ffdshow tryouts project: Discussion & Development

[sarvatt]

sarvatt, buy Zoom Player Pro and set up customized media profiles. That way you can force everything to be decoded and played exactly how you want.

I'm curious about the possibility of adding coreavc as an 'alternative' codec, if it's not complicated, but I really have no time to even look into it. :( Particularly since I typically use it for post-processing everything except HD.

Thank you for the suggestion, I actually have purchased it and use it fine that way in the past for my laptop but I have alot of problems using it as an external video playback program inside mediaportal. I have just gone back to VMR9+vsfilter for now as I think the EVR implementation in mediaportal isn't ready for prime time because it would crash using FFDShow for subtitles loading from raw video sometimes (on coreavc h264 HD files only oddly enough) then work when the program was restarted.

SPDIF passthrough/AC3 output is still broken for me as well in 1437 on vista x86, 1413 is the last one that works. I don't have any builds between 1413 and 1425 to see which one broke it though, but 1425 was the first one that I noticed it.

[cbemoore]

SPDIF passthrough/AC3 output is still broken for me as well in 1437 on vista x86, 1413 is the last one that works. I don't have any builds between 1413 and 1425 to see which one broke it though, but 1425 was the first one that I noticed it.

I'm also seeing this problem with the latest builds.....

[_xxl]

Read more:
http://ffdshow-tryout.sourceforge.net/phpBB2/viewtopic.php?t=478
Is rev 1424 working?Are there other related issues?
Can you please test:
http://rapidshare.com/files/51252887/ffdshow.rar.html

[cc979]

compiling svn 1439 using gcc-4.0.4 i get this error


TffDecoder.cpp: In member function 'void TffdshowDecVideo::ConnectCompatibleFilter()':
TffDecoder.cpp:623: error: expected primary-expression before ')' token
TffDecoder.cpp:623: error: '__uuidof' was not declared in this scope
TffDecoder.cpp: In member function 'void TffdshowDecVideo::DisconnectFromCompatibleFilter()':
TffDecoder.cpp:696: error: expected primary-expression before ')' token
TffDecoder.cpp:696: error: '__uuidof' was not declared in this scope
TffdshowVideoInputPin.cpp: In member function 'virtual long int TffdshowVideoInputPin::CheckMediaType(const CMediaType*)':
TffdshowVideoInputPin.cpp:180: error: expected primary-expression before ')' token
TffdshowVideoInputPin.cpp:180: error: '__uuidof' was not declared in this scope
make: *** [ffdshow_all.o] Error 1


anyone else getting problems ?

[ShadowVlican]

Read more:
http://ffdshow-tryout.sourceforge.net/phpBB2/viewtopic.php?t=478
Is rev 1424 working?Are there other related issues?
Can you please test:
http://rapidshare.com/files/51252887/ffdshow.rar.html
folders #2 and #3 work!!! but #1 doesn't

[_xxl]

Partialy reverted 1424:
Index: TffdshowDecAudio.cpp
===================================================================
--- TffdshowDecAudio.cpp (revision 1438)
+++ TffdshowDecAudio.cpp (working copy)
@@ -796,7 +796,7 @@
pDataOutW[2]=type;
pDataOutW[3]=WORD(size*16);
//pDataOutW[4] = 0x0b77; // AC3 syncword (removed because works only for DTS, not for DD)
- _swab((char*)buf,(char*)&pDataOutW[10],(int)(size*2-2));
+ _swab((char*)buf,(char*)&pDataOutW[4],(int)size);
}
else
{

[cbemoore]

I'd like to try out Albain's latest fixes for the SPDIF passthrough problem (rev 1440), but the latest build available to download is rev 1437. Could I ask clsid or xxl to kick off a nightly build when you have a free moment?

[haruhiko_yamagata]

compiling svn 1439 using gcc-4.0.4 i get this error
Yes, __uuidof is microsoft specific. Rev 1443 is OK. :)

[cc979]

Yes, __uuidof is microsoft specific. Rev 1443 is OK. :)

anyway around it ?

