Media Player Classic Home Cinema (MPC-HC) - DXVA!

[betaking]

命令行 正在创建临时文件“C:\Users\BetaKing\AppData\Local\Temp\RSP00000738561336.rsp”，其内容为
[
/O2 /I "..\..\include" /I "..\filters\BaseClasses" /I "..\..\include\dx" /I "D:\Program Files\Microsoft DirectX SDK (June 2010)\Include" /D "WIN32" /D "NDEBUG" /D "_LIB" /D "_SCL_SECURE_NO_WARNINGS" /D "WINVER=0x0600" /D "_UNICODE" /D "UNICODE" /FD /EHsc /MT /arch:SSE /Fo"C:\mpc-hc\bin\obj\Release_Win32\SubPic\\" /Fd"C:\mpc-hc\bin\obj\Release_Win32\SubPic\\vc90.pdb" /W3 /c /Zi /TP /MP

".\CoordGeom.cpp"

".\SubPicQueueImpl.cpp"

".\SubPicProviderImpl.cpp"

".\SubPicImpl.cpp"

".\SubPicAllocatorPresenterImpl.cpp"

".\MemSubPic.cpp"

".\DX9SubPic.cpp"

".\DX7SubPic.cpp"
]
正在创建命令行“cl.exe @C:\Users\BetaKing\AppData\Local\Temp\RSP00000738561336.rsp /nologo /errorReport:queue”
正在创建临时文件“C:\Users\BetaKing\AppData\Local\Temp\RSP00000838561336.rsp”，其内容为
[
/O2 /I "..\..\include" /I "..\filters\BaseClasses" /I "..\..\include\dx" /I "D:\Program Files\Microsoft DirectX SDK (June 2010)\Include" /D "WIN32" /D "NDEBUG" /D "_LIB" /D "_SCL_SECURE_NO_WARNINGS" /D "WINVER=0x0600" /D "_UNICODE" /D "UNICODE" /FD /EHsc /MT /arch:SSE /Yc"stdafx.h" /Fp"C:\mpc-hc\bin\obj\Release_Win32\SubPic\\SubPic.pch" /Fo"C:\mpc-hc\bin\obj\Release_Win32\SubPic\\" /Fd"C:\mpc-hc\bin\obj\Release_Win32\SubPic\\vc90.pdb" /W3 /c /Zi /TP /MP

".\stdafx.cpp"
]
正在创建命令行“cl.exe @C:\Users\BetaKing\AppData\Local\Temp\RSP00000838561336.rsp /nologo /errorReport:queue”
正在创建临时文件“C:\Users\BetaKing\AppData\Local\Temp\RSP00001A38561336.rsp”，其内容为
[
/OUT:"C:\mpc-hc\bin\lib\Release_Win32\\SubPic.lib" /LIBPATH:"D:\Program Files\Microsoft DirectX SDK (June 2010)\Lib\x86" /IGNORE:4221 /IGNORE:4006 d3d9.lib;d3dx9.lib

"..\..\bin\obj\Release_Win32\SubPic\CoordGeom.obj"

"..\..\bin\obj\Release_Win32\SubPic\DX7SubPic.obj"

"..\..\bin\obj\Release_Win32\SubPic\DX9SubPic.obj"

"..\..\bin\obj\Release_Win32\SubPic\MemSubPic.obj"

"..\..\bin\obj\Release_Win32\SubPic\stdafx.obj"

"..\..\bin\obj\Release_Win32\SubPic\SubPicAllocatorPresenterImpl.obj"

"..\..\bin\obj\Release_Win32\SubPic\SubPicImpl.obj"

"..\..\bin\obj\Release_Win32\SubPic\SubPicProviderImpl.obj"

"..\..\bin\obj\Release_Win32\SubPic\SubPicQueueImpl.obj"
]
正在创建命令行“lib.exe @C:\Users\BetaKing\AppData\Local\Temp\RSP00001A38561336.rsp /NOLOGO”
输出窗口 正在编译...
stdafx.cpp
正在编译...
CoordGeom.cpp
SubPicQueueImpl.cpp
SubPicProviderImpl.cpp
SubPicImpl.cpp
SubPicAllocatorPresenterImpl.cpp
MemSubPic.cpp
DX9SubPic.cpp
DX7SubPic.cpp
正在创建库...
LINK : fatal error LNK1181: 无法打开输入文件“d3d9.lib;d3dx9.lib”
结果 生成日志保存在“file://C:\mpc-hc\bin\obj\Release_Win32\SubPic\BuildLog.htm”
SubPic - 1 个错误，0 个警告
VS2008SP1 CAN NOT compiled last svn

[nevcairiel]

d3dx9.lib should not be used anymore. Its been deprecated with the DX SDK Update February 2005 (over 5 years ago!). Instead, one should load the required functions manually from the d3dx9_xx.dlls as needed, as its done in all other parts of the DX code in the renderers!

This change needs to be reverted or fixed.

JanWillem really needs to pay more attention.

[betaking]

build 3045 is ok!

[nevcairiel]

The change in 3048 caused it. I'll fix it when i get home, unless someone else does it first.

[betaking]

The change in 3048 caused it. I'll fix it when i get home, unless someone else does it first.

:thanks:

[betaking]

build last 3050 no ploberm! to:nevcairiel thanks!

[clsid]

Has this also fixed the black screen issues with older ATI drivers? Can anyone who had that problem confirm?

[Ger]

d3dx9.lib should not be used anymore. Its been deprecated with the DX SDK Update February 2005 (over 5 years ago!). Instead, one should load the required functions manually from the d3dx9_xx.dlls as needed, as its done in all other parts of the DX code in the renderers!

This change needs to be reverted or fixed.

JanWillem really needs to pay more attention.

The change in 3048 caused it. I'll fix it when i get home, unless someone else does it first.

Sigh. Thanks for cleaning that part up in 3050, but 3048 also broke "Save Image..." and "Save Thumbnails..." with default 8-bit surface on ATI. Nvidia unaffected as usual.

xhmikosr build 3046 working fine.
xhmikosr build 3049 and 3050 broken:
GetDIB failed, hr = 8876086c

Has this also fixed the black screen issues with older ATI drivers? Can anyone who had that problem confirm?

Casshern, are you around? Since using X8R8G8B8 fixed the "other" black screen issue I'm actually quite hopeful.

[nevcairiel]

Sigh. Thanks for cleaning that part up in 3050, but 3048 also broke "Save Image..." with default 8-bit surface on ATI. Nvidia unaffected as usual.

xhmikosr build 3046 working fine.
xhmikosr build 3049 and 3050 broken:
GetDIB failed, hr = 8876086c

I knew i should've reverted it completly. Can't trust these commits.
I hope its fixed in 3051, but i have no idea. If it isn't, i'm just going to revert it completely.

[v0lt]

For what the player needed A8R8G8B8? Why can not always use X8R8G8B8? :confused:

[Ger]

I knew i should've reverted it completly. Can't trust these commits.
I hope its fixed in 3051, but i have no idea. If it isn't, i'm just going to revert it completely.

:thanks: Appreciate it. At least it won't be in the trunk for over 200 revisions and 3 months like the broken stuff in 2837.

I'll test 3051 when someone compiles it.

[Casshern]

Casshern, are you around? Since using X8R8G8B8 fixed the "other" black screen issue I'm actually quite hopeful.

Unfortunately, i just updated to catalyst 11.4 unofficial beta (with the 99% load bug fixed) - so i cannot test anymore, as with this catalyst everything works. But its likely that change of surface type would have fixed it for older drivers.

[Ger]

I hope its fixed in 3051, but i have no idea. If it isn't, i'm just going to revert it completely.

I'll test 3051 when someone compiles it.

Tested xhmikosr build 3052 and your fix worked. Both "Save Image" and "Save Thumbnails" are working again. So no need for a full revert then.

[suanm]

Why dassent anyone want or like built-in second subtitle function in mpc-hc?

[ceb]

I use the BE Mod but I don't follow the instructions that ask to change some video decoder settings (the following image). To me everything seems to work right, so what are those instructions for?

http://www.xvidvideo.ru/images/stories/my_pic/new_00005.png

P.S.
Now DXVA works for me, so a fix between 1.5.2.3045 and 1.5.2.3052 solved the problem. Anyway I'm disabling it since my CPU is powerful enough and I've also read that DXVA can cause some artifacts.

[Ger]

I don't think you should interpret those screenshots as instructions. They are just there to show the differences from the regular trunk builds. If you don't need all the additional options, there is no need to use them.

[Damien147]

Something I always forget to mention,after unchecking ''remember last window size'' in options-->player-->history the window doesn't scale to the native res of the video

[ceb]

I don't think you should interpret those screenshots as instructions. They are just there to show the differences from the regular trunk builds. If you don't need all the additional options, there is no need to use them.
OK, thanks! :)

[IanD]

Recently re-installed XP and upgraded to MPC 3043 from a much earlier version and attempted to play a wmv (wmv3 properties), but MPC says it can't render the file.

I have the wmv123 option ticked in the decoder.

It works in an old version of VLC.

Any particular reason why no render in MPC, what am I missing? Normally MPC just works.

[namaiki]

Quite recently someone had a similar/same issue and had to download wmfdist64.exe / wmfdist.exe to get wmv to work.

[IanD]

Quite recently someone had a similar/same issue and had to download wmfdist64.exe / wmfdist.exe to get wmv to work.
Thanks for the memory trigger: I remembered that I had previously installed the WMP11SDK and that fixed things.

However, why is there a wmv1/2/3 decoder option in MPC if it doesn't actually decode wmv3?

[namaiki]

Thanks for the memory trigger: I remembered that I had previously installed the WMP11SDK and that fixed things.

However, why is there a wmv1/2/3 decoder option in MPC if it doesn't actually decode wmv3?

Apparently the WMP thing was required for the splitter, not decoder.

[ryrynz]

I occasionally have MPC pause during video playback. I stream over a wireless network (line of sight 300Mb/s DD-WRT and Tomato firmwares used with no difference) along with ffdshow decoding and post-processing along with madVR. Before I attempt to troubleshoot it I would like to know if anyone has experienced this before? Once I press play the video continues normally.
There's obviously a fairly broad range of possible problems so I'm only looking to see if anyone has experienced this issue and what they found to be the cause, thanks.

[korsan]

I have media files with srt subtitles on a network drive.
I can play them when I connect through windows explorer \\192.168.2.xxx\.
But the problem is that Mplayerc HC cannot detect the srt subtitles automatically when the files are on the network drive. This problem does not occur with Vlc player.
Or do I miss a setting somewhere in mpc-hc?

[JanWillem32]

I've optimized the final pass shader for some very big rendering efficiency improvements for the color management's interpolation and all ditherers.
I've also added file imports+exports to the color management's 3DLUT. Writing the files to the hard disk is very slow, as I set it to prevent fragmentation, but loading it is very fast.
I've made the file naming method (nearly) human readable, please let me know what you think of it. I can also convert the parameters from the "yyy=zzz" descriptions to the indexed numbers for shorter file names.
I didn't add date/revision checking for the installed color profile yet, so delete the 3DLUT files after recalibration manually, and I advise caution with usage on systems with multiple calibrated displays.
As usual, the files are in my folder, along with the source code. I used revision 3052 as a base.

[Tiduz]

all *.TS files instantly crash for me now on .3052 :eek:

[molitar]

What is the latest recommended version for MPC-Home now? Currently using 1.4.2677.0 which I know is quite old now.

[LigH]

http://sourceforge.net/projects/mpc-hc/ offers MPC-HomeCinema.1.5.1.2903.x86.exe as latest more or less "stable" build.

Those at http://www.xvidvideo.ru/media-player-classic-home-cinema-x86-x64/ are more "beta", so I could not recommend any version there more or less than the one from SourceForge.

[molitar]

Thanks LigH

[kennae]

Been using the latest builds for maybe an year now and never had any real problems. So you should test builds from xvideo.ru if you arent in a hurry.

[burfadel]

Been using the latest builds for maybe an year now and never had any real problems. So you should test builds from xvideo.ru if you arent in a hurry.

Its xvidvideo.ru

xvideo.ru looks to be a completely 'different' um, 'media' oriented website...!

[kennae]

Its xvidvideo.ru

xvideo.ru looks to be a completely 'different' um, 'media' oriented website...!

Haha, sorry. Good catch!

[Trey_S]

Hello.
I've tried to get HD video from a Canon DSLR to play with HD-acceleration in MPC-HC but I've so far never gotten it to work. All i get is a black screen when dxva is enabled.

The moveis are H.264? in .mov-containers. Tried updating to the latest nvidia drivers (have a 8600m gt card). Had an older MPC-HC version at first which I updated to the version on the sourceforge page but still the same problem.

Are these videos incompatible with HD-acceleration or am i doing something else wrong? Never had any problems with other HD-videos in DXVA-mode...

Links to video samples (scroll down on the pages):
http://www.dpreview.com/reviews/canoneos600d/page15.asp
http://www.dcresource.com/reviews/canon/eos_rebel_t2i-review/using

Anyone else have the same problem?
Tried again with a clean update of the video drivers to the latest verison, now 270.61, installed the latest mpc-hc and tried with all default settings (EVR custom pres). But still nothing. And other videos in h264 seems to work just fine in dxva mode.

[Ger]

DXVA with boatpan2.mov works for me (ATI).

Maybe you're using the QuickTime framework for mov. Go to Options->Formats, scroll down and change the "QuickTme file" setting to "DirectShow" in the right column (if it says "QuickTime" there).

[Mercury_22]

@ bobdynlan your "MPCHC_SWSCALER_MOD" is crashing MPC-HC with any MPEG2 (i) files
@ alexins please keep only the GUI patch in "BE Mod" and remove the MPCHC_SWSCALER_MOD until bobdynlan can fix it

[Aleksoid1978]

@ bobdynlan your "MPCHC_SWSCALER_MOD" is crashing MPC-HC with any MPEG2 (i) files
@ alexins please keep only the GUI patch in "BE Mod" and remove the MPCHC_SWSCALER_MOD until bobdynlan can fix it

Just disable MPEG2 DXVA internal decoder.

[Mercury_22]

Just disable MPEG2 DXVA internal decoder.

It happens only in BE Mod!
But there is another problem with both official (SVN) version and BE Mod version: the internal MPEG-2 DXVA decoder never gets used only the non DXVA MPEG-2 decoder it's used and if I disable it MPC-HC it's using Microsoft's DTV-DVD Video Decoder (not in my "External Filters" list and with 0x005FFFFF MERIT_NORMAL - 1) although I don't have any video decoders in MPC-HC's "External Filters" list

[nevcairiel]

According to your signature, you have a ATI HD4670 - those do not support DXVA for MPEG-2. Neither do the 5xxx series. The 6xxx series claim to support it, but i have no conclusive evidence that it works in MPC-HC.

[bobdynlan]

@Mercury_22, what would be the point of the stable builds without these issues... :rolleyes:

Anyway, the crash = my fault, as updating the libswscale alone was not enough. But this ffmpeg mpeg2 decoder never played for nvidia cards I've trown at it, until last time I've worked on the whole ffmpeg update - it did play, DXVA-less (on my hw). At least it will have 4.2.2 chroma support that the stand-alone one lacks.

I will finish it when I get back from the countryside escapade. ETA: unknown. :helpful:

[Mercury_22]

According to your signature, you have a ATI HD4670 - those do not support DXVA for MPEG-2. Neither do the 5xxx series. The 6xxx series claim to support it, but i have no conclusive evidence that it works in MPC-HC.
Correct !
Sorry I've forgot no VLD for me :mad: only IDCT :stupid:

But the crash is still there in BE MOD (with MPEG-2 DXVA enable but not used) and it isn't in the official (SVN) version:p

@Mercury_22, what would be the point of the stable builds without these issues... :rolleyes:

Anyway, the crash = my fault, as updating the libswscale alone was not enough. But this ffmpeg mpeg2 decoder never played for nvidia cards I've trown at it, until last time I've worked on the whole ffmpeg update - it did play, DXVA-less (on my hw). At least it will have 4.2.2 chroma support that the stand-alone one lacks.

I will finish it when I get back from the countryside escapade. ETA: unknown. :helpful:

"No problem" :rolleyes:we are waiting for you:p:)

[chapas]

Is there any support for ATI HD6450 and its DVXA support for Divx and MPEG-4 decoder in MPC-HC? Or it involves some modification of the code?

Thanks.

[Fadeout]

Can one have some control on the buffering/loading of files? On a computer I notice stutters at certain times, but if I rewind and replay the stutter is gone, so it's something momentary.

Is there some way to prevent this? Similar problem at the start of the video, it needs a few seconds to settle down and stop stuttering.

[JanWillem32]

@chapas: An MPEG-4 part 2 for DXVA video decoder isn't available yet for integration. If a developer writes one, it can probably be integrated in the future.

@Fadeout: http://forum.doom9.org/showthread.php?p=1487946#post1487946
That's a read-ahead buffer. I believe it's a mixed task for splitters and the main program to build up a buffer to memory. I still don't know what part in the program's code sets the buffer size. It's indeed annoying with high bitrate videos to have a rather small buffer. Blu-ray allows up to 48 megabit per second for transport, for example. A minor interruption while reading at such speed can deplete a small buffer fast.
Note that it's different from the renderer's buffer. That setting is pretty useless. There's no use for more than 3 textures in the buffer, really. This main texture buffer on the GPU is duplicated and linked in system memory. A large texture buffer only uses up memory twice and it takes up effort to keep this buffer controlled. So, I'm thinking of removing this setting from the options screen.

[8GPMPY]

Hello everybody,

I desperately waited 5 days to be able to post here, because after a lot of Google and forum search you seem to be the only experts.

I moved from Windows XP 32-bit to Windows 7 64-bit on the same hardware and now I have the problem that DXVA is working, but lagging. It does not reach the framerate of the movie.

Is it ok to discuss the problem here or should I go for a seperate thread?

Thanks in advance
8GPMPY

[janos666]

JanWillem32: Something is wrong with the dfr3052r version. It has to be some kind of interpolation or clipping problem because ~100% white areas show some coloization artfacts with the absolute colorimetric rendering intent.

[JanWillem32]

@janos666: That's an experimental one I made to test the 16-bit unsigned integer format. I don't exactly know what the problem is, as I expected it to be enough to just take the base 32-bit floating point input data (floating point interval [0, 1], very unlikely to have any undershoots or overshoots in this range), multiply it by 2^16-1 = 65535 and use the standard type-casting to 16-bit unsigned integer (integer interval [0, 65535]). I already generated a memory overshoot-type profile, opened up the profile with a hex editor and verified that the raw data is indeed without offset and ends after the last color sequence, so saving the profiles works properly. Also, the Alpha value in all 3 types of profiles get the correct hex value. So, in the 16-bit unsigned integer profiles a sequence will always be "hh hh hh hh hh hh FF FF" {BGRA}.
Also, isn't it normal with absolute color intent to have a RGB ratio of 97% : 99% : 100% (for example) to display encoded 100% white, without changing it to the whitepoint of the screen (a ratio of 1 : 1 : 1)? It has to sacrifice something in the top RGB ranges to get the correct RGB mix of the encoded whitepoint for encoded 100% white to display.
Anyway, thanks for testing. I'll try out lCMS with direct 16-bit unsigned integer output. I hope that will work better, but in theory, it shouldn't matter.

@8GPMPY: What's your hardware, and what did you use before?

[chapas]

8GPMPY:

I have that problem sometimes with AMD 785G chipset. Sometimes the video lags like hell, but when that happens, even non-DXVA playback lags.

[janos666]

Yes, that would be normal. But I can see vivid magenta and cyan colors on white clothes.

I also played with the 16-bit integer texture format back in the days. It didn't work well for me.

[HoP]

EVR CP doesn't work. I get a black screen with only audio
http://forum.doom9.org/showthread.php?t=160588

i have this problem,too.
anyone can help??
tnx in advance

[JanWillem32]

@janos666: With WP Adapt state to "None", switching between Absolute and the three other intent types is quite a difference. No matter fp32, fp16 or ui16 profiles, all white gets a heavy magenta tint on absolute rendering intent on my monitor. My monitor is set up for general production at the moment, with a WP of x = .314, y = .351 and an absolute gamma of 2.6. So, I'm not surprised by a heavy colorized white output with an absolute intent.
How well do the regular fp32 and fp16 formats work with dfr3052 for you? While on this topic, is loading the profiles from the hard disk fast enough for you?

@HoP: Even I have this problem. I can only replicate it if I use a 32-bit program on a 64-bit system, and it happens with all internal renderers, even overlay fails. I also don't know what causes it.

[janos666]

What software did you use to create your ICM profile?
My WP is close to 6300K as well as yours, so it should work fine unless your profiler is buggy (like many calibration softwares).

I found that it is not a texture format problem but the bug appears when MPC loads the exported textures. It happens with both (float and int) versions. I have artifacts when I start a movie with a previously saved lut3d file. (It's not a surprise but it's different with float and integer formats.)

So the bug is in the texture export or load code. (I don't know why it appears for me only with the absolute mode.)

[roytam1]

I wonder why there is no new MSVC2008 build of standalone filters.
MSVC2010 build of standalone filters are just too bloat when comparing with MSVC2008 builds.

[nevcairiel]

VS2008 is dead, long live VS2010! There aren't even builds of MPC-HC itself with VS08 anymore. :p

Seriously, file size doesn't matter. It has *no* effect on the performance of the player - only difference is the size of the download and the size on disc.

[CruNcher]

Hi initially i searched for interoperability issues between Lav Splitter and MPC-HCs DXVA decoder (Nvidia VP2 Hardware) and i stumbled upon this Progressive H.264 test sample which goes high wire with MPCs DXVA decoder (works fine with Cyberlink DXVA, Lav CUVID, CoreAVC 2.5.5 DXVA/CUDA,Cuda Video Decoder,Arcsoft DXVA,Mainconcept DXVA and several others) with any splitter (the Decoder tries to playback the bitstream @ higher framerate then it actually is) tested with VMR9 renderless.
I also tried several audio decoder combinations always the same result (ffdshow audio,lav audio,mpc ac3 decoder filter)

http://www.mediafire.com/?lpjkmpf174vidi4

[Aleksoid1978]

Hi initially i searched for interoperability issues between Lav Splitter and MPC-HCs DXVA decoder (Nvidia VP2 Hardware) and i stumbled upon this Progressive H.264 test sample which goes high wire with MPCs DXVA decoder (works fine with Cyberlink DXVA, Lav CUVID, CoreAVC 2.5.5 DXVA/CUDA,Cuda Video Decoder,Arcsoft DXVA,Mainconcept DXVA and several others) with any splitter (the Decoder tries to playback the bitstream @ higher framerate then it actually is) tested with VMR9 renderless.
I also tried several audio decoder combinations always the same result (ffdshow audio,lav audio,mpc ac3 decoder filter)

http://www.mediafire.com/?lpjkmpf174vidi4

Play fine with internal splitter + internal Decoder.

