ffdshow).

I found the flag. Haali's media splitter is working perfectly.
Damn, I did check the AM_SAMPLE2_PROPERTIES::dwTypeSpecificFlags of the first and second frames, but didn't check the third...
I've modified the code a bit to set time stamp properly. Now the sync is OK, but it still has two problems. One is interlacing related flags to send to downstream, one is jerkiness caused by unloading and loading dlls during playback. Second problem is caused because ffdshow receives samples with media type attached and then re-initialize codec drivers.
The fix will surely break the code too much that I can't do without adding bugs. I'm going to do it after beta4. Please wait.

[Leak]

Originally Posted by Inventive Software

To quote the enviable Terry Pratchett: it's a Pun, or a "play on words".

I dunno, "Fair enough. Will checkout later." doesn't sound very punny to me... but what do I know?

Anyway, what university are you at that prevents using SVN over teh internets? Unseen University?

[molitar]

Is their a settings guide for improving playback of anime? I have tried a few of the DVD ones but they do not work very well especially with x.264 since playback is too slow with the dvd settings. How can I improve on sharpness with my setup?

WinXP Sp2
AMD 3000+ Processor
1GB Dual Channel DDR Ram

[Inventive Software]

Originally Posted by Leak

I dunno, "Fair enough. Will checkout later." doesn't sound very punny to me... but what do I know?

Anyway, what university are you at that prevents using SVN over teh internets? Unseen University?

Now you mention it.................

My uni kinda restricts installing software on their computers, and it's against the code of conduct. So I'd rather not be kicked off the course thanks to my curiosity in attempting to code something that'll most likely take half my uni life to complete!

I have checked out all bar mpc-hc that were suggested by Haruhiko, plus ffmpeg. TortoiseSVN complains that it can't find it. I'll grab guliverkli instead. I'll have a good look at all sources this weekend.

Ideally, what I'm aiming for is a container selection system, much like ffdshow, that you can pick and choose which containers libavformat parses. If I can get AVI working, I'll consider that a success!!!

[Px]

Originally Posted by molitar

Is their a settings guide for improving playback of anime? I have tried a few of the DVD ones but they do not work very well especially with x.264 since playback is too slow with the dvd settings. How can I improve on sharpness with my setup?

WinXP Sp2
AMD 3000+ Processor
1GB Dual Channel DDR Ram

Upgrade or overclock your CPU....

[Shakey_Jake33]

Originally Posted by molitar

Is their a settings guide for improving playback of anime? I have tried a few of the DVD ones but they do not work very well especially with x.264 since playback is too slow with the dvd settings. How can I improve on sharpness with my setup?

WinXP Sp2
AMD 3000+ Processor
1GB Dual Channel DDR Ram

On such a spec machine, CoreAVC is really the only choice for x264 material.

[blizard]

Originally Posted by molitar

Is their a settings guide for improving playback of anime? I have tried a few of the DVD ones but they do not work very well especially with x.264 since playback is too slow with the dvd settings. How can I improve on sharpness with my setup?

WinXP Sp2
AMD 3000+ Processor
1GB Dual Channel DDR Ram

What resolution are you talking about here? At the highest resolution of 1080p even high end hardware will have trouble to do too much work before rendering it to video card for display.

What kind of video card do you use? It is possible that you could make better use of your video card hardware features and another renderer before you send it to display. There is also the option of changing size on output screen to better fit original resolution from movie before you send picture to screen.

[Leak]

Originally Posted by Inventive Software

I have checked out all bar mpc-hc that were suggested by Haruhiko, plus ffmpeg. TortoiseSVN complains that it can't find it.

Well, of course TSVN can't find it - it's at

https://mpc-hc.svn.sourceforge.net/svnroot/mpc-hc/trunk

np: Bola - Rainslaight (Kroungrine)

[LotharZ]

Originally Posted by molitar

Is their a settings guide for improving playback of anime? I have tried a few of the DVD ones but they do not work very well especially with x.264 since playback is too slow with the dvd settings. How can I improve on sharpness with my setup?

WinXP Sp2
AMD 3000+ Processor
1GB Dual Channel DDR Ram

I'm watching anime at 720p x264 with

Win2003 SP1
AMD Sempron 2800+
480 MB DDR
ffdshow

without problems (CPU - YUV 50% & RGB32 85%).

But if you mean a AMD XP then you'll need to upgrade at least the CPU.

[RaynQuist]

Originally Posted by molitar

Is their a settings guide for improving playback of anime? I have tried a few of the DVD ones but they do not work very well especially with x.264 since playback is too slow with the dvd settings. How can I improve on sharpness with my setup?

WinXP Sp2
AMD 3000+ Processor
1GB Dual Channel DDR Ram

I think he's asking what post-processing he can add without making the video stutter, not how to achieve smooth vanilla playback.

LimitedSharpenFaster(ss_x=1.0)
Turning super sampling off makes it really cheap to run, but you'll get aliasing.

But before that do you have Deband on? Always do deband; it's awesome.

There are also line darkening and line thinning filters but I'm no experience with them. Have you tried all the ffdshow built-in sharpeners?

EDIT: if his rig really has dual channel DDR then it's socket 939, which means he has a K8 (Athlon64-ish) at 1.8GHz

[Inventive Software]

Originally Posted by Leak

Well, of course TSVN can't find it - it's at





np: Bola - Rainslaight (Kroungrine)

Ta!

[molitar]

Originally Posted by blizard

What resolution are you talking about here? At the highest resolution of 1080p even high end hardware will have trouble to do too much work before rendering it to video card for display.

What kind of video card do you use? It is possible that you could make better use of your video card hardware features and another renderer before you send it to display. There is also the option of changing size on output screen to better fit original resolution from movie before you send picture to screen.

Nvidia Geforce 6600GT 128MB
my screen resolution is 1024x768 32bit as default.

[blizard]

Originally Posted by molitar

Nvidia Geforce 6600GT 128MB
my screen resolution is 1024x768 32bit as default.

Do you use Nvidia PureVideo to help you to take advantage of your video cards hardware support in the rendering process? I use Haali Renderer as that renderer will make use of hardware features in processing video before it is sent to display.

As LotharZ have pointed out, you should be able to watch at least 720p with good result. You can also re-size output, to a smaller size for different type of post processing in FFDSHOW where the order from top to bottom will process your video before you send it to renderer. (Be aware that MPC as player can also auto re-size to fill your screen, make adjustment for aspect ration etc so there is many way to do this after you have processed video through FFDSHOW for decoding.) Here the order in the chain of FFDSHOW will matter as a large resolution will demand more CPU power to process video and not necessary give you any better result if it is to be displayed at 1024 x768 pixel. Resize & Aspect is a very good filte in FFSHOW to do this processing, but you will need to tweak it to learn what is the best in different situation. Different type of renderer can also make a difference in how well they wil support your need and hardware. LotharZ do also give you a suggestion to use make use of YUY2 or YV12 as colour compression output to video card as that will need less CPU power then to make use of any option in RGB compression. You will find these setting in "output" for FFDSHOW.

I wouldn't say that your system are outdated yet, but you could gain from an update on your video card before you would need to move on to a complete new system. This is something you will need to consider if you would like to upgrade later so that you will not get an AGP video card, but a PCI express based video card that you could use with your new system. There is also motherboard that could take both AGP and PCI express for video card, but that is outside this threads purpose and you will find other places to guide you on that issue. From what I have understood the newest generation of video card have hardware support that could off load your CPU in processing of H.264/AVC video. The best would be that you let us know exactly what hardware you have with some diagnostic tool.
Google for CPU-Z and let us know exactly what kind of motherboard you use and what kind of CPU you have as AMD 3000+ is not really much of help here.

Don't forget the basic update of all your drivers to the latest that is recommended for your hardware. Look around on other hardware sites for which version they recommend you to update your video cards drivers. Today's driver might not be the best (optimal) for older hardware, so you will need to figure out what is the best for your system of the newest driver version since you have made an update during your hardwares life time.

____
What is the size of your anime videos? Use MediaInfo to find out: http://mediainfo.sourceforge.net/no

[molitar]

Originally Posted by blizard

Do you use Nvidia PureVideo to help you to take advantage of your video cards hardware support in the rendering process? I use Haali Renderer as that renderer will make use of hardware features in processing video before it is sent to display.

As LotharZ have pointed out, you should be able to watch at least 720p with good result. You can also re-size output, to a smaller size for different type of post processing in FFDSHOW where the order from top to bottom will process your video before you send it to renderer. (Be aware that MPC as player can also auto re-size to fill your screen, make adjustment for aspect ration etc so there is many way to do this after you have processed video through FFDSHOW for decoding.) Here the order in the chain of FFDSHOW will matter as a large resolution will demand more CPU power to process video and not necessary give you any better result if it is to be displayed at 1024 x768 pixel. Resize & Aspect is a very good filte in FFSHOW to do this processing, but you will need to tweak it to learn what is the best in different situation. Different type of renderer can also make a difference in how well they wil support your need and hardware. LotharZ do also give you a suggestion to use make use of YUY2 or YV12 as colour compression output to video card as that will need less CPU power then to make use of any option in RGB compression. You will find these setting in "output" for FFDSHOW.

I wouldn't say that your system are outdated yet, but you could gain from an update on your video card before you would need to move on to a complete new system. This is something you will need to consider if you would like to upgrade later so that you will not get an AGP video card, but a PCI express based video card that you could use with your new system. There is also motherboard that could take both AGP and PCI express for video card, but that is outside this threads purpose and you will find other places to guide you on that issue. From what I have understood the newest generation of video card have hardware support that could off load your CPU in processing of H.264/AVC video. The best would be that you let us know exactly what hardware you have with some diagnostic tool.
Google for CPU-Z and let us know exactly what kind of motherboard you use and what kind of CPU you have as AMD 3000+ is not really much of help here.

Don't forget the basic update of all your drivers to the latest that is recommended for your hardware. Look around on other hardware sites for which version they recommend you to update your video cards drivers. Today's driver might not be the best (optimal) for older hardware, so you will need to figure out what is the best for your system of the newest driver version since you have made an update during your hardwares life time.

____
What is the size of your anime videos? Use MediaInfo to find out: http://mediainfo.sourceforge.net/no

No I have not used Purevideo.. but if it helps I'm willing to start using it. I tried Haali Renderer but I didn't get no video on my secondary output (my hdtv). My secondary display is set at 720x480 aka 480i SDTV which happens to the be only setting where I still could zoom out the video some to take care of the overscan issue so I have none at 480i SDTV.

Setup Primary Monitor: 1024x768 (19" Flat Screen CRT)
Secondary Display: 720x480 (1080i HDTV)
Mainboard: MSI K8N-Neo 2 Platinum (Nforce3 Ultra Chipset)
Processor: AMD Athlon 64 3000+
Operating System: Windows XP Pro Sp2 32bit

Video:
Codec : MPEG-4 AVC
Codec/Info : MPEG4 ISO advanced profile
PlayTime : 23mn 58s
Width : 1024 pixels
Height : 576 pixels
Display Aspect ratio : 16/9
Frame rate : 23.976 fps
Title : DCII Ep4 h264
Language : Japanese
AspectRatio : 1.778

Now on the yv12 I do have that checked but all the rest are checked also by default should I uncheck the rest to make sure it uses the YV12?

Yeah I been wanting to update to a new video card but that also requires a new mainboard, ram, processor, and video card so that is something I have to save up for.

[rickardk]

Originally Posted by haruhiko_yamagata

I found the flag. Haali's media splitter is working perfectly.
Damn, I did check the AM_SAMPLE2_PROPERTIES::dwTypeSpecificFlags of the first and second frames, but didn't check the third...
I've modified the code a bit to set time stamp properly. Now the sync is OK, but it still has two problems. One is interlacing related flags to send to downstream, one is jerkiness caused by unloading and loading dlls during playback. Second problem is caused because ffdshow receives samples with media type attached and then re-initialize codec drivers.
The fix will surely break the code too much that I can't do without adding bugs. I'm going to do it after beta4. Please wait.

That's great to hear!!
Thanks alot for looking at this problem. You are doing a fantastic job with ffdshow!

[Kado]

@blizard and @molitar

Purevideo is a hardware video decoder and post-processing technology for geforce series6 and 7, there's also purevideo hd for series8. to use it you need a compatible graphics card, a software decoder that supports it (like cyberlink's one) and a compatible renderer (like EVR or VMR depending on the OS). since Haali renderer does not support DXVA you cant use purevideo with it.
To use it in xp with a geforce 6600gt you'll need the following:
Cyberlink h264/avc decoder with hardware acceleration enabled in the options and vmr9 renderless, make sure you have the latest directx (nov2007 redist) and latest drivers.
to see if you have software or hardware decoding hold "CTRL" then open the decoder options, in profile if you have "SW" it has no hw accel.
That worked for me but since I changed for vista hw accel h264 for my geforce 6800gs does not work because directx can't do the dxva1 to dxva2 translation for h264, or something like that.

I use haali since I don't really need purevideo.

[haruhiko_yamagata]

@Fizick
Thank you for your fix, makeAVIS works much better.
I have added support for float PCM. Please test if all the formats work.

[haruhiko_yamagata]

@Kado
We are waiting for your translations .

[Kado]

Originally Posted by haruhiko_yamagata

@Kado
We are waiting for your translations .

Really? OK, I'm gonna try to make them this weekend then... I was doing it just for fun... Well, "Gambarimasu!"

[Fizick]

haruhiko_yamagata,

IMO it is not enough for correct WAVE_FORMAT_IEEE_FLOAT audio support.
Probably to add:
1. cbSize
2. Fact Chunk

And 32 bit integer PCM should be implemented as WAVE_FORMAT_EXTENSIBLE with a WAVE_FORMAT_PCM subcode

http://www-mmsp.ece.mcgill.ca/Docume...WAVE/WAVE.html

But I am not sure, and I do not know how it is related to Avisynth command "global OPT_UseWaveExtensible=True"

[Delerue]

I'm wondering, is it normal that kernel bob draw this vertical weird line at left of the video?

http://img.photobucket.com/albums/v2...1123032814.jpg

http://img.photobucket.com/albums/v2...1123032821.jpg

BTW, I tested the AC3/DTS SPDIF and it worked very well.

P.S.: attachments approval take too long...

[haruhiko_yamagata]

Originally Posted by Fizick

haruhiko_yamagata,

IMO it is not enough for correct WAVE_FORMAT_IEEE_FLOAT audio support.
Probably to add:
1. cbSize
2. Fact Chunk

And 32 bit integer PCM should be implemented as WAVE_FORMAT_EXTENSIBLE with a WAVE_FORMAT_PCM subcode

http://www-mmsp.ece.mcgill.ca/Docume...WAVE/WAVE.html

But I am not sure, and I do not know how it is related to Avisynth command "global OPT_UseWaveExtensible=True"

Thank you. I'll read the link later.

[haruhiko_yamagata]

Closed caption decoder needs fixing.
We have two choices, update current one or import MPC's CCDecoder. Either way, it requires Unicode. Supporting Windows 9x would be a burden.
Besides, recent updates of runtime libraries from Microsoft no longer support Windows 9x. We don't have time to test on Windows 9x, do we?

Do you agree it's time to drop support for Windows 9x?
Should we release beta4 before fixing CCDecoder and do it as beta4a?

[obieobieobie]

Win9x is dead.

[_xxl]

Do you agree it's time to drop support for Windows 9x?

Yes, please remove support for 9x.

[Leak]

Originally Posted by obieobieobie

Win9x is dead.

Agreed.

[SeeMoreDigital]

Originally Posted by haruhiko_yamagata

Do you agree it's time to drop support for Windows 9x?

Yep... without doubt

[Inventive Software]

Keep a 9x build for 3 months, and compare the number of downloads to the unicode one. If there's not sufficient interest, drop it.

[clsid]

Just let beta4 be the last version that works on Win9x.

[fastplayer]

I agree. After beta4 show no mercy

[haruhiko_yamagata]

OK, l'll drop support for Win9x after beta4.
Thank you for replies.

[molitar]

Originally Posted by Kado

@blizard and @molitar

Purevideo is a hardware video decoder and post-processing technology for geforce series6 and 7, there's also purevideo hd for series8. to use it you need a compatible graphics card, a software decoder that supports it (like cyberlink's one) and a compatible renderer (like EVR or VMR depending on the OS). since Haali renderer does not support DXVA you cant use purevideo with it.
To use it in xp with a geforce 6600gt you'll need the following:
Cyberlink h264/avc decoder with hardware acceleration enabled in the options and vmr9 renderless, make sure you have the latest directx (nov2007 redist) and latest drivers.
to see if you have software or hardware decoding hold "CTRL" then open the decoder options, in profile if you have "SW" it has no hw accel.
That worked for me but since I changed for vista hw accel h264 for my geforce 6800gs does not work because directx can't do the dxva1 to dxva2 translation for h264, or something like that.

I use haali since I don't really need purevideo.

Ok so I need to purchase the Nvidia Purevideo and the Cyberlink h264/avc decoder? Or will coreavc work in place of Cyberlink?

[Jay Bee]

Originally Posted by molitar

Ok so I need to purchase the Nvidia Purevideo and the Cyberlink h264/avc decoder? Or will coreavc work in place of Cyberlink?

No, Nvidia Purevideo decoder is only for MPEG-2. Cyberlink may work but I used to have the same card as you and had very bad experiences with Purevideo. And most x264 files don't work with hardware accel anyway.

[Kado]

What Jay Bee said is true, also your 6600GT only does decoding, there's no post processing like deblocking. So, in the end I would recommend ffdshow try-outs (currently single treaded for h264, multi-treaded version in the future) or if you want to invest a little then CoreAVC pro v1.6 (supports up to quad core cpus), those are both software decoders only. If you have a decent CPU ffdshow is more than enough for 720p streams.

[molitar]

Ok after testing.. I found cyberlink h264 really does suck. Coreavc is tons better. Cyberlink used over 130MB of memory and was still about 10% higher cpu resources even with hardware enabled. Coreavc used only around 60MB of memory for the same video I tested using Graphedt.

Currently working towards upgrading to a new AM2 setup and getting an ATI 2600 XT video card I think it's better for video than the Geforce 8600 GTS video card. Is that true from what I been reading ATI is better than the 8600 GTS? Especially since for gaming I can eventually get a second 2600 XT for a crossfire setup and still have good video and gaming capability.

[fastplayer]

Originally Posted by molitar

Is that true from what I been reading ATI is better than the 8600 GTS?

Make sure to get the GDDR4 version of the XT

[molitar]

Originally Posted by fastplayer

Make sure to get the GDDR4 version of the XT

Thanks for the tip

[Kado]

@molitar

If you're going for ATI I recommend an 3850 or 3870, they have the new decoding and post processing engine for vc-1/wmv,h264 and mpeg2 (directx 10.1 supported) and they are way faster than a 8600gts or a 2600xt, check the guru3d review.

[haruhiko_yamagata]

Originally Posted by Fizick

haruhiko_yamagata,

IMO it is not enough for correct WAVE_FORMAT_IEEE_FLOAT audio support.
Probably to add:
1. cbSize
2. Fact Chunk

And 32 bit integer PCM should be implemented as WAVE_FORMAT_EXTENSIBLE with a WAVE_FORMAT_PCM subcode

http://www-mmsp.ece.mcgill.ca/Docume...WAVE/WAVE.html

But I am not sure, and I do not know how it is related to Avisynth command "global OPT_UseWaveExtensible=True"

I'll add cbSize. Fact Chunk is implemented in the current version.
WAVE_FORMAT_EXTENSIBLE is annoying. AviSynth doesn't give channel mask information AFAIK. There is little merit using WAVE_FORMAT_EXTENSIBLE without proper channel mask information.
WAVSource cannot open files without error if they have WAVE_FORMAT_EXTENSIBLE. DirectShowSource works though. Current WMP can play 32bit integer PCM without WAVE_FORMAT_EXTENSIBLE.
Do you have any media player or DirectShow filter that doesn't play 32bit integer PCM with WAVE_FORMAT_EX?

[thuan]

Yes, the 3850 should be around two times faster than the 2600XT and cheaper than two 2600XT.

[molitar]

Originally Posted by Kado

@molitar

If you're going for ATI I recommend an 3850 or 3870, they have the new decoding and post processing engine for vc-1/wmv,h264 and mpeg2 (directx 10.1 supported) and they are way faster than a 8600gts or a 2600xt, check the guru3d review.

Thanks.. wow it's almost neck to neck with the 2900 XT. That is one thing I could never get use to with ATI the higher video cards aren't always the fastest. While with Nvidia most of the time the higher models are faster.

[foxyshadis]

I had no idea tryouts had ever worked on 9x. Learn something new every day, I guess. =p

[Delerue]

Anyone here?

BTW, this video has wrong AR when opened with FFDShow (v.1632). If you select CoreAVC or use VLC the AR is correct.

[cc979]

i've been having problems compiling svn 1633

any clues thanks ?

yasm.exe -f win32 -DWIN32 -DPREFIX -I../ffmpeg/libavcodec/i386/ -DWIN32 -DUSE32 -Dalloc -Dexec -I../ffmpeg/libavcodec/i386/ -o ../ffmpeg/libavcodec/i386/skl_dct_sse.o ../ffmpeg/libavcodec/i386/skl_dct_sse.asm
yasm.exe -f win32 -DWIN32 -DPREFIX -I../ffmpeg/libavcodec/i386/ -DWIN32 -DUSE32 -Dalloc -Dexec -I../ffmpeg/libavcodec/i386/ -o ../ffmpeg/libavcodec/i386/skl_fdct_mmx.o ../ffmpeg/libavcodec/i386/skl_fdct_mmx.asm
yasm.exe -f win32 -DWIN32 -DPREFIX -I../ffmpeg/libavcodec/i386/ -DWIN32 -DUSE32 -Dalloc -Dexec -I./ -o simple_idct_mmx.o simple_idct_mmx.asm
make[1]: Leaving directory `/home/User/svn/ffdshow-tryout/ffdshow-tryout/src/imgFilters'
make -C codecs
make[1]: Entering directory `/home/User/svn/ffdshow-tryout/ffdshow-tryout/src/codecs'
gcc -c -Itheora -mno-cygwin -mdll -fno-rtti -mthreads -pipe -D_WINGDI_ -DUCLIBCPP -D_GLIBCPP_HAVE_MBSTATE_T -D_WIN32_IE=0x0500 -DARCH_IS_IA32 -DARCH_IS_32BIT -DHAVE_MMX -mmmx -msse -mfpmath=sse -w -DNDEBUG -UDEBUG -DFFDEBUG=0 -I. -I.. -Iuclibc++ -Ibaseclasses -I../baseclasses -IimgFilters -I../imgFilters -Implayer -I../mplayer -Isettings -I../settings -Isettings/filters -I../settings/filters -Icodecs -I../codecs -Isubtitles -I../subtitles -Iconvert -I../convert -Idialog -I../dialog -IaudioFilters -I../audioFilters -Icygwin -I../cygwin -Iffmpeg -I../ffmpeg -Iacm -I../acm -Ixiph -I../xiph -Ifilters -I../filters -Imuxers -I../muxers -I/dx/Include -L/dx/MingLib -ldx9 -O2 -march=pentium-mmx -mtune=i686 -fomit-frame-pointer -finline-functions -finline -frename-registers -fweb -funit-at-a-time -MMD -o ffdshow_codecs_all.o ffdshow_codecs_all.cpp
TaudioCodecTremor.cpp: In member function 'bool TaudioCodecTremor::initVorbis()':
TaudioCodecTremor.cpp:61: error: 'ogg_reference' was not declared in this scope
TaudioCodecTremor.cpp:61: error: expected `;' before 'orId'
TaudioCodecTremor.cpp:62: error: 'ogg_buffer' was not declared in this scope
TaudioCodecTremor.cpp:62: error: expected `;' before 'obId'
TaudioCodecTremor.cpp:70: error: 'obId' was not declared in this scope
TaudioCodecTremor.cpp:74: error: 'orId' was not declared in this scope
TaudioCodecTremor.cpp:85: error: 'obCmt' was not declared in this scope
TaudioCodecTremor.cpp:89: error: 'orCmt' was not declared in this scope
TaudioCodecTremor.cpp:100: error: 'obSetup' was not declared in this scope
TaudioCodecTremor.cpp:104: error: 'orSetup' was not declared in this scope
TaudioCodecTremor.cpp: In member function 'virtual long int TaudioCodecTremor::decode(TbyteBuffer&)':
TaudioCodecTremor.cpp:266: error: 'ogg_reference' was not declared in this scope
TaudioCodecTremor.cpp:266: error: expected primary-expression before ';' token
TaudioCodecTremor.cpp:267: error: 'ogg_buffer' was not declared in this scope
TaudioCodecTremor.cpp:267: error: expected `;' before 'ob'
TaudioCodecTremor.cpp:270: error: 'ob' was not declared in this scope
TaudioCodecTremor.cpp:274: error: expected primary-expression before '||' token
TaudioCodecTremor.cpp:274: error: expected primary-expression before '.' token
TaudioCodecTremor.cpp:275: error: expected primary-expression before '||' token
TaudioCodecTremor.cpp:275: error: expected primary-expression before '.' token
TaudioCodecTremor.cpp:276: error: expected primary-expression before '||' token
TaudioCodecTremor.cpp:276: error: expected primary-expression before '.' token
TaudioCodecTremor.cpp:277: error: expected primary-expression before '||' token
TaudioCodecTremor.cpp:277: error: expected primary-expression before '.' token
TaudioCodecTremor.cpp:278: error: expected primary-expression before '||' token
TaudioCodecTremor.cpp:278: error: expected primary-expression before ';' token
make[1]: *** [ffdshow_codecs_all.o] Error 1
make[1]: Leaving directory `/home/User/svn/ffdshow-tryout/ffdshow-tryout/src/codecs'
make: *** [lib] Error 2

[clsid]

@cc979, it has to do with include directories. Try removing the xiph ones.