[rack04]

Is it ok to install over an older version or should I uninstall before installing the newest version?

[foxyshadis]

Uninstalling removes all of your settings, overwriting won't.

[Leak]

anyway around it ?
Well, using revision 1442+ should do it... *hint hint*

np: Pass Into Silence - Iceblink (Pop Ambient 2006)

[Porsche_fan]

I’ve experienced a minor bug using BSplayer and the aviSynth in ffdshow starting after version 1193. I am not exactly sure which version the issue started occurring because I started downloading Leak’s builds with the frame buffering feature and didn’t resume normal ffdshow downloads until after build 1300 or so, which is when I initially noticed the issue.


The problem is that BSplayer gives a floating point error at the first initial start of trying to play an avi file. The error message gives the option to close BSplayer or continue. If I choose to continue, the player is then able to play any video fine without displaying the error message again and plays as many videos as I like just like normal. The error message won’t resurface until the player is actually closed and is opened again.


This issue doesn’t occur with Media Player Classic and I have tried several versions of BSplayer including the newest and the last version 137.826 which didn’t contain adware.


Also the issue doesn’t occur when aviSynth isn’t selected and it only occurs when some scripts are used in aviSynth. So for example the following script doesn’t produce an error…

LimitedSharpenFaster(ss_x=1.0,ss_y=1.0)

However, when smode=4 is added the error appears.

LimitedSharpenFaster(ss_x=1.0,ss_y=1.0,Smode=4)


I think it may be related somehow to the input colorspaces. I have had a similar error in versions that “worked” if I had not selected all of the input colorspaces in aviSynth. So for example if only YV12 was selected I receive the same floating point error in build 1193. However, in the versions of ffdshow after 1193 version I can’t solve the issue by changing the input colorspaces.


Also there seems to be an issue with resize. When I select resize to 1.25 it resizes much more than 1.25.


Thanks for the effort involved in developing ffdshow...I absolutely love it.:)

[thuan]

Again, subtitle rendering crashes with clsid's ffdshow rev1443 on [SS-Eclipse] Hayate no Gotoku! - 22 (1280x720 h264) [971BE017].mkv. Rev1437 from _xxl works fine. Both were tested with latest MPC by clsid. Here's a sample (http://rapidshare.com/files/52157538/_SS-Eclipse__Hayate_no_Gotoku__-_22__1280x720_h264___971BE017_-001.mkv.html) and my ffdshow settings (http://rapidshare.com/files/52157608/settings.reg.html). It crashes at Hayate's line: "I must work hard for Hinagiku-san's sake, too."
Computer spec: Core 2 Duo, chipset/video card: 945G, Vista32 up to date, using Haali renderer (latest). This is my new OS installation, hope you can reproduce it.

[foxyshadis]

Can you use mkvextract (or one of the guis for it) to demux the ssa subs, then zip them and attach here or to an uploader service?

[fastplayer]

It crashes for me too at 0:28min (XPSP2, rev1443, MPC internal MKV splitter). Here's the ASS file:
[Script Info]
; Script generated by Aegisub v2.00 PRE-RELEASE (SVN r1458, ArchMageZeratuL)
; http://www.aegisub.net
Title: Hayate no Gotoku - 22
ScriptType: v4.00+
WrapStyle: 0
PlayResX: 704
PlayResY: 400
ScaledBorderAndShadow: no
Collisions: Normal
Last Style Storage: hayate
Audio File: ?video
Video File: Hayate22-Timing-Crapraw.avi
Video Aspect Ratio: 0
Video Zoom: 8
Video Position: 20015