[nevcairiel]

Also plays fine with LAV Splitter and MPC DXVA decoder. MPC-HC reports that its playing it at ~47fps (using EVR-CP), but that must be some display thing, it looks perfectly smooth.

[CruNcher]

Interesting a fix for it was to disable Nvidias telecine in the control panel though only for VMR7 it's fixed (though no other DXVA decoder needed that in any DXVA renderer, with or without VMR9-mixer mode for VMR9 renderless)

Yep disabling Nvidias Telecine fixes it in VMR7 with MPC-HC DXVA but for VMR9 it's still existant (XP SP3 G92 VP2, 270.61 WHQL)

Like Nvidias Telecine is still on on VMR9 (forced, with MPC-HCs DXVA decoder) but disabled in the Control Panel, hmm disabling VMR9-Mixer mode seems to brake that force on VMR9 and MPCs-HCs DXVA Telecine it works now like on VMR7 and Nvidias Telecine setting has no impact anymore with VMR9-Mixer mode disabled @ all (turn on off no effect with VMR9-mixer mode disabled).

Nev could you try to disable Nvidias Telecine and see if that brings the framerate back to normal ?

PS: Disabling VMR9-Mixer mode is no real solution as it disables any Hardware Deinterlacing also :(


Also this is my interoperability test result for .ts (All tests are done with VMR9 Renderless)

So there are 2 .ts based issues with MPCs Video Decoder DXVA and Lav Splitter (sample bellow)

http://www.mediafire.com/download.php?61ynvmy0x47n5b3

HDPVR->Video Parsing enabled->MPC-HC DXVA = works, though unsync
HDPVR->Video Parsing disabled->MPC-HC DXVA = fails, black screen
HDVPR->Video Parsing disabled->LAV CUVID,Cyberlink DXVA = works
HDPVR->Video Parsing enabled ->LAV CUVID,Cyberlink DXVA = works, though unsync
HDPVR->MPC-HC Splitter->MPC-HC DXVA = works, stays sync

2nd problem are these H.264 interlaced streams (sample bellow)

http://www.mediafire.com/download.php?0268ha2ucbp39l9

Interlaced TS Stream->MPC-HC Splitter->MPC-HC DXVA = works
Interlaced TS Stream->LAV Splitter->MPC-HC DXVA = fails (no matter Video Parsing enabled/disabled)
Interlaced TS Stream->LAV Splitter->LAV CUVID,Cyberlink DXVA = works (no matter Video Parsing enabled/disabled)

[HoP]

@HoP: Even I have this problem. I can only replicate it if I use a 32-bit program on a 64-bit system, and it happens with all internal renderers, even overlay fails. I also don't know what causes it.
tnx..but my mpc build is x64 and i'm using win7 x64
i havent this problem with potplayer_x64

[HTPC-User]

Hello!

I don´t know if this is the right place to post my issues and questions regarding to MPC-HC. Hopefully it is. :)

My system: Win7 64 Bit, MPC-HC x86, ffdshow tryouts R3562 32bit, EVR CP renderer.

The bug/problem is:
In MPC-HC 1.5.0.2827 DVD menus are fully functional with EVR CP and activated Direct3D fullscreen to avoid tearing (section output).

I tested MPC-HC 1.5.2.3016, 1.5.2.3045, 1.5.2.3052 and 1.5.2.3056 and it is not working properly: trying to navigate with cursor keys the highlighted option always jumps at least 2 entries forward (or backward), pressing enter either nothing happens or a different option which is not highlighted is activated.

If I first select an item with the mouse, I can use the menu with cursor keys afterwards. :eek:

If I deactivate the Direct3D fullscreen mode, then the menu is fully functional, I can navigate with the cursor keys right from the start!

Any further informations needed to fix that bug?

Are there any plans to integrate the open source direct show DVD navigator as an alternative in the near future? This could also improve the use of madshi`s madvr renderer, too, which has problems with the inbuild navigator.

My last question is subtitle related. Is it possible to configure the MPC with his inbuilt features that only forced subtitles are shown? At the moment I can switch subs on, but then they are shown all the time and not only when they are needed.

:thanks:

[8GPMPY]

@8GPMPY: What's your hardware, and what did you use before?
My hardware is a Gigabyte GA-MA78G-DS3H (rev. 1.0) with 780G chipset and Radeon HD3200 onboard.

My old system, where everything was working flawlessly:

- Windows XP 32-bit
- Catalyst 10.9 32-Bit
- MPC-HC 1.3.1824.0 32-bit

I set MPC-HC to store the settings in an INI file, so I can run it in the new OS. And I can still dual-boot my old XP.

My new system which makes problems:

- Windows 7 64-bit
- Catalyst 11.4 64-bit (also tried 10.9 and 11.3)
- MPC-HC 1.5.0.2827 32-bit (also tried 64-bit and later SVN builds)
- plain install, no codec packs (Yuck!)

I started with a fresh install of MPC-HC and went through this tutorial again:
http://imouto.my/watching-h264-videos-using-dxva/

DXVA was working at once, but the video was lagging.
Displaying the stats shows that DXVA is working, but the actual framerate does not match the framerate of the movie.

Then I installed Haali Media Splitter, to see if it helps. I made the necessary config changes in MPC-HC and the tray icon showed, but still no joy and I uninstalled Haali.

Then I tried CoreAVC with ATI support. The tray icon showed that DXVA was working, but the video was still lagging so I uninstalled CoreAVC.

Then I installed SPlayer, because it it said that it works without configuration. This is true and it shows "GPU Accel. + EVR", but is also lagging.

Then I installed a mini app which shows the GPU load and while DXVA there is full load nearly all the time.

Then I installed the latest VLC, which I working quite ok, but also has lags and full GPU load on 1080p stuff.


I have another machine with the following specs, which works as expected:

- GA-P55-USB3 (rev. 1.0)
- ATI Radeon HD5770
- Windows 7 64-bit
- Catalyst 11.2 64-bit
- MPC-HC 1.5.0.2827 32-bit
- plain install, no codec packs (Yuck!)


Any ideas?
What else could I try?

Thanks in advance!


I have that problem sometimes with AMD 785G chipset. Sometimes the video lags like hell, but when that happens, even non-DXVA playback lags.
Very strange, as I had no problems at all on Windows XP with my setup.

[Qaq]

My hardware is a Gigabyte GA-MA78G-DS3H (rev. 1.0) with 780G chipset and Radeon HD3200 onboard.
I have the same mobo. Tried everything with 3200 and have found it too slow for good video playback though. Upgraded to 5450 finally. Now 5450 gives me perfect video playback with best video renderers and VA deinterlasing. All surround sound formats are supported.
Best options you can use with 3200 are VMR7 windowed renderer for XP and EVR (not EVR CP!) for win7. Also, set Catalyst to Best perfomance. Check all the CCC features carefully, disable all "optimizations".
But better buy ATI 5450 or 6450.

[8GPMPY]

But why has everything worked well under Windows XP without a single bottleneck?

[Thunderbolt8]

I have difficulties with my inbuilt .ass subtitles after reinstalling windows. it seems like that the inbuilt subtitle renderer of mpc-hc is always used and not directvobsub, even though the symbol loads up properly, because the subs look a bit smeared and not really in focus (also am using madvr). I have the directvobsub and also the directvobsub (auto-loading version) directshow filter both set to prefer under external filters. but it seems like thats not enough.

so what am I missing to have the subs being displayed by directvobsub and not the mpc internal renderer?

[Qaq]

But why has everything worked well under Windows XP without a single bottleneck?
Want the "bottleneck" in XP? Try VMR9 renderless or madVR. Video rendering is way slower in XP than in win7 for ATI.

[Hera]

Want the "bottleneck" in XP? Try VMR9 renderless or madVR. Video rendering is way slower in XP than in win7 for ATI.

VMR9 provides DXVA on XP. I cannot fathom why it is anything else but the fastest option.

[ForceX]

I have difficulties with my inbuilt .ass subtitles after reinstalling windows. it seems like that the inbuilt subtitle renderer of mpc-hc is always used and not directvobsub, even though the symbol loads up properly, because the subs look a bit smeared and not really in focus (also am using madvr). I have the directvobsub and also the directvobsub (auto-loading version) directshow filter both set to prefer under external filters. but it seems like thats not enough.

so what am I missing to have the subs being displayed by directvobsub and not the mpc internal renderer?

Options > Playback > Auto-load subtitles. Uncheck the box.

Also, it doesn't seem the subtitles show up if you use ffdshow as a raw video filter.

[Thunderbolt8]

when I untick auto load subtitles I wont have any at all, despite the directvobsub icon.

and yes, Im using ffdshow as raw video filter. but it has always been like that in the past and it somehow worked fine.

edit: seems like I got it running now. actually i forgot to set ffdshow to raw video while I had coreavc as preferred filter, so I guess that made the difference. subs look way better now, thanks.

[Qaq]

VMR9 provides DXVA on XP. I cannot fathom why it is anything else but the fastest option.
Because of very simple logic: fast speed - low quality, good quality (and functionality) - low speed. For example: VMR9r has shaders/subs etc support but low rendering speed. VMR7w has fast rendering speed... and nothing else.

[v0lt]

@HTPC-User
Please check these versions: 2990 (http://www.xvidvideo.ru/media-player-classic-home-cinema-x86-x64/media-player-classic-homecinema-x86-x64-1-5-2-2990.html), 2991 (http://www.xvidvideo.ru/media-player-classic-home-cinema-x86-x64/media-player-classic-homecinema-x86-x64-1-5-2-2991.html).

[JanWillem32]

Sorry I'm responding a bit late, a national holiday was taking up my time.

@janos666: I verified that the exported texture data indeed isn't the problem. The problem is the thread safety. I've verified that all initialization threads are started at least 5 times before rendering of the first frame starts. That does lead to some problems, of course. It's most notable on the 2-pass scaling initialization that requires a sequential setup. With multiple threads starting it, it just fails with an error. I'm familiar with C# methods/instructions that can disable multitreading, but I've had no success with changing the renderer behavior yet (C++ code).
My last display profile was made with Argyll CMS v1.3.2 + dispcalGUI v0.6.7.7, and I've been very satisfied with the results so far with both MPC-HC and the tiny bit of design work I've actually done (coding takes up a lot of my already scarce free time).

@HoP: Do you also have the problem with all renderers? For me, even the overlay renderer fails, which isn't even an internal renderer at all.

@HTPC-User: I have that, too. In D3D exclusive mode without VSync, all keystrokes before the first mouse click are just too sensitive. It also happens if you try to seek by chapters or time, pause/play, change the audio or subtitle streams. Everything is done like as if you tap the key or key combination two or three times. I don't know what causes it. I haven't found a bug report on this item on trac yet, so if you could submit one, I would appreciate it.
For the subtitle part, I'm not familiar with the handling of forced subtitles. For one of the movies in my collection, forced subtitles are present on the BD, but I'm quite glad that those are easily disabled.

@8GPMPY: Desktop composition can be a bit of a strain on the GPU and memory. Also, EVR CP and VMR-9 (renderless) are not light renderers. Even on default settings, everything becomes a problem when displaying subtitles on low end GPUs. (Or even newer ones with animated subtitles.)
I'll be patching a lot of things in due time, but I don't have a perfect solution for you at the moment. Just try to relieve the strain from CPU and GPU by disabling some system items and hope for the best.

[Mercury_22]

I can't build ffmpeg x64 any more ! (revision 3062)
$ buildmingwlibs.sh --update --compile
Downloading MinGW64 crt and headers...
patching file mingw-w64-crt/misc/delayimp.c
patching file mingw-w64-crt/misc/mingw_getsp.S
Hunk #1 FAILED at 22.
1 out of 1 hunk FAILED -- saving rejects to file mingw-w64-crt/misc/mingw_getsp.
S.rej
Please help !

[komisar]

:) Mercury_22, mingw-w64 commit half of this patch... patch only delayimp.c

[nevcairiel]

I still wonder why this was necessary in the first place. I can build ffmpeg without any mingw hacking for my LAV Splitter.

[Mercury_22]

:) Mercury_22, mingw-w64 commit half of this patch... patch only delayimp.c

:thanks:

[oddball]

Hey guys. I have an issue. I cannot get MPC-HC (3058) to playback the Birds clip (1080p60fps) from this thread

http://forum.doom9.org/showthread.php?t=159486

in anything other than overlay mode. I use CoreAVC as the decoder. But if I disable it DXVA does not work on that clip. Windows Media Player plays it flawlessly and with hardly any load on the CPU.

If I also playback the 1080p60fps VC1 WMV clip (second one down) from this thread.

http://forum.doom9.org/showthread.php?t=156660

It will not playback smooth either other than in EVR mode when DXVA kicks in. However I get green corruption all over the place. Again if I playback this clip in Windows Media Player it plays back perfectly.

The only clip that plays back smoothly from those 1080p60fps test clips using MPC-HC (even with MadVR as the renderer) is the Girls 1080p60fps clip encoded as h.264 in MKV.

Any ideas? I've tried using Microsoft's DTV decoder and it helps a little but it's still not as fluid as using WMP itself.

EDIT: Also noticed something else. MadVR tries to scale a 1080 clip to 1920x11140 on my 1020x1200 monitor which is weird in itself. I tried changing various scaling options in the Nvidia control panel but it makes no difference. Any idea why it's scaling a genuine 1920x1080p clip (exact dimensions) to 1920x1140? It scaled correctly on my HDTV but still plays back dropping tons of frames.

[VipZ]

This might not be the correct place to post this, but I found a bug in the BE mod version, seeking with the 5sec forward key's aren't working, it will go forward once, then backward again. Seeking backwards works fine.

This is in both x86 and x64 with both MPC-HC and LAV Splitter for MKV.

[hayan]

mpc-hc.exe x86 1.5.2.2990 (XhmikosR)
Total running time: about 2 hrs

Problem: Srceen freeze/crash

Event ID: 4101 Display driver nvlddmkm stopped responding and has successfully recovered.

What you did: Press ← or → key ( "Fast seek" check box is checked)

display mode: Full Srceen (alt-enter)
Media type:H264/AVC (DXVA) (Internal Filter)
EVRCP, EVR Buffers:5
Vsync: VSync and AVSync
GPU control: all unchecked
Resizer: Bilinear
win7 x64 sp1, Fermi driver 266.58

After I rollback to 1.5.2.2976 (XvidVideo.RU)

mpc-hc.exe x86 1.5.2.3023 (XvidVideo.RU)
Total running time: about 2 hrs+

Problem: Srceen freeze/crash

Event ID: 4101 Display driver nvlddmkm stopped responding and has successfully recovered.

What you did: Press ← or → key ( "Fast seek" check box is checked)

display mode: Windows mode
Media type:H264/AVC (DXVA) (Internal Filter)
EVRCP, EVR Buffers:5
Vsync: VSync and AVSync
GPU control: all unchecked
Resizer: Bicubic A=-0.75
win7 x64 sp1, Fermi driver 266.58


back to 1.5.2.2976 (XvidVideo.RU)...

Revision 2988
fixed: AlphaBlt corrupted render state for D3DCREATE_PUREDEVICE devices <-suspicious?


mpc-hc.exe x86 1.5.2.3052 (XvidVideo.RU)
Total running time: about 50 min+

Problem: Srceen freeze/crash

Event ID: 4101 Display driver nvlddmkm stopped responding and has successfully recovered.

What you did: Playing

display mode: Full Srceen (alt-enter)
Media type: mp4 (Internal Filter, H264/AVC (DXVA))
EVRCP, EVR Buffers:5
Vsync: VSync and AVSync
GPU control: all unchecked
Resizer: Bilinear
win7 x64 sp1, DWM(Aero), Fermi driver 270.61


back to 1.5.2.2976 (XvidVideo.RU)...

[oddball]

This is what it looks like when I play the Devil May Cry test video.

http://www.phyrefile.com/image/view/DC8iVJs79YYcn4yD

[CruNcher]

Hey guys. I have an issue. I cannot get MPC-HC (3058) to playback the Birds clip (1080p60fps) from this thread

http://forum.doom9.org/showthread.php?t=159486

in anything other than overlay mode. I use CoreAVC as the decoder. But if I disable it DXVA does not work on that clip. Windows Media Player plays it flawlessly and with hardly any load on the CPU.

If I also playback the 1080p60fps VC1 WMV clip (second one down) from this thread.

http://forum.doom9.org/showthread.php?t=156660

It will not playback smooth either other than in EVR mode when DXVA kicks in. However I get green corruption all over the place. Again if I playback this clip in Windows Media Player it plays back perfectly.

The only clip that plays back smoothly from those 1080p60fps test clips using MPC-HC (even with MadVR as the renderer) is the Girls 1080p60fps clip encoded as h.264 in MKV.

Any ideas? I've tried using Microsoft's DTV decoder and it helps a little but it's still not as fluid as using WMP itself.

EDIT: Also noticed something else. MadVR tries to scale a 1080 clip to 1920x11140 on my 1020x1200 monitor which is weird in itself. I tried changing various scaling options in the Nvidia control panel but it makes no difference. Any idea why it's scaling a genuine 1920x1080p clip (exact dimensions) to 1920x1140? It scaled correctly on my HDTV but still plays back dropping tons of frames.

Wow if i read something like this it makes the impression Microsoft this time did some great work (still on XP), though they had alot of time todo so ;)
Hmm could you try this with PotPlayer look @ your Power Consumption and see if it fixes your playback problems http://forum.doom9.org/showpost.php?p=1497175&postcount=6334 ?
Also does both playback fine with VLC and it's DXVA mode in Vista/7 ?

[oddball]

VLC also gives corruption same as MPC-HC on the Devil May Cry video. Also Birds freezes. This is with video acceleration enabled in VLC. I will try Potplayer in a moment.

[CruNcher]

Nvidia (VPx and Forceware) or AMD/ATI (UVD and Catalyst) ?
Different things i found useful in times debuging MPC-HC is turning of it's Vsync stuff and for Nvidia its Control Panel Telecine stuff, though im not sure how that might cause your problems and you said it started of with 3058 or is a general problem ?

[oddball]

Nvidia GTX560 with latest drivers. PotPlayer plays the Bird clip fine. Still get corruption with Devil May Cry.

BTW Playing back the Birds video with MadVR as renderer is a no show. Continuous dropped frames. Which is odd because the Girls 1080p60fps clip plays absolutely fine with 0 dropped frames.

EDIT: I think the MadVR thing is another issue and hopefully he will look into it. The corruption I wonder if it's a driver problem. Maybe I should go back to the slightly older drivers.

[CruNcher]

Try the Samsung clip it should give you similiar problems :)
The 4 Girls clip (proven to work flawless on VP4 and Win7 with Lav Cuvid) and the Devil May Cry should @ least playback fine with LAV CUVID but that is a workaround no fix for your DXVA problem, might be that the Birds clip and Samsung work fine for you as well with it, but then i wonder how Microsoft avoids these issues in WMP without leaving DXVA :)
If it doesn't work flawless with Lav Cuvid either then i would say the Bird stream is absolutely out of the capabilities of VP2 and VP4 on the parts it stucks bitrate spikes overload the VPx (VBR buffer issue) and then it would be really surprising how it can work flawless in WMP in a DXVA Bitstreaming context, same for PotPlayer. Though i saw allready some crazy things with these 60p streams and that it's not guranteed that if Nvcuvid fails Playing them back DXVA would also depending on the Decoder Renderer combo used, but Birds doesn't want to work over here @ all VP2 same for Cats and 4 Girls works seemy good with 52 fps doesn't reach realtime here (which causes either it to playback @ 52 fps or drop frames depending on the renderer used).

The Devil May Cry works perfect on VP2 with this combo no issues

http://img849.imageshack.us/img849/5338/flawlesscombo.png

[oddball]

So why does Devil May Cry work in WMP with hardly any CPU load, no screen corruption and not in MPC-HC or PotPlayer using EVR? In WMP it plays back flawlessly.

Also the Birds clip plays back smoother in MPC-HC when set to EVR but not quite as smooth as WMP. One other thing. Birds clip freezes when using evr-cp or evr-sync. Oh yeah and MadVR refuses to render the Devil May Cry clip. Hmm.

[CruNcher]

The same issue why IE9/IE10 works better with the GPU and Vista/7 Microsoft hides API stuff as in older times the pay us 10.000 and well tell you something thing ;) , or are just better in how to utilize it efficiently damn they made the OS and the GPU concept (DWM) Driver Model ect ;)

Btw here is such a funny thing of efficiency ;) http://forum.doom9.org/showpost.php?p=1497233&postcount=457

[oddball]

I hate Microsoft now. :P

[nevcairiel]

Its not Microsofts fault, most likely you just have some broken configuration somewhere. Your hardware is perfectly capable to play all that.
I run 270.61 drivers, and everything is just perfect (GTX 570).

btw, the Birds file works perfectly for me, VP4 surely is fast enough for it.

[Mercury_22]

I can't compile ffmpeg x64 using gcc 4.6.0 ...
from ../mingw-w64-crt/gdtoa/dmisc.c:32:
c:/mpc-hc/src/filters/transform/mpcvideodec/ffmpeg_x64/root-i686-pc-mingw32/x86_
64-w64-mingw32/include/float.h:36:2: error: #error The installed header-set does
not support your version of GCC.
c:/mpc-hc/src/filters/transform/mpcvideodec/ffmpeg_x64/root-i686-pc-mingw32/x86_
64-w64-mingw32/include/float.h:37:2: error: #error Please update your header-set
or use GCC 4.5.X.
make[1]: *** [gdtoa/lib64_libmingwex_a-dmisc.o] Error 1
make: *** [all] Error 2
$

Can someone please help me :helpful:

[oddball]

OK I've found that CoreAVC was the problem with the birds clip and MadVR. If I use LAV splitters and CUVID it plays the bird clip without hardly dropping any frames. I can also play the Devil May Cry without screen corruption. However I have another bug that seems to be directly related either to MPC-HC or MadVR 0.61 whereby I get this weird repeating frame effect (It's like it's bouncing between each frome backwards and forwards as it progresses).