[Yong]

System spec:
vista ultimate 32bit
amd althon x2 6000+
ati 2600xt
2gb ram

player config:
mpc hc v1.0.11.0, VMR9 renderless, frame queue off,
ffdshow config, YV12 output colorspace, resize filter(default config), and only 3dnow SIMD extension is activated.

im using ffdshow_rev1632_20071123_clsid_sse_icl10.exe, resize a 640x480 mpeg1/h264 video, by using 1.5 multiplier:
fast/biliner default settings = video b0rked
fast/biliner luma + chroma sharpen 2.00 = normal
bicubic default = normal
bicubic luma/chroma sharpen = crash
point = b0rked or crash
area default = video b0rked
aera luma/chroma 2.00 = normal
gause luma/chormal 2.00 = crash
sinc = crash
lanczos default = b0rked
lanczos luma/chroma 2.00 = normal
spline default/ luma chroma 2.00 = b0rked
simple resize = normal
warped resize = normal
super2xSaI = normal
2xSaI = normal
hq2x = normal
crash means libmplayer.dll crash.

gaussian luma/chorma 2.0 with following resize method:
point = either video b0rked or crash
lanczos = b0rked or crash
gauss = b0rked or crash with certain luma/chroma value
area = b0rked or crash with certain luma/chroma value
fast/bilinear, bicubic = b0rked if luma gaussian > 0.66 or chroma gaussian > 0.67

swscaler gausian blur default setting = b0rked

im just messing around the SIMD setting and accidentally triggered these bugs lol

[haruhiko_yamagata]

Originally Posted by Delerue

Anyone here?

BTW, this video has wrong AR when opened with FFDShow (v.1632). If you select CoreAVC or use VLC the AR is correct.

I can't reproduce the Kernel bob issue.
As for the AR issue, the file plays OK for me. Please make sure 'Set pixel aspect ratio in output media type' is checked (or intermediate state).

[haruhiko_yamagata]

Originally Posted by Yong

System spec:
vista ultimate 32bit
amd althon x2 6000+
ati 2600xt
2gb ram

player config:
mpc hc v1.0.11.0, VMR9 renderless, frame queue off,
ffdshow config, YV12 output colorspace, resize filter(default config), and only 3dnow SIMD extension is activated.

I don't have 3DNow capable CPU.
Does it crash if both MMX and 3DNow! are checked?

[Yong]

Originally Posted by haruhiko_yamagata

I don't have 3DNow capable CPU.
Does it crash if both MMX and 3DNow! are checked?

nope, i think ffdshow will use mmx instead of 3dnow?
hmm looks like mmxext have same problem with 3dnow too..

[haruhiko_yamagata]

Originally Posted by Yong

nope, i think ffdshow will use mmx instead of 3dnow?

I guess 3DNow! code depends on MMX code.

[Yong]

Originally Posted by haruhiko_yamagata

I guess 3DNow! code depends on MMX code.

so turn off mmx will screw up ffdshow?

[clsid]

I am able to reproduce Yong's problem with my generic (non ICL) build. It also happens when resizing with factor 2.0, so target resolution is mod16.

I don't think these bugs will appear in normal use, since then MMX will likely be activated. Perhaps we should remove the 3dnow resizing code? It's useless if it doesn't work correctly. Removing it will reduce filesize.

Or perhaps the swscaler code of libmplayer just needs to be updated.

[haruhiko_yamagata]

Originally Posted by Yong

so turn off mmx will screw up ffdshow?

Yes, please do not turn off MMX. Those check boxes exist for debugging purpose. I don't have 3DNow! capable CPU, so I'm not motivated to fix it.

[haruhiko_yamagata]

Originally Posted by clsid

I am able to reproduce Yong's problem with my generic (non ICL) build. It also happens when resizing with factor 2.0, so target resolution is mod16.

I don't think these bugs will appear in normal use, since then MMX will likely be activated. Perhaps we should remove the 3dnow resizing code? It's useless if it doesn't work correctly. Removing it will reduce filesize.

Or perhaps the swscaler code of libmplayer just needs to be updated.

Nope. 3DNow code works as long as MMX is enabled, doesn't it?

[Inventive Software]

Pop-up box to warn that disabling MMX will enable 3DNow?

[clsid]

If 3dnow does depend on MMX, then the best 'solution' would be to automatically uncheck 3dnow when a user unchecks MMX. But since these settings are for debugging purposes there is not a real need to fix this.

[Yong]

im now compiling mplayer with only 3dnow extension is enabled, see if the libswscale does crash with the same setting

[Yong]

noob attemp failed
disable mmx in ffmpeg wiill cause ffmpeg use x86 asm and c code to do resizing instead of 3dnow.
disable mmx in mplayer will cause compiling error lol

btw u guys should try to update the libmplayer code and see if that will fix the 3dnow problem instead of remove it

Does ffdshow have its own irc channel?

[cc979]

Originally Posted by clsid

@cc979, it has to do with include directories. Try removing the xiph ones.

i dont have any xiph include directories only the mingw-dx one

[clsid]

makefile.inc contains "-Ixiph -I../xiph"
removing that should probably do the trick

let me know if it works, then I can commit the change to SVN

[Delerue]

Originally Posted by haruhiko_yamagata

I can't reproduce the Kernel bob issue.

Hmmm... I think that this is a problem with this video, specially.

Originally Posted by haruhiko_yamagata

As for the AR issue, the file plays OK for me. Please make sure 'Set pixel aspect ratio in output media type' is checked (or intermediate state).

Ah, great. I didn't know that. Thanks and sorry for the wrong bug report.

[haruhiko_yamagata]

Originally Posted by Delerue

Hmmm... I think that this is a problem with this video, specially.

Please send us the sample, if it is possible.

[multiblitz]

Originally Posted by haruhiko_yamagata

I found the flag. Haali's media splitter is working perfectly.
Damn, I did check the AM_SAMPLE2_PROPERTIES::dwTypeSpecificFlags of the first and second frames, but didn't check the third...
I've modified the code a bit to set time stamp properly. Now the sync is OK, but it still has two problems. One is interlacing related flags to send to downstream, one is jerkiness caused by unloading and loading dlls during playback. Second problem is caused because ffdshow receives samples with media type attached and then re-initialize codec drivers.
The fix will surely break the code too much that I can't do without adding bugs. I'm going to do it after beta4. Please wait.

THAT IS THE BEST NEWS OF THE YEAR !!! WOW !

ffdshow-performance on HD-stuff...wonderful !

[Yong]

here is how to make ffvdub crash:
1. open virtualdub and load a video
2. add ffvdub to filter list
3. tick the avisynth filter on ffvdub
4. load any avisynth fucntions or plugins, eg invert()
5. press preview/ok > crash.

heres the details:
An out-of-bounds memory access (access violation) occurred in module 'ffdshow'...
...writing address 00000000...
...while running filter "ffvdub" (FilterSystem.cpp:594).

[haruhiko_yamagata]

Originally Posted by Yong

here is how to make ffvdub crash:
1. open virtualdub and load a video
2. add ffvdub to filter list
3. tick the avisynth filter on ffvdub
4. load any avisynth fucntions or plugins, eg invert()
5. press preview/ok > crash.

heres the details:
An out-of-bounds memory access (access violation) occurred in module 'ffdshow'...
...writing address 00000000...
...while running filter "ffvdub" (FilterSystem.cpp:594).

@Leak
Could you take a look at this?

[Yong]

Sorry, lacks some details, Default setting does not crash when click preview/ok, but if use "buffer back/ahead" with any value would cause virtualdub crash, even without using any avisynth plugins/functions.
im still using clsid icl10 build.

[Mysteron]

This is my first post. I did a search on this forum, but did not find any messages discussing my problem.

I have been using FFDSHOW since 2004, and I enjoy using the program.

However, I seem to have discovered a problem in all FFDSHOW versions dated after October 5. When using the "Automatic" option in the "Levels" filter, the levels in the video are distorted badly when manually seeking forward or backwards while the video file is playing. This problem also occurs when the video has reached the end and loops back to the beginning (in repeat play mode). The current version of FFDSHOW (1642) still has the same problem.

I have had this problem using Media Player Classic 6.4.9.1 and Windows Media Players 10 and 11. This has happened in Vista 32 bit using an Nvidia 8600GT video card and on Windows XP SP2 using an Intel 910GL video chip. Both PCs are Pentium 4 HTs (one is a Sony, the other is a Dell). The probelm occurs regardless of the type of video file that is being played. I have updated all drivers on both PCs, but the problem remains.

I never had this problem with earlier FFDSHOW versions.

Your help is appreciated. Thank you.

[Leak]

Originally Posted by haruhiko_yamagata

@Leak
Could you take a look at this?

That's what I had in mind, yes...

Originally Posted by Yong

Sorry, lacks some details, Default setting does not crash when click preview/ok, but if use "buffer back/ahead" with any value would cause virtualdub crash, even without using any avisynth plugins/functions.
im still using clsid icl10 build.

Oh, wait...

I'm not very surprised - buffering simply can't work when used via VfW/VDub, as those are missing the one thing from DirectShow that buffering makes heavy use of: the possibility of getting passed a frame and *not* returning anything (yet).

It shouldn't crash, though. But I won't guarantee any useful results when I fix the crash...

[pmarco]

Hi, I'm from Italy. In some dvd movies I can read italian subtitles under the signs and writings appearing in a foreign language (only under the signs, not in the dialogues). This works in programs like Nero Showtime.
Anyway, in vlc or wmp+ffdshow, if I select the italian subtitle track, everything is subtitled (signs and dialogues too). It seems I cannot choose to subtitle only the signs. How can I do it, please ? (using wmp+ffdshow)

In some dvds of mine, aspect ratio changes by mistake, only in some chapters of the movie. It goes from 16/9 to 64/27 (and SAR goes from 64/45 to 256/135). Then it comes back to normal. Is it a bug of ffdshow? How can I correct that?

[St Devious]

y cant ffdshow do 2 pass encoding for H.264 ?

[haruhiko_yamagata]

Originally Posted by pmarco

Hi, I'm from Italy. In some dvd movies I can read italian subtitles under the signs and writings appearing in a foreign language (only under the signs, not in the dialogues). This works in programs like Nero Showtime.
Anyway, in vlc or wmp+ffdshow, if I select the italian subtitle track, everything is subtitled (signs and dialogues too). It seems I cannot choose to subtitle only the signs. How can I do it, please ? (using wmp+ffdshow)

What do you mean by 'signs'? Do you mean closed caption?

In some dvds of mine, aspect ratio changes by mistake, only in some chapters of the movie. It goes from 16/9 to 64/27 (and SAR goes from 64/45 to 256/135). Then it comes back to normal. Is it a bug of ffdshow? How can I correct that?

Are you sure that the DVD doesn't have the wrong aspect ratio?

[tjmitchem]

Originally Posted by wolf2009

y cant ffdshow do 2 pass encoding for H.264 ?

It can. It basically consists of doing two 1-pass encodes while making use of the libavcodec/x264 stats functionality.

Select x264 as your ffdshow encoder, then take a look at the "output" section of the dialog. On the first pass, write the stats. On the 2nd, "use" the stats.

Terry

[cc979]

Originally Posted by clsid

makefile.inc contains "-Ixiph -I../xiph"
removing that should probably do the trick

let me know if it works, then I can commit the change to SVN

make[1]: Leaving directory `/home/User/svn/ffdshow-tryout/ffdshow-tryout/src/acm'
make -C audioFilters
make[1]: Entering directory `/home/User/svn/ffdshow-tryout/ffdshow-tryout/src/audioFilters'
gcc -c -mno-cygwin -mdll -fno-rtti -mthreads -pipe -D_WINGDI_ -DUCLIBCPP -D_GLIBCPP_HAVE_MBSTATE_T -D_WIN32_IE=0x0500 -DARCH_IS_IA32 -DARCH_IS_32BIT -DHAVE_MMX -mmmx -msse -mfpmath=sse -w -DNDEBUG -UDEBUG -DFFDEBUG=0 -I. -I.. -Iuclibc++ -Ibaseclasses -I../baseclasses -IimgFilters -I../imgFilters -Implayer -I../mplayer -Isettings -I../settings -Isettings/filters -I../settings/filters -Icodecs -I../codecs -Isubtitles -I../subtitles -Iconvert -I../convert -Idialog -I../dialog -IaudioFilters -I../audioFilters -Icygwin -I../cygwin -Iffmpeg -I../ffmpeg -Iacm -I../acm -Ifilters -I../filters -Imuxers -I../muxers -I/dx/Include -L/dx/MingLib -ldx9 -O2 -march=pentium-mmx -mtune=i686 -fomit-frame-pointer -finline-functions -finline -frename-registers -fweb -funit-at-a-time -MMD -o ffdshow_audiofilters_all.o ffdshow_audiofilters_all.cpp
In file included from ffdshow_audiofilters_all.cpp:13:
TsampleFormat.cpp:22:33: error: vorbis/vorbisformat.h: No such file or directory
TsampleFormat.cpp: In member function 'void TsampleFormat::init(const VORBISFORMAT&)':
TsampleFormat.cpp:110: error: invalid use of undefined type 'const struct VORBISFORMAT'
TsampleFormat.h:4: error: forward declaration of 'const struct VORBISFORMAT'
TsampleFormat.cpp:112: error: invalid use of undefined type 'const struct VORBISFORMAT'
TsampleFormat.h:4: error: forward declaration of 'const struct VORBISFORMAT'
TsampleFormat.cpp: In member function 'void TsampleFormat::init(const VORBISFORMAT2&)':
TsampleFormat.cpp:121: error: invalid use of undefined type 'const struct VORBISFORMAT2'
TsampleFormat.h:5: error: forward declaration of 'const struct VORBISFORMAT2'
TsampleFormat.cpp:122: error: invalid use of undefined type 'const struct VORBISFORMAT2'
TsampleFormat.h:5: error: forward declaration of 'const struct VORBISFORMAT2'
TsampleFormat.cpp:130: error: invalid use of undefined type 'const struct VORBISFORMAT2'
TsampleFormat.h:5: error: forward declaration of 'const struct VORBISFORMAT2'
TsampleFormat.cpp: In member function 'void TsampleFormat::init(const VORBISFORMATILL&)':
TsampleFormat.cpp:139: error: invalid use of undefined type 'const struct VORBISFORMATILL'
TsampleFormat.h:6: error: forward declaration of 'const struct VORBISFORMATILL'
TsampleFormat.cpp:141: error: invalid use of undefined type 'const struct VORBISFORMATILL'
TsampleFormat.h:6: error: forward declaration of 'const struct VORBISFORMATILL'
make[1]: *** [ffdshow_audiofilters_all.o] Error 1
make[1]: Leaving directory `/home/User/svn/ffdshow-tryout/ffdshow-tryout/src/audioFilters'
make: *** [lib] Error 2

i just tried the edit, but it has new error

[St Devious]

Originally Posted by tjmitchem

It can. It basically consists of doing two 1-pass encodes while making use of the libavcodec/x264 stats functionality.

Select x264 as your ffdshow encoder, then take a look at the "output" section of the dialog. On the first pass, write the stats. On the 2nd, "use" the stats.

Terry

Thx , one more q , can it do VBR or only CBR ?

[foxyshadis]

Eh? x264 can't do CBR, not without a lot of coaxing. It only does different VBR & ABR modes.

[haruhiko_yamagata]

Originally Posted by Mysteron

This is my first post. I did a search on this forum, but did not find any messages discussing my problem.

I have been using FFDSHOW since 2004, and I enjoy using the program.

However, I seem to have discovered a problem in all FFDSHOW versions dated after October 5. When using the "Automatic" option in the "Levels" filter, the levels in the video are distorted badly when manually seeking forward or backwards while the video file is playing. This problem also occurs when the video has reached the end and loops back to the beginning (in repeat play mode). The current version of FFDSHOW (1642) still has the same problem.

I have had this problem using Media Player Classic 6.4.9.1 and Windows Media Players 10 and 11. This has happened in Vista 32 bit using an Nvidia 8600GT video card and on Windows XP SP2 using an Intel 910GL video chip. Both PCs are Pentium 4 HTs (one is a Sony, the other is a Dell). The probelm occurs regardless of the type of video file that is being played. I have updated all drivers on both PCs, but the problem remains.

I never had this problem with earlier FFDSHOW versions.

Your help is appreciated. Thank you.

Thank you for report. I have fixed at rev 1648.

[iron2000]

I would like to know does ffdshow use the graphics card?
Or is it more on the cpu and not on having a good graphics card?

[pmarco]

Originally Posted by haruhiko_yamagata

What do you mean by 'signs'? Do you mean closed caption?
Are you sure that the DVD doesn't have the wrong aspect ratio?

Yes, I mean closed captions.

No, the Dvd doesn't have the wrong aspect ratio. It's displayed well in vlc and nero showtime.

Please, don't abandon me.

[haruhiko_yamagata]

Originally Posted by pmarco

Yes, I mean closed captions.

No, the Dvd doesn't have the wrong aspect ratio. It's displayed well in vlc and nero showtime.

Please, don't abandon me.

As for subtitles, you can't disable bitmap subtitle from ffdshow's dialog. All DVD (AFAIK) have menu, you can disable bitmap subtitle through DVD menu. Closed caption still displays and it can be disabled from ffdshow's dialog.

As for the aspect ratio, please send us the sample. Otherwise, the fix is hopeless.
Please try not to violate the copyright.

[pmarco]

Originally Posted by haruhiko_yamagata

Closed caption still displays and it can be disabled from ffdshow's dialog. Are you sure that the DVD doesn't have the wrong aspect ratio?

I don't know, closed caption is not displayed in two dvds of mine. And it's not disabled, of course.

Concerning aspect ratio, I have no idea how to make samples.

I can use other programs to watch dvds, but it's nice to find out.

Thanks

[St Devious]

Originally Posted by foxyshadis

Eh? x264 can't do CBR, not without a lot of coaxing. It only does different VBR & ABR modes.

in ffdshow vfw when u select H.264 , it says mode "One pass - constant bitrate " , is it really CBR or VBR ?

[Leak]

Originally Posted by Yong

Sorry, lacks some details, Default setting does not crash when click preview/ok, but if use "buffer back/ahead" with any value would cause virtualdub crash, even without using any avisynth plugins/functions.
im still using clsid icl10 build.

Fixed in rev. 1650.

VirtualDub was passing me an RGB image with a negative pitch (i.e. the image was stored vertically flipped in memory) and my code then happily allocated a negative amount of memory for the buffers - guess I never tested that case...

Anyway, it'll work now, but buffering ahead in VDub will introduce a delay in the video since it'll return an empty frame until all buffers are filled, and of course the same amount of frames will be missing at the end. But, once again, there's nothing I can do about that...

np: Bola - W.i.K (Soup)

[haruhiko_yamagata]

Originally Posted by pmarco

I don't know, closed caption is not displayed in two dvds of mine. And it's not disabled, of course.

Not all DVDs have closed captions.

[foxyshadis]

Originally Posted by wolf2009

in ffdshow vfw when u select H.264 , it says mode "One pass - constant bitrate " , is it really CBR or VBR ?

Most commonly known as ABR, but it's a form of VBR, yes. Renamed to average bitrate in svn.

[chros]

@haruhiko_yamagata: RealAAC decoder and the LFE Crossover filter is not working together in r1625 , libfaad2 is OK. (Older version is OK, eg: r1523 .)

Tried with Nero AAC LC 5.1 in an mkv container.

Btw: what is the difference between the 2 (libfaad2 and reelaac) ?
The latter sounds better to me ... Am I right ?

Thanks

[clsid]

They are two different decoding libraries. Afaik, libfaad2 is the best one of the two.

[haruhiko_yamagata]

Originally Posted by chros

@haruhiko_yamagata: RealAAC decoder and the LFE Crossover filter is not working together in r1625 , libfaad2 is OK. (Older version is OK, eg: r1523 .)

Tried with Nero AAC LC 5.1 in an mkv container.

Btw: what is the difference between the 2 (libfaad2 and reelaac) ?
The latter sounds better to me ... Am I right ?

Thanks

I can't reproduce. What does "is not working" mean? No sound at all or no sound from sub woofer or ...

[haruhiko_yamagata]

I've just tried to test ffdshow_prebeta4_rev1656_20071029_clsid.exe on Windows 98, but unfortunately my old PC seems to be broken.
Any volunteers, would you test it on Win9x?

[pmarco]

Originally Posted by haruhiko_yamagata

Not all DVDs have closed captions.

I told you: i have two dvds with closed captions (I know for sure) but they're not displayed by ffdshow.

[St Devious]

is it possible to automate 2-pass encoding in ffdshow vfw with xvid and H.264 codecs , or does it have to be done manually ( first running the first pass , then the second ) ?

[Inventive Software]

Originally Posted by haruhiko_yamagata

I've just tried to test ffdshow_prebeta4_rev1656_20071029_clsid.exe on Windows 98, but unfortunately my old PC seems to be broken.
Any volunteers, would you test it on Win9x?

I'm in the process of installing 98SE on my desktop computer. I should have a response by the end of the weekend, if not, next weekend.

[chros]

Originally Posted by haruhiko_yamagata

I can't reproduce. What does "is not working" mean? No sound at all or no sound from sub woofer or ...

Thanks for the quick feedback: so I have found the bug.

It's only present with 5.1 ch AAC source:
- when you use realaac the output channel config is 7 ch instead of 6 !!!!!! (see at the Info & CPU section)
- I'm getting destortion sound because AC3 filter (v1.46) is connetced at the end of the chain to ffdshow, and -as it seems- it doesn't like "7 ch - LFE crossovered" input !

And -as always- thanks again for investigating/fixing the problem ...

Originally Posted by clsid

Afaik, libfaad2 is the best one of the two.

Thanks. Strange, for me realaac is sounding better ...

[Taurus]

Originally Posted by haruhiko_yamagata

I've just tried to test ffdshow_prebeta4_rev1656_20071029_clsid.exe on Windows 98, but unfortunately my old PC seems to be broken.
Any volunteers, would you test it on Win9x?

You can count me in.
I have one machine running win98.
Just PM me or do whatever you like...

[haruhiko_yamagata]

Originally Posted by Taurus

You can count me in.
I have one machine running win98.
Just PM me or do whatever you like...

Thank you very much!
Just opening one file using ffdshow would help much.
Further important testing may be ...

• subtitles (ANSI and Unicode input)
• Launching VFW configuration dialog
• Launching makeAVIS.exe
• preset loading

Of course, if you are interested in these features.

[haruhiko_yamagata]

Originally Posted by chros

Thanks for the quick feedback: so I have found the bug.

It's only present with 5.1 ch AAC source:
- when you use realaac the output channel config is 7 ch instead of 6 !!!!!! (see at the Info & CPU section)
- I'm getting destortion sound because AC3 filter (v1.46) is connetced at the end of the chain to ffdshow, and -as it seems- it doesn't like "7 ch - LFE crossovered" input !

And -as always- thanks again for investigating/fixing the problem ...

Oh, yes. I saw the letters "7ch". I should have found it. Thank you.

[Taurus]

Originally Posted by haruhiko_yamagata

Thank you very much!
Just opening one file using ffdshow would help much.
Further important testing may be ...

[*]subtitles (ANSI and Unicode input)[*]Launching VFW configuration dialog[*]Launching makeAVIS.exe[*]preset loading
Of course, if you are interested in these features.

[*]subtitles (ANSI and Unicode input) not tested, no file available at this time.
[*]Launching VFW configuration dialog | ok
[*]Launching makeAVIS.exe | ok
[*]preset loading | ok

Pretty much I tested on the fly seems stable.
Subtitles in ANSI works on former releases.

[chros]

Originally Posted by haruhiko_yamagata

Oh, yes. I saw the letters "7ch". I should have found it. Thank you.

No !!! Thank You !!!

[Amdh]

I'm New In Here .. May I Ask A Question : Is The Ffdshow H264 Encoder Updated To The Latest x264 Edition, I Mean, On Which x264 Library Build Is It Based ?

I Have Noticed Also That When Decoding With Ffdshow, The Decoder Usualy Fails In Autodetecting The ColorSpace I'm Using .. This Occurs Frequently With RGB32 Encoded Files, He's Decoding Them As If They Were In YV12 Color Space, Is It A Bug Or Is This Normal ?

By The Way, In Case The Above Question Has Been Asked Before Please Forgive Me, I'm Just A Newbie.

[haruhiko_yamagata]

Originally Posted by Taurus

[*]subtitles (ANSI and Unicode input) not tested, no file available at this time.
[*]Launching VFW configuration dialog | ok
[*]Launching makeAVIS.exe | ok
[*]preset loading | ok

Pretty much I tested on the fly seems stable.
Subtitles in ANSI works on former releases.

Thank you very much.
This sample would help testing subtitles. It has UTF-8 SSA subtitles, which may be a bit difficult task for Win98.