[V4+ Styles]
Format: Name, Fontname, Fontsize, PrimaryColour, SecondaryColour, OutlineColour, BackColour, Bold, Italic, Underline, StrikeOut, ScaleX, ScaleY, Spacing, Angle, BorderStyle, Outline, Shadow, Alignment, MarginL, MarginR, MarginV, Encoding
Style: Default,Montara Gothic,32,&H0FFFFFFF,&H0000FFFF,&H00714A40,&H80000000,0,0,0,0,100,100,0,0,1,2,2,2,10,10,10,0
Style: Default-alt,Montara Gothic,32,&H1EFFFFFF,&H0000FFFF,&H00855A7F,&H80000000,0,0,0,0,100,100,0,0,1,2,2,2,10,10,10,0
Style: Note,Arial Rounded MT Bold Eclipse,20,&H00FFFFFF,&H0000FFFF,&H00000000,&H00000000,0,0,0,0,100,100,0,0,1,2,2,8,10,10,10,0

[Events]
Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
Comment: 0,0:00:00.00,0:00:00.00,Default,,0000,0000,0000,,
Comment: 0,0:00:00.00,0:00:00.00,Default,,0000,0000,0000,,
Comment: 0,0:00:00.00,0:00:00.00,Default,,0000,0000,0000,,
Comment: 0,0:00:00.00,0:00:00.00,Default,,0000,0000,0000,,
Comment: 0,0:00:00.00,0:00:00.00,Default,,0000,0000,0000,,
Comment: 0,0:00:00.00,0:00:00.00,Default,,0000,0000,0000,,
Comment: 0,0:00:00.00,0:00:00.00,Default,,0000,0000,0000,,

Dialogue: 0,0:00:04.53,0:00:09.37,Default,Hay,0000,0000,0000,,When watching Hayate no Gotoku, make sure to watch at a good distance away from the TV in a well lit room.
Dialogue: 0,0:00:09.37,0:00:09.99,Default,Nag,0000,0000,0000,,Got it?
Dialogue: 0,0:00:14.57,0:00:16.81,Default,Nag,0000,0000,0000,,He's out training again?
Dialogue: 0,0:00:16.81,0:00:21.70,Default,Mar,0000,0000,0000,,He did say he was going to come up with an ultimate move over summer vacation, so I think he's a little under the gun.
Dialogue: 0,0:00:24.77,0:00:26.38,Default,Nag,0000,0000,0000,,From so early in the morning...
Dialogue: 0,0:00:26.38,0:00:29.14,Default,Nag,0000,0000,0000,,Is he a sixty-year-old boxer in training?
Dialogue: 0,0:00:29.34,0:00:31.65,Default,Hay,0000,0000,0000,,I must work hard for Hinagiku-san's sake, too. {ep20}
Dialogue: 0,0:00:31.65,0:00:34.52,Default,Nag,0000,0000,0000,,Yup, you're right, for Hinagiku's sake, too-{ep20}
Dialogue: 0,0:00:34.52,0:00:35.08,Default,Nag,0000,0000,0000,,Wait. {ep20}
Dialogue: 0,0:00:36.62,0:00:40.38,Default,Mar,0000,0000,0000,,But how he can put full effort into everything is one of his good attributes.
Dialogue: 0,0:00:40.38,0:00:41.64,Default,Nag,0000,0000,0000,,It's not good!
Dialogue: 0,0:00:43.10,0:00:47.46,Default,Nag,0000,0000,0000,,Because summer is about to end,\Nshouldn't we be spending time together?
Dialogue: 0,0:00:48.02,0:00:52.69,Default,Nag,0000,0000,0000,,Imagine what Ro*** must have felt when he lost Adri**!
Dialogue: 0,0:00:53.13,0:00:56.08,Default,Nag,0000,0000,0000,,Seriously, what's on his mind?
Dialogue: 0,0:00:56.08,0:00:57.10,Default,Kla,0000,0000,0000,,Milady.
Dialogue: 0,0:00:58.51,0:01:00.57,Default,Kla,0000,0000,0000,,Ayasaki Hayate at your service!
Dialogue: 0,0:00:19.90,0:00:24.27,Note,,0000,0000,0000,,Shirt: Namidabashi

[thuan]

Can you use mkvextract (or one of the guis for it) to demux the ssa subs, then zip them and attach here or to an uploader service?