[ForceX]

and yes, Im using ffdshow as raw video filter. but it has always been like that in the past and it somehow worked fine.
I figured out that is wasn't working because I had both ffdshow and Directvobsub set to Prefer, but ffdshow should be higher on the list than Dvobsub so video stream passes from decoder to ffdshow to dvobsub to renderer. If dvobsub is higher then video passes from dvobsub to ffdshow to renderer, and subtitles don't show up.

[HoP]

@HoP: Do you also have the problem with all renderers? For me, even the overlay renderer fails, which isn't even an internal renderer at all.

i'm using this configs now:

http://forum.doom9.org/attachment.php?attachmentid=12212&stc=1&d=1304379507

[ageback]

:thanks:
How did you resolve this problem,please?

[Mercury_22]

How did you resolve this problem,please?
Like komisar said I've "patch only delayimp.c " Basically I've modify the "mpchc_Mingw64.patch" to patch only the "delayimp.c" (I've deleted everything after "\ No newline at end of file")

But now won't compile using gcc 4.6.0 (with 4.5.2 it's compiling) ! So ...:confused:

Anyone can help us ? Please :helpful:

P.S. on x86 side everything it's fine with 4.6.0

[betaking]

命令行 正在创建临时文件“C:\Users\BetaKing\AppData\Local\Temp\RSP0000512444516.rsp”，其内容为
[
/O2 /I "..\..\..\..\include" /I "ffmpeg" /I "ffmpeg\libavcodec" /I "ffmpeg\libavcore" /I "ffmpeg\libavutil" /I "..\..\BaseClasses" /I "D:\Program Files\Microsoft DirectX SDK (June 2010)Include" /D "WIN32" /D "NDEBUG" /D "_SCL_SECURE_NO_WARNINGS" /D "WINVER=0x0600" /D "_UNICODE" /D "UNICODE" /FD /EHsc /MT /arch:SSE /Fo"C:\mpc-hc\bin\obj\Release_Win32\MPCVideoDec\\" /Fd"C:\mpc-hc\bin\obj\Release_Win32\MPCVideoDec\\vc90.pdb" /W3 /c /Zi /TP /MP

".\CpuId.cpp"

".\VideoDecOutputPin.cpp"

".\VideoDecDXVAAllocator.cpp"

".\TlibavcodecExt.cpp"

".\MPCVideoDecSettingsWnd.cpp"

".\MPCVideoDecFilter.cpp"

".\MPCFfmpegDec.cpp"

".\ffImgfmt.cpp"

".\DXVADecoderVC1.cpp"

".\DXVADecoderMpeg2.cpp"

".\DXVADecoderH264.cpp"

".\DXVADecoder.cpp"
]
正在创建命令行“cl.exe @C:\Users\BetaKing\AppData\Local\Temp\RSP0000512444516.rsp /nologo /errorReport:queue”
正在创建临时文件“C:\Users\BetaKing\AppData\Local\Temp\RSP0000522444516.rsp”，其内容为
[
/O2 /I "..\..\..\..\include" /I "ffmpeg" /I "ffmpeg\libavcodec" /I "ffmpeg\libavcore" /I "ffmpeg\libavutil" /I "..\..\BaseClasses" /I "D:\Program Files\Microsoft DirectX SDK (June 2010)Include" /D "WIN32" /D "NDEBUG" /D "_SCL_SECURE_NO_WARNINGS" /D "WINVER=0x0600" /D "_UNICODE" /D "UNICODE" /FD /EHsc /MT /arch:SSE /Yc"stdafx.h" /Fp"C:\mpc-hc\bin\obj\Release_Win32\MPCVideoDec\\MPCVideoDec.pch" /Fo"C:\mpc-hc\bin\obj\Release_Win32\MPCVideoDec\\" /Fd"C:\mpc-hc\bin\obj\Release_Win32\MPCVideoDec\\vc90.pdb" /W3 /c /Zi /TP /MP

".\stdafx.cpp"
]
正在创建命令行“cl.exe @C:\Users\BetaKing\AppData\Local\Temp\RSP0000522444516.rsp /nologo /errorReport:queue”
正在创建临时文件“C:\Users\BetaKing\AppData\Local\Temp\RSP0000532444516.rsp”，其内容为
[
/O2 /I "..\..\..\..\include" /I "ffmpeg" /I "ffmpeg\libavcodec" /I "ffmpeg\libavcore" /I "ffmpeg\libavutil" /I "..\..\BaseClasses" /I "D:\Program Files\Microsoft DirectX SDK (June 2010)Include" /D "WIN32" /D "NDEBUG" /D "_SCL_SECURE_NO_WARNINGS" /D "WINVER=0x0600" /D "_UNICODE" /D "UNICODE" /FD /EHsc /MT /arch:SSE /Fo"C:\mpc-hc\bin\obj\Release_Win32\MPCVideoDec\\" /Fd"C:\mpc-hc\bin\obj\Release_Win32\MPCVideoDec\\vc90.pdb" /W1 /c /Zi /TC /MP

".\FfmpegContext.c"
]
正在创建命令行“cl.exe @C:\Users\BetaKing\AppData\Local\Temp\RSP0000532444516.rsp /nologo /errorReport:queue”
正在创建临时文件“C:\Users\BetaKing\AppData\Local\Temp\RSP00005D2444516.rsp”，其内容为
[
/OUT:"C:\mpc-hc\bin\lib\Release_Win32\\MPCVideoDec.lib" /LIBPATH:"C:\mpc-hc\bin\lib\Release_Win32\\" /LIBPATH:"C:\mpc-hc\lib" /IGNORE:4221 libavcodec.lib libgcc.a libmingwex.a

"..\..\..\..\bin\obj\Release_Win32\MPCVideoDec\CpuId.obj"

"..\..\..\..\bin\obj\Release_Win32\MPCVideoDec\DXVADecoder.obj"

"..\..\..\..\bin\obj\Release_Win32\MPCVideoDec\DXVADecoderH264.obj"

"..\..\..\..\bin\obj\Release_Win32\MPCVideoDec\DXVADecoderMpeg2.obj"

"..\..\..\..\bin\obj\Release_Win32\MPCVideoDec\DXVADecoderVC1.obj"

"..\..\..\..\bin\obj\Release_Win32\MPCVideoDec\ffImgfmt.obj"

"..\..\..\..\bin\obj\Release_Win32\MPCVideoDec\FfmpegContext.obj"

"..\..\..\..\bin\obj\Release_Win32\MPCVideoDec\MPCFfmpegDec.obj"

"..\..\..\..\bin\obj\Release_Win32\MPCVideoDec\MPCVideoDecFilter.obj"

"..\..\..\..\bin\obj\Release_Win32\MPCVideoDec\MPCVideoDecSettingsWnd.obj"

"..\..\..\..\bin\obj\Release_Win32\MPCVideoDec\stdafx.obj"

"..\..\..\..\bin\obj\Release_Win32\MPCVideoDec\TlibavcodecExt.obj"

"..\..\..\..\bin\obj\Release_Win32\MPCVideoDec\VideoDecDXVAAllocator.obj"

"..\..\..\..\bin\obj\Release_Win32\MPCVideoDec\VideoDecOutputPin.obj"
]
正在创建命令行“lib.exe @C:\Users\BetaKing\AppData\Local\Temp\RSP00005D2444516.rsp /NOLOGO”
输出窗口 正在编译...
stdafx.cpp
正在编译...
CpuId.cpp
VideoDecOutputPin.cpp
VideoDecDXVAAllocator.cpp
TlibavcodecExt.cpp
MPCVideoDecSettingsWnd.cpp
MPCVideoDecFilter.cpp
MPCFfmpegDec.cpp
ffImgfmt.cpp
DXVADecoderVC1.cpp
DXVADecoderMpeg2.cpp
DXVADecoderH264.cpp
DXVADecoder.cpp
正在编译...
FfmpegContext.c
.\FfmpegContext.c : warning C4819: 该文件包含不能在当前代码页(936)中表示的字符。请将该文件保存为 Unicode 格式以防止数据丢失
正在创建库...
LINK : fatal error LNK1181: 无法打开输入文件“libavcodec.lib”
结果 生成日志保存在“file://C:\mpc-hc\bin\obj\Release_Win32\MPCVideoDec\BuildLog.htm”
MPCVideoDec - 1 个错误，1 个警告
VS2008SP1 CAN NOT compiled last svn

[nevcairiel]

Looks like you just forgot to compile ffmpeg before. Also, your errors are all in chinese. :p

[betaking]

Looks like you just forgot to compile ffmpeg before. Also, your errors are all in chinese. :p

i use build.bat to build! and i see this bat only compile ffmpegX64 not compile ffmpeg x86,and compile svn 3068 is ok! 3069 not compile ffmpeg x86
only one file C:\mpc-hc\bin\obj\Release_Win32\libavcodec_gcc\BuildLog.htm
C:\mpc-hc\bin\lib\Release_x64\libavcodec.lib

[betaking]

生成日志 生成 已启动: 项目: libavcodec_gcc，配置: Release|Win32
命令行 正在创建临时文件“C:\Users\BetaKing\AppData\Local\Temp\BAT00002831722564.bat”，其内容为
[
@echo off

gccbuild.bat

if errorlevel 1 goto VCReportError

goto VCEnd

:VCReportError

echo Project : error PRJ0019: 某个工具从以下位置返回了错误代码: "正在执行生成文件项目操作"

exit 1

:VCEnd
]
正在创建命令行“C:\Users\BetaKing\AppData\Local\Temp\BAT00002831722564.bat”
输出窗口 正在执行生成文件项目操作
make.exe 64BIT=yes -j4
CC libavcodec/aac_ac3_parser.c
CC libavcodec/ac3.c
CC libavcodec/ac3_parser.c
CC libavcodec/ac3dec.c
CC libavcodec/ac3dec_data.c
CC libavcodec/ac3tab.c
CC libavcodec/adpcm.c
CC libavcodec/allcodecs.c
CC libavcodec/avfft.c
CC libavcodec/avpacket.c
CC libavcodec/bitstream.c
CC libavcodec/cabac.c
CC libavcodec/CompilerVersion.c
CC libavcodec/dsputil.c
CC libavcodec/eac3dec.c
CC libavcodec/eac3dec_data.c
CC libavcodec/error_resilience.c
CC libavcodec/faandct.c
CC libavcodec/faanidct.c
CC libavcodec/fft.c
CC libavcodec/flvdec.c
CC libavcodec/golomb.c
CC libavcodec/h263.c
CC libavcodec/h263_parser.c
CC libavcodec/h264.c
CC libavcodec/h263dec.c
CC libavcodec/h264_cabac.c
libavcodec/h264.c: In function 'ff_h264_decode_extradata':
libavcodec/h264.c:820:28: warning: initialization discards 'const' qualifier from pointer target type [enabled by default]
libavcodec/h264.c: In function 'fill_filter_caches':
libavcodec/h264.c:2319:38: warning: initialization from incompatible pointer type [enabled by default]
libavcodec/h264.c:2396:38: warning: initialization from incompatible pointer type [enabled by default]
libavcodec/h264.c:2411:42: warning: initialization from incompatible pointer type [enabled by default]
CC libavcodec/h264_cavlc.c
In file included from libavcodec/h264_dxva.c:25:0,
from libavcodec/h264.c:3187:
libavcodec/dxva.h: At top level:
libavcodec/dxva.h:1175:5: warning: 'NominalRange' is narrower than values of its type [enabled by default]
libavcodec/dxva.h:1176:5: warning: 'VideoTransferMatrix' is narrower than values of its type [enabled by default]
libavcodec/dxva.h:1177:5: warning: 'VideoLighting' is narrower than values of its type [enabled by default]
libavcodec/dxva.h:1178:5: warning: 'VideoPrimaries' is narrower than values of its type [enabled by default]
libavcodec/dxva.h:1179:5: warning: 'VideoTransferFunction' is narrower than values of its type [enabled by default]
CC libavcodec/h264_direct.c
libavcodec/h264_direct.c: In function 'pred_spatial_direct_motion':
libavcodec/h264_direct.c:263:12: warning: assignment from incompatible pointer type [enabled by default]
libavcodec/h264_direct.c:264:12: warning: assignment from incompatible pointer type [enabled by default]
libavcodec/h264_direct.c: In function 'pred_temp_direct_motion':
libavcodec/h264_direct.c:443:12: warning: assignment from incompatible pointer type [enabled by default]
libavcodec/h264_direct.c:444:12: warning: assignment from incompatible pointer type [enabled by default]
CC libavcodec/h264_loopfilter.c
CC libavcodec/h264_parser.c
CC libavcodec/h264_ps.c
CC libavcodec/h264_refs.c
CC libavcodec/h264_sei.c
CC libavcodec/h264idct.c
CC libavcodec/h264dsp.c
CC libavcodec/h264pred.c
CC libavcodec/imgconvert.c
CC libavcodec/huffman.c

[betaking]

libavcodec/imgconvert.c: In function 'av_picture_data_copy':
libavcodec/imgconvert.c:800:19: warning: passing argument 3 of 'av_image_copy' from incompatible pointer type [enabled by default]
./libavcore/imgutils.h:112:6: note: expected 'const uint8_t **' but argument is of type 'uint8_t **'
libavcodec/imgconvert.c: In function 'av_picture_copy':
libavcodec/imgconvert.c:808:19: warning: passing argument 3 of 'av_image_copy' from incompatible pointer type [enabled by default]
./libavcore/imgutils.h:112:6: note: expected 'const uint8_t **' but argument is of type 'uint8_t * const*'
CC libavcodec/intelh263dec.c
CC libavcodec/intrax8.c
CC libavcodec/intrax8dsp.c
CC libavcodec/ituh263dec.c
CC libavcodec/jpegls.c
CC libavcodec/jpeglsdec.c
CC libavcodec/jrevdct.c
CC libavcodec/libamr.c
CC libavcodec/mdct.c
libavcodec/libamr.c: In function 'amr_nb_decode_frame':
libavcodec/libamr.c:166:5: warning: passing argument 2 of 'Decoder_Interface_Decode' discards 'const' qualifier from pointer target type [enabled by default]
libavcodec/amr_float/interf_dec.h:33:6: note: expected 'unsigned char *' but argument is of type 'const uint8_t *'
CC libavcodec/mjpegdec.c
CC libavcodec/mlp_parser.c
CC libavcodec/mjpeg.c
CC libavcodec/mlp.c
CC libavcodec/mlpdsp.c
CC libavcodec/mlpdec.c
CC libavcodec/mpc_helper.c
CC libavcodec/mpeg12.c
CC libavcodec/mpeg12data.c
In file included from libavcodec/mpeg12.c:41:0:
libavcodec/dxva.h:1175:5: warning: 'NominalRange' is narrower than values of its type [enabled by default]
libavcodec/dxva.h:1176:5: warning: 'VideoTransferMatrix' is narrower than values of its type [enabled by default]
libavcodec/dxva.h:1177:5: warning: 'VideoLighting' is narrower than values of its type [enabled by default]
libavcodec/dxva.h:1178:5: warning: 'VideoPrimaries' is narrower than values of its type [enabled by default]
libavcodec/dxva.h:1179:5: warning: 'VideoTransferFunction' is narrower than values of its type [enabled by default]
libavcodec/mpeg12.c: In function 'mpeg_decode_user_data':
libavcodec/mpeg12.c:2128:34: warning: initialization discards 'const' qualifier from pointer target type [enabled by default]
CC libavcodec/mpeg4video.c
CC libavcodec/mpeg4video_parser.c
CC libavcodec/mpeg4videodec.c
CC libavcodec/mpegvideo.c
CC libavcodec/msmpeg4.c
CC libavcodec/msmpeg4data.c
CC libavcodec/nellymoser.c
CC libavcodec/nellymoserdec.c
CC libavcodec/options.c
CC libavcodec/parser.c
CC libavcodec/rv10.c
CC libavcodec/rv30.c
CC libavcodec/rv30dsp.c
CC libavcodec/rv34.c
CC libavcodec/rv40.c
CC libavcodec/rv40dsp.c
CC libavcodec/simple_idct.c
CC libavcodec/sp5xdec.c
CC libavcodec/svq1dec.c
CC libavcodec/svq1.c
libavcodec/sp5xdec.c: In function 'sp5x_decode_frame':
libavcodec/sp5xdec.c:90:5: warning: implicit declaration of function 'av_init_packet' [-Wimplicit-function-declaration]
CC libavcodec/svq3.c
CC libavcodec/utils.c
CC libavcodec/vc1.c
libavcodec/utils.c: In function 'avcodec_encode_audio':
libavcodec/utils.c:539:9: warning: passing argument 4 of 'avctx->codec->encode' discards 'const' qualifier from pointer target type [enabled by default]
libavcodec/utils.c:539:9: note: expected 'void *' but argument is of type 'const short int *'
libavcodec/utils.c: In function 'avcodec_encode_video':
libavcodec/utils.c:556:9: warning: passing argument 4 of 'avctx->codec->encode' discards 'const' qualifier from pointer target type [enabled by default]
libavcodec/utils.c:556:9: note: expected 'void *' but argument is of type 'const struct AVFrame *'
libavcodec/utils.c: In function 'avcodec_decode_video':
libavcodec/utils.c:571:5: warning: implicit declaration of function 'av_init_packet' [-Wimplicit-function-declaration]
libavcodec/utils.c:572:16: warning: assignment discards 'const' qualifier from pointer target type [enabled by default]
libavcodec/utils.c: In function 'avcodec_decode_audio2':
libavcodec/utils.c:616:16: warning: assignment discards 'const' qualifier from pointer target type [enabled by default]
CC libavcodec/vc1dec.c
libavcodec/vc1dec.c: In function 'vc1_decode_init':
libavcodec/vc1dec.c:3055:24: warning: initialization discards 'const' qualifier from pointer target type [enabled by default]
CC libavcodec/vc1data.c
CC libavcodec/vc1dsp.c
libavcodec/vc1dsp.c: In function 'ff_vc1dsp_init':
libavcodec/vc1dsp.c:633:39: warning: assignment from incompatible pointer type [enabled by default]
libavcodec/vc1dsp.c:650:39: warning: assignment from incompatible pointer type [enabled by default]
CC libavcodec/vp3.c
CC libavcodec/vp3dsp.c
libavcodec/vp3.c: In function 'theora_decode_init':
libavcodec/vp3.c:2217:31: warning: passing argument 1 of 'ff_split_xiph_headers' discards 'const' qualifier from pointer target type [enabled by default]
libavcodec/xiph.h:39:5: note: expected 'uint8_t *' but argument is of type 'const uint8_t *'
CC libavcodec/vp5.c
CC libavcodec/vp56.c
CC libavcodec/vp56data.c
CC libavcodec/vp56dsp.c
CC libavcodec/vp56rac.c
CC libavcodec/vp6.c
CC libavcodec/vp6dsp.c
CC libavcodec/vp8.c
CC libavcodec/vp8dsp.c
CC libavcodec/w32thread.c
CC libavcodec/wmv2dec.c
CC libavcodec/wmv2.c
CC libavcodec/xiph.c
CC libavcodec/amr_float/interf_dec.c
CC libavcodec/amr_float/sp_dec.c
CC libavcodec/x86/fdct_mmx.c
CC libavcodec/x86/dsputil_mmx.c
CC libavcodec/x86/fft.c
CC libavcodec/x86/fft_3dn.c
In file included from libavcodec/x86/fft_3dn.c:23:0:
libavcodec/x86/fft_3dn2.c: In function 'ff_imdct_calc_3dn':
libavcodec/x86/fft_3dn2.c:157:5: warning: use of memory input without lvalue in asm operand 4 is deprecated [enabled by default]
CC libavcodec/x86/fft_3dn2.c
CC libavcodec/x86/fft_sse.c
libavcodec/x86/fft_sse.c: In function 'ff_imdct_calc_sse':
libavcodec/x86/fft_sse.c:87:5: warning: use of memory input without lvalue in asm operand 4 is deprecated [enabled by default]
CC libavcodec/x86/h264_intrapred_init.c
libavcodec/x86/fft_3dn2.c: In function 'ff_imdct_calc_3dn2':
libavcodec/x86/fft_3dn2.c:157:5: warning: use of memory input without lvalue in asm operand 4 is deprecated [enabled by default]
CC libavcodec/x86/idct_mmx.c
CC libavcodec/x86/h264dsp_mmx.c
CC libavcodec/x86/idct_mmx_xvid.c
CC libavcodec/x86/idct_sse2_xvid.c
CC libavcodec/x86/mlpdsp.c
CC libavcodec/x86/mpegvideo_mmx.c
libavcodec/x86/mlpdsp.c:44:36: warning: taking address of expression of type 'void' [enabled by default]
libavcodec/x86/mlpdsp.c:44:56: warning: taking address of expression of type 'void' [enabled by default]
libavcodec/x86/mlpdsp.c:45:36: warning: taking address of expression of type 'void' [enabled by default]
libavcodec/x86/mlpdsp.c:45:56: warning: taking address of expression of type 'void' [enabled by default]
libavcodec/x86/mlpdsp.c:46:36: warning: taking address of expression of type 'void' [enabled by default]
libavcodec/x86/mlpdsp.c:46:56: warning: taking address of expression of type 'void' [enabled by default]
libavcodec/x86/mlpdsp.c:47:36: warning: taking address of expression of type 'void' [enabled by default]
libavcodec/x86/mlpdsp.c:47:56: warning: taking address of expression of type 'void' [enabled by default]
libavcodec/x86/mlpdsp.c:48:36: warning: taking address of expression of type 'void' [enabled by default]
libavcodec/x86/mlpdsp.c:49:36: warning: taking address of expression of type 'void' [enabled by default]
libavcodec/x86/mlpdsp.c:49:56: warning: taking address of expression of type 'void' [enabled by default]
libavcodec/x86/mlpdsp.c:50:36: warning: taking address of expression of type 'void' [enabled by default]
libavcodec/x86/mlpdsp.c:50:56: warning: taking address of expression of type 'void' [enabled by default]
libavcodec/x86/mlpdsp.c:51:36: warning: taking address of expression of type 'void' [enabled by default]
CC libavcodec/x86/simple_idct_mmx.c
CC libavcodec/x86/vc1dsp_mmx.c
CC libavcodec/x86/vp56dsp_init.c
CC libavcodec/x86/vp8dsp-init.c
CC libavcore/audioconvert.c
CC libavcore/avcore_utils.c
CC libavcore/imgutils.c
CC libavcore/samplefmt.c
CC libavutil/avstring.c
CC libavutil/cpu.c
libavutil/avstring.c: In function 'av_stristr':
libavutil/avstring.c:57:9: warning: return discards 'const' qualifier from pointer target type [enabled by default]
libavutil/avstring.c:61:13: warning: return discards 'const' qualifier from pointer target type [enabled by default]
CC libavutil/crc.c
CC libavutil/intfloat_readwrite.c
CC libavutil/eval.c
CC libavutil/inverse.c
CC libavutil/log.c
CC libavutil/lfg.c
CC libavutil/mathematics.c
CC libavutil/mem.c
CC libavutil/md5.c
CC libavutil/pixdesc.c
CC libavutil/random_seed.c
CC libavutil/rational.c
CC libavutil/x86/cpu.c
CC libswscale/rgb2rgb.c
CC libswscale/swscale.c
CC libswscale/sww32thread.c
In file included from libswscale/../libavutil/internal.h:37:0,
from libswscale/sww32thread.c:33:
./config.h:53:0: warning: "ASMALIGN" redefined [enabled by default]
libswscale/config.h:49:0: note: this is the location of the previous definition
libswscale/sww32thread.c: In function 'sws_thread_free':
libswscale/sww32thread.c:85:5: warning: implicit declaration of function 'av_freep' [-Wimplicit-function-declaration]
libswscale/sww32thread.c: In function 'sws_thread_init':
libswscale/sww32thread.c:121:5: warning: implicit declaration of function 'av_mallocz' [-Wimplicit-function-declaration]
libswscale/sww32thread.c:121:6: warning: assignment makes pointer from integer without a cast [enabled by default]
CC libswscale/yuv2rgb.c
YASM: Assembling libavcodec/x86/dsputil_yasm.asm
In file included from libswscale/../libavutil/x86_cpu.h:25:0,
from libswscale/yuv2rgb.c:36:
./config.h:53:0: warning: "ASMALIGN" redefined [enabled by default]
libswscale/config.h:49:0: note: this is the location of the previous definition
YASM: Assembling libavcodec/x86/fft_mmx.asm
In file included from libswscale/../libavutil/internal.h:37:0,
from libswscale/swscale.c:70:
./config.h:53:0: warning: "ASMALIGN" redefined [enabled by default]
libswscale/config.h:49:0: note: this is the location of the previous definition
libswscale/swscale.c: In function 'yuv2packedXinC':
libswscale/swscale.c:742:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale.c:742:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale.c:742:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale.c:778:4: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale.c:778:4: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale.c:778:4: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale.c:793:4: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale.c:793:4: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale.c:793:4: warning: assignment from incompatible pointer type [enabled by default]
In file included from libswscale/swscale.c:939:0:
libswscale/swscale_template.c: In function 'yuv2packed2_C':
libswscale/swscale_template.c:1452:1: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1452:1: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1452:1: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1452:1: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1452:1: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1452:1: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1452:1: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1452:1: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1452:1: warning: assignment from incompatible pointer type [enabled by default]
libswscale/yuv2rgb.c: In function 'yuv2rgb_c_32':
libswscale/yuv2rgb.c:177:9: warning: assignment from incompatible pointer type [enabled by default]
libswscale/yuv2rgb.c:177:9: warning: assignment from incompatible pointer type [enabled by default]
libswscale/yuv2rgb.c:177:9: warning: assignment from incompatible pointer type [enabled by default]
libswscale/yuv2rgb.c:181:9: warning: assignment from incompatible pointer type [enabled by default]
libswscale/yuv2rgb.c:181:9: warning: assignment from incompatible pointer type [enabled by default]
libswscale/yuv2rgb.c:181:9: warning: assignment from incompatible pointer type [enabled by default]
libswscale/yuv2rgb.c:185:9: warning: assignment from incompatible pointer type [enabled by default]
libswscale/yuv2rgb.c:185:9: warning: assignment from incompatible pointer type [enabled by default]
libswscale/yuv2rgb.c:185:9: warning: assignment from incompatible pointer type [enabled by default]
libswscale/yuv2rgb.c:189:9: warning: assignment from incompatible pointer type [enabled by default]
libswscale/yuv2rgb.c:189:9: warning: assignment from incompatible pointer type [enabled by default]
libswscale/yuv2rgb.c:189:9: warning: assignment from incompatible pointer type [enabled by default]
libswscale/yuv2rgb.c: In function 'yuv2rgb_c_16':
libswscale/yuv2rgb.c:234:9: warning: assignment from incompatible pointer type [enabled by default]
libswscale/yuv2rgb.c:234:9: warning: assignment from incompatible pointer type [enabled by default]
libswscale/yuv2rgb.c:234:9: warning: assignment from incompatible pointer type [enabled by default]
libswscale/yuv2rgb.c:238:9: warning: assignment from incompatible pointer type [enabled by default]
libswscale/yuv2rgb.c:238:9: warning: assignment from incompatible pointer type [enabled by default]
libswscale/yuv2rgb.c:238:9: warning: assignment from incompatible pointer type [enabled by default]
libswscale/yuv2rgb.c:242:9: warning: assignment from incompatible pointer type [enabled by default]
libswscale/yuv2rgb.c:242:9: warning: assignment from incompatible pointer type [enabled by default]
libswscale/yuv2rgb.c:242:9: warning: assignment from incompatible pointer type [enabled by default]
libswscale/yuv2rgb.c:246:9: warning: assignment from incompatible pointer type [enabled by default]
libswscale/yuv2rgb.c:246:9: warning: assignment from incompatible pointer type [enabled by default]
libswscale/yuv2rgb.c:246:9: warning: assignment from incompatible pointer type [enabled by default]
libswscale/yuv2rgb.c: In function 'yuv2rgb_get_func_ptr':
libswscale/yuv2rgb.c:470:21: warning: return from incompatible pointer type [enabled by default]
libswscale/yuv2rgb.c:471:21: warning: return from incompatible pointer type [enabled by default]
libswscale/yuv2rgb.c:472:21: warning: return from incompatible pointer type [enabled by default]
libswscale/yuv2rgb.c:473:21: warning: return from incompatible pointer type [enabled by default]
libswscale/yuv2rgb.c:478:21: warning: return from incompatible pointer type [enabled by default]
libswscale/yuv2rgb.c:479:21: warning: return from incompatible pointer type [enabled by default]
libswscale/yuv2rgb.c:480:21: warning: return from incompatible pointer type [enabled by default]
libswscale/yuv2rgb.c:481:21: warning: return from incompatible pointer type [enabled by default]
libswscale/yuv2rgb.c: In function 'yuv2rgb_c_init_tables':