[haruhiko_yamagata]

Originally Posted by Amdh

I Have Noticed Also That When Decoding With Ffdshow, The Decoder Usualy Fails In Autodetecting The ColorSpace I'm Using .. This Occurs Frequently With RGB32 Encoded Files, He's Decoding Them As If They Were In YV12 Color Space, Is It A Bug Or Is This Normal ?

What kind of file is the RGB32 encoded one? Is it H.264?

[Amdh]

Yei Friend It's H.264 In An AVI Container, I Usualy Use CoreAVC To Decode It And It's Working Well Since Ever, However I Just Tried To Use Ffdshow Decoder And I Discovered That He's Decoding It As YV12 ..

[clsid]

You can configure the output colorspace in ffdshow options.

[haruhiko_yamagata]

Originally Posted by Amdh

Yei Friend It's H.264 In An AVI Container, I Usualy Use CoreAVC To Decode It And It's Working Well Since Ever, However I Just Tried To Use Ffdshow Decoder And I Discovered That He's Decoding It As YV12 ..

H.264 encode RGB32? It's impossible.

[Amdh]

Impossible .. Why Friend ? I'm Doing That Since 2 Years, It's Working Well As For The Encoding And Decoding, I'm Even Encoding In RGB32 With Ffdshow H264 Encoder .. When Decoding With CoreAVC Pro The Decoder Shows That He's Decoding In RGB32. So Am I Doing Something Wrong .. ? If Yes Please Tell Me How To Correct That .. ?

[max-pain]

Any chance for a newer 64 bit build?

[clsid]

Originally Posted by Amdh

Impossible .. Why Friend ? I'm Doing That Since 2 Years, It's Working Well As For The Encoding And Decoding, I'm Even Encoding In RGB32 With Ffdshow H264 Encoder .. When Decoding With CoreAVC Pro The Decoder Shows That He's Decoding In RGB32. So Am I Doing Something Wrong .. ? If Yes Please Tell Me How To Correct That .. ?

Encoding into RGB32 is impossible for H.264. But I think what you mean is encoding RGB32 into H.264 format and decoding H.264 to RGB32.

Like I said before, go to ffdshow video decoder configuration, on the output page uncheck all colorspaces except for RGB32. Then ffdshow will output RGB32.

[Leak]

Originally Posted by Amdh

So Am I Doing Something Wrong .. ?

Well, for one screenshots of the right and wrong decoding results would be nice, as well as a small sample...

[Amdh]

Thanks For The Info Friends .. I've Tried That Decoding Tip And It Worked Well .. Ffdshow Was Able To Decode RGB32 Corectly, Anyway, If It's Impossible To Encode RGB32 In H264 Then I Was Surely Doing Something Wrong, Gonna Try Some Other Color Spaces, Any Suggestions About That ?

[St Devious]

Originally Posted by wolf2009

is it possible to automate 2-pass encoding in ffdshow vfw with xvid and H.264 codecs , or does it have to be done manually ( first running the first pass , then the second ) ?

Help !!

[clsid]

Originally Posted by Amdh

Thanks For The Info Friends .. I've Tried That Decoding Tip And It Worked Well .. Ffdshow Was Able To Decode RGB32 Corectly, Anyway, If It's Impossible To Encode RGB32 In H264 Then I Was Surely Doing Something Wrong, Gonna Try Some Other Color Spaces, Any Suggestions About That ?

You are understanding it wrong. The H.264 format has NOTHING to do with colorspaces.

And please stop using annoying fonts.

[LoRd_MuldeR]

Originally Posted by wolf2009

is it possible to automate 2-pass encoding in ffdshow vfw with xvid and H.264 codecs , or does it have to be done manually ( first running the first pass , then the second ) ?

I think this can not be automated, due to limitations in VFW itself. As far as I know, the VFW interface does not support 2-Pass encoding at all! In fact the 2-Pass encoding with all VFW Codecs I'm aware of is simply "emulated" with 2 separate encodes of the same video. Of course the VFW Codec has to be reconfigured between the two encodes, because it needs to know which run is the first and which run is the second pass. But from the point of view of the VFW interface (or the host application) those are still two independent encodes. You might want to give Avidemux a try, which does not use VFW Codecs and is able to automate 2-Pass encoding with both, x264 and Xvid:

http://mulder.dummwiedeutsch.de/home...jects#avidemux

[St Devious]

Originally Posted by LoRd_MuldeR

You might want to give Avidemux a try, which does not use VFW Codecs and is able to automate 2-Pass encoding with both, x264 and Xvid:

http://mulder.dummwiedeutsch.de/home...jects#avidemux

i already use avidemux , i wanted to render files in vegas so i asked the q .

[Taurus]

Originally Posted by haruhiko_yamagata

Thank you very much.
This sample would help testing subtitles. It has UTF-8 SSA subtitles, which may be a bit difficult task for Win98.

No chance on win98
The letters are burked and language subs are missing

[clsid]

That is to be expected since Win9x does not support unicode. Most important thing is that it doesn't crash.

[Amdh]

OK .. I've understood the thing .. thanks !

[Taurus]

Originally Posted by clsid

That is to be expected since Win9x does not support unicode. Most important thing is that it doesn't crash.

Yeah, real stable OS
Running in the same Rig since 1998.

[haruhiko_yamagata]

Originally Posted by Taurus

No chance on win98
The letters are burked and language subs are missing

Thank you. Then we can anounce that beta4 does not support subtitles in Win9x.
I may try to fix the problem for beta4a or later if I have time.

[Inventive Software]

Hold it right there... Win98 has a unicode DLL that can be downloaded IIRC.... is there still a chance that it can still be supported?

[haruhiko_yamagata]

Originally Posted by Inventive Software

Hold it right there... Win98 has a unicode DLL that can be downloaded IIRC.... is there still a chance that it can still be supported?

hmm... It may work. ffdshow may have to detect the dll and treat Win98 as Unicode OS.

[Liisachan]

You guys might want to use another clip to test Unicode SSA subs on Win98 if you test it at all. The reason is, mewmew vorbis has the Encoding parameter 0 for all subs. For instance, the Russian has:

Style: Ichigo,Verdana,32,...30,30,15,0,0

The last '0' is supposed to be lfCharSet, and if it is 204 (RUSSIAN_CHARSET), Win98 may like it more. That's how it was when I tested it long ago (in 2003), on Win98 with vsfilter. I couldn't re-test it to confirm it, as I don't have win98 anymore (or do I... somewhere...) Back then, I thought that last 0 (or \fe) was just a place holder for a Unicode SSA, since the file is in Unicode and CharSet is irrelevant (Haali said the same thing too, Mosu a Linux uers even said he didn't understand the meaning of \fe at all and he's right, charset is Window specific), but apparently it may be useful for VSFilter. (Gabest's ass-specs.doc has it about that param "When the file is Unicode, this field is useful during file format conversions." btw). ffdshow may be totally different than vsfilter so this might not be the problem, but when I tested it, mewmew vorbis + win98 + vsfilter (or mpc) worked more or less iff I changed that CharSet param 0 to an appropriate value.

SSA should have 204 as that parameter if it is in Russian, or 134 if it's in simplified chinese, etc etc. That is basically so even now that SSA/ASS is in Unicode. For instance, even in Unicode, even on Windows XP, Simp Chinese SSA doesn't work correctly without that magic number 134 (or actually 1, DEFAULT_CHARSET will do too) w/ VSFilter/MPC. (Subs may be readable but CreateFont fails and so not styled) Yet, mewmew vorsbis has every Encoding parameter 0.
So, altho ffdshow's algo might be totally different, I'd say mewmew vorbis may not be a good clip for testing in this case.

Also note that even on Windows XP, some languages are not displayed correctly unless you have East Asian Language support installed, or complicated text support like RtoL.

Here's another test clip that has proper Encoding parameters in ASS. Win98 might like it better (tho I'm quite skeptical anyway~)

[haruhiko_yamagata]

Liisachan, thank you for the info.
I've installed Virtual PC 2007, Windos 98 work on it .
Thank you, Taurus, I can reproduce the problem.

I've changed the source code of ffdshow to use Microsoft Layer for Unicode on Windows 98. It's better now, still crashes though.
I'm learning how to use remote debugger. It's quite .

[Taurus]

Originally Posted by Inventive Software

Hold it right there... Win98 has a unicode DLL that can be downloaded IIRC.... is there still a chance that it can still be supported?

Originally Posted by haruhiko_yamagata

hmm... It may work. ffdshow may have to detect the dll and treat Win98 as Unicode OS.

Thats what I have tried at first attempt.
Unicode assistance for Win98 is installed on this machine for a long time.
http://www.microsoft.com/downloads/d...displaylang=en

@haruhiko_yamagata
You're faster...
If you want me to test another build...
Count on me

[haruhiko_yamagata]

Originally Posted by Taurus

Thats what I have tried at first attempt.
Unicode assistance for Win98 is installed on this machine for a long time.
http://www.microsoft.com/downloads/d...displaylang=en

Layer for Unicode is used only if calling program (ffdshow) is build to use it.

[Taurus]

Originally Posted by haruhiko_yamagata

Layer for Unicode is used only if calling program (ffdshow) is build to use it.

I know
Here is another pic from the clip Liisachan provided.

[sillKotscha]

just a question about "good old" mov format again

I have found some mov videos - they can be found/ downloaded here - which ffdshow fails to play... sound and video

is it a ffdshow issue or a splitter issue??

Quicktime Player plays them just fine...

EDIT: sorry, found the problem... it is the internal mp4/mov source filter of MPC - disable it and the files play just fine... I can see that the quicktime movie parser is used instead...

EDIT II: who's working/ developing the mp4/mov source filter... I'd like to ask for a possible fix - as those files a the only files I have not working with this splitter...

[clsid]

The filter is open source. But there currently is nobody who is working on it. You could ask Casimir666 if he can take a look at your samples.

[sillKotscha]

thx for your answer

[Inventive Software]

Can somebody confirm that interlaced H.264 decoding is broken? I've uploaded a sample here: http://rapidshare.com/files/73670684...laced_test.mkv. It's an x264 encode with --interlaced and --crf 20. That's it. Basically, the decoder gives a grey garbled image. I've made it mod-32 in height and width, but that hasn't done anything. I don't have another H.264 decoder to test it with so I can't determine if it's the encoder or decoder that's the problem.

EDIT: I've tested with ffplay on Windows, and it's ffdshow that's not decoding correctly. ffplay has no trouble, so something's missing in the libavcodec H.264 decoder you guys have.

EDIT 2: My bad. Was using rev 1620, updated to rev 1656, and that seems to have fixed it. I gotta concentrate more!

[Harukalover]

Some bugs I've encountered in latest build of ffdshow tryouts. (ffdshow_prebeta4_rev1656_20071029_clsid.exe was the one tested) All of them dealing with lossless audio.

First is a flac issue. In the particular sample it skips all the way to the end of the file as soon as it hits around 4 seconds in. Here it is for testing:
http://dnous.nekonation.org/Harukalover/flac_skip.mkv

This doesn't happen if I use CoreFLAC to decode instead of libavcodec in ffdshow.

Second issue involves tta audio.

http://www.cccp-project.net/beta/tes...Audio_Test.mkv

Download that and switch to the 2ch tta track. You'll notice if using libavcodec to decode it. That there is a large desync of audio. You can definitely notice this at 00:28 in where the crowd comes in really late.

After that switch to the 6ch tta track. You should notice a mix of speedup/slowdown of the video. Especially at 00:12 in which it will just stall the video for a bit.

Both of those tta issues are gone when using the official tta decoder from True Audio's website instead of libavcodec.

[haruhiko_yamagata]

@Harukalover
Thank you for report. I can reproduce both of the problems.

[haruhiko_yamagata]

Originally Posted by Taurus

I know
Here is another pic from the clip Liisachan provided.

OK, now I've fixed the word wrapping issue and crash issue. I've updated the installer to install unicows.dll on Windows 9x. I'll commit when ready.

getTextExtentExPointW seems to be buggy on Layer for Unicode. It requires 2 * strlen instead of strlen for alpDx array size.
Nevertheless, Layer for Unicode is useful. I think we don't have to drop Windows 9x support for a while.

[Taurus]

Originally Posted by haruhiko_yamagata

I think we don't have to drop Windows 9x support for a while.

Thanks!

[Thunderbolt8]

Originally Posted by Inventive Software

Can somebody confirm that interlaced H.264 decoding is broken? I've uploaded a sample here: http://rapidshare.com/files/73670684...laced_test.mkv. It's an x264 encode with --interlaced and --crf 20. That's it. Basically, the decoder gives a grey garbled image. I've made it mod-32 in height and width, but that hasn't done anything. I don't have another H.264 decoder to test it with so I can't determine if it's the encoder or decoder that's the problem.

EDIT: I've tested with ffplay on Windows, and it's ffdshow that's not decoding correctly. ffplay has no trouble, so something's missing in the libavcodec H.264 decoder you guys have.

EDIT 2: My bad. Was using rev 1620, updated to rev 1656, and that seems to have fixed it. I gotta concentrate more!

afaik galapagos HD DVD VC-1 interlaced at 29.97 is still not playable with ffdshow, no picture.

[Inventive Software]

I have tested with WMV9 AP, encoded as interlaced, decoded with libavcodec, and that was fine...

[vlada]

Try this sample: http://www.megaupload.com/de/?d=7QIPXP6I or http://www.sendspace.com/file/lvlvtv.

[demoniccannon]

helloooooooooooo guys, i finally managed to register an account here because i have a few bugs and suggestions to say.
1. h.264 lossless encoding simply doesn't work. whenever i encode with it in virtualdub, it pops "cannot start video compression: an unknown error occurred(error code -100)". but i remember an older version of ffdshow worked, which turned out to be beta2.
2. newer versions of ffdshow crash with certain .ass subtitles, usually when displaying several lines at different positions, for example on line at the top and the other at the bottom simultaneously. older versions such as beta3 don't crash, but display subtitles at wrong positions instead. well i see you claimed ass/ssa support experimental so i have to live with it, but i hope you would keep working on it.
3. what about adding support to 2pass/3pass h.264 encoding? since it's supported by x264.
4. who are the guys working on libavcodec? can decoding speed be even further improved? although it's pretty fast compared to other codecs, it still stretches my poor old computer out.
5. i hear you are going drop support for win9x, that's a sad news because many people still keep their ancient computers and win98 as a backup platform to run certain old games and applications, and there will be no effective codecs to play various media types on those computers. so if you plan to do that, i hope there would least be a relatively complete and stable final release for win98 before that.
last, dispite a few flaws, ffdshow is still my favorite codec so keep on good work!
and btw the ffdshow forum sux. i have tried to register at least 3 accounts using different mail addresses, but none of them received the "activation mail"!

[haruhiko_yamagata]

Originally Posted by demoniccannon

2. newer versions of ffdshow crash with certain .ass subtitles, usually when displaying several lines at different positions, for example on line at the top and the other at the bottom simultaneously. older versions such as beta3 don't crash, but display subtitles at wrong positions instead. well i see you claimed ass/ssa support experimental so i have to live with it, but i hope you would keep working on it.

Thank you for testing and report. Could you send me a sample? As far as I know, it works.

5. i hear you are going drop support for win9x, that's a sad news because many people still keep their ancient computers and win98 as a backup platform to run certain old games and applications, and there will be no effective codecs to play various media types on those computers. so if you plan to do that, i hope there would least be a relatively complete and stable final release for win98 before that.

I'm working on it.

[Inventive Software]

I got a suggestion for the SVN changelog... when you put "updated libavcodec", maybe a note to say what ffmpeg SVN version you used would be good so we can track changes more. "updated libavcodec" just seems too general in a way.

[vio_man]

I've seen this decoder test (I know this is not 0-day news) and I would like to know if the ffdshow's decoding speed has been improved since this test?

[demoniccannon]

Originally Posted by haruhiko_yamagata

Thank you for testing and report. Could you send me a sample? As far as I know, it works.

hello haruhiko_yamagata, i m sorry i can't send you the crashing sample because it's 300mb large to send, also it's mkv encapsulation so it's difficult to cut the crashing part and keep the crashing ass in sync time.
btw: the sample never crashed with subtitles turned off, so i m sure it's caused by ass handling.

[haruhiko_yamagata]

Originally Posted by demoniccannon

hello haruhiko_yamagata, i m sorry i can't send you the crashing sample because it's 300mb large to send, also it's mkv encapsulation so it's difficult to cut the crashing part and keep the crashing ass in sync time.
btw: the sample never crashed with subtitles turned off, so i m sure it's caused by ass handling.

I don't need the video and audio tracks of the file. Please extract the subtitle track.

• Download and install mkvtoolnix.
• Launch cmd.exe and goto the directory where the file exists.
• Type "mkvmerge -i filename.mkv" to get the track number of the subtitle.
• Type "mkvextract tracks filename.mkv 3:ass_crash.ass", if the track number of the subtitle is 3.

Please send me ass_crash.ass and tell me dimension of the video and where it chrashes.

[albain]

I have posted an updated build of FFDShow 64 bit to sourceforge.

Can someone tell me if it is working ?

Thanks

Damien BT

[Px]

Originally Posted by albain

I have posted an updated build of FFDShow 64 bit to sourceforge.
Can someone tell me if it is working ?
Thanks
Damien BT

Check in Vdub x64 and MPC-HC x64 under WinXP x64 on variety of files (DivX, Xvid, Mpeg2, H.264, both SD and HD up to 1080p), works fine

[albain]

Originally Posted by Px

Check in Vdub x64 and MPC-HC x64 under WinXP x64 on variety of files (DivX, Xvid, Mpeg2, H.264, both SD and HD up to 1080p), works fine

OK thanks !

[PatchWorKs]

Originally Posted by Px

Check in Vdub x64 and MPC-HC x64 under WinXP x64 on variety of files (DivX, Xvid, Mpeg2, H.264, both SD and HD up to 1080p), works fine

+1 (Win x64/MPC-HC & DV files)

[clsid]

Beta 4 has officially been released.

[fastplayer]

Originally Posted by clsid

Beta 4 has officially been released.

Great!
I'll update the links and changelog ASAP.

[Delerue]

Originally Posted by clsid

Beta 4 has officially been released.

Weird... This video here uses dual-thread when decoded by 1681 xxl build, but it uses only one if decoded by your beta 4 1685 build. Any idea?

[clsid]

That is because the EXPERIMENTAL multithreading patch for H.264 decoding has NOT yet been applied to the official ffdshow code. Please read the opening post of this topic before you start asking when that will happen.

[Delerue]

Originally Posted by clsid

That is because the EXPERIMENTAL multithreading patch for H.264 decoding has NOT yet been applied to the official ffdshow code. Please read the opening post of this topic before you start asking when that will happen.

Hmmmm... I'm not asking for a 'when'. I only thought that is weird that an earlier build doesn't have something that an old build has. Sorry, man.

[multiblitz]

Originally Posted by haruhiko_yamagata

I found the flag. Haali's media splitter is working perfectly.
Damn, I did check the AM_SAMPLE2_PROPERTIES::dwTypeSpecificFlags of the first and second frames, but didn't check the third...
I've modified the code a bit to set time stamp properly. Now the sync is OK, but it still has two problems. One is interlacing related flags to send to downstream, one is jerkiness caused by unloading and loading dlls during playback. Second problem is caused because ffdshow receives samples with media type attached and then re-initialize codec drivers.
The fix will surely break the code too much that I can't do without adding bugs. I'm going to do it after beta4. Please wait.

Now that Beta 4 is out: PLEEEEEAAAASE, Please give us the possibilty to playback HD-DVD with ffdshow...to het the ultimate home-cinema experience....THX a lot

[Inventive Software]

We're waiting on menu support for that miracle to happen, and a *proper* E-AC3 decoding patch, along with TrueHD and DTS-HD decoders. All of these are quite a long way off. E-AC3 is preliminarily a dodgy patch in ffmpeg, TrueHD requires the actual decoder, DTS-HD hasn't been reverse-engineered yet.

[Inventive Software]

My digital camera videos with FourCC SP54 are unable to be decoded by ffdshow, where they were a couple of years ago. It's the Sunplus MJPEG codec, and they used various FourCCs for different cameras. The only one I could spot that was similar was SP5X. Any chance of SP54 being added to the list? If memory serves correctly, FourCCs SP51, SP55, SP56 and SP57 were also used...

[Liisachan]

Hi,
Is this a known limitation or am I doing something wrong?
- Xvid and DivX with b-frames get decoder delay when decoded by ffdshow VFW 'libavcodec' but not by 'Xvid'.

If it's a known limitation, I'd prefer that ffdshow VFW uses 'Xvid' by default for Xvid/DivX. Or if it's fixable that'd be the best of course.

- Phenomenon: If Xvid or DivX video in AVI with b-frames is decoded by ffdshow 'libavcodec' via VfW (e.g. on VirtualDub),
the results get some kind of decoder delay:
* Frame 0 is first displayed as null (empty black) even if it is not originally empty black.
* Seeking via keyframes doesn't work correctly. VirtualDub apparently knows the correct frame number
for keyframes, but not getting the correct image at that frame number in presentation order. Apparently 1 (or 2) off.
* The results of seeking around the keyframe is different depending on whether you're seeking forward or seeking back.

This happened when I happened to uninstall Xvid/DivX and ffdshow was the only decoder avilable.

- If 'Xvid' is selected in the ffdshow VfW conf, the pb is gone.
Also, there is no such pb if Xvid or DivX codecs are used to decode it instead of ffdshow.
- ffdshow 'libavcodec' via DirectShow seems okay too.
- Tested both ffdshow beta 3 and 4.

Thanks for any suggestions

[JarrettH]

Congrats on beta 4 you guys

[Yong]

ffdshow OSD screw up when playback this mjpeg video
http://y0ngc6.googlepages.com/DungeonSiegeIITrailer2.7z
btw thanks, ffdshow is getting better and better now, i think we are close to 1.0

[clsid]

Originally Posted by Inventive Software

My digital camera videos with FourCC SP54 are unable to be decoded by ffdshow, where they were a couple of years ago. It's the Sunplus MJPEG codec, and they used various FourCCs for different cameras. The only one I could spot that was similar was SP5X. Any chance of SP54 being added to the list? If memory serves correctly, FourCCs SP51, SP55, SP56 and SP57 were also used...

Can you upload a sample file? I prefer to test playback when I add new FourCCs.

[multiblitz]

Originally Posted by Inventive Software

We're waiting on menu support for that miracle to happen, and a *proper* E-AC3 decoding patch, along with TrueHD and DTS-HD decoders. All of these are quite a long way off. E-AC3 is preliminarily a dodgy patch in ffmpeg, TrueHD requires the actual decoder, DTS-HD hasn't been reverse-engineered yet.

Well, let's start small: Simply by getting rid of the 24/30 fps bug, which does not allow to have ffdshow in dthe filer-chain for video. For Audio, the sonic codecs work OK for the moment.

The topic becomes obviously hotter now as cyberlink has erase the option to playback HD-DVD from harddisk in PDVD ultra. So, many people are looking now for alternatives. I personally found that a hood sharpening and noise-reduction is still important with HD-stuff...plus: PDVD in XP only works with overlay, not VMR9. So a good quality video-chain with ffdshow in would help a lot and later than aiudio and we have perfection...

[Blkbird]

The "Subtitles > Letterbox" option of the video decoder erroneously applies cropping instead of letterboxing if the specified AR is wider than the AR of the given video.

For example, if the letterbox AR is set to 8:5 (which is the common AR for most widescreen monitors nowadays), then a 16:9 video gets properly letterboxed top and bottom, however a 4:3 video isn't correctly letterboxed left and right, but instead gets cropped top and bottom. Since this happens both with Zoom Player and Media Player Classic, I'm pretty sure it's a problem with ffdshow and not with the player software.

I discovered this bug after recently (and finally) dumping my old CRT monitor and replacing it with a LCD one. I'm very surprise to find - through searching in this forum - that this bug has actually been reported year ago, but still isn't fixed yet. I would have thought a lot of users are affected by the bug, and I believe it shouldn't even be difficult to fix.

Also, while we at it, could someone please add an option to have vertical letterboxing - i. e. top and bottom, if the video is wider than the screen - applied asymmetrically? I assume most people choose to have their subtitle displayed on the bottom, and by having the letterboxing applied only to the bottom, there will be more empty space for the subtitle, so it remains more reliably off the video.

[haruhiko_yamagata]

Originally Posted by Liisachan

Hi,
Is this a known limitation or am I doing something wrong?
- Xvid and DivX with b-frames get decoder delay when decoded by ffdshow VFW 'libavcodec' but not by 'Xvid'.

If it's a known limitation, I'd prefer that ffdshow VFW uses 'Xvid' by default for Xvid/DivX. Or if it's fixable that'd be the best of course.

Thank you. I can reproduce. I saw the black screen many times, I wasn't careful enough to find a bug.

[haruhiko_yamagata]

Originally Posted by Yong

ffdshow OSD screw up when playback this mjpeg video
http://y0ngc6.googlepages.com/DungeonSiegeIITrailer2.7z
btw thanks, ffdshow is getting better and better now, i think we are close to 1.0

Thank you. Confirmed.

[Inventive Software]

Originally Posted by clsid

Can you upload a sample file? I prefer to test playback when I add new FourCCs.

Ask and ye shall receive... within reason!

http://rapidshare.com/files/74954695/bgp_0041.avi

[fastplayer]

Can somebody please check out the new "Speed Racer" trailer?
All I get with MPC/beta4 is 100% CPU load and weird noise. VLC plays the trailer without breaking a sweat...
Download [85.3MB, H.264/SOWT]

[Liisachan]

Originally Posted by haruhiko_yamagata

Thank you. I can reproduce.