Here you are http://rapidshare.com/files/52238614/sub.ass.html.

[B.F.]

Postprocessing doesn't work with FLV4 in new ffdshow builds.

[SBeaver]

Hope to see the eac3 decoder in ffdshow soon :)

btw, why is there no realtime DTS encoder in ffdshow? Is that a licensing issue or does it eat too much cpu?

[_xxl]

eac3 decoder
What is new?Can't be decoded by AC3 decoder?

[haruhiko_yamagata]

Again, subtitle rendering crashes with clsid's ffdshow rev1443 ...
Sorry for the bug. fixed at rev 1444.

[Mc Onyx]

What is new?Can't be decoded by AC3 decoder?

He means EAC3 a.k.a. Dolby Digital Plus, which is the codec used on HDDVD and Blu-Ray and it's not decodable by old AC3 decoders. FFMPEG is working on the decoder and it's present i nightly builds of FFMPEG and Mplayer. It would be great if someone could compile a ffdshow test build with the decoder for testing, if it's possible. A lot of people would be very happy to be able to decode EAC3 through Dshow with opensource decoder. You can find more info in this thread.

http://forum.doom9.org/showthread.php?t=129050

[tongyang]

Hello, I apologize if this is not the right place to post this.

http://www.megaupload.com/?d=HSNZ3DFV

[Hauu~].Higurashi.no.Naku.Koro.ni.Kai.08.(H264.Vorbis)[ED850938].mkv

I have a problem with the subtitle positioning of the file linked above (139MB). Most of the videos I have work very well except this one. I have included screenshots on what ffdshow shows and what it should show with directvobsub.

Wrong:

http://img263.imageshack.us/img263/1521/higurashi1bpj4.th.jpg (http://img263.imageshack.us/my.php?image=higurashi1bpj4.jpg)http://img263.imageshack.us/img263/6480/higurashi2bfo2.th.jpg (http://img263.imageshack.us/my.php?image=higurashi2bfo2.jpg)http://img211.imageshack.us/img211/9900/higurashi3bno2.th.jpg (http://img211.imageshack.us/my.php?image=higurashi3bno2.jpg)http://img505.imageshack.us/img505/685/higurashi4bxd7.th.jpg (http://img505.imageshack.us/my.php?image=higurashi4bxd7.jpg)http://img505.imageshack.us/img505/213/higurashi5bbm4.th.jpg (http://img505.imageshack.us/my.php?image=higurashi5bbm4.jpg)http://img211.imageshack.us/img211/8839/higurashi6bcs8.th.jpg (http://img211.imageshack.us/my.php?image=higurashi6bcs8.jpg)

What it should show:

http://img505.imageshack.us/img505/5148/higurashi1auy9.th.jpg (http://img505.imageshack.us/my.php?image=higurashi1auy9.jpg)http://img263.imageshack.us/img263/5662/higurashi2aoi0.th.jpg (http://img263.imageshack.us/my.php?image=higurashi2aoi0.jpg)http://img211.imageshack.us/img211/6501/higurashi3aaj0.th.jpg (http://img211.imageshack.us/my.php?image=higurashi3aaj0.jpg)http://img211.imageshack.us/img211/9589/higurashi4awp0.th.jpg (http://img211.imageshack.us/my.php?image=higurashi4awp0.jpg)http://img263.imageshack.us/img263/7398/higurashi5aah1.th.jpg (http://img263.imageshack.us/my.php?image=higurashi5aah1.jpg)http://img211.imageshack.us/img211/8898/higurashi6aoo1.th.jpg (http://img211.imageshack.us/my.php?image=higurashi6aoo1.jpg)

Hope what I provided is enough. Feel free to ask for more details... (this is using revision 1431 by the way)

[thuan]

I've seen some of this kind of inconsistency between ffdshow sub renderer and VSFilter too. But I am lazy to make a screenshot comparison like you.