[betaking]

libswscale/yuv2rgb.c:576:20: warning: assignment from incompatible pointer type [enabled by default]
libswscale/yuv2rgb.c:579:17: warning: assignment from incompatible pointer type [enabled by default]
libswscale/yuv2rgb.c:580:17: warning: assignment from incompatible pointer type [enabled by default]
libswscale/yuv2rgb.c:581:17: warning: assignment from incompatible pointer type [enabled by default]
libswscale/yuv2rgb.c:603:20: warning: assignment from incompatible pointer type [enabled by default]
libswscale/yuv2rgb.c:606:17: warning: assignment from incompatible pointer type [enabled by default]
libswscale/yuv2rgb.c:607:17: warning: assignment from incompatible pointer type [enabled by default]
libswscale/yuv2rgb.c:608:17: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c: In function 'yuv2packed1_C':
libswscale/swscale_template.c:1655:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1655:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1655:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1655:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1655:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1655:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1655:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1655:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1655:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1657:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1657:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1657:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1657:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1657:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1657:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1657:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1657:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1657:3: warning: assignment from incompatible pointer type [enabled by default]
In file included from libswscale/swscale.c:970:0:
libswscale/swscale_template.c: In function 'yuv2packed2_MMX':
libswscale/swscale_template.c:1452:1: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1452:1: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1452:1: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1452:1: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1452:1: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1452:1: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1452:1: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1452:1: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1452:1: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c: In function 'yuv2packed1_MMX':
libswscale/swscale_template.c:1655:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1655:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1655:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1655:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1655:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1655:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1655:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1655:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1655:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1657:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1657:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1657:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1657:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1657:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1657:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1657:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1657:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1657:3: warning: assignment from incompatible pointer type [enabled by default]
libavcodec/x86/fft_mmx.asm:74: warning: section flags ignored on section redeclaration
In file included from libswscale/swscale.c:970:0:
libswscale/swscale_template.c: In function 'swScaleI_MMX':
libswscale/swscale_template.c:3126:46: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast]
libswscale/swscale_template.c:3133:50: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast]
In file included from libswscale/swscale.c:980:0:
libswscale/swscale_template.c: In function 'yuv2packed2_MMX2':
libswscale/swscale_template.c:1452:1: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1452:1: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1452:1: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1452:1: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1452:1: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1452:1: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1452:1: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1452:1: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1452:1: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c: In function 'yuv2packed1_MMX2':
libswscale/swscale_template.c:1655:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1655:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1655:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1655:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1655:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1655:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1655:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1655:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1655:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1657:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1657:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1657:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1657:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1657:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1657:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1657:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1657:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1657:3: warning: assignment from incompatible pointer type [enabled by default]
In file included from libswscale/swscale.c:980:0:
libswscale/swscale_template.c: In function 'swScaleI_MMX2':
libswscale/swscale_template.c:3126:46: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast]
libswscale/swscale_template.c:3133:50: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast]
In file included from libswscale/swscale.c:990:0:
libswscale/swscale_template.c: In function 'yuv2packed2_3DNow':
libswscale/swscale_template.c:1452:1: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1452:1: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1452:1: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1452:1: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1452:1: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1452:1: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1452:1: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1452:1: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1452:1: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c: In function 'yuv2packed1_3DNow':
libswscale/swscale_template.c:1655:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1655:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1655:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1655:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1655:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1655:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1655:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1655:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1655:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1657:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1657:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1657:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1657:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1657:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1657:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1657:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1657:3: warning: assignment from incompatible pointer type [enabled by default]
libswscale/swscale_template.c:1657:3: warning: assignment from incompatible pointer type [enabled by default]
In file included from libswscale/swscale.c:990:0:
libswscale/swscale_template.c: In function 'swScaleI_3DNow':
libswscale/swscale_template.c:3126:46: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast]
libswscale/swscale_template.c:3133:50: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast]
YASM: Assembling libavcodec/x86/h264_chromamc.asm
YASM: Assembling libavcodec/x86/h264_deblock.asm
YASM: Assembling libavcodec/x86/h264_idct.asm
YASM: Assembling libavcodec/x86/h264_intrapred.asm
YASM: Assembling libavcodec/x86/h264_weight.asm
YASM: Assembling libavcodec/x86/vc1dsp_yasm.asm
YASM: Assembling libavcodec/x86/vp3dsp.asm
YASM: Assembling libavcodec/x86/vp56dsp.asm
YASM: Assembling libavcodec/x86/vp8dsp.asm
YASM: Assembling libavcodec/x86/x86util.asm
AR ../../../../../bin/lib/Release_x64/libavcodec.lib
结果 生成日志保存在“file://C:\mpc-hc\bin\obj\Release_Win32\libavcodec_gcc\BuildLog.htm”
libavcodec_gcc - 0 个错误，221 个警告

[SamuriHL]

For the love of God use pastebin!!!

[nevcairiel]

:stupid: Use pastebin :eek:

[betaking]

:stupid: Use pastebin :eek:

can you try compiled last svn？？

[SamuriHL]

I already did for x86 a little while ago. I had no problems.

[Mercury_22]

I already did for x86 a little while ago. I had no problems.

Can you compile the x64 ? (with 2010SP1 and 4.6.0 http://xhmikosr.1f0.de/index.php?folder=dG9vbHM=)

[betaking]

16>------ 已启动生成: 项目: libavcodec_gcc, 配置: Release Win32 ------
16>正在执行生成文件项目操作
16>make.exe 64BIT=yes -j4
12>FontInstaller.cpp
why make 64BIT? i use vc compile32BIT!

[SamuriHL]

I haven't tried x64 cause I don't use it. I'm also still using 4.5.2, not 4.6.0. Haven't seen the need to change it as it's working still. I am using 2010SP1, though. I can try the x64 build with what I have if that helps?

[Mercury_22]

I haven't tried x64 cause I don't use it. I'm also still using 4.6.0. Haven't seen the need to change it as it's working still. I am using 2010SP1, though. I can try the x64 build with what I have if that helps?

Yes it helps cause you have 4.6.0 !
I'm just interested if you can compile the ffmpeg x64

Edit : you've edit your post ? you have 4.6.0 or 4.5.2 ?

[SamuriHL]

Yes it helps cause you have 4.6.0 !
I'm just interested if you can compile the ffmpeg x64

See my update. Typo. Doing too many things at once. I DON'T have 4.6.0 is what I meant to type.

[Mercury_22]

See my update. Typo. Doing too many things at once. I DON'T have 4.6.0 is what I meant to type.
:) ok
Still if it's not too much trouble can you build it and let us know how it went ?

[SamuriHL]

Yea, I kicked it off a bit ago. :) Figured why not if it might help.

[SamuriHL]

MPC-HC's compilation started on Tue 05/03/2011-10:10:38.97
and completed on Tue 05/03/2011-10:27:04.80

bin10\MPC-HomeCinema.1.5.2.3069.x64.MSVC2010.exe

So, yea, I built the 64 bit version no problem.

[Mercury_22]

MPC-HC's compilation started on Tue 05/03/2011-10:10:38.97
and completed on Tue 05/03/2011-10:27:04.80

bin10\MPC-HomeCinema.1.5.2.3069.x64.MSVC2010.exe

So, yea, I built the 64 bit version no problem.

Ok thanks!
Let us know when you update to 4.6.0 http://xhmikosr.1f0.de/index.php?folder=dG9vbHM=

[nevcairiel]

This obsession with always upgrading to the latest gcc will only cause you guys problems.

Like it did today. :p

[SamuriHL]

I'm not sure I understand why upgrading to 4.6.0 makes any difference. Is it going to generate some super optimized code that'll run on my vic 20? :D In any case, I grabbed 4.6.0 full and am recompiling the x64 build with it. I, otoh, backed up my WORKING 4.5.2 MSYS dir. ;)

[Mercury_22]

I'm not sure I understand why upgrading to 4.6.0 makes any difference. Is it going to generate some super optimized code that'll run on my vic 20? :D In any case, I grabbed 4.6.0 full and am recompiling the x64 build with it. I, otoh, backed up my WORKING 4.5.2 MSYS dir. ;)

I'm just interested if you can update / patch / compile ffmpeg x64 :thanks:

Also, you just need to backup your x64 libgcc.a and libmingwex.a and you can use them with 4.6.0

[SamuriHL]

I'm just interested if you can update / patch / compile ffmpeg x64 :thanks:

Also, you just need to backup your x64 libgcc.a and libmingwex.a and you can use them with 4.6.0

Um, all I'm doing is a full build of what's in SVN with 4.6.0. I haven't patched or touched anything. And by full build, I mean completely clean. It's the only way I build MPC-HC as I've had problems doing incremental builds in the past and would rather take longer to build it right the first time than fight with it. :) Seems to be building fine so far, but, it's not done. I build on my core2duo laptop, so, it's not especially quick. :)

[SamuriHL]

67>MPCVideoDec.lib(utils.o) : warning LNK4217: locally defined symbol toupper im
ported in function ff_toupper4
67>MPCVideoDec.lib(w32thread.o) : warning LNK4217: locally defined symbol _begin
threadex imported in function avcodec_thread_init
67>MPCVideoDec.lib(sww32thread.o) : warning LNK4217: locally defined symbol _beg
inthreadex imported in function sws_thread_init
67>MPCVideoDec.lib(log.o) : warning LNK4217: locally defined symbol _vsnprintf i
mported in function av_log_default_callback
67>MPCVideoDec.lib(log.o) : warning LNK4217: locally defined symbol __iob_func i
mported in function av_log_default_callback
67>MPCVideoDec.lib(interf_dec.o) : warning LNK4049: locally defined symbol __iob
_func imported
67>MPCVideoDec.lib(sp_dec.o) : warning LNK4049: locally defined symbol __iob_fun
c imported
67>MPCVideoDec.lib(mdct.o) : error LNK2001: unresolved external symbol ___chkstk
_ms
67>MPCVideoDec.lib(mpegvideo.o) : error LNK2019: unresolved external symbol ___c
hkstk_ms referenced in function ff_update_duplicate_context
67>MPCVideoDec.lib(error_resilience.o) : error LNK2001: unresolved external symb
ol ___chkstk_ms
67>MPCVideoDec.lib(dsputil_mmx.o) : error LNK2001: unresolved external symbol __
_chkstk_ms
67>MPCVideoDec.lib(rv34.o) : error LNK2001: unresolved external symbol ___chkstk
_ms
67>c:\dev\projects\mpc-hc\bin10\mpc-hc_x64\mpc-hc64.exe : fatal error LNK1120: 1
unresolved externals
67>
67>Build FAILED.
67>
67>Time Elapsed 00:01:44.96
========== Build: 65 succeeded, 1 failed, 0 up-to-date, 1 skipped ==========


**ERROR: Build failed and aborted!**

Well, there you have it. 4.6.0 = broken. I'm not about to run crazy patches or any other nonsense to get it to build. I see no value in updating to a new compiler. Unless someone can explain to me what the value is other than "but it's newer and shiny." I'm reverting to 4.5.2.

[Mercury_22]

67>MPCVideoDec.lib(utils.o) : warning LNK4217: locally defined symbol toupper im
ported in function ff_toupper4
67>MPCVideoDec.lib(w32thread.o) : warning LNK4217: locally defined symbol _begin
threadex imported in function avcodec_thread_init
67>MPCVideoDec.lib(sww32thread.o) : warning LNK4217: locally defined symbol _beg
inthreadex imported in function sws_thread_init
67>MPCVideoDec.lib(log.o) : warning LNK4217: locally defined symbol _vsnprintf i
mported in function av_log_default_callback
67>MPCVideoDec.lib(log.o) : warning LNK4217: locally defined symbol __iob_func i
mported in function av_log_default_callback
67>MPCVideoDec.lib(interf_dec.o) : warning LNK4049: locally defined symbol __iob
_func imported
67>MPCVideoDec.lib(sp_dec.o) : warning LNK4049: locally defined symbol __iob_fun
c imported
67>MPCVideoDec.lib(mdct.o) : error LNK2001: unresolved external symbol ___chkstk
_ms
67>MPCVideoDec.lib(mpegvideo.o) : error LNK2019: unresolved external symbol ___c
hkstk_ms referenced in function ff_update_duplicate_context
67>MPCVideoDec.lib(error_resilience.o) : error LNK2001: unresolved external symb
ol ___chkstk_ms
67>MPCVideoDec.lib(dsputil_mmx.o) : error LNK2001: unresolved external symbol __
_chkstk_ms
67>MPCVideoDec.lib(rv34.o) : error LNK2001: unresolved external symbol ___chkstk
_ms
67>c:\dev\projects\mpc-hc\bin10\mpc-hc_x64\mpc-hc64.exe : fatal error LNK1120: 1
unresolved externals
67>
67>Build FAILED.
67>
67>Time Elapsed 00:01:44.96
========== Build: 65 succeeded, 1 failed, 0 up-to-date, 1 skipped ==========


**ERROR: Build failed and aborted!**

Well, there you have it. 4.6.0 = broken. I'm not about to run crazy patches or any other nonsense to get it to build. I see no value in updating to a new compiler. Unless someone can explain to me what the value is other than "but it's newer and shiny." I'm reverting to 4.5.2.

OK :thanks:
EDIT: And if it's not too much after you revert to 4.5.2 please delete the x64 libgcc.a and libmingwex.a (in C:\mpc-hc\lib64) and try to rebuild them and let me know if it works cause it seems to me that first time you didn't update them :thanks: for your help

[SamuriHL]

No problem. Just remember that when upgrading to newer compilers, backing up your working config is probably the best idea. :) I still have 4.6.0 extracted in case this gets fixed and there's some reason to move to it, but, I'm sticking with what works for now.

[betaking]

i found why can not compiler last svn!
gccbuild.bat is build ffmpegx86 why this is bat build x64
TITLE "make.exe 64BIT=yes %make_args% %*"
ECHO make.exe 64BIT=yes %make_args% %*
make.exe 64BIT=yes %make_args% %*
EXIT /B
and i Modified below, the re-compile without any problem!
TITLE "make.exe %make_args% %*"
ECHO make.exe %make_args% %*
make.exe %make_args% %*
EXIT /B

[betaking]

gccbuild_2010.bat no problem
TITLE "make.exe VS2010=yes %make_args% %*"
ECHO make.exe VS2010=yes %make_args% %*
make.exe VS2010=yes %make_args% %*
EXIT /B

[hoborg]

Hi
This sample (https://rapidshare.com/files/460405805/Vier-001.mkv) will crash MPC-HC DXVA decoder on ~22s.
There seems to be problem in video itself, but decoder should handle it, not crash i think.

[Ger]

This might not be the correct place to post this, but I found a bug in the BE mod version, seeking with the 5sec forward key's aren't working, it will go forward once, then backward again. Seeking backwards works fine.

This is in both x86 and x64 with both MPC-HC and LAV Splitter for MKV.

Any improvement if you disable "Fast seek (on keyframe)" in Options->Tweaks?

[VipZ]

Any improvement if you disable "Fast seek (on keyframe)" in Options->Tweaks?

Nope, I already tried that option and sadly didn't do anything.

[molitar]

Latest MPC-Home has been modified in the dumbest way! No wonder people can't get DXVA to work with corecodec the person who did it set all the filters to internal and it now does not use any of FFDShow by default! dumb.. so if your having issues.. go back to the older version that works properly and has not been set to use all internal filters.. otherwise your going to be guessing what filters need to be deactivated.

[tetsuo55]

Latest MPC-Home has been modified in the dumbest way! No wonder people can't get DXVA to work with corecodec the person who did it set all the filters to internal and it now does not use any of FFDShow by default! dumb.. so if your having issues.. go back to the older version that works properly and has not been set to use all internal filters.. otherwise your going to be guessing what filters need to be deactivated.Internal filters have been enabled for years. Its only codecpacks that provide modified mpc-hc builds with different default settings.