You're very welcome.

Another thing, for anyone who is in charge of the web pages. The change log here still says "Added before beta1... WavPack (lossless)"...
http://ffdshow-tryout.sourceforge.net/html/en/
...which is misleading and confusing since WavPack support was dropped. Maybe not saying about WavPack at all would be better. Either that or adding a note like "(WavPack support has been temporarily removed due to technical problems)"?

[clsid]

Originally Posted by Inventive Software

Ask and ye shall receive... within reason!

http://rapidshare.com/files/74954695/bgp_0041.avi

FourCCs have been added in rev1688

[fastplayer]

Originally Posted by Liisachan

Either that or adding a note like "(WavPack support has been temporarily removed due to technical problems)"?

If there are no objections, I'll change the docs.

[clsid]

Originally Posted by fastplayer

Can somebody please check out the new "Speed Racer" trailer?
All I get with MPC/beta4 is 100% CPU load and weird noise. VLC plays the trailer without breaking a sweat...
Download [85.3MB, H.264/SOWT]

Sound is ok with Haali's splitter, which detects the audio as PCM.
Noise and high cpu confirmed with Gabest's splitter.

[fastplayer]

Originally Posted by clsid

Sound is ok with Haali's splitter, which detects the audio as PCM.
Noise and high cpu confirmed with Gabest's splitter.

Thanks for confirmation.
"Uncompressed audio --> 16bit integer" is the correct configuration for this trailer though, right?

[Inventive Software]

Originally Posted by clsid

FourCCs have been added in rev1688

I owe you a drink at some point. Thanks very much for that!

[haruhiko_yamagata]

Originally Posted by fastplayer

If there are no objections, I'll change the docs.

Thank you. Please remove the line.

[haruhiko_yamagata]

Originally Posted by Liisachan

Hi,
Is this a known limitation or am I doing something wrong?
- Xvid and DivX with b-frames get decoder delay when decoded by ffdshow VFW 'libavcodec' but not by 'Xvid'.

If it's a known limitation, I'd prefer that ffdshow VFW uses 'Xvid' by default for Xvid/DivX. Or if it's fixable that'd be the best of course.

Unfortunately it seems to be the spec of libavcodec. The fix would be hard.

@clsid : Could you update the installer so that Xvid is used as VFW decoder (of course if Xvid is packed in the package)?

[haruhiko_yamagata]

Originally Posted by Blkbird

The "Subtitles > Letterbox" option of the video decoder erroneously applies cropping instead of letterboxing if the specified AR is wider than the AR of the given video.

For example, if the letterbox AR is set to 8:5 (which is the common AR for most widescreen monitors nowadays), then a 16:9 video gets properly letterboxed top and bottom, however a 4:3 video isn't correctly letterboxed left and right, but instead gets cropped top and bottom. Since this happens both with Zoom Player and Media Player Classic, I'm pretty sure it's a problem with ffdshow and not with the player software.

I discovered this bug after recently (and finally) dumping my old CRT monitor and replacing it with a LCD one. I'm very surprise to find - through searching in this forum - that this bug has actually been reported year ago, but still isn't fixed yet. I would have thought a lot of users are affected by the bug, and I believe it shouldn't even be difficult to fix.

Thank you I can reproduce.

Also, while we at it, could someone please add an option to have vertical letterboxing - i. e. top and bottom, if the video is wider than the screen - applied asymmetrically? I assume most people choose to have their subtitle displayed on the bottom, and by having the letterboxing applied only to the bottom, there will be more empty space for the subtitle, so it remains more reliably off the video.

Please use resize and borders for this advanced configuration. "Dividing borders" options have been added for your purpose. It is possible to use those settings for letterboxing. Is it a good idea?

[Blkbird]

Originally Posted by haruhiko_yamagata

Thank you I can reproduce.

Great! Hope it can be fixed soon.

Originally Posted by haruhiko_yamagata

Please use resize and borders for this advanced configuration. "Dividing borders" options have been added for your purpose. It is possible to use those settings for letterboxing. Is it a good idea?

The way I see it, "Resize & aspect > Borders" sets an unconditional borders - which isn't what I want. I want the letterboxing only to apply if the video AR is not equal to my screen AR. And I want the config to be universal, so I don't have to change it when I watch a different video.

But I think you've indirectly raised a valid point - whether the "Letterbox" option should belong to "Subtitles", or is it logically more consistent to move it somewhere under "Resize".

This is what I'd like to have:

1. The video canvas of the media player should always have the same AR as my screen at all times, letterboxing in case the video has a different AR.

2. Keep the letterboxing to the minimum, meaning do it either horizontally or vertically (or none as all of course, if the video AR is already equal to the screen AR), depending on the AR difference, but never both.

3. In case of vertical letterboxing, apply only to the bottom side and keep the video all the way up.

[haruhiko_yamagata]

Originally Posted by Blkbird

The way I see it, "Resize & aspect > Borders" sets an unconditional borders

Nope, this is not correct.
Resize to your screen size and select "Keep original aspect ratio". Check "Process pixel aspect ratio internally". Inside/Outsize borders 0%. Dvide vertical borders to bottom 100% or whatever you like. Goto Reize settings page and select Luma method which is suitable for your CPU power. If your CPU is powerful enough use Bicubic or Lanczos and output in RGB32 for better quality.

In the settings above, borders are added only when necessary.

[Yong]

uh one more crash, didnt test it carefully
same clip:
http://y0ngc6.googlepages.com/DungeonSiegeIITrailer2.7z
ffdshow.ax crash if using ffdshow avisynth filter.

[haruhiko_yamagata]

Originally Posted by multiblitz

Now that Beta 4 is out: PLEEEEEAAAASE, Please give us the possibilty to playback HD-DVD with ffdshow...to het the ultimate home-cinema experience....THX a lot

I'm working on it .

[Thunderbolt8]

Originally Posted by multiblitz

Now that Beta 4 is out: PLEEEEEAAAASE, Please give us the possibilty to playback HD-DVD with ffdshow...to het the ultimate home-cinema experience....THX a lot

same for blu-rays please

[Liisachan]

Originally Posted by haruhiko_yamagata

Unfortunately it seems to be the spec of libavcodec. The fix would be hard.

@clsid : Could you update the installer so that Xvid is used as VFW decoder (of course if Xvid is packed in the package)?

I did some more tests. Not surprisingly, libavcodec has the same problem with other MPEG4 Part2 codecs, such as 3ivx, too. But selecting 'Xvid' doesn't help here and I see no reason to change this one:
Other MPEG4 -- libavcodec
while these appear to be reasonable, fixing the pb.
Xvid -- Xvid
DivX 4/5/6 -- Xvid

The pb doesn't happen as long as you have Xvid and/or DivX installed, when ffdshow is not used in VFW (VirtualDub etc). Those who use VirtualDub are very likely to have Xvid or DivX, so I'd say this problem is not that critical in reality. A hidden glitch that usually doesn't surface, until for some reason you need to do frame-accurate editing of video via ffdshow VfW.

Thanks again anyway~

[Blkbird]

Originally Posted by haruhiko_yamagata

Resize to your screen size and select "Keep original aspect ratio". Check "Process pixel aspect ratio internally". Inside/Outsize borders 0%. Dvide vertical borders to bottom 100% or whatever you like.

Looks incredibly complicated, but it's indeed somehow working. Now I'm wondering why "Subtitles > Letterboxing" is there in the first place because it's fully redundant to this set of settings. Anyway, I would have never figured it out by myself since I don't know what a lot of those options really mean.

BTW, do you really want me to "resize to screen size" as opposed to "resize to screen AR"? Because if the player canvas is resized to screen size, then the player window would overflow the screen.

Also, who is normally doing the fullscreen resizing, the video renderer? Would the resize by ffdshow be significantly better then, say, VMR9?

Originally Posted by haruhiko_yamagata

Goto Reize settings page and select Luma method which is suitable for your CPU power. If your CPU is powerful enough use Bicubic or Lanczos and output in RGB32 for better quality.

What is this "Lanczos" method really? And what do you mean by "output in RGB32"? - Should I disable all output colorspaces except for RGB32?

My CPU is an Athlon 64 X2, so yes, it's fairly fast.

[squid_80]

Originally Posted by Liisachan

I did some more tests. Not surprisingly, libavcodec has the same problem with other MPEG4 Part2 codecs, such as 3ivx, too. But selecting 'Xvid' doesn't help here and I see no reason to change this one:

If the bitstream isn't packed, you're going to have delay issues no matter what codec is used for decoding.

[Liisachan]

Originally Posted by squid_80

If the bitstream isn't packed, you're going to have delay issues no matter what codec is used for decoding.

Of course it is packed for Xvid/divx. That's default, isn't it? And the user data is getting "p" like "DivX503b1393p" in Xvid too, for example.

That's about Xvid and divx. I don't know about 3ivx. I just used it for testing; had never used it before, and probably will never use it for anything real. Since 3ivxVfWCodec.dll itself cannot decode its own format frame-accurately, we cannot blame libavcodec here.

[haruhiko_yamagata]

Originally Posted by Blkbird

Looks incredibly complicated, but it's indeed somehow working. Now I'm wondering why "Subtitles > Letterboxing" is there in the first place because it's fully redundant to this set of settings. Anyway, I would have never figured it out by myself since I don't know what a lot of those options really mean.

Yes, it's incredibly complicated. Here is a document. Not perfect though.

BTW, do you really want me to "resize to screen size" as opposed to "resize to screen AR"? Because if the player canvas is resized to screen size, then the player window would overflow the screen.

"resize to screen AR" is OK too. "resize to screen size" is for full screen mode, i.e. home theater purpose.

Also, who is normally doing the fullscreen resizing, the video renderer? Would the resize by ffdshow be significantly better then, say, VMR9?

Nope, but if you want better quality for subtitles, it is best to use software resize before subtitles and output in RGB32.

And what do you mean by "output in RGB32"? - Should I disable all output colorspaces except for RGB32?

Yes. The document. The help button on the dialog page works.

My CPU is an Athlon 64 X2, so yes, it's fairly fast.

I hope that's fast enough, but the settings described here is quite heavy.

[Liisachan]

Originally Posted by squid_80

If the bitstream isn't packed, you're going to have delay issues no matter what codec is used for decoding.

PS: I think you're right about 3ivx. I just found that this 3ivx AVI is not frame-accurate even when it is played via DirectShow., and cannot be used for anything that requires good sync (eg softsubbing, karaoke).
Among ffdshow 'Other MPEG4', 'libavcodec' works relatively better for me than 'Xvid', both in VFW and DS, for this 3ivx thing, which however you cannot seek frame-accurately anyway.

Then again, I don't believe there are many people who use 3ivx in AVI. Xvid and DivX are much more common, and 3ivx is not free either.

Anyway, 'libavcodec' VFW is getting the decoder delay even if the bitstream is 'packed', while 'Xvid' supports that hack; that's why I'd like to suggest 'Xvid' as default in Xvid and DivX.

[clsid]

The installer has already been modified to use xvidcore by default for VFW decoding.

Why should libavcodec be used for 'Other MPEG4'? XviD/DivX encoded video can also fall under this type, since the choice is simply made based on the FourCC.

[SeeMoreDigital]

Do we really need to separate all the MPEG-4.2 "format" types now anyway?

Why not offer just one, like there is for "H.264/AVC" and call it something like "MPEG-4 ASP". Complete with the "disabled, libavcodec and Xvid" decoder options?


Cheers

[clsid]

Yes, they are needed because not everyone uses ffdshow for all MPEG-4 ASP variants.

[Blkbird]

Originally Posted by haruhiko_yamagata

Yes. The document.

After turning off all YUV output formats, I only get blank screen with all the XviD videos (ffdshow build 1673, icl, x86, unicode).

[SeeMoreDigital]

Originally Posted by clsid

Yes, they are needed because not everyone uses ffdshow for all MPEG-4 ASP variants.

More fools them....

[haruhiko_yamagata]

Originally Posted by Blkbird

After turning off all YUV output formats, I only get blank screen with all the XviD videos (ffdshow build 1673, icl, x86, unicode).

It doesn't seem to be a bug of ffdshow.
To which filter ffdshow is connected?
Updating the video driver may help.

[Liisachan]

Originally Posted by clsid

The installer has already been modified to use xvidcore by default for VFW decoding.

Why should libavcodec be used for 'Other MPEG4'? XviD/DivX encoded video can also fall under this type, since the choice is simply made based on the FourCC.

in my tests, libavcodec works better for 3ivx in seeking, while xvid is better for xvid. the stand alone xvid decoder does not even support 3ivx officially btw

[Liisachan]

Originally Posted by SeeMoreDigital

Why not offer just one, like there is for "H.264/AVC" and call it something like "MPEG-4 ASP". Complete with the "disabled, libavcodec and Xvid" decoder options?

Maybe one of the typical situations is, when you have Xvis or DivX or other codecs installed and you want them decode things (everything they can decode or their own formats), but you want ffdshow to decode the other formats. Many users use Xvid standalone codec for Xvid in VFW (if they used ffdshow def. there would be a lot of bug reports by now) or maybe also in DS, but they don't have 3ivx so they still need ffdshow in some case when they happen to need to test a 3ivx clip. Or similarly some ppl may prefer Divx standalone decoder.

Plus, altho I'm not sure about this, a specific standalone decder may have too high merit, and ffdshow might have to use really hight merit for that specific format to be enabled, as a workaround. If ffdshow has a very high merit for MPEG-4 Part2 generally, there might be some kind of bad side effects. A very hight merit is supposed to be a workaround only for a specific codec for specific fourccs. I kind of remember milan says something about this, about DivX... but I may be wrong.

Personally, I can live with it if MPEG-4 Part 2 becomes one item. (But don't call part2 mepg4 "ASP". It's a term used mainly only in doom9 forums)

[SeeMoreDigital]

Originally Posted by Liisachan

Personally, I can live with it if MPEG-4 Part 2 becomes one item. (But don't call part2 mepg4 "ASP". It's a term used mainly only in doom9 forums)

Agreed...

I much prefer to see terms such as: MPEG-4 Part-2 or MPEG-4.2. And MPEG-4 Part-10 or MPEG-4.10, being used....

[FFWD]

[edit]
potential flame-bait removed.

[haruhiko_yamagata]

Originally Posted by FFWD

Is there any reason why the CCCP guys deny they use ffdshow tryouts, while they actually ARE using it? Why the bashing?

btw: they haven't published their patches for the latest beta.

Everyone, please just ignore them. I don't want flame here.
And it's best not to read that link.

[Liisachan]

@SeeMoreDigital
"DivX, Xvid, etc."
would be easier to understand for many users. And btw, when I said, '(but don't call part2 mepg4 "ASP")', I meant in the menu of ffdshow. I have nothing against the term "ASP" itself. It should be ok to call MPEG-4 Part2 "ASP" here in this forum as a kind of convention, because they/we/you know what it means.

[haruhiko_yamagata]

Fix: If ffdshow was used for decoding VC-1 video, then framerate got changed from 23,976 fps to 29.97 fps.

Now it should work as long as you decode VC-1 in ffdshow. I don't have many samples, please send me a sample that doesn't work.
Unfortunately, DMO wrapper doesn't send field 1 repeat flag if it is connected to ffdshow as raw video processor. Besides it sets wrong time stumps, which is much worse than nothing.
I don't know what trick it uses to present at the correct time when it is connected to VMR. Perhaps it doesn't set time stumps and send field 1 repeat flags if it is connected to VMR (just a random guess though).

[Thunderbolt8]

Originally Posted by haruhiko_yamagata

Now it should work as long as you decode VC-1 in ffdshow. I don't have many samples, please send me a sample that doesn't work.

the remuxed video from galapagos HD DVD made trouble, vc-1 with libavcodec didnt give out a picture at all, only sound. had to switch back to wmv9 for that. the video is 1080i at 29.97 fps, guess this combination was too tricky. unfortunately my samples at sendspace or megaupload already got deleted :S

[Shakey_Jake33]

This isn't an essential thing, and I don't know how difficult it would be to do, but I have both the 32-bit and 64-bit builds of ffdshow installed (no real need I know, but I do occasionally fiddle with Vista Media Center x64 and mpc x64), and whatever settings/changes I make for one build, it affects the other (for example, if I disable H.264 in the 32-bit version, it disables it in the 64-bit version too).

Any chance they can be made completely separate from each other at some point?

(same goes for mpc actually, settings changed in one affect the other)

[squid_80]

Originally Posted by Shakey_Jake33

Any chance they can be made completely separate from each other at some point?

Why would you want that?

[check]

Is there any reason why the CCCP guys deny they use ffdshow tryouts, while they actually ARE using it? Why the bashing?

You are trying to stir up trouble. We like and endorse tryouts. You can see that because we use it in our packs and explicitly name the project during our installation. Tryouts is awesome and the developers are awesomer for continuing to keep ffdshow up to date.

The only thing said in that thread was that we didn't include beta3 in the pack. beta3 is rev1324. We include r1355. Therefore, we don't include beta3.

btw: they haven't published their patches for the latest beta.

We don't need to, patches must only be published for releases. Betas are for testing by CCCP staff and other interested users.

Please, let me say it again. We Heart Tryouts.

[Shakey_Jake33]

Originally Posted by squid_80

Why would you want that?

Because in some instances, I'd want to apply some settings to one build but not the other (e.g. I prefer to use CoreAVC for H.264 in the 32-bit build, but ffdshow in the 64-bit build. Another example is I prefer to use ffdshow for Mpeg2 in the 32-bit built, but the Microsoft decoder in the 64-bit build). My current workaround is to set CoreAVC to 'Prefer' in mpc but...
As I say, it's not essential, it'd just be handy.

[clsid]

I agree that 64-bit builds should not write to the 32-bit registry hive.

[haruhiko_yamagata]

Originally Posted by clsid

I agree that 64-bit builds should not write to the 32-bit registry hive.

OK, let's do that in the future.
By the way, has anyone tried MinGW - Minimalist GNU for Windows 64 ?

[haruhiko_yamagata]

Originally Posted by Thunderbolt8

the remuxed video from galapagos HD DVD made trouble, vc-1 with libavcodec didnt give out a picture at all, only sound. had to switch back to wmv9 for that. the video is 1080i at 29.97 fps, guess this combination was too tricky. unfortunately my samples at sendspace or megaupload already got deleted :S

Thank you for testing. libavcodec doesn't handle interlaced VC-1 contents AFAIK. Please use wmv9 for real use.

[Thunderbolt8]

is there a possibility to change this in the future ? dont like to do without llibavcodec's quality

[haruhiko_yamagata]

Originally Posted by Thunderbolt8

is there a possibility to change this in the future ? dont like to do without llibavcodec's quality

As you know, we just depend on FFMPEG. Let's look forward to it　.

[ACrowley]

Mhh.. i made a Test with (Haali-)libavcodec ,ffdshow build 1695 clsid , with VC1 1080p HDDDVD (Doom)

Its still 29.97FPs and i got no Picture

[clsid]

A sample file would be helpful.

[Thunderbolt8]

might be able to make a sample from galapagos again in 1-3 days

[Blkbird]

Originally Posted by haruhiko_yamagata

It doesn't seem to be a bug of ffdshow.
To which filter ffdshow is connected?
Updating the video driver may help.

You're right. It seems the problem only occurs if I set my player (Zoom Player) to use the "overlay mixer".

[Kurtnoise]

@devs: what about a mpc sv8 support from lavc ? There is a splitter now...

[rickardk]

Originally Posted by ACrowley

Yeah...is think ist a Multithreading Issue.
But WMVideoDecoder pushes both Cores ...maybe not efficient enough on Athlon Systems. I
dont know Cyberlink VC1 Decoder works perfect and fully smooth on VC1...to bad that we cant use the Decoder outside PowerDVD

With SonicVC1Decoder and SonicCinemasterVideoDecoder4.3 and WMVideoDecoderDMO its all the same.
Video starts playing smooth with ~24fps and after a few seconds the fps goes down to ~15 and it starts stuttering

Also the WMVideoDecoder cant use DXVA on VC1...DXVA works nice on WM9 MP

So far i Know ,the Problem only exist for A64 Users...all Intel CoreDuo User "i" know are having no Problems

No Rpoblme for my because I dont watch the untouched VC1, i always reencode it to 1080p x264 with German and English Audio.
However...im going to by to replace my A64 X2 5000 with a Intel Quad this Year

@RichardK
As i told you :
To get alltime 23.976 simply remove the Pulldown from HDDVD VC1 with vc1conv .
The remux it into your TS .
For MKV :
1. gdsmux (haali) muix EVO VC1 into mkv. You only get 29.97 FPS !! So youve to remux it :
2.mkvmerge : remux it with 23.976 Timecode
So you have solved you 29.97 fps Problem

I have given up on waiting for a solution on this 29.97 problem (I hoped it could be solved on splitter or decoder level).

So I will remove pulldown from all my HD DVD remuxes. I know how to extract the vc-1 es from my ts (using tsremux). But how do I replace it with the new stream created by v1conv (without pulldown).

Do I have to rebuild the TS with all audio streams seperatly extracted. Or can I mux the new vc-1 es into my old ts. Replacing the old vc1 es. Is there any tool for this?

[ACrowley]

cant help you cause i dont use VC1 in TS , SRY

[_xxl]

DirectX Video Acceleration Specification for H.264/AVC Decoding:

http://www.microsoft.com/downloads/d...ng=en#filelist

DirectX® Video Acceleration Specification for Windows Media Video® v8, v9 and vA Decoding (Including SMPTE 421M "VC-1")

http://www.microsoft.com/downloads/d...ng=en#filelist

Quick Details
Version: 1
Date Published: 12/14/2007
Language: English

[Inventive Software]

Grabbed with thanks! I'll take a look when I got some free time this weekend.

[Thunderbolt8]

Originally Posted by _xxl

http://www.microsoft.com/downloads/d...ng=en#filelist

http://www.microsoft.com/downloads/d...ng=en#filelist

in how far will we be able to benefit from that when we use ffdshow with libavcodec for both (h264 and vc-1) for playback? will this work together?

[RadScorpion]

Originally Posted by Kurtnoise13

@devs: what about a mpc sv8 support from lavc ? There is a splitter now...

Well. there is more than just a splitter

http://blog.monogram.sk/janos/2007/1...decoder-v0900/

... more to come. Any testers, reporters, hints, ideas are welcome.

Igor

[clsid]

Originally Posted by RadScorpion

Well. there is more than just a splitter

http://blog.monogram.sk/janos/2007/1...decoder-v0900/

... more to come. Any testers, reporters, hints, ideas are welcome.

Igor

I hope this filter will eventually also support the older Musepack formats. Combining this filter with the Radlight Musepack filter doesn't work in WMP.

Removing the dependency on the VC2005 redist by static linking would be useful too.

[RadScorpion]

Hm... the older radlight filter works quite fine here along with the new splitter/decoder filters :-\. Yes. SV7 support is on the todo list.

As for the static linking, I prefer smaller files and dynamic linking, but you're right. Having both won't hurt anyone.

[clsid]

The two filters work fine for me in Media Player Classic. But your filter doesn't work properly in WMP as a result of the regkey "HKCR\Media Type\Extensions\.mpc", which is required for the RadLight filter. That key forces WMP to use the RadLight filter for all .mpc files. Removing this key makes your filter work, but then the other one fails.

But once you add SV7 support, the RadLight filter is obsolete

[RadScorpion]

Oh. I see now. there have been two versions of the old MPC filter. One that was (FileSrc+Split+Dec) in one filter, and the other one could connect to file source filter. So I assume you are using the first one.
Ok. Thanks for testing... gonna take a close look at the SV7 stuff.

[clsid]

That explains it. I have got rlmpcdec.ax 1.0.0.4. Any idea where I can get the other version of the filter?

[RadScorpion]

I think it's part of latest radlight build... at least that's what graphedit said.
In case someone was interested I've uploaded the filter here :
http://blog.monogram.sk/janos/downlo...rs/RLMPCDec.ax

[Thunderbolt8]

got a problem with ffdshow's h264decoder/codec (libavcodec problem?): when I remux the 1st episode from the band of brothers HD DVD (AVC) then I get rainbow frames, funny coloured pictures (it looks already that way in the original .evo, when I open it with mpc & ffdshow). all the other episodes are fine so far, though. tested this episode with coreavc and no rainbow frames to be seen in there.
I can provide you with a sample, just tell me where to post the link and if the sample has to be from the original .evo, demuxed raw stream (evodemux) or remux to .mkv and if it matters if the sample is taken from the mid of the file or has to be from the beginning.

edit: heres a sample (video only, remux to .mkv) http://www.sendspace.com/file/yml19z

and heres another 50mb sample of the original .evo so you can remux and demux the video stream yourself and see if this could help
http://www.sendspace.com/file/iexd6k

[_xxl]

Mplayer is working?

[Thunderbolt8]

dont have mplayer, in how far is it different from mpc?

[clsid]

It does not use DirectShow. But more importantly it uses the libavcodec library, like ffdshow does. So in case of decoding problems, it is useful to know if the same problem also occurs with a recent SVN build of mplayer. If the problem also occurs with mplayer, then the bug is likely to be within libavcodec.