And Higurashi is a great series since Monster about this kind of thing hehe.

[Milvus]

I think this example show that différent implementations of ass/ssa renderers can have significant differences in the final result. It's the same kind of problems than with HTML rendering in different browsers : specs are complex, incompletly implemented, with lot of versions...

I hope in the future there will be work to harmonize all this, because if not, we wil never be able to use softsubs a there true potential.

[Kado]

I've tested it with casimir's mpc, and it is confirmed.
Using mpc's internal renderer the subs appear in proper place and with ffdshow's (rev1448) they don't and there are some different colors as well (3rd image).
however ffdshow subs seems to blend more to the video witch is nice (maybe some translucent effect?, 2nd image).
Also, I don't get that incorrect character from image 2 (between "original work" and "original story")

[haruhiko_yamagata]

@tongyang
Thank you. I'll test that sample. At least I can reproduce some of the issues.

@Milvus
I have just implemented font rendering engine. SSA/ASS in ffdshow is experimental and pre-alpha status. They are just bugs or not implemented. It's not time to mention the compatibility.

[Leak]

Haruhiko,

when I turn the OSD on (using the default font settings) with rev. 1451 it crashes on me at _fontPrepareOutline_sse2 + 0x1e - going back to 1450 fixes this.

Setting the outline width for the OSD to 0 works for a few frames (with stray black lines appearing above the text) but then it still crashes at the exact same location.

np: Burnt Friedman & The Nu Dub Players - It's Thunder (Just Landed)

[Warpman]

however ffdshow subs seems to blend more to the video witch is nice (maybe some translucent effect?, 2nd image).


That is because mpcs internal renderer doubles the rendering resolution by default and creates much sharper subtitles, vsfilter can do that, too.

[haruhiko_yamagata]

Haruhiko,

when I turn the OSD on (using the default font settings) with rev. 1451 it crashes on me at _fontPrepareOutline_sse2 + 0x1e - going back to 1450 fixes this.

Setting the outline width for the OSD to 0 works for a few frames (with stray black lines appearing above the text) but then it still crashes at the exact same location.

np: Burnt Friedman & The Nu Dub Players - It's Thunder (Just Landed)Oops, release build crashes. Constant data was not aligned. I have fixed at rev 1452.

[haruhiko_yamagata]

[Hauu~].Higurashi.no.Naku.Koro.ni.Kai.08.(H264.Vorbis)[ED850938].mkv

I have a problem with the subtitle positioning of the file linked above (139MB).
Positioning problems have been fixed at rev 1452-1453. :thanks:

[Leak]

Oops, release build crashes. Constant data was not aligned. I have fixed at rev 1452.
I'm afraid that didn't do it...

I checked out rev. 1452, made a fresh export into an empty directory, compiled everything there and - same result.

It doesn't matter if I do a debug or release build, by the way.

It still crashes in this line:
026CFA16 F3 0F 7E 09 movq xmm1,mmword ptr [ecx]
with an access violation... :(

For what's it worth - I put a
DPRINTF(_l("%i/%i,%i/%i"),x,_dx,y,_dy);
in front of the call to fontPrepareOutline_sse2 in line 323 and it crashed at "201/288,21/22". so it almost made it to the end - I guess you're just a few bytes short there...

*time passes*

Yeah, changing line 229 to
bmp[0]=(unsigned char*)aligned_calloc3(extra_dx,extra_dy+1,16,16);
prevents the crash, but it surely isn't correct either.

Also, is it just me, or has the distance between the OSD lines increased?

Hope this helps... :)

EDIT: I'm too sleepy to do another compile, but looking at the changes I guess rev. 1453 doesn't affect the above...

np: Burnt Friedman & Jaki Liebezeit - The Sticks (Secret Rhythms 2)

[tongyang]

Positioning problems have been fixed at rev 1452-1453. :thanks:

Wow, that was fast. Thanks!!