[robpdotcom]

I thought one of the beauties of MPC-HC was supposed to be that you don't really need any external filters/codecs.:confused:

I guess there are just so many codecpacks that bundle MPC-HC with ffdshow that people assume they are part of the same program.

[molitar]

Default settings for mpc before that were enabled were.

cda
cdxa
directshow media
dvd video title
dvd2avi
fli
flac
flv
realmedia
shoutcast
realaudio
ps2 audio
qt pcm
real video

them were the defaults.. now the new even after resetting everything is all filters enabled which breaks coreavc and ffdshow causing issues with playback. dropped the older version 1.4.2677.0 back and deleted the ini and the filter settings were those above.. deleted ini and dropped newer version 1.5.1.2903 and all filters were checked and ffdshow and coreavc stopped working because mpc was trying to handle all the rendering. Now thing is with haali media splitter dxva mode does not really work in mpc home.. but with the new coreavc I set it to EVR and with haali I get the red coreavc showing dxva is working properly. So to use the newer version I had to go uncheck just about every filter to get it to work properly.

[Octo-puss]

Latest MPC-Home has been modified in the dumbest way! No wonder people can't get DXVA to work with corecodec the person who did it set all the filters to internal and it now does not use any of FFDShow by default! dumb.. so if your having issues.. go back to the older version that works properly and has not been set to use all internal filters.. otherwise your going to be guessing what filters need to be deactivated.
I thought corecodec and ffdshow were two completely different things.

[tetsuo55]

@Molitar: reset the registry settings: http://sourceforge.net/apps/trac/mpc-hc/wiki/Troubleshooting_guidelines

[namaiki]

Default settings for mpc before that were enabled were
...
dropped the older version 1.4.2677.0 back and deleted the ini and the filter settings were those above..

molitar, those are CCCP defaults, not MPC-HC defaults. CCCP's MPC-HC is a custom build.

[Mercury_22]

Since :) Mercury_22, mingw-w64 commit half of this patch... patch only delayimp.c why no one made the update to the SVN ?

P.S. If a patch it's really needed here (http://www.multiupload.com/SSDOCXF4VM) it is (I've just deleted everything after "\ No newline at end of file")

I'm still waiting for some help about compiling the x64 with gcc 4.6.0 (xhmikosr alexins ? :) )

[clsid]

The MSVC2010 builds (and now also MSVC2008 after latest windows updates) of MPC and the standalone filters are about 1.2MB larger than in the past. This is caused by changes that MS made to MFC code that results in a lot of extra (unneeded) code being linked into the builds. That sucks. But fortunately I have been able to implement a fix for this problem, based on this (http://tedwvc.wordpress.com/2010/05/27/how-to-make-small-statically-linked-mfc-exes-in-visual-c-2010/) blog post.

Everyone please test the following build to see if there are any negative side effects:
http://www.mediafire.com/file/apb97r929cy3ltb/mpc_3062_mfcfix.rar

It contains MPC r3062 plus a couple of the standalone filters.

You will notice that MPC is much smaller now. That is partly due to the fix, partly due to the fact that most of the internal filters are not included in this build. I repeat, the missing internal filters is not a bug. It is intentional and unrelated to the MFC fix. I simply don't need/use them.

[madshi]

@clsid, is there an easy way to compile MPC without the internal filters?

[betaking]

TO CLSID.THANKS! if we test ok? can you put this patch to svn?

[roytam1]

The MSVC2010 builds (and now also MSVC2008 after latest windows updates) of MPC and the standalone filters are about 1.2MB larger than in the past. This is caused by changes that MS made to MFC code that results in a lot of extra (unneeded) code being linked into the builds. That sucks. But fortunately I have been able to implement a fix for this problem, based on this (http://tedwvc.wordpress.com/2010/05/27/how-to-make-small-statically-linked-mfc-exes-in-visual-c-2010/) blog post.

Everyone please test the following build to see if there are any negative side effects:
http://www.mediafire.com/file/apb97r929cy3ltb/mpc_3062_mfcfix.rar

It contains MPC r3062 plus a couple of the standalone filters.

You will notice that MPC is much smaller now. That is partly due to the fix, partly due to the fact that most of the internal filters are not included in this build. I repeat, the missing internal filters is not a bug. It is intentional and unrelated to the MFC fix. I simply don't need/use them.

Yeah it look nice to me, thanks.:)

[clsid]

@clsid, is there an easy way to compile MPC without the internal filters?Yes, very easy! See src\apps\mplayerc\internal_filter_config.h

can you put this patch to svn?If no bugs are found and the other devs don't object, then of course it will get committed.

[v0lt]

The problem with the built-in software MPEG-2 decoder. (MPEG-2 video 4:2:2)
http://i039.radikal.ru/1105/09/fa4cda307ecbt.jpg (http://radikal.ru/F/i039.radikal.ru/1105/09/fa4cda307ecb.jpg.html)
sample 1 (http://www.mediafire.com/?u68ko5qreaoxxn6)
422_sample_from_DVB.mpeg (http://samples.multimedia.cx/MPEG2/mpeg422/422_sample_from_DVB.mpeg)
With ffdshow (libavcodec and libmpeg2) is no problem.

[SamuriHL]

Yes, very easy! See src\apps\mplayerc\internal_filter_config.h


Now that rocks! :thanks:

[SamuriHL]

Ok, maybe :stupid: but if I change the header file to 0's instead of 1's for the internal filters, my compile breaks. Am I doing it wrong? The defines are still needed right? And we are supposed to set them to 0 for filters we don't want to include? Sorry, I'm not having the best of days with compiling. :thanks:

[clsid]

1 = include, 0 = don't include

[SamuriHL]

That's what I thought. When I set them all to don't include, the build fails.

[clsid]

Try with at least one enabled. That should fix the zero sized array errors.

[SamuriHL]

Yea, I tried with all the ffmpeg ones enabled. Didn't seem to help. I didn't get a chance to trap the errors to see what was breaking, but, as soon as I put them all back to 1 it built successfully. :confused:

[clsid]

Will commit a fix for it in a moment.

[SamuriHL]

Thank you. That's much appreciated!

[clsid]

Btw, here is another small tweak to reduce size:
Index: src/apps/mplayerc/PPageFileMediaInfo.h
===================================================================
--- src/apps/mplayerc/PPageFileMediaInfo.h (revision 3078)
+++ src/apps/mplayerc/PPageFileMediaInfo.h (working copy)
@@ -22,7 +22,7 @@

#pragma once

-#define USE_MEDIAINFO_STATIC
+//#define USE_MEDIAINFO_STATIC

// CPPageFileMediaInfo dialog

[SamuriHL]

That removes the mediainfo static lib. What functionality does that impact?

[clsid]

MediaInfo is then no longer embedded. Instead you must put mediainfo.dll in the MPC folder if you want internal MediaInfo functionality (File->Properties->MediaInfo).

[SamuriHL]

Oh. I never use that anyway, so, that's ok. Nice. :) I discovered that yes, you do need at least one internal filter still else you get 0 array errors. Oops. :D

[SamuriHL]

SIGH. Nope. I'm still getting this kind of stuff:


60>MpaDecFilter.cpp(1568): error C2065: 'm_ps2_state' : undeclared identifier
60>MpaDecFilter.cpp(1568): error C2228: left of '.sync' must have class/struct/u
nion
60> type is ''unknown-type''
60>MpaDecFilter.cpp(1570): error C2065: 'm_ps2_state' : undeclared identifier
60>MpaDecFilter.cpp(1570): error C2228: left of '.sync' must have class/struct/u
nion
60> type is ''unknown-type''
60>MpaDecFilter.cpp(1575): error C2065: 'm_ps2_state' : undeclared identifier
60>MpaDecFilter.cpp(1575): fatal error C1003: error count exceeds 100; stopping
compilation
60>
60>Build FAILED.


Tons of errors above that saying pretty much the same thing.

[clsid]

Fixed that too.

[SamuriHL]

Thanks. Will try again. :)

[SamuriHL]

P.S. I haven't pulled yet, but, have you run into this yet?


67>c:\dev\projects\mpc-hc\src\filters\./transform/MpaDecFilter/MpaDecFilter.h(13
0): error C2143: syntax error : missing ';' before '*'
67>c:\dev\projects\mpc-hc\src\filters\./transform/MpaDecFilter/MpaDecFilter.h(13
0): error C4430: missing type specifier - int assumed. Note: C++ does not suppor
t default-int
67>c:\dev\projects\mpc-hc\src\filters\./transform/MpaDecFilter/MpaDecFilter.h(13
0): error C4430: missing type specifier - int assumed. Note: C++ does not suppor
t default-int
67>c:\dev\projects\mpc-hc\src\filters\./transform/MpaDecFilter/MpaDecFilter.h(13
1): error C2143: syntax error : missing ';' before '*'
67>c:\dev\projects\mpc-hc\src\filters\./transform/MpaDecFilter/MpaDecFilter.h(13
1): error C4430: missing type specifier - int assumed. Note: C++ does not suppor
t default-int
67>c:\dev\projects\mpc-hc\src\filters\./transform/MpaDecFilter/MpaDecFilter.h(13
1): error C4430: missing type specifier - int assumed. Note: C++ does not suppor
t default-int
67>c:\dev\projects\mpc-hc\src\filters\./transform/MpaDecFilter/MpaDecFilter.h(13
2): error C2143: syntax error : missing ';' before '*'
67>c:\dev\projects\mpc-hc\src\filters\./transform/MpaDecFilter/MpaDecFilter.h(13
2): error C4430: missing type specifier - int assumed. Note: C++ does not suppor
t default-int
67>c:\dev\projects\mpc-hc\src\filters\./transform/MpaDecFilter/MpaDecFilter.h(13
2): error C4430: missing type specifier - int assumed. Note: C++ does not suppor
t default-int

[SamuriHL]

Brilliant. The last update did it. Thank you so much for fixing this!

[SamuriHL]

WOOPS. It wasn't done. :(


67>..\..\thirdparty\mfc\afxglobals.cpp(44): error C2039: 'm_hinstD2DDLL' : is no
t a member of 'AFX_GLOBAL_DATA'
67> C:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\atlmfc\inclu
de\afxglobals.h(111) : see declaration of 'AFX_GLOBAL_DATA'
67>..\..\thirdparty\mfc\afxglobals.cpp(45): error C2039: 'm_hinstDWriteDLL' : is
not a member of 'AFX_GLOBAL_DATA'
67> C:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\atlmfc\inclu
de\afxglobals.h(111) : see declaration of 'AFX_GLOBAL_DATA'
67>..\..\thirdparty\mfc\afxglobals.cpp(47): error C2143: syntax error : missing
';' before '*'
67>..\..\thirdparty\mfc\afxglobals.cpp(47): error C4430: missing type specifier
- int assumed. Note: C++ does not support default-int
67>..\..\thirdparty\mfc\afxglobals.cpp(47): error C2039: 'm_pDirect2dFactory' :
is not a member of 'AFX_GLOBAL_DATA'
67> C:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\atlmfc\inclu
de\afxglobals.h(111) : see declaration of 'AFX_GLOBAL_DATA'
67>..\..\thirdparty\mfc\afxglobals.cpp(47): error C4430: missing type specifier
- int assumed. Note: C++ does not support default-int
67>..\..\thirdparty\mfc\afxglobals.cpp(48): error C2143: syntax error : missing
';' before '*'
67>..\..\thirdparty\mfc\afxglobals.cpp(48): error C4430: missing type specifier
- int assumed. Note: C++ does not support default-int
67>..\..\thirdparty\mfc\afxglobals.cpp(48): error C2039: 'm_pWriteFactory' : is
not a member of 'AFX_GLOBAL_DATA'
67> C:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\atlmfc\inclu
de\afxglobals.h(111) : see declaration of 'AFX_GLOBAL_DATA'
67>..\..\thirdparty\mfc\afxglobals.cpp(48): error C4430: missing type specifier
- int assumed. Note: C++ does not support default-int
67>..\..\thirdparty\mfc\afxglobals.cpp(49): error C2143: syntax error : missing
';' before '*'
67>..\..\thirdparty\mfc\afxglobals.cpp(49): error C4430: missing type specifier
- int assumed. Note: C++ does not support default-int
67>..\..\thirdparty\mfc\afxglobals.cpp(49): error C2039: 'm_pWicFactory' : is no
t a member of 'AFX_GLOBAL_DATA'
67> C:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\atlmfc\inclu
de\afxglobals.h(111) : see declaration of 'AFX_GLOBAL_DATA'
67>..\..\thirdparty\mfc\afxglobals.cpp(49): error C4430: missing type specifier
- int assumed. Note: C++ does not support default-int
67>..\..\thirdparty\mfc\afxglobals.cpp(51): error C2039: 'm_pfD2D1MakeRotateMatr
ix' : is not a member of 'AFX_GLOBAL_DATA'
67> C:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\atlmfc\inclu
de\afxglobals.h(111) : see declaration of 'AFX_GLOBAL_DATA'
67>..\..\thirdparty\mfc\afxglobals.cpp(51): error C2146: syntax error : missing
';' before identifier 'm_pfD2D1MakeRotateMatrix'
67>..\..\thirdparty\mfc\afxglobals.cpp(51): error C4430: missing type specifier
- int assumed. Note: C++ does not support default-int
67>..\..\thirdparty\mfc\afxglobals.cpp(51): error C4430: missing type specifier
- int assumed. Note: C++ does not support default-int
67>..\..\thirdparty\mfc\afxglobals.cpp(53): error C2039: 'm_bD2DInitialized' : i
s not a member of 'AFX_GLOBAL_DATA'
67> C:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\atlmfc\inclu
de\afxglobals.h(111) : see declaration of 'AFX_GLOBAL_DATA'
67>..\..\thirdparty\mfc\afxglobals.cpp(915): error C3861: 'ReleaseD2DRefs': iden
tifier not found
67>..\..\thirdparty\mfc\afxglobals.cpp(1463): error C2039: 'InitD2D' : is not a
member of 'AFX_GLOBAL_DATA'
67> C:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\atlmfc\inclu
de\afxglobals.h(111) : see declaration of 'AFX_GLOBAL_DATA'
67>..\..\thirdparty\mfc\afxglobals.cpp(1463): error C2065: 'D2D1_FACTORY_TYPE' :
undeclared identifier
67>..\..\thirdparty\mfc\afxglobals.cpp(1463): error C2146: syntax error : missin
g ')' before identifier 'd2dFactoryType'
67>..\..\thirdparty\mfc\afxglobals.cpp(1463): error C2059: syntax error : ')'
67>..\..\thirdparty\mfc\afxglobals.cpp(1464): error C2143: syntax error : missin
g ';' before '{'
67>..\..\thirdparty\mfc\afxglobals.cpp(1464): error C2447: '{' : missing functio
n header (old-style formal list?)
67>..\..\thirdparty\mfc\afxglobals.cpp(1521): error C2039: 'ReleaseD2DRefs' : is
not a member of 'AFX_GLOBAL_DATA'
67> C:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\atlmfc\inclu
de\afxglobals.h(111) : see declaration of 'AFX_GLOBAL_DATA'
67>..\..\thirdparty\mfc\afxglobals.cpp(1530): error C2227: left of '->Release' m
ust point to class/struct/union/generic type
67> type is 'int *'
67>..\..\thirdparty\mfc\afxglobals.cpp(1536): error C2227: left of '->Release' m
ust point to class/struct/union/generic type
67> type is 'int *'
67>..\..\thirdparty\mfc\afxglobals.cpp(1542): error C2227: left of '->Release' m
ust point to class/struct/union/generic type
67> type is 'int *'
67> winmdi.cpp
67> AppSettings.cpp
67> AuthDlg.cpp
67>c:\dev\projects\mpc-hc\src\apps\mplayerc\AppSettings.h(442): error C2229: cla
ss 'CAppSettings' has an illegal zero-sized array
67>c:\dev\projects\mpc-hc\src\apps\mplayerc\AppSettings.h(555): error C2229: cla
ss 'CAppSettings' has an illegal zero-sized array
67> BaseGraph.cpp


I guess I didn't have enough patience. :D

[SamuriHL]

Yea, this isn't good. Even reverting the header to include all the filters, the build is still broken on a clean full build.

[clsid]

Perhaps your visual studio isn't up-to-date? You need SP1.

[SamuriHL]

Oh son of a! Sometimes I don't like this laptop. I watched WU update it to SP1 a few weeks ago. What I DIDN'T apparently notice is that the damn update failed. Ok, off to fix VC2010 and get it up to SP1. This blows!

[molitar]

@Molitar: reset the registry settings: http://sourceforge.net/apps/trac/mpc-hc/wiki/Troubleshooting_guidelines

Thanks did that.. and someone stated them are CCCP defaults but them are the same defaults in the original MPC non cinema version because I tested dropped in another folder and tested and the defaults were just about the same as CCCP pack except for two more items checked in that list. But like I stated dropping the new version in rewrote all the defaults to being all checked. Good to know how to clear registry of what it changes when first ran..

[betaking]

CLSID.thansk you work but when i install last MSVC2008 sp1 after latest windows updates i can not compiler last svn!
输出窗口 正在编译...
OggFile.cpp
stdafx.cpp
OggSplitter.cpp
.\OggSplitter.cpp(697) : error C2668: “abs”: 对重载函数的调用不明确
C:\Program Files\Microsoft Visual Studio 9.0\VC\include\math.h(539): 可能是“long double abs(long double)”
C:\Program Files\Microsoft Visual Studio 9.0\VC\include\math.h(491): 或 “float abs(float)”
C:\Program Files\Microsoft Visual Studio 9.0\VC\include\math.h(487): 或 “double abs(double)”
C:\Program Files\Microsoft Visual Studio 9.0\VC\include\math.h(485): 或 “long abs(long)”
C:\Program Files\Microsoft Visual Studio 9.0\VC\include\stdlib.h(380): 或 “int abs(int)”
试图匹配参数列表“(REFERENCE_TIME)”时
结果 生成日志保存在“file://C:\mpc-hc\bin\obj\Release_Win32\OggSplitter\BuildLog.htm”
OggSplitter - 1 个错误，0 个警告
输出窗口 正在编译...
stdafx.cpp
正在编译...
MatroskaFile.cpp
MatroskaMuxer.cpp
.\MatroskaMuxer.cpp(667) : error C2668: “abs”: 对重载函数的调用不明确
C:\Program Files\Microsoft Visual Studio 9.0\VC\include\math.h(539): 可能是“long double abs(long double)”
C:\Program Files\Microsoft Visual Studio 9.0\VC\include\math.h(491): 或 “float abs(float)”
C:\Program Files\Microsoft Visual Studio 9.0\VC\include\math.h(487): 或 “double abs(double)”
C:\Program Files\Microsoft Visual Studio 9.0\VC\include\math.h(485): 或 “long abs(long)”
C:\Program Files\Microsoft Visual Studio 9.0\VC\include\stdlib.h(380): 或 “int abs(int)”
试图匹配参数列表“(REFERENCE_TIME)”时
结果 生成日志保存在“file://C:\mpc-hc\bin\obj\Release_Win32\MatroskaMuxer\BuildLog.htm”
MatroskaMuxer - 1 个错误，0 个警告
PPageFileInfoDetails.cpp
.\PPageFileInfoDetails.cpp(158) : error C3861: “StrFormatByteSizeW”: 找不到标识符
Playlist.cpp
.\Playlist.cpp(271) : error C3861: “StrCmpLogicalW”: 找不到标识符
.\MainFrm.cpp(4972) : error C3861: “StrFormatByteSizeW”: 找不到标识符
.\MainFrm.cpp(11702) : error C3861: “StrCmpLogicalW”: 找不到标识符
结果 生成日志保存在“file://C:\mpc-hc\bin\obj\Release_Win32\mpc-hc\BuildLog.htm”
mpc-hc - 4 个错误，18 个警告
PS: vs2010sp1 compiler is ok!

[SamuriHL]

My MSVC 2010 install is borked. :( I'm gonna have to completely uninstall and reinstall it. I got the SP1 DVD image from MSDN and it also failed to install and is in the process of rolling back. SIGH. I guess I'll get back to building this at some point in the near future. cough cough. :(

[SamuriHL]

WOO HOO!

MPC-HC's compilation started on Wed 05/04/2011-22:28:00.20
and completed on Wed 05/04/2011-22:48:12.47

bin10\MPC-HomeCinema.1.5.2.3080.x86.MSVC2010.exe

Hurray. That's with only 1 filter selected. Thank you, clsid! Looks like I'm back in business.

[clsid]

This patch might fix the MSVC2008 build:

Index: include/stdafx_common_afx.h
===================================================================
--- include/stdafx_common_afx.h (revision 3078)
+++ include/stdafx_common_afx.h (working copy)
@@ -1,19 +1,15 @@
// turns off MFC's hiding of some common and often safely ignored warning messages
#define _AFX_ALL_WARNINGS

-#include <afxwin.h> // MFC core and standard components
-#include <afxext.h> // MFC extensions
-
-#include <afxdisp.h> // MFC Automation classes
-
-#ifndef _AFX_NO_OLE_SUPPORT
-#include <afxdtctl.h> // MFC support for Internet Explorer 4 Common Controls
+#if (_MSC_VER < 1600)
+#include <afx.h>
#endif
-#ifndef _AFX_NO_AFXCMN_SUPPORT
-#include <afxcmn.h> // MFC support for Windows Common Controls
-#endif // _AFX_NO_AFXCMN_SUPPORT

+#include <afxwin.h> // MFC core and standard components
+
+#if (_MSC_VER >= 1600)
#include "afxpriv.h"
#include "afxole.h"
#include "..\src\mfc\oleimpl2.h"
#include "..\src\mfc\afximpl.h"
+#endif
Index: include/stdafx_common_afx2.h
===================================================================
--- include/stdafx_common_afx2.h (revision 0)
+++ include/stdafx_common_afx2.h (revision 0)
@@ -0,0 +1,22 @@
+// turns off MFC's hiding of some common and often safely ignored warning messages
+#define _AFX_ALL_WARNINGS
+
+#include <afxwin.h> // MFC core and standard components
+#include <afxext.h> // MFC extensions
+#include <afxdisp.h> // MFC Automation classes
+
+#ifndef _AFX_NO_OLE_SUPPORT
+#include <afxdtctl.h> // MFC support for Internet Explorer 4 Common Controls
+#endif
+
+#ifndef _AFX_NO_AFXCMN_SUPPORT
+#include <afxcmn.h> // MFC support for Windows Common Controls
+#endif
+
+#include "afxole.h"
+
+#if (_MSC_VER >= 1600)
+#include "afxpriv.h"
+#include "..\src\mfc\oleimpl2.h"
+#include "..\src\mfc\afximpl.h"
+#endif
Index: src/apps/mplayerc/stdafx.h
===================================================================
--- src/apps/mplayerc/stdafx.h (revision 3078)
+++ src/apps/mplayerc/stdafx.h (working copy)
@@ -26,7 +26,7 @@
#include "../../../DSUtil/SharedInclude.h"
#include "../../../../include/stdafx_common.h"
#undef _ATL_CSTRING_EXPLICIT_CONSTRUCTORS
-#include "../../../../include/stdafx_common_afx.h"
+#include "../../../../include/stdafx_common_afx2.h"
#include "../../../../include/stdafx_common_dshow.h"

#include <afxdlgs.h>
Index: src/filters/transform/VSFilter/stdafx.h
===================================================================
--- src/filters/transform/VSFilter/stdafx.h (revision 3078)
+++ src/filters/transform/VSFilter/stdafx.h (working copy)
@@ -22,6 +22,6 @@

#include "../../../DSUtil/SharedInclude.h"
#include "../../../../include/stdafx_common.h"
-#include "../../../../include/stdafx_common_afx.h"
+#include "../../../../include/stdafx_common_afx2.h"
#include "../../../../include/stdafx_common_dshow.h"
#include "../../../DSUtil/DSUtil.h"

Please also test MSVC2010 build after this patch. I haven't got the time right now to do full builds to test.