[Thunderbolt8]

ill see if I can get mplayer working here then.
where should I download it from, can it be from here: http://esby.free.fr/CelticDruid/mirror/mplayer/ ?
if yes, do I have to take the "neutral" version (=non athloxp, prescott, p4 etc.; got a c2d)?

edit: aerf its a command line tool, how can I get a frontend for it ? :S

edit²: went to mplayerhq.hu and downloaded the "MPlayer 1.0rc2 Windows GUI file" from there and the funny frames are still in it with this version. so if this was a valid test, yes, it looks funny as well with mplayer.

[clsid]

You can find a newer version here:
http://mulder.dummwiedeutsch.de/home...ojects#mplayer

[Thunderbolt8]

got the celtic druid version from december, 3rd running with mpui, same funny picture with it

[Thunderbolt8]

Originally Posted by clsid

You can find a newer version here:
http://mulder.dummwiedeutsch.de/home...ojects#mplayer

funny picture with this version as well

[toytown]

I tried the clip with coreavc/powerdvd/ffdshow, only ffdshow had the problem viewing it.

Interesting though, if i opened the clip and immediately jumped to say 10 seconds in, then the clip played fine without the rainbow affect going through there faces.

[Thunderbolt8]

not in my case, no matter whether I played the evo or the remuxed .mkv and jumped somewhere, the effect was always present

[toytown]

I just tried it again with windows media player and media player classic, it definetely works it i seek immediately before it starts showing the first frame.

[Thunderbolt8]

dont seem to be able to get that, it starts too fast before I can hit space for pause. best I could get so far with mpc was 1 drawn frame

[toytown]

I guess what i was trying to add, is that, it can decode it properly from x frames in, but there is something in the first frames that is an issue

[Thunderbolt8]

hm I muxed the file again and enabled file splitting after 20mb. so I tried to play the 2nd file then, because the very first frame of the big video is then skipped of course, but the problem still existed.

[Inventive Software]

Originally Posted by Thunderbolt8

funny picture with this version as well

See if you can find a recent ffplay binary, and try it with that.

[Thunderbolt8]

im not familiar with ffplay and mplayer and that at all so I dont really know where to look for it and which files I need :S

[clsid]

Just be patient and wait until the FFmpeg developers have implemented support for the entire H.264 standard.

In the meantime, you could use for example CoreAVC.

[Thunderbolt8]

I'd rather prefer to die instead, coreavc's PQ sucks :P
then I will stick to powerdvd just for this episode until a solution has been found.

btw what does it mean 'for the entire H.264 standart', is there something missing, I thought h/x264 was already fully implemented?

[EpheMeroN]

Just got a new Core 2 Duo laptop running Vista Home Premium.

Was wondering if there are recommended versions for a) Vista, and b) that are compiled for performance with a Core 2 Duo processor?

I just downloaded "ffdshow_beta4_rev1685_20071206_clsid.exe" off the SourceForge page.

[_xxl]

Was wondering if there are recommended versions for a) Vista, and b) that are compiled for performance with a Core 2 Duo processor?

Only handwritten code make a diference in speed.

I thought h.264 was already fully implemented

Interlaced pictures + spatial direct mode is not implemented.
Start a donation for this issue?

[clsid]

Originally Posted by EpheMeroN

Just got a new Core 2 Duo laptop running Vista Home Premium.

Was wondering if there are recommended versions for a) Vista, and b) that are compiled for performance with a Core 2 Duo processor?

I just downloaded "ffdshow_beta4_rev1685_20071206_clsid.exe" off the SourceForge page.

The version that you have works on all versions of Windows and is suitable for all kinds of processors. There is no 'optimized' version that will perform better on your processor.

[Thunderbolt8]

Originally Posted by _xxl

Interlaced pictures + spatial direct mode is not implemented. Start a donation for this issue?

would this actually help to speed things up?

[_xxl]

Some good news:

http://mpc-hc.svn.sourceforge.net/vi...v&revision=346

would this actually help to speed things up?

You could support ffmpeg developers for hard work.

[clsid]

Idea:
A whitelist/blacklist for DirectShow filters that ffdshow is allowed to connect with in a graph.

I think this may be particularly useful when raw video is enabled in ffdshow.

Any opinions?

[Inventive Software]

Anything as long as Cyberlink gets screwed over.

[molitar]

Originally Posted by _xxl

Some good news:

http://mpc-hc.svn.sourceforge.net/vi...v&revision=346

So how do these files work? And how do they work? These are to use the hardware acceleration built into our video cards correct? How will it compare to software like CoreAVC?

[Thunderbolt8]

Originally Posted by _xxl

You could support ffmpeg developers for hard work.

well I would even do so, knowing that I wouldnt be the only one then so that when actually most people who use ffdshow spent some money a larger sum would come together, which may really be a big support then. apart from that I guess in case i'd be the only one my money wouldnt do much :/

[Ryokurin]

Originally Posted by molitar

So how do these files work? And how do they work? These are to use the hardware acceleration built into our video cards correct? How will it compare to software like CoreAVC?

At least when it comes to hardware acceleration, the last time I used CoreAVC it was actually slower in hardware than using its software decoder.

[EpheMeroN]

After installation of the latest stable ffdshow tryout, I noticed it handles the video decoding for webcams w/ Windows Live Messenger. This is causing random crashes for the messenger. What format am I looking for in the config to disable ffdshow decoding?

[clsid]

Options -> DirectShow control

Add "msnmsgr.exe" to the blacklist.
Remove "msnmsgr.exe" from the whitelist.

[haruhiko_yamagata]

Originally Posted by EpheMeroN

After installation of the latest stable ffdshow tryout, I noticed it handles the video decoding for webcams w/ Windows Live Messenger. This is causing random crashes for the messenger. What format am I looking for in the config to disable ffdshow decoding?

Please install the latest Nightly build and try again. Rev 1687 fixes not only Windows Movie Maker.

[Thunderbolt8]

got another AVC sample where ffdshow libavcodec seems to have a problem. in seconds 1-5 you can see that the video is a bit off along a horizontal line, it doesnt really fit (well in every 2nd frame the picture is off there as it seems).

I included both, rawstream and .mkv remux. though the .mkv file is not taken from this exact piece of rawstream here, but cut from the complete .mkv I made from the complete rawstream, because the mux from this raw sample here results in mpc crashes when I try to open it (guess I hit some open frame or something like that, I cut with a hex editor).

I included both, because when trying to open the raw sample with mpc / ffdshow you wont be able to recognize anything like picture, the problem only becomes evident when opening the .mkv file with mpc / ffdshow. but coreavc gives a picture for both, the rawstram and the .mkv file and doesnt show the problem in either of them, so its basically to ensure that the problem is not caused by the demuxing of the rawfile or remuxing to .mkv process.

http://www.sendspace.com/file/ysw3qv


edit: is this problem also due because of those h264 things which are not yet implemented, or is this just a "normal bug" ?

[haruhiko_yamagata]

@Thunderbolt8
The .mkv video plays fine for me.

[Thunderbolt8]

with which version? im using xxl rev1703 and have those problems from seconds 1-5 of the sample

[haruhiko_yamagata]

Oh, I see. Probably it is fixed at rev 1711.

[Valeron]

kostya submitted the RV30/40 decoder core to ffmpeg recently, when will ffdshow get support of decoding those streams?

[Thunderbolt8]

Originally Posted by haruhiko_yamagata

Oh, I see. Probably it is fixed at rev 1711.

it is, I dled xxl rev1716 and its fine now (and also another problem with another video is fixed, too). thanks!

[clsid]

Originally Posted by Valeron

kostya submitted the RV30/40 decoder core to ffmpeg recently, when will ffdshow get support of decoding those streams?

The RV30/40 code in FFMPEG is still incomplete.

[Blight]

I just came upon a weird issue,
It seems ffdshow is not accepting the "{6134706D-0000-0010-8000-00AA00389B71}" sub-type "mp4a" fourcc (no matching media type error) from the Nero MP4 splitter (tested with v7 of nero).

I dont think this is nero white/black listing filters as the gabest realaudio decoder does accept a connection and does decode the audio.

[clsid]

I assume this is just plain AAC audio?

[gordonwatts]

Hi,
I've got a Sony SR7 which produces the .mts (or .m2ts) files. Using the latest version of ffdshow tryouts (rev 1716 by xxl) or beta4 does not work. But in a funny way.

First: I'm using Vista x32, 2 gig, old Pentium M 2GHz processor - so barely enough to keep up, probably. Also have the most recent version of the Haali splitter.

Tried in WMP11: The time line doesn't move at all. No CPU is being used; just sits forever. Icons for Haali splitter, ffdshow audio, and ffdshow video show up in the task bar. If I use the video codec config settings to disable the H264 decoding, then sound plays just fine.

Tried MPC. No sound at all (the sound options are all disabled in the menus). But video does play -- but only at half speed. The Haali splitter and ffdshow video icons show up in the sys tray.

Tried mplayer - the version released on 10/12. I get continuous errors: "PAFF interlacing not supported..." - I assume that means 10/12 was too early for the added PAFF support to make it - I'm not sure which of the codecs I should update to enable it (nor where to get the new codec).

A 60 MB test can be found here, if the problem isn't know:
http://d0.phys.washington.edu/~gwatts/00024.zip

I have not yet tried CoreAVC, but will later in the week.

Many thanks in advance,

Cheers,
Gordon.

[haruhiko_yamagata]

Originally Posted by gordonwatts

Hi,
I've got a Sony SR7 which produces the .mts (or .m2ts) files. Using the latest version of ffdshow tryouts (rev 1716 by xxl) or beta4 does not work. But in a funny way.

Beta4 and the latest build by clsid plays the video (no audio). My CPU (Core2duo 2.66GHz) can't catch up though. Here is where I want multithreading, but unfortunately xxl's rev 1703/1716 doesn't work. This is the first sample I have had problem that is xxl's build specific.

CoreAVC works with audio.

// EDIT
Audio plays with MPC's built-in splitter and libavcodec-H.264 decoder.
libavcodec debug output says "Interlaced pictures + spatial direct mode is not implemented".

[thuan]

Works fine with Nero 8 h264 decoder, too on my Core 2 e4300.

[Thunderbolt8]

Originally Posted by haruhiko_yamagata

Beta4 and the latest build by clsid plays the video (no audio). My CPU (Core2duo 2.66GHz) can't catch up though. Here is where I want multithreading, but unfortunately xxl's rev 1703/1716 doesn't work.

do you mean the multithreading support for 1716 xxl is broken? im also using this version atm and when I play avc stuff and check in task manager then the load is distributed evenly on both cores for me.

[clsid]

The multi-threading path that xxl uses does not work properly with PAFF interlacing. But PAFF is quite rare in H.264 files, so that shouldn't be a big problem.

[haruhiko_yamagata]

Originally Posted by Thunderbolt8

do you mean the multithreading support for 1716 xxl is broken? im also using this version atm and when I play avc stuff and check in task manager then the load is distributed evenly on both cores for me.

libavcodec debug output says "Interlaced pictures + spatial direct mode is not implemented" for the file. So you're lucky if you get any pictures. We can't blame the multithreading patch in this case.

By the way, Haali's media splitter doesn't set any "Sync point" (= I-frame) information (IMediaSample::IsSyncPoint()) on the samples sent to ffdshow for the file.

On the other hand, the sample here have too many sync points. All the samples are labeled as sync point.

Current ffdshow's implementation depends on the sync point information that is attached to the samples. Splitters have to set this information.
Of course it is possible to know if the samples are I-frames after decoding. These two samples are suggesting that we have to reconsider current splitter dependent implementation.

Index: codecs/TvideoCodecLibavcodec.cpp
===================================================================
--- codecs/TvideoCodecLibavcodec.cpp	(revision 1719)
+++ codecs/TvideoCodecLibavcodec.cpp	(working copy)
@@ -321,7 +321,7 @@
 }
 HRESULT TvideoCodecLibavcodec::decompress(const unsigned char *src,size_t srcLen0,IMediaSample *pIn)
 {
- if (codecId==CODEC_ID_H264 || codecId==CODEC_ID_FFV1) // libavcodec can crash or loop infinitely when first frame after seeking is not keyframe
+ if (/* codecId==CODEC_ID_H264 || */ codecId==CODEC_ID_FFV1) // libavcodec can crash or loop infinitely when first frame after seeking is not keyframe
   if (!wasKey)
    if (pIn && pIn->IsSyncPoint()==S_OK)
     wasKey=true;
Index: ffmpeg/libavcodec/mpegvideo.c
===================================================================
--- ffmpeg/libavcodec/mpegvideo.c	(revision 1719)
+++ ffmpeg/libavcodec/mpegvideo.c	(working copy)
@@ -945,6 +945,8 @@
 
     if(s->pict_type != I_TYPE && (s->last_picture_ptr==NULL || s->last_picture_ptr->data[0]==NULL) && !s->dropable){
         av_log(avctx, AV_LOG_ERROR, "warning: first frame is no keyframe\n");
+        if (s->codec_id == CODEC_ID_H264 || s->codec_id == CODEC_ID_FFV1)
+            return -1;
         assert(s->pict_type != B_TYPE); //these should have been dropped if we don't have a reference
         goto alloc;
     }

This patch fixes both of the two files above. Is this safe to add?

[gordonwatts]

Originally Posted by haruhiko_yamagata

Beta4 and the latest build by clsid plays the video (no audio). My CPU (Core2duo 2.66GHz) can't catch up though. Here is where I want multithreading, but unfortunately xxl's rev 1703/1716 doesn't work. This is the first sample I have had problem that is xxl's build specific.

Shoot -- saw references earlier that said the multi-threaded build and PAFF encoding were not compatible. However, I am on a single core machine for now (traveling) and beta4, xxl, and cslid all behave the same with me.

BTW -- I do get audio, but not in MPC. In WMP11 the audio plays fine if I disable the H264 decoding in the ffdshow video codec config page.

Audio plays with MPC's built-in splitter and libavcodec-H.264 decoder.
libavcodec debug output says "Interlaced pictures + spatial direct mode is not implemented".

How do you use the internal splitter? I tried removing the Haali splitter and MPC refused to play claiming I might be missing some codecs...

Thanks a lot for looking at this! And let me know when I can take down the sample.

[gordonwatts]

Originally Posted by haruhiko_yamagata

libavcodec debug output says "Interlaced pictures + spatial direct mode is not implemented" for the file. So you're lucky if you get any pictures. We can't blame the multithreading patch in this case.

Right -- that is what I was afraid of. I take it that means it is not likely that libavcodec will be able to correctly render video from this camera for some time yet.

BTW -- how do you look at the debug output from libavcodec? Is that a special build, or is that with mplayer?

[clsid]

To enable an internal filter of MPC, go to: Options -> Internal Filters.

[gordonwatts]

Originally Posted by clsid

To enable an internal filter of MPC, go to: Options -> Internal Filters.

Thanks, now that makes sense - now I can repro the WMP11 behavior in MPC now by turning off the TS splitter. This particular difference between the WMP and MPC w/internal splitter is due to the bug (and workaround) that haruhiko_yamagata posted above, correct?

Why doesn't MPC internal splitter pick up the audio?

Another funny thing I noticed was GSpot fails to display any codec for video or audio when I load this file, but it will generate a render chain (which is the chain I assume is being used by WMP11). I've never seen GSpot both not display a video/audio codec and still generate a render chain.

Thanks!

[clsid]

GSpot is only able to analyze the contents of a few container formats (AVI/MPEG/MP4/FLV). The rendering functionality is completely separated from that. GSpot constructs a graph like any other DirectShow application would do.

[haruhiko_yamagata]

Originally Posted by gordonwatts

Why doesn't MPC internal splitter pick up the audio.

The audio is OK with MPC internal splitter. With Haali's media splitter, because video doesn't start playing, the filter graph is "paused" forever (the filter graph is paused until first video frame is delivered to renderer), the audio does not start.

[Blight]

clsid:
I believe it's AAC audio, yes.

[Blight]

I have a request if you're up for it,
Can you instance the ffdshow Subtitle support into a separate instance (CLSID) so that it can be used without any other post-processing effects that may be enabled in ffdshow?

This is similar to the RAW mode filter instance, just for subtitles...

The reason behind this request is to give users an additional subtitle rendering choice that doesn't require them to decode the video or conflict if they already use the RAW mode for other processing.

[haruhiko_yamagata]

Originally Posted by Blight

I have a request if you're up for it,
Can you instance the ffdshow Subtitle support into a separate instance (CLSID) so that it can be used without any other post-processing effects that may be enabled in ffdshow?

This is similar to the RAW mode filter instance, just for subtitles...

The reason behind this request is to give users an additional subtitle rendering choice that doesn't require them to decode the video or conflict if they already use the RAW mode for other processing.

I planed to do this after I finished the subtitle filter. It's a long way.
But yes, if you are requesting this, why not .
I'm going to do this in the near future (but not soon). Please wait.

[DeepBeepMeep]

Originally Posted by haruhiko_yamagata

I planed to do this after I finished the subtitle filter. It's a long way.
But yes, if you are requesting this, why not .
I'm going to do this in the near future (but not soon). Please wait.

That's a very neat idea! I second Blight's request.

If I am not mistaken, this would probably mean that we could get in Zoomplayer DXVA decoding AND subtitles at the same time!!! Being able to take advantage of hardware decoding in zoomplayer would be a very nice christmas/newyear's gift.

Thanks in advance for your help haruhiko_yamagata.

[sheppaul]

Originally Posted by DeepBeepMeep

If I am not mistaken, this would probably mean that we could get in Zoomplayer DXVA decoding AND subtitles at the same time!!! Being able to take advantage of hardware decoding in zoomplayer would be a very nice christmas/newyear's gift.

Probably, it's impossible to get DxVA support with that subtitle renderer too. It's another implementation using the subtitle parser of ffdshow just like vsfilter. There should be a native support in viewpoint of player to get DxVA support with a subtitle displaying. If it is possible, it is a better idea to improve vsfilter than to reinvent wheel.

[clsid]

No, that is incorrect. Using a "standalone" subtitle filter will work the same as using ffdshow for processing raw video, just with less options. It is still an additional filter in the graph between decoder and renderer. That means DXVA is not possible. The only way to have both DXVA and subtitles is to use DXVA API calls to overlay the subtitles onto the video (in graphics card memory). And I don't think that Haruhiko has plans for implementing that.

[sheppaul]

I just wonder if a subtitle renderer could exist in form of standalone filter while enabling DxVA support. If it was possible, vsfilter should have coded in that way.

[clsid]

VSFilter was written long before DXVA came into existance.

But this whole DXVA discussion is pointless, because seriously I doubt that ffdshow will get DXVA capabilities. The whole design of ffdshow is simply incompatible with the limitations that DXVA would impose on it. It would be much easier to build a whole new filter from the ground up, that is specifically designed to make use of DXVA.

[clsid]

Originally Posted by Blight

I just came upon a weird issue,
It seems ffdshow is not accepting the "{6134706D-0000-0010-8000-00AA00389B71}" sub-type "mp4a" fourcc (no matching media type error) from the Nero MP4 splitter (tested with v7 of nero).

I dont think this is nero white/black listing filters as the gabest realaudio decoder does accept a connection and does decode the audio.

Added at rev1724. Please test.

[DeepBeepMeep]

Originally Posted by clsid

VSFilter was written long before DXVA came into existance.

But this whole DXVA discussion is pointless, because seriously I doubt that ffdshow will get DXVA capabilities. The whole design of ffdshow is simply incompatible with the limitations that DXVA would impose on it. It would be much easier to build a whole new filter from the ground up, that is specifically designed to make use of DXVA.

The idea here is not to implement DXVA in ffdshow, it is to be able to use existing hardware decoders (e.g: Cyberlink AVC decoder)with the subtitle engine (vsfilter?) of ffdshow.

To do that the subtitles should not be part of a software postprocessing filter that comes after the decoding but rather generated in "parallel" and muxed at the end using VMR9. This is done in mplayer classic and seems to work quite well.

Whenever a filter is located after a hardware decoder, the latter falls back to software mode, this obviously doesn't happen when the subtitles is merged at the very end at the graphic card level.

[SBeaver]

I would just like to add some helpful information regarding ac3/dts passthrough in Vista x64.
I have been bugging people around forums about this issue but had somehow missed the solution which is available here:
http://support.microsoft.com/kb/936225

This hotfix makes the old way of spdif passthrough from XP and older work under Vista x64 too, so now everyone can have spdif passthrough from 32bit media players (and ffdshow 32bit, naturally).
Hotfix is available from places like thehotfixshare and the likes, and Microsoft too if you can bother calling them up.

[leeperry]

Hi there,

I was wondering if there was some dualcore build >1604 available ? by XXL maybe ?

thanks,

[Thunderbolt8]

http://sourceforge.net/projects/ffdshow-tryout/

afaik all xxl builds from then supported multithreading

[fastplayer]

Originally Posted by Thunderbolt8

http://sourceforge.net/projects/ffdshow-tryout/

afaik all xxl builds from then supported multithreading

Only builds that were posted in this thread by XXL had the patch applied. Search in his post history for the latest MT build.

Edit: I stand corrected

[Thunderbolt8]

atm im xxl's 1716 built, which was not mentioned here. but when I open AVC stuff and check via task manager I can see the load being distributed evenly on both cores.

[leeperry]

ok thanks! the latest from this link seems to be dual-core enabled :
http://sourceforge.net/project/showf...kage_id=214245

BTW, I got a more annoying problem

I run KMPlayer with YV12 output from ffdshow, and YUY2 output to "EVR custom" on XP SP2 with an ATI card.

I use this version :
http://mapage.noos.fr/cguyome/EVR/EVR_For_XP.zip

when I don't have ffdshow in the directshow filter chain, everything's fine.....but as soon as I add it to decode or to resize, it will randomly freeze the image but the sound is still running

a friend of mine has exactly the same problem with MPC in RGB32 with a nvidia card.......so we have managed to narrow down the problem to ffdshow

it was working fine for the both of us with Haali's Renderer.....maybe some of you programmers can give me a debug version so I can give a log of the problem ?

Thank you for your help!

[_xxl]

"EVR custom" on XP SP2

with nvidia 7900 and latest mpc-hc doesn't work for me.
http://rapidshare.com/files/79392113/mplayerc.rar.html

[leeperry]

it works perfectly in KMPlayer, and the 3D depth of the image is a lot better than Haali's Renderer.

my friend is using it with "MPC casimir"

http://tibrium.neuf.fr/

Support for EVR (Enhanced Video Renderer) :
EVR is a new video renderer available with Windows Vista. It could be used on Windows XP if .Net Framework 3.0 is installed. MPC support two type of EVR :
EVR : video is displayer by Microsoft presenter. Internal subtitles, Pixels Shaders, and D3D fullscreen are not supported with this version
EVR custom presenter : the rendering part is made by MPC, almost the same way it's done with VMR9 renderless. Internal subtitles, Pixels Shaders and D3D fullscreen options are working in this mode.

any chance that you give me a debug version of ffdshow so I can pin-point the freezing problem ?

Thanks,

[clsid]

MPC casimir = MPC Home Cinema = mpc-hc

The "official" method to get EVR on XP is to install .NET 3.0.

[leeperry]

well yeah, or you can use my package

http://mapage.noos.fr/cguyome/EVR/EVR_For_XP.zip

anyhow, can we try to fix the freezing problem please ?

the problem is very easy to reproduce with x264 720p mkv and coreavc.

sometimes 5 seconds after skipping, sometimes 20 mins......damn annoying

[Kado]

Originally Posted by _xxl

with nvidia 7900 and latest mpc-hc doesn't work for me.
http://rapidshare.com/files/79392113/mplayerc.rar.html

Do you have a change log for that build?
If I recall correctly casimir was trying to enable DXVA for EVR custom as well.

[EDIT]I have just tested and EVR custom presenter now supports DXVA! Tried with a ts file with a mpeg2 stream and Cyberlink Video/SP decoder.[/EDIT]

[fastplayer]

Originally Posted by Kado

Do you have a change log for that build?

http://mpc-hc.svn.sourceforge.net/vi...runk/?view=log

[iron2000]

Does ffdshow have problems with AAC?
I played a movie with AAC sound with ffdshow and it sounds robotic but it sounds ok with MPC's built-in filter.

It seems to only play stereo AAC and has problem with 6 channel AAC.
Maybe its this version as I don't remember problems with AAC with the previous ffdshow version.

For the EVR for XP leeperry posted, it has a few property pages.
I tried to set the video processor but it doesn't remember the setting and the options are weird.
Anyway is EVR something like the next generation VMR and how does it compare with Haali's Renderer?

[leeperry]

iron : you have to use the regular "EVR", then when your options are set, use "EVR C/A".

yes it looks much better than any other renderer for me and many other ppl.

yet ffdshow still freezes randomly with coreavc and that's really ANNOYING

I'm going to try an old version of ffdshow from march 2007 to see if it still freezes......then I'll try from june 2006....maybe that will help

[iron2000]

Select a VideoProcessor, close MPC, play video, MPC will freeze.
I think I'll stick to Haali's until theres some documentation on the stuff you can tweak in this EVR.

Now I'm more concerned about the AAC problem.

[leeperry]

I don't use MPC, so I can't really help.

but MPC HC is supposed to work fine.

KMP does.......except for the damn ffdshow random freezes...

[_xxl]

KMP does.......except for the damn ffdshow random freezes...

Isn't KMP build with most ffdshow filters?