[chros]

I tried to playback a dts-wav file (*.wav) with MPC and ffdshow v1381 , but no success ...
Filters: Wave Parser , ffdshow

in ffdshow codecs: DTS in wav is checked (nothing happens)
AC3filter can playback it (if I check PCM at system tab)

Anyone confirm it ? Or do I need something to set?
If it's a bug, can someone tell a working version number?

Thanks

[haruhiko_yamagata]

I'm afraid that didn't do it...

I checked out rev. 1452, made a fresh export into an empty directory, compiled everything there and - same result.
Thank you. I hope rev 1454 is OK.

[Yong]

guys, do i need to use msvc 2005 to compile ffdshow?
i've tried compile it with gcc4.2.1(from mingw.org), everything is fine(not very sure though :p) but the OSD eats alot of cpu usage.
especially the dynamic one, eg frame type, cpu load, frame mean quantizer and so on.
static one is fine though, and shadow is disabled.
using rev1450 here, tried sourceforge build(xxl rev1425) which doesnt have cpu usage problem :(
btw my cpu is celeron 1.2ghz :D

update: tried sourceforge build by xxl rev1453, same problem, can anyone confirm this?
i think it isnt gcc problem, may be due to recently updated OSD code?
time to get new cpu:p

[vortex_hl]

update: tried sourceforge build by xxl rev1453, same problem, can anyone confirm this?
i think it isnt gcc problem, may be due to recently updated OSD code?

i can confirm that. ffdshow_rev1431_20070821_clsid_sse_icl10.exe and earlier versions are fine. after rev1434 osd slowdown bug appears. maybe "vertical scaling" code cause this.

[cyberbeing]

Also, I don't get that incorrect character from image 2 (between "original work" and "original story")

FFDshow just isn't displaying the Unicode character correctly. It must be displaying the subs with an ANSI encoding when it should be using the UTF-8 encoding of the subtitle.

[cc979]

guys, do i need to use msvc 2005 to compile ffdshow?
i've tried compile it with gcc4.2.1(from mingw.org), everything is fine(not very sure though :p) but the OSD eats alot of cpu usage.
especially the dynamic one, eg frame type, cpu load, frame mean quantizer and so on.
static one is fine though, and shadow is disabled.
using rev1450 here, tried sourceforge build(xxl rev1425) which doesnt have cpu usage problem :(
btw my cpu is celeron 1.2ghz :D

update: tried sourceforge build by xxl rev1453, same problem, can anyone confirm this?
i think it isnt gcc problem, may be due to recently updated OSD code?
time to get new cpu:p

i only compile it with gcc-4.0.4

[Leak]

Thank you. I hope rev 1454 is OK.
Confirmed - rev. 1455 doesn't crash anymore. But the speed, as mentioned above, is still abysmal even in a release build... :(

Also, "Fast rendering" only renders the very first OSD line when checked - I don't think it's supposed to do that?

np: Reinhard Voight - Ready For Take Off (Kompakt Total 6 (Disc 1))

[ShadowVlican]

thanks for addressing the SPDIF issue i pointed out back a couple pages

but the way you have implemented it is non-intuitive:
http://img27.picoodle.com/img/img27/9/9/3/f_tempm_38d3851.jpg (http://www.picoodle.com/view.php?img=/9/9/3/f_tempm_38d3851.jpg&srv=img27)
revision 1453

the option "Maximize SPDIF compatibility" should not be part of AC3 output (should not be greyed out) because it also affects AC3/DTS passthrough, where ffdshow shouldn't be touching the AC3/DTS stream at all

"Maximize SPDIF compatibility" should not be bound to the AC3 checkbox

i get around this little quirk by enabling AC3, checking on "Maximize SPDIF compatibility", then checking off AC3

[Yong]

i only compile it with gcc-4.0.4
do you have the osd high cpu usage problem?

i can confirm that. ffdshow_rev1431_20070821_clsid_sse_icl10.exe and earlier versions are fine. after rev1434 osd slowdown bug appears. maybe "vertical scaling" code cause this.
thx for confirmation;)

[DMD]

Up to the version ffdshow_rev1381_20070729, ffdshow audio works in correct way getting in exit SPDIF the digital stream.
But from the following releases this doesn't happen, yet the setting is the same.