[betaking]

thanks can you put this patch to svn? thanks!

[SamuriHL]

It needs to be fully tested with both a 2008 and 2010 compile first. That's what clsid was asking people to do. I'm off to bed so I can't test the 2010 part of that right now.

[SamuriHL]

Ok, clsid, I was able to build it again with no filters now. That's so awesome! Thanks for doing this for us. madshi's likely to be quite happy, as well. I wasn't able to test your patch to fix the 2008 build with 2010 but if no one's done it by the morning I'll take a look.

[Aleksoid1978]

http://aleksoid.tosei.ru/Dist/mpc-hc.exe.3083.7z - Only .exe
http://aleksoid.tosei.ru/Dist/mpc-hc.3083.7z - Complete set.

[betaking]

test clsid patch still have
PPageFileInfoDetails.cpp
.\PPageFileInfoDetails.cpp(158) : error C3861: “StrFormatByteSizeW”: 找不到标识符
Playlist.cpp
.\Playlist.cpp(271) : error C3861: “StrCmpLogicalW”: 找不到标识符
.\MainFrm.cpp(4972) : error C3861: “StrFormatByteSizeW”: 找不到标识符
.\MainFrm.cpp(11702) : error C3861: “StrCmpLogicalW”: 找不到标识符
结果 生成日志保存在“file://C:\mpc-hc\bin\obj\Release_Win32\mpc-hc\BuildLog.htm”
mpc-hc - 4 个错误，18 个警告

[nevcairiel]

Perhaps your visual studio isn't up-to-date? You need SP1.

While installing SP1 is certainly a good idea for everyone, breaking compilation for people that don't have it is just horrible.

I should really switch players, MPC-HC is a lost cause.. no developers, and enough people that claim to be developers breaking things left and right and arguing pointless semantics..

LAV Player, anyone?

[ryrynz]

While installing SP1 is certainly a good idea for everyone, breaking compilation for people that don't have it is just horrible.

I should really switch players, MPC-HC is a lost cause.. no developers, and enough people that claim to be developers breaking things left and right and arguing pointless semantics..

LAV Player, anyone?

Personally I have no problems with MPC-HC although I do think it's interface is severely outdated.
I'm in agreement with you regarding MPC-HC, If you'd like to develop for a media player probably best making your own.
Potplayer seems to be the best choice for a free media player right now.

[mindbomb]

isn't making a player too large an undertaking for a single person?

that being said, i would probably use lav player

[nevcairiel]

isn't making a player too large an undertaking for a single person?

Would probably take a while, but other devs would be welcome to join/help. :rolleyes:

BTW, i'm not saying i will do it. Maybe if some other active developers express some interest, there could be formed some kind of plan, but for now its just a pipe-dream. :p

[madshi]

At the speed you're currently developing you'll soon run out of work, so maybe a new player is not a bad idea... ;) Personally, I don't have enough time for developing or helping to develop a player, too much other work to do.

[ForceX]

isn't making a player too large an undertaking for a single person?

that being said, i would probably use lav player
Not really, AFAIK KMPlayer and later PotPlayer was developed by a single person and you can see that PotPlayer is being feverishly updated now, you just have to dedicate your complete life to that, that's all. Admittedly, a lot of the base for both KMP and PotP seems to be based on the work of MPC(-HC) and I doubt these players would have been this far if not for the existence of MPC. While MPC development does seem to have stagnated a bit, it is still pretty advanced in terms of good video playback capability and still setting standards for others to follow.

As a non-funded open source program, it is as good as it gets.

[nevcairiel]

I'm not talking about the player itself, i still use it and i think its fine. I'm mostly talking about decisions and actions by the people currently trying to maintain it. I, however, have no ambitions to try to maintain and enhance it myself, because the code is just horribly ugly.

Anyhow, this really doesn't belong here.

[neoufo51]

While installing SP1 is certainly a good idea for everyone, breaking compilation for people that don't have it is just horrible.

I should really switch players, MPC-HC is a lost cause.. no developers, and enough people that claim to be developers breaking things left and right and arguing pointless semantics..

LAV Player, anyone?
If you make a LAV Player, I will drop MPC-HC in an instant, and so will the majority of people who use MPC-HC. You'd also win over the people who use Zoomplayer and PotPlayer because they want something high quality that isn't MPC-HC, and your player would be the way to go, without question.

All you need to do is create a player that is simple, minimalistic, high quality, built from scratch and MPC-HC would just die in a matter of months. Sorry to anybody who loves MPC-HC but that's just the truth. It will be a lot of work, but I don't think anybody here wants something fancy-looking.

[tetsuo55]

We don't "need" a new player. Most of the bugs that exist are in filters. The GUI is really the only thing most people want changed.

The reason the project is moving so slow is because we lack a motivated, dedicated developer. Someone who is willing and able to spend 20-30 hours a week on MPC-HC.(all the dev activity you do see is all filter based)

Nevcairiel is right about the source being difficult to maintain. A dedicated dev would focus mostly on rewriting the actual player into a new clean codebase as mpc-hc2.0.

[nevcairiel]

A dedicated dev would focus mostly on rewriting the actual player into a new clean codebase as mpc-hc2.0.

Which is, essentially, a whole new player. :cool:
Loads of other players already use the same filters as MPC-HC (which i'm in the process to replace, anyhow :p)

Here is basically a concise list how i see this would work, if ever.
- A new base player (GUI, configuration, filter management, DirectShow graph building)
- A new, easier and universal way to render subtitles - preferably as a DirectShow filter, not directly hacked into the client.
- MASSIVE cleanup in the renderers (i mean actual cleanup, maybe even rewriting of huge parts, not hacking new "features" in there)
- Evaluation of filters. LAV can mostly replace internal source filters, it can fully replace the internal audio decoder, video is planned - DXVA1/2 decoders need evaluation.

The first step is the most important, just get some new player going that works with stand-alone filters, with standalone renderers, like default VMR9/EVR and madVR. Then cleanup the custom renderers from MPC-HC and try to reuse EVR-CP and VMR9(renderless). This would already give you proper playback of everything.
Next up would be subtitles. Thats probably a more complicated task and will require cooperation with the renderers.

And as the final touch, select open-source filters that should be bundled with the new player.

[tetsuo55]

Yeah, for maintainability we need a new player.
Will that be mpc-hc2.0, lavplayer, or something else, i don't know

[CruNcher]

Potplayer is in those regards another MPC-HC shares the same codebase though with a lot of additions (especially the skinning engine and the enhanced filter management), only still unique is Zoomplayer, but many already prefer Potplayer also Gom Player and i dunno but i think their are more MPC-HC based Player out their especially in the Asian direction. :)
And that is really sad the Potplayer guy could at least contribute back and normally he would have to under the GPL, testuo55 did you ever talked with him about this he seems to also have problems with ffmpeg and comply with the GPL their ?

However, the The KMPlayer authors have published an official comment on their own forum that denies the alleged GPL violation by Gabest.
Does somebody has a link to the written statement ?

[Daiz]

- A new, easier and universal way to render subtitles - preferably as a DirectShow filter, not directly hacked into the client.

Ever heard of VSFilter? :rolleyes:

[tetsuo55]

- A new, easier and universal way to render subtitles - preferably as a DirectShow filter, not directly hacked into the client.
- MASSIVE cleanup in the renderers (i mean actual cleanup, maybe even rewriting of huge parts, not hacking new "features" in there)
Neither of those are part of the core player imho, but filters. Granted these also need a rewrite and dedicated devs/

@Cruncher, afaik all the forks except splayer violate GPL. And to my knowledge none of them send us patches.

[nevcairiel]

Ever heard of VSFilter? :rolleyes:

VSFilter is crap. Its super slow, a simple text sub can make your video playback crawl - and it renders subs on the image before scaling and color space expansion.

Not to mention its buggy as hell, and gets sub positions wrong half the time.

Neither of those are part of the core player imho, but filters. Granted these also need a rewrite and dedicated devs

Indeed they are not, but for a full playback experience, they are needed. As you see above, only the first point is "core" player, everything else is stuff needed for a full playback package.

[Daiz]

VSFilter is crap. Its super slow, a simple text sub can make your video playback crawl - and it renders subs on the image before scaling and color space expansion.

Not to mention its buggy as hell, and gets sub positions wrong half the time.

While I agree that VSFilter is slow due to being poorly optimized and only single-threaded, it's not so slow that "a simple text sub" could make your video playback crawl unless you were using a Pentium II or something for playback. Neither does it get sub positions wrong half the time. You need to do quite heavy transform effects with ASS to make VSFilter slow down into unplayable territory, and those aren't something you could even do with simpler/unstyled subtitle formats.

But there isn't really much choice in terms of subtitle renderers. It's either VSFilter or libass. And libass doesn't have a DirectShow filter. That's certainly something I'd like to see, though. But while libass is a lot faster than VSFilter, it has also has some quirky rendering issues not present in VSFilter...

[tetsuo55]

While I agree that VSFilter is slow due to being poorly optimized and only single-threaded, it's not so slow that "a simple text sub" could make your video playback crawl unless you were using a Pentium II or something for playback. Neither does it get sub positions wrong half the time. You need to do quite heavy transform effects with ASS to make VSFilter slow down into unplayable territory, and those aren't something you could even do with simpler/unstyled subtitle formats.

But there isn't really much choice in terms of subtitle renderers. It's either VSFilter or libass. And libass doesn't have a DirectShow filter. That's certainly something I'd like to see, though. But while libass is a lot faster than VSFilter, it has also has some quirky rendering issues not present in VSFilter...VSfilter/internal renderer can cause framedrops from even the simplest subtitles. Its a major design flaw in the core that is not related to CPU time.

[nevcairiel]

While I agree that VSFilter is slow due to being poorly optimized and only single-threaded, it's not so slow that "a simple text sub" could make your video playback crawl unless you were using a Pentium II or something for playback. Neither does it get sub positions wrong half the time.

I recently fixed two bugs in VSFilter, and while testing i noticed all this.

I have a 1080p MKV with a very simple SRT subtitle, and displaying it just kills playback - and this is on a i7. Maybe its SRT thats so horribly inefficient, i don't have any real content with ASS subs, except samples for testing.

The bug i fixed was actually a positioning problem of PGS subs, half of them just showed up in the top left corner. Due to its weird code i was unable to fix the underlying issue, so i just fixed it in PGS - i don't doubt other subs still show up in the corner.

Not to mention that VSFilter suffers from the same code quality problems MPC-HC does (hey, its the same project), so recycling it is not really so easy.
If one were to write a new subtitle filter, there would at least be the possibility to look into re-using some of the subtitle rendering logic (after cleanups), as i realize that most fan-sub content is actually designed for VSFilter.

[zhfi]

test clsid patch still have
PPageFileInfoDetails.cpp
.\PPageFileInfoDetails.cpp(158) : error C3861: “StrFormatByteSizeW”: 找不到标识符
Playlist.cpp
.\Playlist.cpp(271) : error C3861: “StrCmpLogicalW”: 找不到标识符
.\MainFrm.cpp(4972) : error C3861: “StrFormatByteSizeW”: 找不到标识符
.\MainFrm.cpp(11702) : error C3861: “StrCmpLogicalW”: 找不到标识符
结果 生成日志保存在“file://C:\mpc-hc\bin\obj\Release_Win32\mpc-hc\BuildLog.htm”
mpc-hc - 4 个错误，18 个警告


try this patch while compile with VC2008:

Index: include/stdafx_common.h
===================================================================
--- include/stdafx_common.h (revision 3084)
+++ include/stdafx_common.h (working copy)
@@ -8,6 +8,10 @@
#define VC_EXTRALEAN // Exclude rarely-used stuff from Windows headers
#endif

+#ifndef NO_DSHOW_STRSAFE
+#define NO_DSHOW_STRSAFE
+#endif
+
#include <SDKDDKVer.h>

#define _ATL_CSTRING_EXPLICIT_CONSTRUCTORS // some CString constructors will be explicit

[Daiz]

I have a 1080p MVK with a very simple SRT subtitle, and displaying it just kills playback

Interesting. Was this with the internal VSFilter of MPC-HC or an external VSFilter? Also, can you cut a sample of the SRT track that I could try on a random 1080p video?

The bug i fixed was actually a positioning problem of PGS subs, half of them just showed up in the top left corner. Due to its weird code i was unable to fix the underlying issue, so i just fixed it in PGS - i don't doubt other subs still show up in the corner.

Admittedly I haven't tried to render any PGS subtitles with VSFilter (I don't think the external one even supports them?), but that's mainly because I OCR/convert all of my subtitles into ASS for superior subtitle quality.

If you were to write a new subtitle renderer, it'd definitely need to support ASS - if it didn't, it'd be a major step back from VSFilter/libass, even if it had a lot less bugs and was more stable. I for one would not be willing to give up all the features that ASS has to offer and neither would many others.

[nevcairiel]

Interesting. Was this with the internal VSFilter of MPC-HC or an external VSFilter? Also, can you cut a sample of the SRT track that I could try on a random 1080p video?

External, internal didn't seem to suffer from the problem.

Here is the sample i tested on, it has both PGS and SRT
http://www.mediafire.com/?ucid8fzk282jm8b


Admittedly I haven't tried to render any PGS subtitles with VSFilter (I don't think the external one even supports them?), but that's mainly because I OCR/convert all of my subtitles into ASS for superior subtitle quality.

It does. Make sure you use a build from the mpc-hc standalone filters package, not the original vsfilter from gabest's old project (last updated years ago)

Oh btw, If i look at the bitmap subs of the forced subs in Avatar, i don't think you could get the same quality, or anything close, without the original Font. Blu-ray PGS are already pre-rendered at 1080p, they look pretty good to me. The "normal" subs (not forced) look a bit worse though, like some standard font - but i only really want forced subs.

OCR'ing DVD subs can only improve their quality, i guess.


If you were to write a new subtitle renderer, it'd definitely need to support ASS - if it didn't, it'd be a major step back from VSFilter/libass, even if it had a lot less bugs and was more stable. I for one would not be willing to give up all the features that ASS has to offer and neither would many others.

Of course it would aim to support as many subtitle formats as possible. Possibly even allowing both libass and vsfilter backends for rendering.

Reminder: This is just thinking here, there are no concrete plans for coding anything.

[Daiz]

Oh btw, If i look at the bitmap subs of the forced subs in Avatar, i don't think you could get the same quality, or anything close, without the original Font.

Well, considering that they're using Papyrus (http://en.wikipedia.org/wiki/Papyrus_%28typeface%29) for it, it wouldn't be particularly hard to use the exact same font... and personally, a big reason to OCR it would be to change that travesty of a font to something actually pleasant to look at. And from what I've seen of BD subtitles in general, they seem to either use quite basic fonts or alternatively cock it up completely by doing something as stupid as using a serif font for the main dialogue. Or having no subtitle borders, only a dropshadow. Or both. And the list goes on...

[clsid]

While installing SP1 is certainly a good idea for everyone, breaking compilation for people that don't have it is just horrible.
Compiling should now work again with 2010 RTM.

Do you have any idea which code (or project settings) in the standalone filters might trigger the MFC bloat inclusion? Your filters don't suffer from it, which means it should be avoidable. A clean solution is of course preferred over the current workaround.

[nevcairiel]

Well, considering that they're using Papyrus (http://en.wikipedia.org/wiki/Papyrus_%28typeface%29) for it, it wouldn't be particularly hard to use the exact same font... and personally, a big reason to OCR it would be to change that travesty of a font to something actually pleasant to look at. And from what I've seen of BD subtitles in general, they seem to either use quite basic fonts or alternatively cock it up completely by doing something as stupid as using a serif font for the main dialogue. Or having no subtitle borders, only a dropshadow. Or both. And the list goes on...


I don't usually look at subs, i really hate them. I only watch content if i understand it - alien languages are of course the exception to this.
The forced subs in Avatar were actually "ok" to read during watching - IMHO of course - the normal subs for the english parts are as ugly as ever.


Do you have any idea which code (or project settings) in the standalone filters might trigger the MFC bloat inclusion? Your filters don't suffer from it, which means it should be avoidable. A clean solution is of course preferred over the current workaround.

My filters don't statically link MFC, i rather require the VS2010 runtime to be installed.

[Daiz]

I don't usually look at subs, i really hate them.

Well, if you're a German I can see why you would have an attitude like that :) In comparison to your experience with the alien language subtitles in Avatar, my thoughts during the first lines of alien dialogue were "Are they seriously using Papyrus as the font? That's goddamn cheap!" Personally I prefer to watch all media that I consume in its original language with subtitles. I also work with subtitles and subtitling on a weekly basis, so subtitle quality, formats, their features and rendering are quite important things for me and something I pay close attention to.

[clsid]

I recently fixed two bugs in VSFilter, and while testing i noticed all this.

I have a 1080p MKV with a very simple SRT subtitle, and displaying it just kills playback - and this is on a i7. Maybe its SRT thats so horribly inefficient, i don't have any real content with ASS subs, except samples for testing.

The bug i fixed was actually a positioning problem of PGS subs, half of them just showed up in the top left corner. Due to its weird code i was unable to fix the underlying issue, so i just fixed it in PGS - i don't doubt other subs still show up in the corner.

Not to mention that VSFilter suffers from the same code quality problems MPC-HC does (hey, its the same project), so recycling it is not really so easy.
If one were to write a new subtitle filter, there would at least be the possibility to look into re-using some of the subtitle rendering logic (after cleanups), as i realize that most fan-sub content is actually designed for VSFilter.
MPC-HC by default uses insane subtitle settings on quadcores (and higher). It enables "allow animation while buffering" which pretty much just eats CPU cycles. It is supposed to help with tearing issues, but its just a terrible code hack. I think it should be disabled or just removed entirely.
To get animation, use subpicture buffer of zero. For best performance (without correct animation) use non-zero buffer.

The biggest flaw in the subtitle renderer is that is uses the animation code path for every type of subtitle, even static ones like srt and vobsub. Fixing that would give a nice performance boost.

[clsid]

My filters don't statically link MFC, i rather require the VS2010 runtime to be installed.I forgot to mention that I modified them to be static to check out the size increase. The added size was in the order of 100KB, so pretty small.

[nevcairiel]

Well, if you're a German I can see why you would have an attitude like that :)

How is that?

I also don't watch content in german, unless its a german production. I watch it in english, as thats the language most the content i watch was produced in, come to think of it, nearly all of it, i guess.

PS:
Sometimes ignorance is bliss - i just enjoy a movie, and don't wonder about which font was used for the forced subs. :D

[cca]

MPC-HC by default uses insane subtitle settings on quadcores (and higher). It enables "allow animation while buffering" which pretty much just eats CPU cycles. It is supposed to help with tearing issues, but its just a terrible code hack. I think it should be disabled or just removed entirely.
To get animation, use subpicture buffer of zero. For best performance (without correct animation) use non-zero buffer.

The biggest flaw in the subtitle renderer is that is uses the animation code path for every type of subtitle, even static ones like srt and vobsub. Fixing that would give a nice performance boost.

Hack or not, with a buffer length of zero it's almost impossible to get subtitles without introducing Tearing. That's why this was introduced and it does the job, I watch fansubed anime for years now and the majority of them use ASS animated subtitles, I have tested those settings over time.

I do agree that it shouldn't be used for SRT and VobSubs though, no reason to.

[madshi]

Where does tearing come from in a subtitle renderer? Normally it's a video renderer problem...

[nevcairiel]

I forgot to mention that I modified them to be static to check out the size increase. The added size was in the order of 100KB, so pretty small.

It happens only if you use some function that actually causes those objects to be pulled in. Apparently i don't. My way of creating the properties pages is quite different to how MPC-HC filters do it, which may explain it (only place with alot of MFC usage).

Not sure what function exactly triggers it though, may be something in the MPC-HC base classes, DSUtil or whatever.

If its only 100Kb, i might even consider turning that on by default, and make those people shut up that don't have the runtime installed. :p

[clsid]

Ideally buffering should just work with animation as well, without special workarounds. Only VFR could be a bit problematic since the frame rate needs to be known/guessed, that's probably why it wasn't implemented that way.

[Daiz]

How is that?

I also don't watch content in german, unless its a german production. I watch it in english, as thats the language most the content i watch was produced in, come to think of it, nearly all of it, i guess.

Oh. Well, I'd have expected you to watch stuff in german seeing as they dub pretty much everything around there.


PS:
Sometimes ignorance is bliss - i just enjoy a movie, and don't wonder about which font was used for the forced subs. :D

Just like how many people just enjoy the movie, and don't fuss over the extreme smoothing, low resolution and compression artefacts present in the XviD rip they are watching, right ;)

EDIT: So I took a look at the Avatar subtitle sample, and here's what I found out. The subtitles indeed seem to cause some dropped frames (around 8-10, according to MPC-HC's statistics). Converting the SRT to ASS didn't change the situation at all. But then I made an interesting discovery: I turned off CUDA decoding in CoreAVC so that it'd use the CPU to decode the video, and now the subtitles don't cause any lag at all.