[leeperry]

I have disabled them all

here's my KMP filtergraph for 720p mkv :

[Primary Filtergraph]
0) - KMP Matroska Reader
1) - CoreAVC Video Decoder
2) - KMP AVI<->AC3/DTS Transform
3) - ffdshow raw video filter (to do lanczos resize)
4) - Sonic Cinemaster® Audio Decoder 4.2
5) - KMP Video Transform
6) - Enhanced Video Renderer(C/A)
7) - DirectSound: SB X-Fi Audio [C080]

If I take off the "ffdshow raw video filter", no more freezes..

and same problem for a friend of mine using MPC HC+EVR+COREAVC..

there's infos about EVR here:

http://msdn2.microsoft.com/en-us/library/ms703040.aspx

http://msdn2.microsoft.com/en-us/lib...07(VS.85).aspx

[DeepBeepMeep]

Originally Posted by leeperry

I have disabled them all

here's my KMP filtergraph for 720p mkv :

[Primary Filtergraph]
0) - KMP Matroska Reader
1) - CoreAVC Video Decoder
2) - KMP AVI<->AC3/DTS Transform
3) - ffdshow raw video filter (to do lanczos resize)
4) - Sonic Cinemaster® Audio Decoder 4.2
5) - KMP Video Transform
6) - Enhanced Video Renderer(C/A)
7) - DirectSound: SB X-Fi Audio [C080]

If I take off the "ffdshow raw video filter", no more freezes..

and same problem for a friend of mine using MPC HC+EVR+COREAVC..

there's infos about EVR here:

http://msdn2.microsoft.com/en-us/library/ms703040.aspx

http://msdn2.microsoft.com/en-us/lib...07(VS.85).aspx

How do you insert ffdshow raw filter inside a KMP graph? There are options to change the video decoders but I haven't found a way to add a postprocessor such as ffdshow. Thanks for your help.

[DeepBeepMeep]

Originally Posted by leeperry

I have disabled them all

here's my KMP filtergraph for 720p mkv :

[Primary Filtergraph]

If I take off the "ffdshow raw video filter", no more freezes..

If you don't use the KMP prostprocessor filter. You can remove it from the chain and gain a big speed boost. To do that you have to to turn on the "super speed mode".

[leeperry]

well I don't use the ffdshow options of KMP, but I do use the "basic" sharpen filter

to force ffdshow raw filter go to /decoder usage/custom filter manager/ then add ffdshow and tick "use forcibly"

that really bugs me to go back to Haali's Renderer with its crappy BT601/709 management and its not so great 3D depth

[iron2000]

I solved my AAC problem.
It seems like the problem occurs on checking the "enable multichannel processing" option in the Winamp2 page.

[yonta]

Starting from around build 1704(Build 1703 doesn't), turning OSD on when playing mpeg1/2 video files crashes the player.
I mean any mpeg files from vcd to 1080i/p.

edit: just updated to clsid's icl10 bulid 1732, and osd works again.

[leeperry]

could my freezing problem be related to the "Queue output samples" being disabled ?

I'm gonna turn nuts anytime soon with these random freezes, and that fantastic PQ thanks to EVR

[ACrowley]

Just want to say ive no Problmes anymore with 1080p H(x)264and ffdshowVideoDecoder.

I had Problems with choppy playback and constant Audio Delay.

I could solved ALL my Problems with my new ATI HD 3870 Card!

- ffdshow 1080p H264 plays now abolutly smooth
-VC1 from HDDVD plays now smooth . I wasnt able to play it before

All these Problmes was caused by my old AGP x800xt Card

[Shakey_Jake33]

It's bizzare that upgrading the video card would help that tbh, since ffdshow doesn't use GPU acceleration. The renderer possibly uses it for resize/scaling though.

[haruhiko_yamagata]

Originally Posted by leeperry

could my freezing problem be related to the "Queue output samples" being disabled ?

Thank you. I can reproduce using MPC Homecinema.
It's a bug in IFilterGraph::FindFilterByName (DirectShow or the OS).
I gave up using that API. I'm implementing alternative method.

[leeperry]

that's good news, arigato haruhiko

what's an ETA on having a fixed ffdshow version ?

I will also ask the KMP coder if he can update his internal ffdshow routines with yours...will you make the source code available ?

just FYI, it freezes exactly the same way if I enable the "Queue output samples" option.

[Thunderbolt8]

can someone please explain me whats with version number 1733:

"H.264: disable multithreading for spatial direct mode plus interlacing.
fix rev 1726."

so is multithreading for some h264 modes disabled now? whats the reason for that, some display errors? and then 1734 fixes this again ? :S (im puzzled)

[_xxl]

Interlaced pictures + spatial direct mode is not implemented in ffmpeg so it doesn't work.

[ACrowley]

Originally Posted by Shakey_Jake33

It's bizzare that upgrading the video card would help that tbh, since ffdshow doesn't use GPU acceleration. The renderer possibly uses it for resize/scaling though.

yes...however, other Decoders like cyberlink, are now decoding much more faster.
VC1 decoding with WMVideodecoder is 100% smooth with low CPU Usage

I wasnt able to play VC1 from BR/HD with my old AGP x800 card

[Px]

Originally Posted by Shakey_Jake33

It's bizzare that upgrading the video card would help that tbh, since ffdshow doesn't use GPU acceleration.

Near year ago I made some tests with GPU Acceleration, software competitors was CoreAVC and ffdshow. CPU usage on new videocards (7600GT/7900GT/X1300/X1650/X1900) was near the same, but on 6600 it was higher by 10-15% (not only with ffdshow, bul also with PowerDVD and CoreAVC).....

[leeperry]

Originally Posted by ACrowley

yes...however, other Decoders like cyberlink, are now decoding much more faster.
VC1 decoding with WMVideodecoder is 100% smooth with low CPU Usage

I wasnt able to play VC1 from BR/HD with my old AGP x800 card

yeah same here.

PCI-E is a lot smoother than AGP for HD

[Rectal Prolapse]

Idea:
A whitelist/blacklist for DirectShow filters that ffdshow is allowed to connect with in a graph.

I think this may be particularly useful when raw video is enabled in ffdshow.

Any opinions?

clsid, I think this is a great idea! I hacked an older version of ffdshow nearly a year ago to prevent RAW ffdshow from loading when using CoreAVC. That way, I could play DVDs with ffdshow's upscaling and AVISynth filters, yet also play h.264/AVC 1080p movies without the overhead of ffdshow. Profiles are useless to me, because if ffdshow loads, it still has the processor overhead.

[ACrowley]

Originally Posted by leeperry

yeah same here.

PCI-E is a lot smoother than AGP for HD

Yeah.. it think its the AGP Bus.

Mus be around 15-20% more Perfomance with all Decoders on HD 1080p (Software Mode) with my HD3870 PCI-E compared with the old X800XT AGP

[Thunderbolt8]

@haruhiko_yamagata

I tested to remux galapagos (1080i, different fps rate) to .mkv again after the changes you made, but I still dont get a picture with libavcodec, only with wmv9. is this supposed to be correct atm?

[fano]

I'm trying to hear a x264 video file with dts audio in my old ac3 only receiver so I'm setting in codecs configuration of ffdshow audio for dts: libdts, mixer is on Dolby Prologic II (other settings doesn't give 2.0-->5.1 conversion) and in output Ac3 with 640 Kbs of bitrate... but I don't hear nothing form my speakers and on the receiver display I see only Dolby Prologic (that is its default when nothing is received or its PCM stereo...)

If my file is mp3 or aac instead it works the audio is converted in Ac3 5.1 and the receiver display show : Dolby Digital & Dolby prologic...

My ffdshow version is: ffdshow-tryout rev. 1530 of 10/14/2007
is this a bug in this version and I would have to upgrade?

Thank for your help,
fano

[Shakey_Jake33]

You shouldn't be setting the Mixer to Pro Logic II, you should be setting it to 'same as input', and setting the output to AC3 (S/PDIF Encode Mode).

Basically what happens is ffdshow decompresses the DTS file to 5.1 PCM. By setting to 'same as input' and 'AC3', this 5.1 PCM input gets recompressed as AC3 and sent to your receiver as Dolby Digital.
Your current setup of setting Pro Logic II means the 5.1 PCM is getting converted to Pro Logic II, which is then being encoded as AC3. Pro Logic II is not 'true' 5.1 in that it's basically a 2.0 channel signal with the additional channels 'mapped' and flagged into the sound, so you will still be getting 5.1, but of inferior quality.

I'm not sure of if this is actually going to solve your problem, but it's something I felt I should point out.

btw you should really update to the newest ffdshow tryouts, there's no reason not to!

[fano]

I've tried it but the dts sound doesn't work...
I'm always in doubt to use Dolby Prologic II in the mixer setting
but if I don't use this I don't lose something with 2 channel audio?
(I've tried now and it seems that I have the same DD + Dolby Prologic light on... my rear speakers sound I suppose they are interpolated by my receiver not by FFdshow).

I've tried to use two preset condition:

Stereo Expanded: if the number of channels is 2 and the
Mixer does Dolby Prologic II
Multichannel: if the number of channels is 5;6;7 the mixer is set "same as input" (maybe I can uncheck it too? It is not the same as "

I've checked Automatic Preset loading but it always loads the selected item... it is not automatic

I've done further testing and maybe the problem is X264+dts a dts-wav (audio only) file and the dts track on a DVD are converted in Ac3 5.1...

Thanks,
fano

[haruhiko_yamagata]

Originally Posted by Thunderbolt8

@haruhiko_yamagata

I tested to remux galapagos (1080i, different fps rate) to .mkv again after the changes you made, but I still dont get a picture with libavcodec, only with wmv9. is this supposed to be correct atm?

Yes, please use wmv9 in ffdshow.

[haruhiko_yamagata]

Originally Posted by fano

I'm trying to hear a x264 video file with dts audio in my old ac3 only receiver so I'm setting in codecs configuration of ffdshow audio for dts: libdts, mixer is on Dolby Prologic II (other settings doesn't give 2.0-->5.1 conversion) and in output Ac3 with 640 Kbs of bitrate... but I don't hear nothing form my speakers and on the receiver display I see only Dolby Prologic (that is its default when nothing is received or its PCM stereo...)

Goto Codecs page, select DTS and uncheck "Use SPDIF when AC3 output set".

[Blight]

clsid:
the ffdshow audio decoder still doesn't connect to the nero mp4 splitter on, tested clsid rev 1732 (subtypes "{6134706D-0000-0010-8000-00AA00389B71}" and "{000000FF-0000-0010-8000-00AA00389B71}").


haruhiko_yamagata:
It would be helpful if you could split whatever currently exists for the ffdshow subtitle filter into a separate filter configuration as it would allow me to start integrating support for it sooner rather than later.

With regards to DXVA. I believe that if you output the right format with the subtitle filter, it can be merged into the image with DXVA enabled using VMR9 or EVR. I tested this to work on vista (DXVA2) and EVR/VMR9 while playing DVD subs.

All Developers:
Is there some sort of API specs I can find on how to programmatically control ffdshow? I'm using Delphi, but a C header file to a COM object would be fine.

[clsid]

Originally Posted by Blight

clsid:
the ffdshow audio decoder still doesn't connect to the nero mp4 splitter on, tested clsid rev 1732 (subtypes "{6134706D-0000-0010-8000-00AA00389B71}" and "{000000FF-0000-0010-8000-00AA00389B71}").

Could you upload that splitter and PM it to me?

[haruhiko_yamagata]

Originally Posted by Blight

haruhiko_yamagata:
It would be helpful if you could split whatever currently exists for the ffdshow subtitle filter into a separate filter configuration as it would allow me to start integrating support for it sooner rather than later.

I'm working on it. Do you prefer "whatever currently exists" rather than subtitles only? In that case, what is the difference between raw and subtitle filter?
Do you need anything as codec private?
Are format changes such as letter boxing or resizing allowed?

With regards to DXVA. I believe that if you output the right format with the subtitle filter, it can be merged into the image with DXVA enabled using VMR9 or EVR. I tested this to work on vista (DXVA2) and EVR/VMR9 while playing DVD subs.

It's a good news. Is ffdshow going to receive black image that has the same size as the video and connected to a sub-picture input pin of the renderer?

All Developers:
Is there some sort of API specs I can find on how to programmatically control ffdshow? I'm using Delphi, but a C header file to a COM object would be fine.

Sorry, there is no documentation. Please ask anything.
Step.1 Read IffdshowBase.h, IffdshowDec.h, IffdshowDecVideo.h and ffdshow_constants.h
Step.2 Query interface on IBaseFilter.
putParam is the most basic function to control ffdshow. Please note that preset settings in ffdshow is initialized during connection. So you would have to connect both of ffdshow's input and output pins before using these API.

Here's sample code. From TtrayIcon.cpp line 507.

   char_t subflnm[MAX_PATH];
   GetMenuString(hm,cmd,subflnm,MAX_PATH,MF_BYCOMMAND);
   deci->putParamStr(IDFF_subTempFilename,subflnm);
   deci->putParam(IDFF_isSubtitles,1);
   deci->putParam(IDFF_subShowEmbedded,0);

ffdshow writes to registry everything that are modified through API. I can add API to stop writing the changes, but it may not be the best idea. It will drop any changes that users make on purpose. My idea is to create a new preset for subtitle-raw-filter.

[georgevalkov]

I'm sure the GPU can deal with cropping, resizing and other tasks pretty fast (that's what it is made for) ;-). The GPU can also do more than one tasks in parallel. I wonder how far can one go, if he decides to use the GPU for running computational tasks? For instance calculate 2+2=4 or 4|1=5 using the GPU.

I don't know much about DXVA, but perhaps it is possible to implement custom video format encoding or decoding running on the GPU.

Video filtering would probably be possible too: like deinterlacing, Gamma, Brightness, Contrast, denoising, video format conversion: YUY2 - UYVY - YV12 - RGB...

Just think of how many effects are used in Gaming. There is no need to output the image to screen: e. g. an application can pass data and instructions to the GPU, it will process the data and return the result to the application.

[clsid]

Maybe if you are willing to donate a couple million dollars to the ffdshow project, then we could hire a team of professional programmers to rewrite the whole ffdshow codebase.

This is an open source project ran by volunteers, not a professional game studio.

Those who wish to discuss DXVA stuff, please start another topic. This is to wrong place for that.

[Blight]

clsid:
The filter comes with NERO 7.8.5, I'm aware it's not the most recent version, I wouldn't even have used this filter if not for the haali splitter not recognizing the audio format.


haruhiko_yamagata:
I didn't mean that you should split out another RAW filter... You said you felt the subtitle code isn't finished yet, so I was just asking you to expose the subtitle filter, finished or otherwise so I can start integration.

With DXVA you wont be getting any image, just the subtitle data and then you expose an image. You can easily check it out with the Microsoft MPEG2 decoder that comes with Vista (I believe its a licensed powerdvd decoder) and a DVD Graph. The video decoder has a subtitle output which can be connected to the renderer's secondary pins. You can check out the exact format it outputs by creating such a graph.

With regards to the programmable access to ffdshow, I was thinking of doing color control in ffdshow if the hardware doesn't support it (or if the user chooses it).

Altough for subtitles, I need to find a way to manually specify a subtitle file instead of it working automatically as several Zoom Player features require it and a way to set the subtitle vertical position override.

Also, is there an ffdshow wiki in existance or planned? I'm pretty involved in the media field and even I dont understand what some of the settings do... Better documentation is required and a wiki is a good place to start.

[clsid]

I have no plans to install Nero 7 just to get a filter. So unless someone can send me the filter, it will be low priority for me.

A wiki is planned. We just need someone who is willing to set it up.

[leeperry]

OMG haruhiko_yamagata has fixed the freezing problem with EVR on XP already

http://ffdshow-tryout.svn.sourceforg...&revision=1736

but where can I get a SSE2 bin version for that please ?

and is it only fixed with MPC or with all players including KMPlayer ?

thanks again!!

[foxyshadis]

Forget SSE2/P4 builds or anything else like that. Generic builds will be fastest or within a fraction of a percentage point of the performance of bothering to make 10 different builds for every possible instruction set. ffdshow builds have been generic-only for a while now.

[leeperry]

ok cool, I hope clsid will make a night build soon, and that the fix is not MPC only

[haruhiko_yamagata]

Originally Posted by leeperry

and is it only fixed with MPC or with all players including KMPlayer ?

thanks again!!

The code is MPC specific. It won't fix the issue with KMPlayer (not tested yet).

[leeperry]

ahhhhhhhhhhh well I'll rename KMPlayer to mplayerc.exe

it's the same exact problem as with MPC.........random freezes when using ffdshow + EVR C/A

why not fixing the bug for good ?!

[clsid]

Another sign that KMPlayer is really violating GPL. I know the author denies that, but that is what liars do. Rewriting C++ code to Delphi is like translating a book into another language.

[haruhiko_yamagata]

Originally Posted by Blight

haruhiko_yamagata:
I didn't mean that you should split out another RAW filter... You said you felt the subtitle code isn't finished yet, so I was just asking you to expose the subtitle filter, finished or otherwise so I can start integration.

OK, it's easy to implement.

With DXVA you wont be getting any image, just the subtitle data and then you expose an image. You can easily check it out with the Microsoft MPEG2 decoder that comes with Vista (I believe its a licensed powerdvd decoder) and a DVD Graph. The video decoder has a subtitle output which can be connected to the renderer's secondary pins. You can check out the exact format it outputs by creating such a graph.

It would not be a easy work. Current ffdshow doesn't accept subtitle as a primary input.
If you connect to ffdshow with raw video (NV12 or anything) and send empty image, current ffdshow would work.

Altough for subtitles, I need to find a way to manually specify a subtitle file instead of it working automatically as several Zoom Player features require it

The sample code I posted above would do it.

and a way to set the subtitle vertical position override.

It's "Vertical position:" in the dialog. So please search "Vertical position:" in ffdshow.rc. You'll find "IDC_TBR_SUB_POSY" and "IDC_LBL_SUB_POSY". Next, search "IDC_TBR_SUB_POSY" in ffdshow source code. You'll fine "IDFF_subPosY". Then the code will be

    IffdshowBase *deci;
    int posy;

    // prepare deci and posy

    deci->putParam(IDFF_subPosY, posy);

Please note that these interface are designed to be used in ffdshow. But it should be possible to use them from applications.
They are subject to change. But most important ones such as putParam won't change.

[leeperry]

Originally Posted by clsid

Another sign that KMPlayer is really violating GPL. I know the author denies that, but that is what liars do. Rewriting C++ code to Delphi is like translating a book into another language.

well, is it my fault

sue the sucker or sumthing

I'm just an end-user..........that would very much like ffdshow not to freeze randomly when used together with EVR

and EVR is also used on Vista, so you want ffdshow to freeze on Vista ?

making the fix only work with mplayerc.exe is not very kind of you

I hope renaming KMPlayer to "mplayerc.exe" will do the trick

any plan for a nightly rls of 1736 please ?

[haruhiko_yamagata]

Originally Posted by haruhiko_yamagata

My idea is to create a new preset for subtitle-raw-filter.

This may not be the best idea. If new preset is required, IffdshowDec::

STDMETHOD (setActivePreset)(const tchar *name,int create) PURE;

is available.

[Blight]

clsid:
I understand... it was a rare issue to begin with.

haruhiko_yamagata:
I'll be keeping an eye on the change logs to see when the subtitle filter is split off so I can start integration

[fano]

Originally Posted by haruhiko_yamagata

Goto Codecs page, select DTS and uncheck "Use SPDIF when AC3 output set".

Now it works

The "Use SPDIF when AC3 output set" is on the Ac3 section in my ffdshow (not DTS)... it's a little confusing apparently I have to disable Ac3 SPDIF passthought decoding to have DTS conversion...

I have used the term apparently because in reality the
SPDIF passthought continues to work... I have set as AC3 decoder SPDIF and in this way ffdshow is disabled and the audio is directly send to the receiver...

Now the spontaneous and evil question:
What is the utility of "Use SPDIF when AC3 output set"?
There is already the SPDIF "codec" for this...

I've solved the preset problems too...
I've 3 presets now:

AC3: only Spdif in codec section is set the rest is at default
used only when the codec is "spdif"
Multichannel: only if the numer of channel is 3;4;5;6;7
(I've to elencate in this way... I've tried > 2
but multichannel is not recognized...)
the audio is resampled to 48 Khz if is lower then 48
(soundcard limit)
in mixer is ported to 5.1
(I hope only if it isn't already 5.1)
and converted in Ac3 5.1
Default: all other audio (1/2 channels for exclusion) is
converted to Ac3 plus Dolby Prologic II

Thanks for your help it is really appreciated,
fano

[haruhiko_yamagata]

Originally Posted by fano

What is the utility of "Use SPDIF when AC3 output set"?

albain added the feature. The thread where the feature was discussed.

[leeperry]

ok thanks for the 1737 nightly build clsid

I've just tried it with KMPlayer(renamed to mplayerc.exe), and the random freezing problem still occurs with EVR C/A

if I disable ffdshow, it doesn't freeze....and a friend of mine is also using KMPlayer+EVR on XP w/o ffdshow w/o any freeze

[leeperry]

oh well I give up on EVR

noone cares to fix the issue with ffdshow and KMPlayer, and the "Direct3D 9 YUV Surface" renderer in KMPlayer looks just as good, without the random freezes

Hopefully when I'll upgrade to Vista in a few years, these nasty bugs will be fixed

[clsid]

Maybe you should be a little more patient

[leeperry]

yeah I guess, but random freezes turn me nuts

[fano]

Originally Posted by haruhiko_yamagata

albain added the feature. The thread where the feature was discussed.

If I understand well this options means "passthrough if (in output)
Ac3 is set" that is output directly the DTS stream?

A standard that my shi**y receiver can't understand so no sound

The strange thing is that a DTS on mpeg and dts on wav file that I tried was corretly converted in Ac3 with that option set
It seems this setting is unified... in my case instead I want
the Ac3 untoched (libac52 or spdif is set is indifferent? + the Ac3 checked)
and DTS converted... It seems I can't do it...

Anyway I've solved whit my presets and I'm happy now

fano

[SeeMoreDigital]

Just to confirm...

I've just installed "ffdshow_rev1737_20080101_clsid.exe". And both (6Ch) .AC3 and .DTS pass-thru' is working perfectly via SPDIF to my surround sound receiver.

So is "real-time" 6Ch AAC-LC/HE-to-AC3 (640Kbps) transcoding

[haruhiko_yamagata]

Originally Posted by fano

in my case instead I want
the Ac3 untoched (libac52 or spdif is set is indifferent? + the Ac3 checked)
and DTS converted... It seems I can't do it...

You can do so if you select S/PDIF for AC3, libdts for DTS and check AC3 in output page.

I'm skeptical if we need "Use SPDIF when AC3 output set" option though.

[SeeMoreDigital]

Originally Posted by haruhiko_yamagata

I'm skeptical if we need "Use SPDIF when AC3 output set" option though.

Agreed...

Selecting "S/PDIF" via the drop-down menu should be enough...

While we're on the subject of the SPDIF output. A couple of years ago Pioneer started manufacturing amplifiers with on-board WMA9pro decoders. Could the ability of being able to pass an WMA9pro data-stream via SPDIF be added to FFdshow?

I can't test this myself because I don't have one of the afore mentioned amplifiers but there must be others on the forum who have!

Anyway, it's just a thought....

[fano]

Originally Posted by haruhiko_yamagata

You can do so if you select S/PDIF for AC3, libdts for DTS and check AC3 in output page.

Yes I'm doing this and works well...

Originally Posted by haruhiko_yamagata

I'm skeptical if we need "Use SPDIF when AC3 output set" option though.

In my opinion is a little confusing, too...

fano

[vlada]

Originally Posted by clsid

The multi-threading path that xxl uses does not work properly with PAFF interlacing. But PAFF is quite rare in H.264 files, so that shouldn't be a big problem.

I don't agree. PAFF is very common in TV broadcasting and also all AVCHD camcorders I've seen use it.

[clsid]

Haruhiko has adapted the patch to automatically disable multithreading when PAFF interlaced content is encountered.

[tjmitchem]

Originally Posted by vlada

PAFF is very common in TV broadcasting

As in every single one of Dish Network's HD channels

[haruhiko_yamagata]

Originally Posted by Blight

haruhiko_yamagata:
I'll be keeping an eye on the change logs to see when the subtitle filter is split off so I can start integration

I have committed as rev 1741. It's simply a duplication of raw-filter. Only one difference is it always accept all the raw video formats.
Please tell me what you need.

One thing I noticed during my tests: If ffdshow is inputing subtitle from a input pin, subtitle file reader is disabled. This is OK, but if there are more than two ffdshow in the graph, one ffdshow accepts DirectShow input pin and other instances of ffdshow read from files. You would have to check if you have multiple instances of ffdshow and if yes, you would have to disable subtitle filter for other ffdshow.

[Blight]

haruhiko_yamagata:
Where is the download link for this build?

I need it to behave as closely as possible to DirectVobSub as i'm going to give the users the option to choose between the two. And by behave closely, I mean the filter connection aspect of the filter.

It would also be good if no filter setting other than the subtitle settings (and any connection-related settings) are exposed in the subtitle filter property page.

An exact duplicate of the RAW filter would cause confusion with users who only want to configure the subtitle portion.

[haruhiko_yamagata]

Originally Posted by Blight

I need it to behave as closely as possible to DirectVobSub as i'm going to give the users the option to choose between the two. And by behave closely, I mean the filter connection aspect of the filter.