These are the screens in sequence (double clik to magnify): http://img178.imageshack.us/img178/7229/settingffdshowaudioij0.jpg

Excuse I had forgotten me the screen of the setting of the output
http://img104.imageshack.us/img104/92/screenoutlz4.jpg (http://imageshack.us)


The test that the setting is correct, is that the "monitor level" some volume is correctly visualized.
What can you prevent the correct transfer the exits spdif of the digital stream?
can you give me some help? :helpful::(

Cordially I thank
Domenico

[cbemoore]

Up to the version ffdshow_rev1381_20070729, ffdshow audio works in correct way getting in exit SPDIF the digital stream.
But from the following releases this doesn't happen, yet the setting is the same.
Domenico

In the "output" section, try selecting "Maximise SPDIF compatibility".

Chris

[haruhiko_yamagata]

guys, do i need to use msvc 2005 to compile ffdshow?
i've tried compile it with gcc4.2.1(from mingw.org), everything is fine(not very sure though :p) but the OSD eats alot of cpu usage.
especially the dynamic one, eg frame type, cpu load, frame mean quantizer and so on.
static one is fine though, and shadow is disabled.
using rev1450 here, tried sourceforge build(xxl rev1425) which doesnt have cpu usage problem :(
btw my cpu is celeron 1.2ghz :D

update: tried sourceforge build by xxl rev1453, same problem, can anyone confirm this?
i think it isnt gcc problem, may be due to recently updated OSD code?
time to get new cpu:p
Thanks, I can reproduce. Swscaler, which I used for scaling, is slower than the original code. When I try bilinear, it's better but still slower. I'll try to use original code when it's not too complicated (vertical scaling, blur).

[haruhiko_yamagata]

Also, "Fast rendering" only renders the very first OSD line when checked - I don't think it's supposed to do that?
Confirmed. I have broken it when I have implemented baseline of the font. OK, it's not too difficult to fix.

[DMD]

In the "output" section, try selecting "Maximise SPDIF compatibility".

Thanks for the jewels suggestions
Problem resolved with the release rev1443_20070829_clsid
But with the release rev1437_20070823_clsid_sse_icl10 doesn't work because the selection doesn't exist "Maximize SPDIF Compatibility" :confused:

rev1437_20070823_clsid_sse_icl10
http://img525.imageshack.us/img525/8320/1437xv4.jpg (http://imageshack.us)

rev1443_20070829_clsid
http://img405.imageshack.us/img405/1920/1443ks9.jpg (http://imageshack.us)

[cbemoore]

Thanks for the jewels suggestions
Problem resolved with the release rev1443_20070829_clsid
But with the release rev1437_20070823_clsid_sse_icl10 doesn't work because the selection doesn't exist "Maximize SPDIF Compatibility" :confused:

Exactly.

Rev 1381 works correctly, but doesn't support high-bitrate DTS streams.

Rev 1437 supports high-bitrate DTS streams, but this makes changes to the SPDIF signal which some receivers can't handle.

Therefore, rev 1443 contains a checkbox (maximise SPDIF compatibility) which allows you to toggle between the original behaviour (which doesn't support high-bitrate DTS streams) and the updated behaviour (which doesn't work with all receivers).

Does that make sense?

[Ginsonic]

Hi guys,

I see a strange effect, when I switch to german language in ffdshow. After restarting the player and displaying ffdshow's filter properties, the property window has shrunk on the right side and many items are no longer acessible (window is not resizable). After switching back to english language and restarting the player again, everything is allright.
BTW: Many thanks for Your marvellous work !

[DMD]

@ cbemoore

A release is needed that is compatible with all the receivers, it doesn't have sense install a release for every model of receiver.