[nevcairiel]

Just like how many people just enjoy the movie, and don't fuss over the extreme smoothing, low resolution and compression artefacts present in the XviD rip they are watching, right ;)
Well its too late for me there, but at least the playback chain i can optimize, so as long as the content is actually good enough, i might make it through the movie happy. :D

[ForceX]

Does somebody has a link to the written statement ?
http://www.kmplayer.com/forums/showthread.php?t=141

I think one of their chief argument was that KMP is written in Delphi, not C, hence there is no chance of there being copied code from MPC, and KMP only links against Gabest's filters, hence it isn't violation of GPL.

MPC-HC by default uses insane subtitle settings on quadcores (and higher). It enables "allow animation while buffering" which pretty much just eats CPU cycles. It is supposed to help with tearing issues, but its just a terrible code hack. I think it should be disabled or just removed entirely.
To get animation, use subpicture buffer of zero. For best performance (without correct animation) use non-zero buffer.
When using animated subtitles, using sub picture buffering is the only way I can get smooth playback, setting it to 0 makes playback horribly stuttery due to crazy frame dropping during animation, even though I have CPU cycles to spare. I'd rather have high CPU usage with smooth playback rather than idle CPU with crappy playback. I really suggest not changing it. MPC-HC is THE recommended player for fansubbed anime these days and they have just started using animated softsubs for karaoke.

[JanWillem32]

The reason of the slow subtitle renderer is mainly because of the font handling. I've been trying to get it corrected, so it can use 8-bit RGBA instead of 4-bit RGBA, handle multithreading better and use a cleaner composing method. It's still a bit hard to get it working, but at least I've disabled mipmap generation for some parts (still in testing phase). A true upgrade would be to move to DirectX 10/11 native font rendering and composing. (DirectX 9 lacks a lot font rendering technology.) That does involve quite a lot of work to get both hardware-level rendering (for DirectX 10 or newer generation GPU's) and software rendering paths working. Also, the OS requirement is an obstacle. I fully understand the need for supporting Windows XP, as it's still an officially supported OS.
At least on the allocator-presenter parts it seems to be okay, as the blending stage for main video, subtitles, OSD and the stats screen is clean and efficient for what I've seen in the code of EVR CP, Sync and VMR-9 (renderless).

[clsid]

http://www.kmplayer.com/forums/showthread.php?t=141

I think one of their chief argument was that KMP is written in Delphi, not C, hence there is no chance of there being copied code from MPC, and KMP only links against Gabest's filters, hence it isn't violation of GPL.


When using animated subtitles, using sub picture buffering is the only way I can get smooth playback, setting it to 0 makes playback horribly stuttery due to crazy frame dropping during animation, even though I have CPU cycles to spare. I'd rather have high CPU usage with smooth playback rather than idle CPU with crappy playback. I really suggest not changing it. MPC-HC is THE recommended player for fansubbed anime these days and they have just started using animated softsubs for karaoke.
The whole KMP issue has been discussed many times before in the past. Translating code to a different language does not make it your own unique work. Plus it has been proven several times in the past that KMP had the same specific bugs as MPC.

The stuttering issues are partly due to the renderer changes that were made in MPC-HC. People were already enjoying smooth playback of fansubs with the old MPC.

[ForceX]

The reason of the slow subtitle renderer is mainly because of the font handling. I've been trying to get it corrected, so it can use 8-bit RGBA instead of 4-bit RGBA, handle multithreading better and use a cleaner composing method. It's still a bit hard to get it working, but at least I've disabled mipmap generation for some parts (still in testing phase). A true upgrade would be to move to DirectX 10/11 native font rendering and composing. (DirectX 9 lacks a lot font rendering technology.) That does involve quite a lot of work to get both hardware-level rendering (for DirectX 10 or newer generation GPU's) and software rendering paths working. Also, the OS requirement is an obstacle. I fully understand the need for supporting Windows XP, as it's still an officially supported OS.
At least on the allocator-presenter parts it seems to be okay, as the blending stage for main video, subtitles, OSD and the stats screen is clean and efficient for what I've seen in the code of EVR CP, Sync and VMR-9 (renderless).
The newest generation web browsers are using Direct2D and DirectWrite to offload font rendering completely on the GPU, which greatly speeds up the rendering quality and performance on those platforms. Unfortunately it's only available on updated Vista and Win 7 (damn MS). If you can use that for Vista+ OSes and use a fallback for XP, I think at least Vista+ can get a huge speedup in subtitle handling.

[cyberbeing]

MPC-HC by default uses insane subtitle settings on quadcores (and higher). It enables "allow animation while buffering" which pretty much just eats CPU cycles. It is supposed to help with tearing issues, but its just a terrible code hack. I think it should be disabled or just removed entirely.
To get animation, use subpicture buffer of zero. For best performance (without correct animation) use non-zero buffer.

Please don't remove this from the standalone VSFilter.dll. For whatever reason, I've found that the standalone VSFilter.dll doesn't have the CPU usage problem that the internal MPC-HC VSFilter has with that buffered animation hack. That feature being enabled by default in VSFilter.dll has been a lifesaver during the past year for a few of the more stupid uses of slow ASS features I've run into. To my surprise, it actually seems to work very well with karaoke, moving text, fades, and other advanced text effects in VSFilter.dll, without causing flickering, and it actually makes some previously unplayable segments, playable on my oldish AMD X2 4800+ PC.

On another note, more testing is probably needed, but VSFilter.dll appears to get a reduction in CPU usage and a decent performance boost on difficult samples, by using Profile-guided optimization (PGO). The only tricky part about using PGO, is finding good representative samples to use for manual profiling step of the instrumented dll, since you can potentially make things worse.

[clsid]

VSFilter.dll does not even have that option.

[clsid]

If its only 100Kb, i might even consider turning that on by default, and make those people shut up that don't have the runtime installed. :pSizes of the filters were still tiny compared to the FFmpeg dlls. Static linking by default would be helpful for portability. Certainly when they eventually replace the internal ones in MPC ;)

[ceb]

@HoP & JanWillem32
I had your same problem with EVR CP but a more recent version solved it. Try a version after r3043, I think it's because of the fix "YV12 Chroma Upsampling for ATI", it's the only one that makes sense to me. Sorry for the late answer HoP (I had forgotten about the other thread :p).

Anyway, I see that there are a lot of people who enjoy watching anime fansub like me. :D
I watch both soft and hard sub, what are the best quality settings regardless of performance?
I leave everything at default except "Maximum Texture Resolution" changed to "Desktop" (improves subtitle quality of softsubs) and remove the tick from "Round up to power of two".
Also, I noticed that the shaders (in my case Sharpen Complex 2) are applied to the subtitles (only hardsubs). Is there a way to avoid this? I guess it's not possible since they're part of the video but perhaps there is a way...

[clsid]

Hardsubs are part of the video. So do the math...

[cyberbeing]

VSFilter.dll does not even have that option.

Correct, there is no option, but for whatever reason it seems to be enabled automatically when using pre-buffering in VSFilter.dll. No clue when exactly that change happened in the MPC-HC branch (VSFilterMod?), but karaoke, moving text, fading, etc work with pre-buffering in the standalone VSfilter.dll possibly for the past year or so.

As I mentioned above, it's helpful and seems to offer a performance boost, so be careful not to disable it in the standalone dll if you make any changes to the MPC-HC internal filter. It's one of the major advantages the MPC-HC VSFilter has over the Guliverkli2 VSFilter, and has allowed me to leave pre-buffering enabled full-time, without any apparent repercussions.

[jos99]

Any advice on how to force MPC-HC to use the MPEG2 DXVA decoder for MPEG2 .ts TV recordings, it always seem to default to non DXVA decoder here (1.5.2.3085)?



Recent recordings of BBC HD (UK) give sound but no video, using internal or external codecs/splitters in MPC-HC - sample 22MB
Open a new ticket.

OK, done here (https://sourceforge.net/apps/trac/mpc-hc/ticket/829) #829.
Thanks for whoever fixed this, all BBC HD and BBC One HD recordings now seem to work smoothly with MPC-HC inbuilt H.264 decoder and TS splitter:thanks:.

[Mercury_22]

Any advice on how to force MPC-HC to use the MPEG2 DXVA decoder for MPEG2 .ts TV recordings, it always seem to default to non DXVA decoder here (1.5.2.3085)?

Thanks for whoever fixed this, all BBC HD and BBC One HD recordings now seem to work smoothly with MPC-HC inbuilt H.264 decoder and TS splitter:thanks:.

See this According to your signature, you have a ATI HD4670 - those do not support DXVA for MPEG-2. Neither do the 5xxx series. The 6xxx series claim to support it, but i have no conclusive evidence that it works in MPC-HC.
Correct !
Sorry I've forgot no VLD for me :mad: only IDCT :stupid:

But the crash is still there in BE MOD (with MPEG-2 DXVA enable but not used) and it isn't in the official (SVN) version:p



"No problem" :rolleyes:we are waiting for you:p:)
So does your card support DXVA VLD for mpeg2 ?

[jos99]

See this

So does your card support DXVA VLD for mpeg2 ?
9800GT, DXVA works if I set Microsoft DTV DVD MPEG2 decoder as preferred external filter, but if I don't do that MPC-HC just defaults to non DXVA internal MPEG2 decoder.

[lukanz]

i have encodec one file with Xmedia Recode and try to watch with MPC-HC but there is no VIDEO & AUDIO using a fresh installed MPC-HC but adding CoreAVC i'm able to watch the movie...anyone else having problem with this movie and MPC-HC

here is the sample movie : http://www23.zippyshare.com/v/51283375/file.html

here the mediainfo :

General
Complete name : Test.mp4
Format : MPEG-4
Format profile : Base Media
Codec ID : isom
File size : 15.5 MiB
Duration : 1mn 53s
Overall bit rate : 1 146 Kbps
Encoded date : UTC 2011-05-05 18:05:55
Tagged date : UTC 2011-05-05 18:05:55
Writing application : Lavf52.106.0

Video
ID : 1
Format : AVC
Format/Info : Advanced Video Codec
Format profile : Baseline@L3.0
Format settings, CABAC : No
Format settings, ReFrames : 1 frame
Codec ID : avc1
Codec ID/Info : Advanced Video Coding
Duration : 1mn 53s
Bit rate mode : Variable
Bit rate : 1 024 Kbps
Width : 640 pixels
Height : 360 pixels
Display aspect ratio : 16:9
Frame rate mode : Constant
Frame rate : 30.000 fps
Color space : YUV
Chroma subsampling : 4:2:0
Bit depth : 8 bits
Scan type : Progressive
Bits/(Pixel*Frame) : 0.148
Stream size : 13.7 MiB (89%)
Writing library : x264 core 114 r1900
Encoding settings : cabac=0 / ref=1 / deblock=1:0:0 / analyse=0x1:0x111 / me=hex / subme=5 / psy=1 / psy_rd=0.00:0.00 / mixed_ref=0 / me_range=16 / chroma_me=1 / trellis=0 / 8x8dct=0 / cqm=0 / deadzone=21,11 / fast_pskip=1 / chroma_qp_offset=0 / threads=12 / sliced_threads=0 / nr=0 / decimate=1 / interlaced=0 / constrained_intra=0 / bframes=0 / weightp=0 / keyint=250 / keyint_min=25 / scenecut=40 / intra_refresh=0 / rc_lookahead=40 / rc=abr / mbtree=1 / bitrate=1024 / ratetol=1.0 / qcomp=0.60 / qpmin=10 / qpmax=51 / qpstep=4 / vbv_maxrate=1500 / vbv_bufsize=1500 / nal_hrd=none / ip_ratio=1.00 / aq=1:1.00
Encoded date : UTC 2011-05-05 18:05:55
Tagged date : UTC 2011-05-05 18:05:55

Audio
ID : 2
Format : AAC
Format/Info : Advanced Audio Codec
Format profile : LC
Codec ID : 40
Duration : 1mn 53s
Bit rate mode : Variable
Bit rate : 126 Kbps
Channel(s) : 2 channels
Channel positions : Front: L R
Sampling rate : 44.1 KHz
Compression mode : Lossy
Stream size : 1.71 MiB (11%)
Encoded date : UTC 2011-05-05 18:05:55
Tagged date : UTC 2011-05-05 18:05:55

[HTPC-User]

Hi!
@HTPC-User
Please check these versions: 2990 (http://www.xvidvideo.ru/media-player-classic-home-cinema-x86-x64/media-player-classic-homecinema-x86-x64-1-5-2-2990.html), 2991 (http://www.xvidvideo.ru/media-player-classic-home-cinema-x86-x64/media-player-classic-homecinema-x86-x64-1-5-2-2991.html).
2990 seems to be working.
2991 seems to be broken.

So hopefully someone can now locate the bug and fix it!



@HTPC-User: I have that, too. In D3D exclusive mode without VSync, all keystrokes before the first mouse click are just too sensitive. It also happens if you try to seek by chapters or time, pause/play, change the audio or subtitle streams. Everything is done like as if you tap the key or key combination two or three times. I don't know what causes it. I haven't found a bug report on this item on trac yet, so if you could submit one, I would appreciate it.


How and where? :)
The link to the bug reports in the first posting doesn't work for me.

Concerning the forced subtitles I got recently the hint to try LAV filters. As soon as possible I will try that, but perhaps has someone an idea how to only show forced subtitles in MPC-HC (with ffdshow) without additional filters or programs.
Manually switching them on and off when needed is quit annoying. :)

:thanks:

[mindbomb]

9800GT, DXVA works if I set Microsoft DTV DVD MPEG2 decoder as preferred external filter, but if I don't do that MPC-HC just defaults to non DXVA internal MPEG2 decoder.

9800gt doesnt support vld. it supports idct, which the microsoft decoder does support.

[SamuriHL]

Something is wrong with the latest build:


67>c:\dev\projects\mpc-hc\src\thirdparty\MediaInfo\MediaInfo\../../../../include
/stdafx_common_afx2.h(1): error C2059: syntax error : '<<'
67>c:\dev\projects\mpc-hc\src\thirdparty\MediaInfo\MediaInfo\../../../../include
/stdafx_common_afx2.h(24): error C2059: syntax error : '=='
67>C:\Program Files (x86)\Microsoft SDKs\Windows\v7.0A\include\windowsx.h(19): e
rror C2143: syntax error : missing ';' before '{'
67>C:\Program Files (x86)\Microsoft SDKs\Windows\v7.0A\include\windowsx.h(19): e
rror C2447: '{' : missing function header (old-style formal list?)
67>
67>Build FAILED.
67>


I've reinstalled MSVC 2010 Premium on my machine and updated it to SP1. So.....yea.

EDIT: Nevermind. Looks like a merge conflict caused by applying clsid's patch from last night before going to bed. I've reverted the file and am going to build again.

[Aleksoid1978]

i have encodec one file with Xmedia Recode and try to watch with MPC-HC but there is no VIDEO & AUDIO using a fresh installed MPC-HC but adding CoreAVC i'm able to watch the movie...anyone else having problem with this movie and MPC-HC

here is the sample movie : http://www23.zippyshare.com/v/51283375/file.html

here the mediainfo :

Play fine with internal splitter and audio/video decoder.

[lukanz]

@ Aleksoid1978

i played around and find out the reason was with Output set to EVR Custom Pres* there is no video playback and no sound...but with Output set to System Default it work...

i'm on win7_x64 and geforce9800 + mpc-hc_x32 1.5.2.3085

[namaiki]

Disable DXVA.

[Aleksoid1978]

@ Aleksoid1978

i played around and find out the reason was with Output set to EVR Custom Pres* there is no video playback and no sound...but with Output set to System Default it work...

i'm on win7_x64 and geforce9800 + mpc-hc_x32 1.5.2.3085

I use - EVR Custom, Win 7 x64, MPC-HC x86 - and all play fine.
But - by video is Intel HD.

[namaiki]

I use - EVR Custom, Win 7 x64, MPC-HC x86 - and all play fine.
But - by video is Intel HD.

I tested on Intel 4500MHD and 9600M GT.
Plays without issue on Intel but just black screen on 9600M GT.

[JanWillem32]

@HTPC-User (and any others interested): trac is the main support center for submitting bug reports, commenting on other reports, and there are even some guides to read. http://sourceforge.net/apps/trac/mpc-hc

[betaking]

yestoday , i install vs2010+SP1 compile is ok! thanks clsid & all dev!

[ceb]

Hardsubs are part of the video. So do the math...
Yeah, I added it later. Perhaps for you and the others it's obvious but for someone like me who knows little it's not. :D
Any suggestions?

Anyway, I see that there are a lot of people who enjoy watching anime fansub like me. :D
I watch both soft and hard sub, what are the best quality settings regardless of performance?
I leave everything at default except "Maximum Texture Resolution" changed to "Desktop" (improves subtitle quality of softsubs) and remove the tick from "Round up to power of two".

[bjd]

Just noticed that somwhere between .3058 and .3085 seeking via a web browser is broken. Sorry have not got any compiled versions between to test and isolate futher. All other web browser functions seem fine.

[mike23]

When I start to play an *.mpg with MPC-HC under 64bit Win7 then at first a popup appears:

"THe installed DirectX End-User Runtime is outdated. please download and install the June 2010 release or newer in order for MPC-HC to function properly"

Hmm, I have already DirectX 11 installed. At least this is shown in dxdiag.exe

So why does MPC-HC not detect this fact?

Mike

[mike23]

I can play an *.mpg successfully with MPC-HC here on 64bit Win7 but the movie fails with other well known wide spread players (e.g.VLC). It is sometimes stuttering.

In order to ensure that my friends and colleagues can play the movie as well I would like to "repair" the *.mpg.

Can I load it in MPC and save it WITH re-indexing it?
Video codec is MPEG-1

Mike

[nevcairiel]

Just do what it tells you to, don't question the popup!

But seriously, if it shows that, you're missing a component. The part of DX that MPC-HC uses is *NOT* part of the standard DirectX, not 9, not 10, and not 11. If a game needs it, they usually bring it with them, but MPC-HC just asks you to download it yourself.
And to clear up something else: DX11 is *NOT* a replacement for DX9. Alot of applications still use DX9, including MPC-HC, so you need to keep that updated as well - which the Runtime update will do.

[namaiki]

For further information about the DirectX Runtimes:
http://www.google.com/search?sourceid=chrome&ie=UTF-8&q=directx+runtime#sclient=psy&hl=en&safe=off&biw=1280&bih=709&source=hp&q=site:microsoft.com+directx+runtime&aq=f&aqi=&aql=f&oq=&pbx=1&bav=on.2,or.r_gc.r_pw.&fp=e7386a9c3eba4fdd

See the first two entries.

[jos99]

9800gt doesnt support vld. it supports idct, which the microsoft decoder does support.
K, thanks for the explanation, I could have sworn MPEG2 DXVA used to work though, maybe not:D!

[kikker]

Anyone have an idea as to what would create such blockiness and green smearing when I switch to full screen? I just put together a new system, and I've tried several different output formats, all of which result in the same thing.

Is there a setting I'm overlooking possibly? I'm playing back primarily MKVs for HD material on a HDTV. Playing the same material within a window works just fine.

I have latest AMD video drivers for my 3xxx card.

Thanks...

[pankov]

Guys,
I'm not using MPC-HC as my main player but I use it from time to time to check if a given problem is player or filter related and I try to replicate the same graph in both MPC-HC and ZoomPlayer and I usually am able to do it.
Recently I've noticed a problem that looks very strange to me - if I add ffdshow audio decoder as preferred and disable the internal decoders for AAC/AC3/DTS and start playback of a video file it plays fine but if I drop another file on MPC-HC instead of stopping the first one and starting the new it directly starts it so I have two files playing ... or at least the audio of two and no video. Often the player crashes afterwards but not always.
Any body got a clue what am I doing wrong?
This is just after settings reset and adding a few external filters.
Here is my .ini file in case it's as simple as changing an option that I've missed
http://www.mediafire.com/?wk07btim5ks6p9d

[v0lt]

MPC-HomeCinema 1.5.2.3089 (x86) (http://www.mediafire.com/?xfgh33iws56bh)
Updated mpciconlib.dll. Redraw all the icons, change the order.
Need to reassociate extensions with icons.
The easiest way to do this:
1. Go to "Options ...-> Formats", uncheck "With icons" and click "Apply"
2. Checkbox "With icons" back and click "Apply" again
3. Close the "Options".


PS: 4-bit 16x16 icons and try a little blur for 32-bit 16x16 icons.
http://i.imgur.com/pD3C2.png

[Aleksoid1978]

Hi
This sample (https://rapidshare.com/files/460405805/Vier-001.mkv) will crash MPC-HC DXVA decoder on ~22s.
There seems to be problem in video itself, but decoder should handle it, not crash i think.

Fix in rev. 3090

[pawlol83]

After updating the MPC home cinema version of 3058 (x86) I have a problem with using the program from the WEB interface. I mean a problem with the inability to scroll through the video progress slider.

The problem occurs in the rev. 3088, 3089

Please refer to the problem. Thanks in advance.

[betaking]

to all dev how to remove DataExecutionPrevention form mpc-hc code? i have chnage release.vsprops DataExecutionPrevention="2" to DataExecutionPrevention="0"
but after compler mpc-hc still use dep!
<?xml version="1.0" encoding="windows-1250"?>
<VisualStudioPropertySheet
ProjectType="Visual C++"
Version="8.00"
Name="release"
>
<Tool
Name="VCCLCompilerTool"
Optimization="2"
PreprocessorDefinitions="_SCL_SECURE_NO_WARNINGS"
RuntimeLibrary="0"
DebugInformationFormat="3"
BufferSecurityCheck="true"
EnableEnhancedInstructionSet="1"
AdditionalOptions="/MP"
/>
<Tool
Name="VCLinkerTool"
GenerateDebugInformation="true"
OptimizeReferences="2"
EnableCOMDATFolding="2"
LargeAddressAware="2"
RandomizedBaseAddress="2"
DataExecutionPrevention="0"
/>
</VisualStudioPropertySheet>

[v0lt]

After updating the MPC home cinema version of 3058 (x86) I have a problem with using the program from the WEB interface. I mean a problem with the inability to scroll through the video progress slider.

The problem occurs in the rev. 3088, 3089
IMHO, the problem appeared in version 3072. There are some very strange changes

Fixed in 3094.

[clsid]

MSVC2010 uses common.props for settings.