I'm afraid it's too much work. ffdshow is not designed that way. It's entire new design, new filter. ffdshow would have to respond to any changes of the video on the primary input pin of the renderer.
ffdshow require raw video as input. Please send ffdshow black images, if ffdshow is connected to sub-picture input pin of the renderer. Of course images with pictures are welcome if ffdshow is connected to the primary input pin of the renderer.
Preferred input color spaces of subtitle filter are YV12 and RGB32 (any other color spaces will be accepted and converted internally). ffdshow will render in better quality if the output color space is RGB32.

It would also be good if no filter setting other than the subtitle settings (and any connection-related settings) are exposed in the subtitle filter property page.

OK, it's easy.

[clsid]

Would it not be better to create a whole new standalone filter (like VSFilter) instead adding lots of 'workarounds' to ffdshow?

[_xxl]

Would it not be better to create a whole new standalone filter (like VSFilter) instead adding lots of 'workarounds' to ffdshow?

Wouldn't be easy to fix and improve VSFilter?

[Px]

Originally Posted by _xxl

Wouldn't be easy to fix and improve VSFilter?

As its already done with ffdshow and mpc

[clsid]

That would even be better.

[Delerue]

I want to point that VSFilter (aka DirectVobSub) is coming in a DLL file. But if you rename it to an OCX file and register ('regsvr32 file.ocx'), you'll get a great performance increase. I discovered this accidentally when I was comparing the performance of MPC vs VLC.

[clsid]

Your performance measuring methods must be flawed.

[Delerue]

Originally Posted by clsid

Your performance measuring methods must be flawed.

I doubt it. Everything began when I discovered that DirectVobSub uses more CPU than the MPC internal subtitle, because both share the same subtitle engine. So I downloaded the old VobSub (I stop using it because of compatibility problems), and to my surprise it did a better job, I mean, used less CPU. But, hey, VobSub also uses the same basic source of the DirectVobSub. The only difference between them is the file type: Ocx (VobSub) vs Dll (DirectVobSub). So I renamed the DirectVobSub Dll to an Ocx file and registered it. Voilà! The performance increased, getting equal to the VobSub.

I tested with a x264 1080p video before multi-thread patch was implemented in FFDShow. In my C2D @ 3 GHz there's a scene where the CPU usage freezes in 50% (so full load). After my 'hack', the CPU usage fell almost 5%. If you doubt, try for yourself.

[clsid]

I have tested it and as expected it makes absolutely no difference at all what the file extension is.

[check]

Originally Posted by clsid

I have tested it and as expected it makes absolutely no difference at all what the file extension is

I can also confirm that sanity prevails and the extension makes no difference.

[Delerue]

Hmmm... I tried again and, well, you're right. I can see my mistake. First, when I said 'ocx' I actually mean 'ax', of course. Second, in fact there's a difference in performance betweeen VobSub and DirectVobSub (as you can see below), but they use different extensions, and that's probably why I made the mistake about the performance vs extension.

My Windows is in portuguese, so "último" means 'last', "médio" means 'average', "mínimo" means 'mininum' and "máximo" means 'maximum/peak'.

Wtihout subtitle:


With VobSub 2.23:


With DirectVobSub 2.7:


I tested for only 1:19 seconds, but is enough to see a 3% between VobSub and DirectVobSub in the average, and almost 10% in the maximum/peak.

Sorry for my mistake, and keep going.

[leeperry]

guys, does fffdshow work with EVR on Vista ?
or does it freeze randomly like it does on XP ?

I might just install Vista if that can make ffdshow work..because EVR looks a lot better than Haali's Renderer.

Thanks,

[clsid]

Install .NET 3.5 and you have EVR on XP.

[leeperry]

hehehe, I do have EVR on XP but it's randomly freezing with ffdshow.

hence my question "does ffdshow work fine with EVR on Vista ? without freezing that is"

[_xxl]

Don't post 100 times the same issue with KMplayer + EVR on XP. Maybe I should blacklist KMplayer.

[leeperry]

I'm asking a simple question, whether ffdshow and EVR don't freeze on Vista.

a friend of mine running MPC HC + EVR + ffdshow on XP also encounters random freezes.......if he disables ffdshow, no more freezes.

maybe you should also blacklist mplayerc.exe all at once ?

I guess the XP EVR+ffdshow bug will never be fixed, I understand that....but what about Vista EVR + ffdshow ? does it "work" w/o freezes ?

I'm asking end users here, no need to be jumpy XXL

[sheppaul]

Or blacklist leepery

Why are you using evr on xp? It's not one of official renderers in xp: It's not guaranteed to work without hassles.

[leeperry]

yeah that's what I thought...hence I just wanna know if ffdshow works fine on Vista with EVR so I don't install Vista for nothing

[multiblitz]

Funny discussion on EVR.

BUT the real great news is that haruhiko_yamagata started the re-work of ffdshow to play back HD-DVD / Blu-Ray in 24p with audio and video in sync, see:

https://sourceforge.net/tracker/?fun...roup_id=173941

THIS IS PHANTASTIC NEWS !!!

PowerDVD is no longer able to play back from HArd-Disk; PDVD anyhow has only a reasonable picture quality on a real big screen if you use it with EVR under Vista (under XP it enables only overlay), but than you have other troubles like no re-clock anymore, no cutsom-timings with 8600GT, etcetc.

So a high-quality playback solution with post-processing through ffdshow with video and audio in sync is exactly what we need these days.

I wil start testing the latest version this evening, looking really forward to it.

[leeperry]

Originally Posted by multiblitz

PowerDVD is no longer able to play back from HArd-Disk

well

I took a look at the previous 3319f patch, and found that the change in playback behaviour (i.e. not being able to play movies from you harddrive) was caused by changes in one single file, the \Navfilter\CLBDROMNav.ax file. The solution was simply to take the "CLBDROMNav.ax" file from the 3304 Build and overwrite the new one, and voilá, playback works as before

Sadly the new 3516 patch however, makes changes in PowerDVD.exe to totally grey out the "ok" box when wanting to play from the harddrive. You can change it back to normal without reinstalling the old version, by replacing the following of the 3516 files, by these 3304 files;

Navfilter\CLBDROMNav.ax
PowerDVD.exe
*.imp (10 files)

Now you should have some of the benefits/improvements of the new patch, while maintaing your ability to play from the harddrive.

[multiblitz]

I read that and all the ISO-ideas to out-smarten PDVD, but as Blue-Ray seems to win with the announcement of Warner yesterday, BD+ and PDVD got a new dimension of importance and in the latest version you can't even playback certain BD+ titles anymore when you have them through AnyDVD on your media-server as PDVD only plys them back, if it realizes the full-blown Copy-Protection is untouched....so....WE REALLY need an alternative way to playback HD-DVD, /Blu-Ray...now more than ever. FFDSHOW would be a wonderful solution.

[leeperry]

well yeah, HD commercial discs support in anything else than PDVD would be awesome, and a good reason for me to buy a combo drive.

the next winDVD and Arcsoft softwares are supposed to offer full blown HD support.

getting support from ffdshow is awesome, but decoding 1920*1080 content without hardware acceleration will require some bad ass CPU in order to avoid tearing.......

[_xxl]

libavcodec/motion_est_template.c:1120: warning: C99 inline functions are not supported; using GNU89
libavcodec/motion_est_template.c: In function 'ff_epzs_motion_search':
libavcodec/motion_est_template.c:1120: internal compiler error: Segmentation fault
Please submit a full bug report,
with preprocessed source if appropriate.

I can't compile with cc979 MinGW GCC 4.2.1 libavcodec.
If I exit MinGW then enter and run "make" again it worked.

[Yong]

are u using gcc version that compiled by cc979?

btw ffdshow video encoder crash for me too:

will post the detail tomorrow if someone want to fix it

[Jay Bee]

I just tested VC-1 performance in newest build (rev1751) and libavcodec performance has increased a lot, it's smoother on one core than wmv9 decoder is on two cores on my machine! :up: Just one small bug: when I open and play a file (VC-1 in TS) the a/v synch is off until I seek in the file at least once, then it's fine. Only libavcodec, not wmv9.

[Delerue]

Originally Posted by Jay Bee

I just tested VC-1 performance in newest build (rev1751) and libavcodec performance has increased a lot, it's smoother on one core than wmv9 decoder is on two cores on my machine! :up: Just one small bug: when I open and play a file (VC-1 in TS) the a/v synch is off until I seek in the file at least once, then it's fine. Only libavcodec, not wmv9.

I tested here with this video. Using VC-1 by 'Libavcodec' I got an average of 8.8% of CPU usage and a peak of 13.2%, against 'WMVideo Decoder DMO' with 19.1% of average and a peak of 24.2%. Impressive!

[leeperry]

is there an easy way to reach 1.85 in the aspect ratio option of ffdshow WITH THE MOUSE ?

I can only reach 1.84 or 1.86...

maybe you could make it slower if right button is pushed or sumthing ?

I remember I asked for this feature ages ago, but that'd still be great to have 4 buttons like 2.35 / 1.85 / 1.78 / 1.33

I've installed the updated EVR.dll from Vista SP1 on XP SP2, and MPC HC seems to work just fine with ffdshow 1757......so I think I will drop KMPlayer

I'll run more tests and report!

I've uploaded the updated EVR for XP here :
http://www.badongo.com/file/7196279

[fastplayer]

Use the arrow keys.

[leeperry]

yes thanks for the reply!

I figured it out before you answered

but I use it on my HCPC, so making it slower if right button is pushed or sumthing would be really handy

[sheppaul]

Originally Posted by Delerue

I tested here with this video. Using VC-1 by 'Libavcodec' I got an average of 8.8% of CPU usage and a peak of 13.2%, against 'WMVideo Decoder DMO' with 19.1% of average and a peak of 24.2%. Impressive!

http://www.citizeninsomniac.com/WMV/

Check if post-processors are enabled by default.

[Delerue]

Originally Posted by sheppaul

http://www.citizeninsomniac.com/WMV/

Check if post-processors are enabled by default.

Great program, but all options in the decoder tab are off.

[Jay Bee]

Originally Posted by Delerue

I tested here with this video. Using VC-1 by 'Libavcodec' I got an average of 8.8% of CPU usage and a peak of 13.2%, against 'WMVideo Decoder DMO' with 19.1% of average and a peak of 24.2%. Impressive!

Playing this (rapidshare) I get total_average/total_peak/core1_peak of 38%/46%/84% with libavcodec vs 80%/89%/95% with WMVideo Decoder DMO'.

If ffdshow could do some multithreading it would absolutely smoke everything else out there. Right now Cyberlink is the best with 58%/60%/72%.

EDIT: Huh? Looks like libavcodec already is pretty good at multithreading but for some reason it doesn't kick in right away. This is CPU usage for each core playing the above file three times in a row:


What's going on here? Anyway to debug why it runs in one thread first and then later in two?

EDIT2: The only way I found to make both cores work is to load and start playing a file, press stop and press play again. Works in both ZP and MPC.

EDIT3: Updated percentages are 39%/45%/68%. So ffdshow would be the best if the a/v synch and multithreading bugs could be fixed.

[Delerue]

Originally Posted by Jay Bee

Playing this (rapidshare) I get total_average/total_peak/core1_peak of 38%/46%/84% with libavcodec vs 80%/89%/95% with WMVideo Decoder DMO'.

Your sample doesn't play here. It crashes MPC. Doesn't matter if I use Libavcodec or M$ default decoder. SMPlayer gets sound, but no image.

Originally Posted by Jay Bee

EDIT2: The only way I found to make both cores work is to load and start playing a file, press stop and press play again. Works in both ZP and MPC.

There's no multi-thread. The stop-and-play-pseudo-multi-thread is a common behaviour, but if you'll check, you'll see that CPU usage never pass the 50%.

[_xxl]

Libavcodec vc-1 is not multithreaded.

[Jay Bee]

What is perfmon showing me then? Is there a better way to check core usage?

Delerue:
Make sure you disable the MPC internal TS splitter and use Haali instead.

[Px]

Originally Posted by Delerue

Your sample doesn't play here. It crashes MPC. Doesn't matter if I use Libavcodec or M$ default decoder. SMPlayer gets sound, but no image.

Plays fine with mpc-hc, no crashes....

Originally Posted by Delerue

There's no multi-thread. The stop-and-play-pseudo-multi-thread is a common behaviour, but if you'll check, you'll see that CPU usage never pass the 50%.

Perfmon screen attached....

[_xxl]

Read more:

http://forum.doom9.org/showthread.ph...514#post924514

[Jay Bee]

I see. So Windows is somehow splitting the decoding onto both cores but it doesn't help performance? Or is perfmon just outright BS and it's really just one thread running on one core?

And the only way to check if a program is multithreading is by checking if it can make total CPU go considerably over 50%? Anyone know a better way?

[cc979]

Originally Posted by _xxl

I can't compile with cc979 MinGW GCC 4.2.1 libavcodec.
If I exit MinGW then enter and run "make" again it worked.

does it compile with the unpatched gcc-4.2.2 ?

[cc979]

Originally Posted by Yong

are u using gcc version that compiled by cc979?

btw ffdshow video encoder crash for me too:

will post the detail tomorrow if someone want to fix it

i think the encoder part was bugged when testing with gcc-4.2 pre-releases(svn) quite a while ago, not sure if anything was done about it

[Leak]

Originally Posted by Jay Bee

I see. So Windows is somehow splitting the decoding onto both cores but it doesn't help performance? Or is perfmon just outright BS and it's really just one thread running on one core?

Actually, it's one thread running on whichever core Windows gives it during scheduling (which can happen up to several hundred or thousand times a second). That's why you get load on both cores - Windows bounces the thread(s) around between your cores unless you set the process CPU affinity to one core only (using Task Manager, for instance).

If more than one core (i.e. more than one thread) is being utilized the "CPU load" column in the Task Manager will go above 50% - that's the best way to find out whether a process uses more than one core.

np: Tocotronic - Imitationen (Kapitulation)

[Jay Bee]

I see, learned something new. Anyway, even without multithreading libavcodec vc-1 is still very impressive and should be good enough for playback of HD files without dxva on most modern CPUs. The tests I did were with 2.7 GHz X2.

[Inventive Software]

Yep, but it's not complete until interlacing's added, and for that we gotta wait for ffmpeg to implement it.

[Jay Bee]

I'm not hoping for that anymore, even AVC decoders still have problems with interlacing and that's a format that has been around for ages now, since European HDTV started. Interlaced VC-1 is nonexistent in comparison.

[Kurosu]

VC-1optimized code has been around for 6 months but only got into the CVS around a month ago. From this post you can see improvements. From those values, only one function accounting for 18% of the decoding time is left to optimize in the case of this sample (YMMV).

[krisq]

Hi guys, I've search the thread and didn't found anything about my problem so here goes. I've installed the latest build by clsid (rev1751 20070105). It was a jump from pretty old version for me, which worked flawlessly BTW
After the update, Adobe After Effects started to act funny on me, it can't initiliaze correctly. It gives me an error at startup ('Bad digit in number'). I've seen such errors before in other apps and it all goes to differences between "," and "." in regional settings of OS. I'm using polish Windows XP SP2 and here default separator for numbers is "," which apparently is causing problems. When I uninstall ffdshow or change to "." in regional settings, everything is back to normal, AFX starts without problem.
I hope you can help me with this because I don't want to revert to old version of ffdshow nor change separator "," to "."

BTW I'm 100% sure it's because of ffdshow, it's started happening the same day and after uninstall everything's back to normal.

[fastplayer]

Then blacklist the Adobe app:
DirectShow control --> Don't use ffdshow in: --> <enter EXE filename>

[haruhiko_yamagata]

@krisq
Could you pin-point the revision which the problem starts? Many old builds are available for download.
I don't have Adobe After Effects nor Polish version of Windows. I'm trying to recall the changes related to locale but nothing come to me.

[krisq]

Originally Posted by fastplayer

Then blacklist the Adobe app:
DirectShow control --> Don't use ffdshow in: --> <enter EXE filename>

Yeah, that was the first thing i did
Unfortunately it didn't help, error is still there

Originally Posted by haruhiko_yamagata

@krisq
Could you pin-point the revision which the problem starts? Many old builds are available for download.
I don't have Adobe After Effects nor Polish version of Windows. I'm trying to recall the changes related to locale but nothing come to me.

I'll try.

[Delerue]

Can anyone confirm a bug with high frame rate videos? With this sample I got three hard stutters with MPC (clsid january build) and FFDShow _xxl 1757, although the CPU usage has an average of 18%. The problem seems to be with the libavcodec version, because with CoreAVC or MPC-HC internal decoder there's no problem (except for the fact that MPC seems to limit the FPS to the actual refresh rate, even with vsync off). Also, VLC and SMPlayer can run just fine.

[krisq]

Originally Posted by haruhiko_yamagata

@krisq
Could you pin-point the revision which the problem starts? Many old builds are available for download.
I don't have Adobe After Effects nor Polish version of Windows. I'm trying to recall the changes related to locale but nothing come to me.

Ok, apparently something has changed between clsid's rev1431 and rev1487. 1431 works flawlessly, 1487 throws out an error in AFX.

[Yong]

Originally Posted by cc979

i think the encoder part was bugged when testing with gcc-4.2 pre-releases(svn) quite a while ago, not sure if anything was done about it

so which ffdshow build have gcc3 or 4.0/4.1 compiled libavcodec? i would like to test it to see wether it is gcc or libavcodec bugged
i got gcc 3.4.6, 4.0.3 and 4.2.2,
but none of them work for me as they will throw an weird error message when using them

[fastplayer]

Originally Posted by Delerue

Can anyone confirm a bug with high frame rate videos?

I get no stutters at all with MPC+ffdshow. The file just plays back slow with <40% CPU load.
VLC on the other side uses 95-100% and plays the file faster than MPC with some micro-stutters. Probably because of the CPU load hitting 100% from time to time...
I'm using clsid's latest MPC and ffdshow versions.

[clsid]

Originally Posted by krisq

Ok, apparently something has changed between clsid's rev1431 and rev1487. 1431 works flawlessly, 1487 throws out an error in AFX.

Could you also test 1437?
Then I suspect that the bug was introduced in revision 1434. Since that revision there also is a weird compatibility problem with a game called "Silent Hunter 3". In that case blacklisting also doesn't help.

Does the ',' versus '.' problem apply to the decimal symbol or the the digit grouping symbol in the regional options? Or both?

[haruhiko_yamagata]

Originally Posted by Delerue

Can anyone confirm a bug with high frame rate videos? With this sample I got three hard stutters with MPC (clsid january build) and FFDShow _xxl 1757, although the CPU usage has an average of 18%. The problem seems to be with the libavcodec version, because with CoreAVC or MPC-HC internal decoder there's no problem (except for the fact that MPC seems to limit the FPS to the actual refresh rate, even with vsync off). Also, VLC and SMPlayer can run just fine.

It's smooth for me. Perhaps it's your video driver. Frame rate that is higher than refresh rate may not be supported.

[Thunderbolt8]

thanks for that black border brightness fix! that brightness was suddenly there although I was pretty sure this wasnt the case in former days and already wondered why this was the case.

[haruhiko_yamagata]

Originally Posted by clsid

Could you also test 1437?
Then I suspect that the bug was introduced in revision 1434. Since that revision there also is a weird compatibility problem with a game called "Silent Hunter 3". In that case blacklisting also doesn't help.

Does the ',' versus '.' problem apply to the decimal symbol or the the digit grouping symbol in the regional options? Or both?

It must be 1434. I have committed the fix.

@krisq
Please test and let us know when next Nightly build is uploaded.

[clsid]

A new build is online.

[haruhiko_yamagata]

Originally Posted by clsid

A new build is online.

[krisq]

Originally Posted by haruhiko_yamagata

It must be 1434. I have committed the fix.

@krisq
Please test and let us know when next Nightly build is uploaded.

Originally Posted by clsid

A new build is online.

Thank you so much, guys!!
rev1763 is working perfectly

[cc979]

Originally Posted by Yong

so which ffdshow build have gcc3 or 4.0/4.1 compiled libavcodec? i would like to test it to see wether it is gcc or libavcodec bugged
i got gcc 3.4.6, 4.0.3 and 4.2.2,
but none of them work for me as they will throw an weird error message when using them

gcc-4.0.4 is what i use mostly, which encoder seems to cause problem ?

[cc979]

ffdshow-rev1763.7z
md5: E5358DEF6DFB6D775074C9D24C2759E9
http://www.speedyshare.com/600685999.html

gcc-4.0.4 and gcc-4.2.1

[LoRd_MuldeR]

Originally Posted by Yong

so which ffdshow build have gcc3 or 4.0/4.1 compiled libavcodec? i would like to test it to see wether it is gcc or libavcodec bugged
i got gcc 3.4.6, 4.0.3 and 4.2.2,
but none of them work for me as they will throw an weird error message when using them

The "official" builds by clsid have gcc 4.2.2 compiled libavcodec.dll (according to the version info window)

[Delerue]

Originally Posted by fastplayer

I get no stutters at all with MPC+ffdshow. The file just plays back slow with <40% CPU load.
VLC on the other side uses 95-100% and plays the file faster than MPC with some micro-stutters. Probably because of the CPU load hitting 100% from time to time...
I'm using clsid's latest MPC and ffdshow versions.

Originally Posted by haruhiko_yamagata

It's smooth for me. Perhaps it's your video driver. Frame rate that is higher than refresh rate may not be supported.

Hmmm... I finally discovered. The problem appears when you set the render to 'Overlay' or 'VMR7 Windowed' (worst on that). 'VMR9 Windowed' or any 'Renderless' doesn't have any stutter.

Haruhiko, the FPS limit is gone with 'Overlay' render. And all the other decoders play fine, without stutter. Do you really think it's a problem with Catalyst 7.11?

Thanks

[Shakey_Jake33]

Slightly off topic (and I have asked this in other topics too, but I figure some people in this topic may know).
Does ffdshow get it's AC3 Encode Mode (to send 5.1 AAC over SPDIF as AC3) code from ffmpeg? I've been assuming so, and have been trying to work out how to use it in ffmpeg!/mplayer

[LigH]

AC3 encoding is one part; S/P-DIF output is another. You seem to refer to the latter?

[haruhiko_yamagata]

Originally Posted by Delerue

Hmmm... I finally discovered. The problem appears when you set the render to 'Overlay' or 'VMR7 Windowed' (worst on that). 'VMR9 Windowed' or any 'Renderless' doesn't have any stutter.

Haruhiko, the FPS limit is gone with 'Overlay' render. And all the other decoders play fine, without stutter. Do you really think it's a problem with Catalyst 7.11?

Thanks

Sorry, it's VMR7 then. CoreAVC have the same behavior. It's not a bug of ffdshow.

[haruhiko_yamagata]

Originally Posted by Shakey_Jake33

Does ffdshow get it's AC3 Encode Mode (to send 5.1 AAC over SPDIF as AC3) code from ffmpeg?

Yes, it does.

[Yong]

Originally Posted by cc979

ffdshow-rev1763.7z
md5: E5358DEF6DFB6D775074C9D24C2759E9
http://www.speedyshare.com/600685999.html

gcc-4.0.4 and gcc-4.2.1

hmm
gcc 4.0.4 crash immediately after encoding start, with mpeg2/4

gcc 4.2.1 show "assertion failed!" with mpeg1/2

i was try to trascode some mpeg4 file to mpeg1/2/4 with ffdshow for testing.

just tried clsid build, same error i got with 4.2.1, mpeg 4 will crash if disable qpel.

[clsid]

There also are some encoding bugs when compiling ffdshow.ax with Visual Studio, so it is not just GCC's fault. Of course, those could be different bugs than the ones you are having.

[Yong]

Originally Posted by clsid

There also are some encoding bugs when compiling ffdshow.ax with Visual Studio, so it is not just GCC's fault. Of course, those could be different bugs than the ones you are having.

hmm but cc979 compiled ffdshow are pure gcc build.

i post the video encoding settings some other infos here incase u guys wan to fix it
system spec:
amd x2 6000+
windows vista
2gb ram.
virtualdub 1.7.7 build 28312/release.
ffdshow rev 1763 msvc 2008 x86 unicode build.

here are the encoding setting,
fast recompress mode,
Encoder: mpeg2, one pass, 2201kbps
max i frame interval: 300, min: 30
bframes enabled, max consecutive bframe: 3, dynamic, closed gop,
number of threads: 2

macroblock decision mode: rate distortion, always try no motion, RD optimization for CBP,
ME/subpixel/macroblock/prepass compare function: SAD with chroma.
EPZS/prepass diamon size: adaptive with 3,
ME pre-pass: always.

Quant type: mpeg with trellis, MPEG2 DC precision: 11bits

Luminance/Darkness/Border masking enabled with default value, normalize adaptive quant.

i dont use it for real encoding, just for testing,
will try to narrow down the problem

[cc979]

i've just tested my build of ffdshow-1763 (gcc-4.2.1) with windows xp, encoding with virtualdub using mpeg2 or mpeg4 is fine, maybe its a vista/virtualdub bug

[Delerue]

Originally Posted by cc979

i've just tested my build of ffdshow-1763 (gcc-4.2.1) with windows xp, encoding with virtualdub using mpeg2 or mpeg4 is fine, maybe its a vista/virtualdub bug

I confirm; 1763 and XP too. The only problem (my problem?) is with x264 multi-thread encode. I can't use the other core.

[Yong]

hmm, the "assertion failed!" message is gone after change the encoding thread to 1 multithread lavc encoding will crash.
but x264 multithreaded encoding work for me lol.

[draggoon01]

i'm trying to use ffdshow for subtitles (srt files) but when i go full screen the text is blurry/jaggie. how can i fix this?

[Yong]

resize ur video to higher resolution, or use VMR7/9/EVR or haali video renderer.

[leeperry]

just wanted to point out that all issues with ffdshow & MPC HC + EVR on XP seem to be fixed.

I've watched several movies w/o any freezes using this package, with updated DLL's of EVR from Vista SP1 :
http://www.badongo.com/fr/file/7196279

Keep up the good work

[Atak_Snajpera]

Everything above ffdshow_rev1724_20071225_clsid does not decode correctly this ts


http://x264.nl/h.264.samples/force.p...e=./luxe.hd.ts

[wiak]

just a stupid question will there be a auto update future?, hehe now that ffdshow gets new builds nearly every day, i want to be up to date to keep my upcoming AVC+HE-AAC with LATM dvb-t setup stable

[Delerue]

Originally Posted by Atak_Snajpera

Everything above ffdshow_rev1724_20071225_clsid does not decode correctly this ts

http://x264.nl/h.264.samples/force.p...e=./luxe.hd.ts

No problem here. XP SP2 + build 1763 + MPC clsid january version + VMR7 Windowed. The only 'problem' is the high CPU demand, hehehe.

[Bathrone]

Hi folks Searches havent found answers so can I please ask:

1. Does ffdshow hook into the DXVA acceleration on Vista with suitable GPUs?

2. When the tryouts FAQ says that xvid and h.264 encoding is multi-threaded, does it actually mean multithreaded decoding or can ffdshow somehow be used for encoding?

[Delerue]

Originally Posted by Bathrone

1. Does ffdshow hook into the DXVA acceleration on Vista with suitable GPUs?

Try the MPC Home Cinema at www.codecs.com.

Originally Posted by Bathrone

2. When the tryouts FAQ says that xvid and h.264 encoding is multi-threaded, does it actually mean multithreaded decoding or can ffdshow somehow be used for encoding?

FFDShow can encode and decode xVid/x264 in multi-thread. But the x264 decode has some limitations; can't work in multi-thread with recently x264 encodes.

[ACrowley]

Originally Posted by Delerue

Try the MPC Home Cinema at www.codecs.com.


FFDShow can encode and decode xVid/x264 in multi-thread. But the x264 decode has some limitations; can't work in multi-thread with recently x264 encodes.

With latest CABAC Multithreading Patch it works for x264 too , so far its CABAC encoded. Latest XXL Builds are already patched with it.
Its working perfect multithreaded on all my x264 Files

~60% CPU Load for 1080p x264 12Mbps and ~85% for BluRay H264

[Leak]

Originally Posted by Bathrone

can ffdshow somehow be used for encoding?

Well, ffdshow comes with both an "ffdshow video encoder" DirectShow video compressor filter and an old-fashioned "ffdshow video codec" for VfW, so - yes?

np: Kid606 - I Miss You (Resilience)

[clsid]

Originally Posted by ACrowley

With latest CABAC Multithreading Patch it works for x264 too , so far its CABAC encoded. Latest XXL Builds are already patched with it.
Its working perfect multithreaded on all my x264 Files

~60% CPU Load for 1080p x264 12Mbps and ~85% for BluRay H264

All recent builds have it, since the patch was committed to SVN.

[clsid]

Originally Posted by Atak_Snajpera

Everything above ffdshow_rev1724_20071225_clsid does not decode correctly this ts

http://x264.nl/h.264.samples/force.p...e=./luxe.hd.ts

Does it work if you change the number of decoding threads in ffdshow options to 1?

[fastplayer]

Originally Posted by wiak

just a stupid question will there be a auto update future?

I'd look at auto-update from another perspective:
Who's gonna provide and pay for the bandwidth?

[Inventive Software]

Sourceforge?

[clsid]

I would prefer to continue putting our efforts into improving existing ffdshow functionality, instead of adding non-essential stuff like auto-update.

[Delerue]

Originally Posted by ACrowley

With latest CABAC Multithreading Patch it works for x264 too , so far its CABAC encoded. Latest XXL Builds are already patched with it.
Its working perfect multithreaded on all my x264 Files

~60% CPU Load for 1080p x264 12Mbps and ~85% for BluRay H264

Well, I told that because of this. What you think?

[leeperry]

Originally Posted by clsid

I would prefer to continue putting our efforts into improving existing ffdshow functionality, instead of adding non-essential stuff like auto-update.

how about some AR buttons for easy access so I don't need to use the virtual keyboard on my PCHC to reach 1.85 ?

just 4 buttons : 1.33 / 1.78 / 1.85 / 2.35

[Atak_Snajpera]

Does it work if you change the number of decoding threads in ffdshow options to 1?

Yes! If I change to 1 thread sample plays without problems. BTW I'm using Vista.

[fastplayer]

Originally Posted by leeperry

how about some AR buttons for easy access so I don't need to use the virtual keyboard on my PCHC to reach 1.85 ?

just 4 buttons : 1.33 / 1.78 / 1.85 / 2.35

And I want my 1.25, 1.0, 0.75, 0.5 and 0.001 button...

[leeperry]

your movies have weird AR sir

well, I can't reach 1.85 with the mouse, only 1.84 or 1.86.....I need to use the keyboard arrows to reach it

if buttons are a no go, how about making it slow down if right button is pushed or sumthing ?

[ACrowley]

@Delerue

Originally Posted by _xxl

yes and cabac / reconstruction parallelism, the code decodes up to 128 macroblocks in one thread while doing prediction+idct+deblock of the previously decoded 128 blocks in another thread.

Since this Patch i can play all x264/H264 1080p totally smooth on my X2@5000
Its a huge Improvement imho. Both Cores are running with ~max 85% ( BluRay AVC)
Its faster then Cyberlink in Software Mode and not far from CoreAVC
Thats the latest Patch what im talking about. Before this Patch it was only "the one Slice per Thread" Solution

[Keepitsimple]

Perhaps time for next offficial beta release then?

[ImAhNoBoDy]

Anyone notice when you install ffdshow-rev1763_20080108_xxl Directshow filters does not work in Graphedit? When I try to get into the list it gives the error "DirectShow SDK Filter Graph Editor has stopped working."

I uninstalled ffdshow and I was able to get all the filters under Directshow Filters. I reinstalled graphedit and ffdshow, same issue. Not sure if it's a bug or not.

[clsid]

Perhaps you have an old version of GraphEdit? I have never had such problems with GE build 061102.

[ImAhNoBoDy]

I have Graphedit DirectX 10.0 - Build: 060822. I think that's the most recent version I could find. But when I look at your build, your's look more recent. When did yours come out?

[clsid]

I don't remember where I got it. Maybe the latest DirectX SDK. Windows SDK 6.0 contains build 061030.

[ImAhNoBoDy]

After searching the forums for a while I found this http://forum.doom9.org/showthread.php?t=104234

It's Build 061030. Still the same issue. If I uninstall FFDShow, the filters under Directshow filter will show up without a problem. The last version I had of FFDShow was around September and thought I'll upgrade but now Graphedit crashes when I try to pull up directshow filters. I can pull up any other filters just fine except directshow filters. Not sure what else to do.

[clsid]

GE 061030 also works fine for me.

Do you have Vista? If so, see if disabling UAC makes a difference.

[ImAhNoBoDy]

Yes I do have Vista. UAC has been turned off from day one =). I only think it's this verison of FFDShow considering it was working with the September release. Which version of FFDShow are you using? Are you using yours or xxl's? Whatever version you have I'll uninstall mines and install the one you have and see if I have the same problem.

[clsid]

I am using my own builds.

There is a huge archive of (old) builds on the sourceforge website:
http://sourceforge.net/project/showf...ease_id=436746

[leeperry]

the opposite would have been quite funny

Originally Posted by clsid

I am using xxl's builds.

[Inventive Software]

Nice joke leeperry!

I think I found a bug in ffdshow's MPEG-2 decoder.

*breathe*

MPEG-2 video in Matroska with the Aspect Ratio set doesn't display correctly in Media Player Classic 6.4.9.1 or WMP11 (the only 2 media players I have). I tried both libmpeg2 and libavcodec, both wouldn't give me a correct aspect ratio (though ffdshow's Info panel gives the correct SAR and DAR), and I have no idea whether it's the splitter (Haali Media Splitter) or the decoder because using MPC's internal MPEG-2 decoder it displays correctly.

AAAAAAAAAAAARRRRRRRRRRRRRRGGGGGGGGGGGGHHHHHHHHH!

*breathe calmly*

I cut a 30 second MKV sample that I'm uploading [CUT uploaded, link: http://rapidshare.com/files/83628092/Monzasplit-001.mkv]. The video in question is 352x576 anamorphic 4:3. This kinda defeats my posting this in a way, but if I set the AR manually in MPC, it displays correctly, but what's the point of having to set it manually when Matroska has the option and it works with every other freakin' video I've muxed so far?!

Can ya tell I'm just a bit highly strung?

[iron2000]

I have a question.
In Dscaler, ffdshow appears in the filters menu.
So does ffdshow work in Dscaler?

I played with the settings and one of it cause Dscaler to not show and UI.
Turning the filter on and off seem to not show any difference.

[ImAhNoBoDy]

Thanks clsid, I downloaded your most recent build ffdshow_rev1771_20080113_clsid.exe and the same thing happened. I downloaded ffdshow_rev1738_20080101_xxl.exe and directshow filters are showing without crashing. Keep up the good work you guys.

[Delerue]

Everytime I install a new build, the VC-1 codec change from 'libavcodec' to 'wmv9'. Is it normal?

[haruhiko_yamagata]

Originally Posted by Inventive Software

MPEG-2 video in Matroska with the Aspect Ratio set doesn't display correctly in Media Player Classic 6.4.9.1 or WMP11 (the only 2 media players I have). I tried both libmpeg2 and libavcodec, both wouldn't give me a correct aspect ratio (though ffdshow's Info panel gives the correct SAR and DAR), and I have no idea whether it's the splitter (Haali Media Splitter) or the decoder because using MPC's internal MPEG-2 decoder it displays correctly.

I can reproduce. The fix shouldn't be too hard.

Can ya tell I'm just a bit highly strung?

Nice report.

[haruhiko_yamagata]

Originally Posted by leeperry

how about some AR buttons for easy access so I don't need to use the virtual keyboard on my PCHC to reach 1.85 ?

just 4 buttons : 1.33 / 1.78 / 1.85 / 2.35

I have a simple question, why do you have to specify aspect ratio manually?

[clsid]

Originally Posted by Delerue

Everytime I install a new build, the VC-1 codec change from 'libavcodec' to 'wmv9'. Is it normal?

I'll improve the installer to allow users to choose the VC-1 decoder.

[leeperry]

Originally Posted by haruhiko_yamagata

I have a simple question, why do you have to specify aspect ratio manually?

because several codecs can't achieve perfect AR, because it needs to be possible to divide the width and height by 16.

so for 2.40 HD movies, you would need to achieve 1280*533, but most ppl encode to 1280*528....which gives an AR of 2.42

same for 1.85 and SD movies, you would require 720*389 but they are encoded in 720*384....which gives an AR of 1.875

in KMPlayer my simple fix was to output from ffdshow without any AR, and then choose it in the player.

but with MPC, you only have a choice of 4:3 or 16/9

so my only option to get perfect AR is to "fix" it with ffdshow, and it's pretty annoying that you can't reach 1.85 with the mouse....only 1.84 or 1.86

I have to use the keyboard arrows to reach 1.85

and I believe I'm not the only guy in the world that wants to watch his movies with perfect AR

actually 6 buttons with the most common AR would be awesome.....alternatively, you could just give an option to set them to whatever the user wants.

1.33(4:3)
1.66(old movies or movies such as http://www.imdb.com/title/tt0107688/ )
1.78(16/9)
1.85(most DVD's use it)
2.35(cinemascope)
2.40(narrow cinemascope, used on HD discs mostly)

all possible cases are covered, if the AR is different, you can just check OAR(Original Aspect Ratio)

or if that's too much trouble, if you could make the AR selection slower if the right button is pushed or something, so 1.85 can be easily selected

[Inventive Software]

Narrowing the range of the AR slider might be a more practical option. Who has an AR of 5:1 (5.00) or 1:10 (0.10) these days anyway? Going from 0.1 to 3 would make things nicer I think.

[Thunderbolt8]

1.85 is more common with US movies, while 1.78 and 1.66 are/were for europe or also asia. so its more a choice of studios or a cultural habit than the difference of being on dvd or HD. the change from 2.35 to 2.40 was made due to technical reasons (which ones I dont know though, just read that somewhere in a film study book)

[Thunderbolt8]

that VC-1 decoder choice implemented in the installing process was a really useful change. it was already getting a bit anoying to change to libavcodec each time manually after each new version.
thanks for that!

edit: can someone please explain what that kakaoke change for ass subs is?

[SeeMoreDigital]

Originally Posted by Inventive Software

I think I found a bug in ffdshow's MPEG-2 decoder.

*breathe*

MPEG-2 video in Matroska with the Aspect Ratio set doesn't display correctly in Media Player Classic 6.4.9.1 or WMP11 (the only 2 media players I have). I tried both libmpeg2 and libavcodec, both wouldn't give me a correct aspect ratio (though ffdshow's Info panel gives the correct SAR and DAR), and I have no idea whether it's the splitter (Haali Media Splitter) or the decoder because using MPC's internal MPEG-2 decoder it displays correctly.

I have just tried doing the same using "ffdshow_rev1771_20080113_clsid.exe" and MediaPlayer Classic (with its internal Matroska selected).

For me, if I set FFdshow's "MPEG-2" decoder filter to "libavcodec", MPC is unable to connect to the decoder. However, if I set FFdshow's "MPEG in AVI" decoder filter to "libavcodec", MPC plays and displays the MPEG-2 video correctly.


Cheers

[leeperry]

Originally Posted by Inventive Software

Narrowing the range of the AR slider might be a more practical option. Who has an AR of 5:1 (5.00) or 1:10 (0.10) these days anyway? Going from 0.1 to 3 would make things nicer I think.

indeed, good idea.

as long as 1.85 is easily reachable, that is

[LoRd_MuldeR]

Originally Posted by clsid

All recent builds have it, since the patch was committed to SVN.

Does that mean the "CABAC Multithreading" patch was committed to the ffmpeg/libavcodec SVN and will be included in recent MPlayer builds as well?
Or was the patch "only" committed to ffdshow's own SVN? Couldn't find any suitable entry on the ffmpeg/libavcodec SVN log...

[clsid]

It has not been committed to FFmpeg SVN. The patch is still experimental.
xxl has recently committed it to ffdshow SVN.

[LoRd_MuldeR]

Originally Posted by clsid

It has not been committed to FFmpeg SVN. The patch is still experimental.
xxl has recently committed it to ffdshow SVN.

I see. thanks for clearing that up.

[Yong]

Audio filter bug: winamp2 filter
"click to select winamp 2 directoty..." button is not working, it doesnt pop up a windows and let me choose directory, when configure it offline.
it does work for me when im playing video/audio, but still, it doesnt show up the plugin(Signal processing studio DSP, yes i select winamp5 as dir, not the plugin dir)

[_xxl]

Can't compile ffvdub with MSVC2003.

[Delerue]

Originally Posted by clsid

I'll improve the installer to allow users to choose the VC-1 decoder.

Thanks, man.

[Inventive Software]

@devs: Any news on the MPEG-2 decoder bug I reported a couple days ago?

[foxyshadis]

Originally Posted by leeperry

because several codecs can't achieve perfect AR, because it needs to be possible to divide the width and height by 16.

I'd think it would be less work if you watch things more than once to just set the AR of the file itself, via remuxing or mpeg4modifier, so you don't have to hassle with always setting ffdshow's.

[SeeMoreDigital]

Originally Posted by leeperry

because several codecs can't achieve perfect AR, because it needs to be possible to divide the width and height by 16.

Well this is how I do it when generating anamorphic MPEG-4 encodes: -

http://forum.doom9.org/showthread.ph...65#post1034465

[leeperry]

Originally Posted by foxyshadis

I'd think it would be less work if you watch things more than once to just set the AR of the file itself, via remuxing or mpeg4modifier, so you don't have to hassle with always setting ffdshow's.

well I'm not gonna remux my files when one click in ffdshow can fix the problem

anyhow, outputting without AR from ffdshow and setting in the AR in the player is even easier, because it seems that noone wants to make setting the AR(like 1.85) easier in ffdshow...I already asked for this feature a few years ago

and anyway MPC HC is a fantastic player, but it doesn't work fine at 48Hz(required for my videoprojector), it's giving a lot of judder and duplicated frames....everything's fine at 60 or 72.....and KMPlayer doesn't work with EVR w/o random freeezes on XP

so I'm trying Zoom Player atm because it also supports EVR......hopefully it will work fine at 48Hz and will let me choose the AR

Originally Posted by SeeMoreDigital

Well this is how I do it when generating anamorphic MPEG-4 encodes: -

http://forum.doom9.org/showthread.ph...65#post1034465

thanks for the tip!
but I'm already having a hard time getting this damn computer to output a good picture, so I can't be hassled with all the commandline arguments of x264

some ppl really master the x264 encoding(who said ESIR? ) so I'll leave it to them

[blizard]

Could anybody explain what is the difference between CLSIDs Nightly build generic and ICL10? I have an Athlon X2 which should have support for MMX, SSE2, SSE3 and even x86-64 ;-) but I don't know which if those builds that would be the best for this CPU. From release notes I can not find out more then there is different compilers and some are said to be supporting SSE.

I dl CLSIDs generic build (r. 1771) and looks like it will use SSE, so what is the point with ICL10 build which is claimed to be for SSE? From what I can see it Mightly build looks stable, but is there any big difference from beta4a build which is called stable?

(On a side note: Thanks for your build CLSID and all other compilers for FFDSHOW tryout!)

[foxyshadis]

ICL is just a better compiler than MSVC, which usually translates into faster code that takes advantage of newer CPUs. This only makes any difference if you use filters that don't already have hand-optimizations (most of them), which are still better than compilers most of the time. It makes zero difference for decoding most formats.

[fastplayer]

Originally Posted by blizard

I dl CLSIDs generic build (r. 1771) and looks like it will use SSE, so what is the point with ICL10 build which is claimed to be for SSE? From what I can see it Mightly build looks stable, but is there any big difference from beta4a build which is called stable?

All this has been asked and discussed a gazillion times. Search is your friend...
Anyway, in addition to what foxy said, read the FAQ:
http://ffdshow-tryout.sourceforge.net/html/en/faq.htm

Nightly builds sometimes break some existing feature(s) or add new features that are not working 100%. The devs are very fast fixing 'em though. So in the end, it does not matter which build you use.

[clsid]

The ICL builds require at least a SSE capable CPU. The generic builds only require MMX. However, ALL builds contain hand-written assembly code with SSE/SSE2/3dnow instructions, which will get used if and only if your CPU supports it.

The difference in builds only applies to the ffdshow.ax file. All builds use GCC for compiling libavcodec. So the decoding performance is EQUAL for all builds. The only situation where the ICL build might give a bit better performance is when you make use of the internal filters in ffdshow. And even then, there is only a gain for some of the filters.

[haruhiko_yamagata]

@Inventive Software, leeperry
Please be patient. We are receiving a lot of bug reports and requests.

[leeperry]

Originally Posted by haruhiko_yamagata

@Inventive Software, leeperry
Please be patient. We are receiving a lot of bug reports and requests.

Yeah sorry Haruhiko, I'm just really disappointed that MPC HC doesn't work properly at 48Hz....which is the most used refresh rate on videoprojectors.....because it outputs perfect 24FPS movies

for all your help, and your great great work

if you could make selecting 1.85 AR easier by making the AR slider range shorter, that'd be totally awesome!

[blizard]

Originally Posted by clsid

The ICL builds require at least a SSE capable CPU. The generic builds only require MMX. However, ALL builds contain hand-written assembly code with SSE/SSE2/3dnow instructions, which will get used if and only if your CPU supports it.

The difference in builds only applies to the ffdshow.ax file. All builds use GCC for compiling libavcodec. So the decoding performance is EQUAL for all builds. The only situation where the ICL build might give a bit better performance is when you make use of the internal filters in ffdshow. And even then, there is only a gain for some of the filters.

Thanks CLSID, fastplayer and foxyshadis for your explanation. I did see that page you linked to before I posted, fastplayer, but I am not a compiler so it is not easy to understand all choices in type of compiles and what those will mean for end user. Searching for an answer is not an easy task when some threads are around 100 pages and search will only point to first thread and not where in those thread there is an actual search term to be found.

[fastplayer]

Originally Posted by blizard

Searching for an answer is not an easy task when some threads are around 100 pages and search will only point to first thread and not where in those thread there is an actual search term to be found.

Ever heard of google.com?
Add "site:doom9.org" as a search parameter plus what you want to actually search for on the D9 forums.
You want to search for "ffdshow ICL generic difference" for example. Here's what you get:
http://www.google.com/search?hl=en&q...te%3Adoom9.org

[fastplayer]

@clsid:
Your 1787 build is Unicode-only now? No Win98 support anymore, right?

[cc979]

rev.1787 i get some errors when compiling with normal gcc-4.0.4:

gcc -c -mno-cygwin -mdll -fno-rtti -mthreads -pipe -D_WINGDI_ -DUCLIBCPP -D_GLIBCPP_HAVE_MBSTATE_T -D_WIN32_IE=0x0500 -DARCH_IS_IA32 -DARCH_IS_32BIT -DHAVE_MMX -mmmx -w -DNDEBUG -UDEBUG -DFFDEBUG=0 -I. -I.. -Iuclibc++ -Ibaseclasses -I../baseclasses -IimgFilters -I../imgFilters -Implayer -I../mplayer -Isettings -I../settings -Isettings/filters -I../settings/filters -Icodecs -I../codecs -Isubtitles -I../subtitles -Iconvert -I../convert -Idialog -I../dialog -IaudioFilters -I../audioFilters -Icygwin -I../cygwin -Iffmpeg -I../ffmpeg -Iacm -I../acm -Ifilters -I../filters -Imuxers -I../muxers -I/dx/Include -L/dx/MingLib -ldx9 -O2 -march=pentium-mmx -mtune=i686 -fomit-frame-pointer -finline-functions -finline -frename-registers -fweb -funit-at-a-time -MMD -o ffdshow_all.o ffdshow_all.cpp
TffdshowBase.cpp:47: error: specialization of 'Tinterface* TffdshowBase::getBaseInterface() [with Tinterface = IffdshowBaseA]' after instantiation
TffdshowDec.cpp:36: error: specialization of 'Tinterface* TffdshowDec::getDecInterface() [with Tinterface = IffdshowDecA]' after instantiation
TffDecoder.cpp: In member function 'virtual void TffdshowDecVideo::getChapters()':
TffDecoder.cpp:1848: error: cannot convert 'char_t*' to 'const wchar_t*' for argument '1' to 'int wcscmp(const wchar_t*, const wchar_t*)'
TffDecoder.cpp:1849: error: cannot convert 'char_t*' to 'const wchar_t*' for argument '1' to 'int wcscmp(const wchar_t*, const wchar_t*)'
TffDecoder.cpp:1850: error: cannot convert 'char_t*' to 'const wchar_t*' for argument '1' to 'int wcscmp(const wchar_t*, const wchar_t*)'
TffDecoder.cpp:1851: error: cannot convert 'char_t*' to 'const wchar_t*' for argument '1' to 'int wcscmp(const wchar_t*, const wchar_t*)'
TffDecoder.cpp:1852: error: cannot convert 'char_t*' to 'const wchar_t*' for argument '1' to 'int wcscmp(const wchar_t*, const wchar_t*)'
TffDecoder.cpp:1873: error: no matching function for call to 'IAMExtendedSeeking::GetMarkerName(int&, char_t**)'
D:/msys/1.0/dx/Include/qnetwork.h:258: note: candidates are: virtual long int IAMExtendedSeeking::GetMarkerName(long int, OLECHAR**)
TffDecoder.cpp:1883: error: cannot convert 'char_t*' to 'OLECHAR*' for argument '1' to 'void SysFreeString(OLECHAR*)'
TffDecoder.cpp: In member function 'virtual long int TffdshowDecVideo::GetMarkerName(long int, OLECHAR**)':
TffDecoder.cpp:1960: error: cannot convert 'char_t*' to 'OLECHAR*' in assignment
TffDecoder.cpp:1961: error: no matching function for call to 'strcpy(OLECHAR*&, const char_t*&)'
d:/mingw/bin/../lib/gcc/i686-pc-mingw32/4.0.4/../../../../include/string.h:45: note: candidates are: char* strcpy(char*, const char*)
char_t.h:45: note:                 wchar_t* strcpy(wchar_t*, const wchar_t*)
make: *** [ffdshow_all.o] Error 1

[haruhiko_yamagata]

Originally Posted by cc979

rev.1787 i get some errors when compiling with normal gcc-4.0.4:

I asked albain to fix this.
fastplayer, clsid had to give up ANSI build because of these compilation errors.

[Inventive Software]

98 support was dropped towards the end of 2007.

[clsid]

Originally Posted by Inventive Software

98 support was dropped towards the end of 2007.

And was restored quickly thereafter.

My unicode only build is temporarily until the compilation errors for the ANSI build are fixed.