[betaking]

MSVC2010 uses common.props for settings.
OK!
change
</ClCompile>
<Link>
<OptimizeReferences>true</OptimizeReferences>
<EnableCOMDATFolding>true</EnableCOMDATFolding>
<LargeAddressAware>true</LargeAddressAware>
<LinkIncremental>false</LinkIncremental>
to
</ClCompile>
<Link>
<OptimizeReferences>true</OptimizeReferences>
<EnableCOMDATFolding>true</EnableCOMDATFolding>
<LargeAddressAware>true</LargeAddressAware>
<DataExecutionPrevention>false</DataExecutionPrevention>
<LinkIncremental>false</LinkIncremental>
dep is remove

[Mercury_22]

Revision 3095 it's not working !
"...No file to patch..."
My working solution it's to delete this lines if [ -f mingw-w64-crt/misc/mingw_getsp.S ];
then
rm mingw-w64-crt/misc/mingw_getsp.S
fifrom buildmingwlibs.sh and this lines Index: mingw-w64-crt/misc/mingw_getsp.S
===================================================================
--- mingw-w64-crt/misc/mingw_getsp.S (revision 2509)
+++ mingw-w64-crt/misc/mingw_getsp.S (working copy)
@@ -22,14 +22,14 @@
#endif
ret

-.globl __MINGW_USYMBOL(longjmp)
- .def __MINGW_USYMBOL(longjmp); .scl 2; .type 32; .endef
-__MINGW_USYMBOL(longjmp):
+.globl _longjmp
+ .def _longjmp; .scl 2; .type 32; .endef
+_longjmp:
#ifdef _WIN64
xorl %eax,%eax
movl %eax, (%rcx)
- leaq __MINGW_USYMBOL(fctMsvcrtLongJmp)(%rip), %rax
+ leaq _fctMsvcrtLongJmp(%rip), %rax
jmpq *(%rax)
#else
- jmp *__MINGW_USYMBOL(fctMsvcrtLongJmp)
+ jmp *_fctMsvcrtLongJmp
#endif
from mpchc_Mingw64.patch

P.S. To use gcc 4.6.0 for x64 too you need to replace the "C:\MPC-HC\src\filters\transform\MPCVideoDec\ffmpeg_x64\root-i686-pc-mingw32\x86_64-w64-mingw32\include" folder with the one from your "C:\MSYS\mingw\x86_64-w64-mingw32\include"

EDIT: why don't we just add libgcc.a and libmingwex.a for both x86 and x64 to the SVN and make Everybody Happy ?

[hajj_3]

remember window position seems to have stopped working for me, it was working in MPC-HomeCinema 1.5.2.3033 (x64) but i just tried build MPC-HomeCinema 1.5.2.3093 (x64) and it won't remember the window position, i've tried unchecking and re-checking the box but it doesn't fix it.

Can someone else confirm this please.

[Aleksoid1978]

remember window position seems to have stopped working for me, it was working in MPC-HomeCinema 1.5.2.3033 (x64) but i just tried build MPC-HomeCinema 1.5.2.3093 (x64) and it won't remember the window position, i've tried unchecking and re-checking the box but it doesn't fix it.

Can someone else confirm this please.

Try this build - http://aleksoid.tosei.ru/Test/mpc-hc.exe.position.7z

If don't help - can make desktop screenshot before close MPC-HC, and after open(when restore position break).

[tetsuo55]

OK!
change
</ClCompile>
<Link>
<OptimizeReferences>true</OptimizeReferences>
<EnableCOMDATFolding>true</EnableCOMDATFolding>
<LargeAddressAware>true</LargeAddressAware>
<LinkIncremental>false</LinkIncremental>
to
</ClCompile>
<Link>
<OptimizeReferences>true</OptimizeReferences>
<EnableCOMDATFolding>true</EnableCOMDATFolding>
<LargeAddressAware>true</LargeAddressAware>
<DataExecutionPrevention>false</DataExecutionPrevention>
<LinkIncremental>false</LinkIncremental>
dep is removeI still don‘t understand why you need to disable it?
You do understand that any problems are caused by either broken filters, drivers or virisses/trojans/worms....

[betaking]

I still don‘t understand why you need to disable it?
You do understand that any problems are caused by either broken filters, drivers or virisses/trojans/worms....

I disable only for mpc-hc! not disable for other app! because i diy a codecpack for some chinese people. if not disable dep for mpc-hc! mpc-hc be crash. Mine is a separator or an old version of the decoder causes, but they do not know, they are only concerned with stability and compatibility, because some of the decoder has stopped development of alternative solutions can not be found!

[hajj_3]

Try this build - http://aleksoid.tosei.ru/Test/mpc-hc.exe.position.7z

If don't help - can make desktop screenshot before close MPC-HC, and after open(when restore position break).

That build doesn't fix it. I don't know what this means: "and after open(when restore position break)"

[Aleksoid1978]

That build doesn't fix it. I don't know what this means: "and after open(when restore position break)"

Make desktop screenshot before you close MPC-HC, and after open again.

[madshi]

I still don‘t understand why you need to disable it?
You do understand that any problems are caused by either broken filters, drivers or virisses/trojans/worms....
One good reason is that the DScaler IVTC Mod settings dialog crashes with DEP enabled. And that's the best MPEG2 decoder for movies.

[pankov]

Guys,
I'm not using MPC-HC as my main player but I use it from time to time to check if a given problem is player or filter related and I try to replicate the same graph in both MPC-HC and ZoomPlayer and I usually am able to do it.
Recently I've noticed a problem that looks very strange to me - if I add ffdshow audio decoder as preferred and disable the internal decoders for AAC/AC3/DTS and start playback of a video file it plays fine but if I drop another file on MPC-HC instead of stopping the first one and starting the new it directly starts it so I have two files playing ... or at least the audio of two and no video. Often the player crashes afterwards but not always.
Any body got a clue what am I doing wrong?
This is just after settings reset and adding a few external filters.
Here is my .ini file in case it's as simple as changing an option that I've missed
http://www.mediafire.com/?wk07btim5ks6p9d

Guys,
I don't want to be PITA but can anybody help me with this?
Am I the only one having this problem?
I've tried different versions of MPC and FFDShow (including the latest ones) I still can't make it work.
:(

[CruNcher]

One good reason is that the DScaler IVTC Mod settings dialog crashes with DEP enabled. And that's the best MPEG2 decoder for movies.

Hmm isn't dscaler outdated no DXVA support also ?


Each and everyone has their own view about what's best in a certain area.

I guess that here applies as well ;)

[hajj_3]

Make desktop screenshot before you close MPC-HC, and after open again.

wow, it worked :) But when i do it with the installed version it doesn't work. Is it possible to fix the bug so that no-one experiences this?

[namaiki]

Guys,
I don't want to be PITA but can anybody help me with this?
Am I the only one having this problem?
I've tried different versions of MPC and FFDShow (including the latest ones) I still can't make it work.
:(

Does this require both LAV Splitter and LAV CUVID decoder to be used to reproduce? (For one, I cannot use LAV CUVID)

[betaking]

One good reason is that the DScaler IVTC Mod settings dialog crashes with DEP enabled. And that's the best MPEG2 decoder for movies.

Yes.you are right! and windows media player not use dep too!

[nevcairiel]

One good reason is that the DScaler IVTC Mod settings dialog crashes with DEP enabled. And that's the best MPEG2 decoder for movies.

Only the configuration dialog crashes, the decoder itself works just perfectly for me.

[betaking]

Only the configuration dialog crashes, the decoder itself works just perfectly for me.

windows xp sp2 have dep! but mpc-hc not crashes!:eek:

[CruNcher]

betaking DEP is not enabled by default for all applications, only windows system related files on Pre Vista/7 you have to force it to be sure it wouldn't crash.

[pankov]

Does this require both LAV Splitter and LAV CUVID decoder to be used to reproduce? (For one, I cannot use LAV CUVID)
no, it happens with every splitter/decoder I've tried (both internal and external (Haali/CoreAVC/ffdshow video)). The only thing that breaks it is ffdshow audio decoder.
:(

Edit:
I've also noticed that it doesn't matter if I drag&drop the file or use the "File->Open File" menu.
Also I've tried using EVR-CP instead of madVR and I noticed that it tries to display both videos too - a few frames from the first opened file and a few (but less) from the second one ... and when I press "stop" only the second one stops and the original continues to play.

[tin3tin]

Has opening of Avisynth scripts been broken? MPC-HC seems to crash when I try to open .avs files.

[betaking]

betaking DEP is not enabled by default for all applications, only windows system related files on Pre Vista/7 you have to force it to be sure it wouldn't crash.

no!mpc-hc's code use DEP enabled by default!

[CruNcher]

no!mpc-hc's code use DEP enabled by default!

Ahh yes sorry its indeed enabled by default so no crash on dscalers option dialog also here on XP SP3

there is though some other oddity with auto creating a graph and dscaler involved it clashes with ffdshow over here somehow, i have to block ffdshows video decoder (even with Mpeg-2 or raw rendering disabled) or it crashes it (not very nice) in MPC-HC. Though i wont give my hand into the fire what the actual cause for the ffdshow.ax crash is when dscaler is involved (to complex auto chain).

[madshi]

Hmm isn't dscaler outdated no DXVA support also ?
Who care about DXVA for *MPEG2*?? Any CPU released in the last 20 years can decode HD MPEG2 just fine. Ok, maybe not 20 years, but you get the picture.

DScaler might be outdated. But DScaler IVTC Mod is not outdated at all, it's still the best MPEG2 decoder for movies because it's the most reliable way to do IVTC. When using DXVA decoding/deinterlacing, you get 30p or 60p, but not 24p. So you'll have the typical 3:2 motion stutter.

Only the configuration dialog crashes, the decoder itself works just perfectly for me.
Yep, same for me.

[CruNcher]

Yeah i see what this Mod is for somekind of realtime software decimate for Hard Telecined stuff so practically a very automated way for handling rare non flagged streams :)


When using DXVA decoding/deinterlacing, you get 30p or 60p, but not 24p. So you'll have the typical 3:2 motion stutter.

It depends on the content and the Decoder but yeah if flags are missing you are lost anyways and i see the use of this Dscaler Mod their,unless you have a hardware chain that can cope with that in realtime.
Though Nvidias Hardware IVTC (at least @ a playback scenario) should be as efficient in cadence detection without flags on the renderer it's interesting that it fails for you even on Vista/7 EVR and you have to fallback to a Software solution such as this Dscaler Mod.
Not sure though but i guess Nvidia doesn't support your renderer yet and its questionable if they ever will so most probably it would be useful their too if you can live with the cpu usage ;)

[madshi]

Yeah i see what this Mod is for somekind of realtime software decimate for Hard Telecined stuff so practically a very automated way for handling rare non flagged streams :)
"Rare"? You must be joking?! None of the MPEG2 broadcasts that I've seen were properly flagged from start to end.

It depends on the content and the Decoder but yeah if flags are missing you are lost anyways and i see the use of this Dscaler Mod their,unless you have a hardware chain that can cope with that in realtime.
Though Nvidias Hardware IVTC (at least @ a playback scenario) should be as efficient in cadence detection without flags on the renderer it's interesting that it fails for you even on Vista/7 EVR and you have to fallback to a Software solution such as this Dscaler Mod.
To my best knowledge, NVidia's hardware IVTC results in 30p, just like ATI's does. The hardware IVTC is used to put the correct fields together, but it does *not* remove duplicate fields, so the output is 30p and not 24p.

[robpdotcom]

"Rare"? You must be joking?! None of the MPEG2 broadcasts that I've seen were properly flagged from start to end.

Much of the telecined ATSC material isn't flagged at all, so DScaler is the only option.

[CruNcher]

"Rare"? You must be joking?! None of the MPEG2 broadcasts that I've seen were properly flagged from start to end.


To my best knowledge, NVidia's hardware IVTC results in 30p, just like ATI's does. The hardware IVTC is used to put the correct fields together, but it does *not* remove duplicate fields, so the output is 30p and not 24p.

Strange i always got the impression it drops to much frames if it becomes active and thus why i gave it up and turned it off.

Much of the telecined ATSC material isn't flagged at all, so DScaler is the only option.

do you have such a sample with a film/video switch non flagged where it beats everything else ?

[tetsuo55]

Libavcodec and DXVA MPEG2 decoding should be mostly identical in quality (there was a time when Dscaler could beat the others for image quality).

But i see the discussion is not about the decoding of the individual frames, but the deinterlacing methods applied to the decoded stream.

Isn't correct detection and subsequent deinterlacing simply a missing(or broken) feature in ffmpeg?

[madshi]

Strange i always got the impression it drops to much frames if it becomes active and thus why i gave it up and turned it off.
It works very well for pure movie content.

But i see the discussion is not about the decoding of the individual frames, but the deinterlacing methods applied to the decoded stream.
Correct.

Isn't correct detection and subsequent deinterlacing simply a missing(or broken) feature in ffmpeg?
I don't know if ffmpeg has proper IVTC functionality built in. If so, no DirectShow filter seems to use it.

[CruNcher]

Where do i get a Mpeg-2 sample Bitstream with missing flags and film/video switching :) ?

[laserfan]

Where do i get a Mpeg-2 sample Bitstream with missing flags and film/video switching :) ?
What tool(s) can be used to identify these flags?

[CruNcher]

immediately to my mind comes Dgindex status display to me :)
also when you playback with Mplayer (ffmpeg) you should see state switches in the command line, which means nothing else that there was a flag indicating it.

[hoborg]

Fix in rev. 3090

:thanks:

[tetsuo55]

Ok i have been told that there is nothing wrong with ffmpeg.
The post-processor/Deinterlace should be able to detect if the frames are or are not telecined and change to IVTC mode when they do.

What happens when you take a correctly flagged file and feed it to nvidia or ATI with vector adaptive and pulldown detection enabled?

[madshi]

Ok i have been told that there is nothing wrong with ffmpeg.
The post-processor/Deinterlace should be able to detect if the frames are or are not telecined and change to IVTC mode when they do.
I don't know if that's true. Maybe it is. If so, this specific ffmpeg functionality does not seem to be made use of in either MPC-HC or ffdshow.

What happens when you take a correctly flagged file and feed it to nvidia or ATI with vector adaptive and pulldown detection enabled?
Probably depends on the decoder. But anyway, it's not really important how correctly flagged files behave, since most MPEG2 broadcasts are *not* correctly flagged, as far as I've seen...

[tetsuo55]

Well once we know that other deinterlacers are capable of correct IVCT it might be trivial to add the flags on the fly using similar logic as in dscaler.

Also dscaler might be able to be made compatible with DEP, does it get compiled with msvc? update the projects to vs2010 and it should get compiled as a DEP compatible filter.

[madshi]

Well once we know that other deinterlacers are capable of correct IVCT it might be trivial to add the flags on the fly using similar logic as in dscaler.
Yeah, that might be a job for ffdshow?

IVTC is a complex topic, though: Detecting the NTSC/ATSC 3:2 movie cadence and removing it to get from 60i to 24p is relatively easy. Doing 50i -> 25p is much more difficult. And then there are also various other cadences, e.g. for Anime content. The question is if ffmpeg supports all that, and with which quality/reliability.

Also dscaler might be able to be made compatible with DEP, does it get compiled with msvc? update the projects to vs2010 and it should get compiled as a DEP compatible filter.
Yeah, it might be possible to recompile/fix the DScaler IVTC Mod. I've no time for that, unfortunately...

[tetsuo55]

Here is my suggestion:
1. Find the sourcecode from the mod as written by kzeuh (he wrote 2 mods, the 2nd one would be preferable)
2. Get Dscaler source from the sourceforge page (might have to go back a few revisions, as the mod was made against release 6 and not 8)
3. Update the project files to vs2010, compile, fix warnings etc.. (and once you finish it should be DEP compatible)
4. Update the mod so it works with release 8 to get the most profit.
5. Add the IVTC detection code to ffdshow (should result in the same or better results, the biggest problem is that the current mod assumes film always, it does not seem to be "detection" but rather forced IVTC)


@ 50i to 25p (Thats not a problem at all imho, since 50 and 25 are multiples of eachother)

[nevcairiel]

You can get IVTC with ffdshow if you setup a simple AviSynth script inside to do it - that works pretty good, but its of course not the fastest solution.

Oh and on the ffmpeg topic:
avcodec is a decoder. It'll give you decoded fields, and all the flags it found in the bitstream.
In addition, it also includes libavfilter, which is a post-processing framework, but its still very early in its development cycle. It does include a yadif deinterlacer, but no special IVTC logic that i'm aware of.
In any case, ffdshow and other ffmpeg-based decoders usually only use avcodec to decode the fields, any post processing is then done outside of ffmpeg itself.

While this would be a job for avfilter, one day, its currently not supported by ffmpeg (and not a "bug").

[laserfan]

Where do i get a Mpeg-2 sample Bitstream with missing flags and film/video switching :) ?

What tool(s) can be used to identify these flags?

immediately to my mind comes Dgindex status display to me :)
also when you playback with Mplayer (ffmpeg) you should see state switches in the command line, which means nothing else that there was a flag indicating it.
Hmmm guess I dunno how to interpret; looked at both DGIndexNV Info window and debug=true when feeding avs to Vdub, and don't know how to find pulldown flags.

But thanks for turning me-on to Mplayer--it plays my source smoothly where no other player (incl MPC-HC) does!

[tetsuo55]

After reading the .txt files included with the dscaler mod it seems to me that the mod is hardcoded for IVTC, thats why it works so well, it simply treats all content as telecined.

The search algorithm used should easily be portable to ffdshow and/or usable through avisynth.

So there are 3 options:
-Recompile the mod to work with DEP (but get stuck with 5+year old decoder)
-Update the mod and then recompile it (and get stuck with a slightly less old decoder)
-Use FFdshow (and get the latest decoder)

The mod is said to also have a big overhead for the detection phase, and less overhead for the blindly decimate phase.

[bigzaj]

I'm having an issue with MKV playback with 5.1 audio. On my 32-bit machine when playing back a file it correctly identifies the audio as 640kbps 6-channel AC3. On my Win 7 64-bit machine it tells me "unidentified audio" and plays back in 2.1 stereo. Any thoughts on what this could be, do I need to configure the built in AC3 filter to 3 front + 2 rear with LFE?

[madshi]

@ 50i to 25p (Thats not a problem at all imho, since 50 and 25 are multiples of eachother)
No no. It *is* a problem. If you put the wrong 50i fields together you'll get heavy combing artifacts. Doing 60i -> 24p is easy because some fields are duplicates. Detecting duplicates is not so difficult, and once you've detected them you already know the exact telecine pattern. With 50i there are no duplicate fields. So finding out which fields belong together is more difficult. Especially if you want to do automatic video/film mode detection, too, which is extra difficult with 50i content.

[tetsuo55]

I guess i don't understand the problem enough then, or we're talking about different things.

In Pal land we either film at 25fps or speed up 24 to 25fps.
The you either show every frame twice, add a soft telecine flag or have actual interlacing.

Regardless of which one it is, it looks to me that the solution is simple, if its interlaced just deinterlace, else just treat it as 50fps progressive.
Now the 24 to 25 fps speedup is something that some people still want to solve, but thats what reclock is for.

But i should stop talking about it from google-search, and ask you for a specific example.

[VipZ]

I have finally got around to compile MPC-HC and got it working perfectly with MingGW 4.6.1 for both x64 and x86.

I like the current BE mod for MPC-HC and would like to compile this as well, but applying the .patch from xvidvideo.ru makes the build go very wrong :p

Does anyone have advice on this?

Thanks

[madshi]

I guess i don't understand the problem enough
Agreed. Here comes a short crash course:

There are 2 kinds of interlaced contents:
(1) film content (filmed in 24p or 25p, then telecined to 60i/50i)
(2) video content (filmed natively interlaced)

You do not "deinterlace" (1). That's the wrong technology. If you "deinterlace" native film content, e.g. by using yadif, you're damaging image quality. The *only* proper way to treat (1) is to do IVTC. That applies to both 60i NTSC/ATSC content and to 50i SD/HD PAL content. Ideally, IVTC results in 24p for NTSC/ATSC movies and in 25p for PAL movies.

For (2) you can't use IVTC, there you have to use a video mode deinterlacer like yadif which (when used correctly) will output 60p for NTSC/ATSC content and 50p for PAL content.

IVTC for PAL movies means that you have to detect which 2 interlaced fields belong together. Once you found that out you simply weave them back together. The weaving is the easy part. Reliably finding out which fields belong together is hard. Finding out whether PAL content is native video or film is even more difficult.

[tetsuo55]

Ok thanks, I also read the wiki page for it.

It made clear that "extra" frames are inserted which have to be removed.
And whatever is left behind has to be de-interlaced into a progressive image. I read elsewhere that there is also a telecine where the remaining frames are already progressive and can be displayed as-is, not sure if that exists in the wild.

The wiki page also made clear that most telecine detectors use image comparison to detect repeating frames, and use that to detect the cadence pattern to decimate with.

[nevcairiel]

the nvidia post-processor does reproduce the original frames perfectly (at least i have not seen any artifacts anywhere), it just does not decimate the extra frames.
If i could write an algorithm for decimation into LAV CUVID .. but i'm really not that deep into video processing..

[madshi]

It made clear that "extra" frames are inserted which have to be removed.
It's extra fields, not frames. And that is only true for NTSC/ATSC, but not for PAL. There are no extra fields for PAL.

And whatever is left behind has to be de-interlaced into a progressive image.
Whatever is left has to be *weaved* together, but you need to know which fields belong together, otherwise you'll get the typical weave combing artifacts.

I read elsewhere that there is also a telecine where the remaining frames are already progressive and can be displayed as-is, not sure if that exists in the wild.
HD DVD video tracks are encoded this way. Basically they're encoded progressively, but there are flags in the video stream telling the decoder to interlace after decoding. Broadcasts sometimes have sections which are encoded like this, too, but usually it's only sections, not the whole movie.

The wiki page also made clear that most telecine detectors use image comparison to detect repeating frames, and use that to detect the cadence pattern to decimate with.
Yes, but as I said, repeating fields/frames do not occur with PAL content, which is why PAL IVTC is more difficult.

[G_M_C]

Madshi, i seem to remember an avisynth plugin called RePal, that tries to 're-pal' telicined etc. sources. Maybe the source code of the plugin is still available, and you can use it to see how it's detection algorithms work.

[madshi]

I know how these algorithms work.

[Andy o]

pankov and I are having the same trouble (http://forum.doom9.org/showthread.php?p=1499647#post1499647) with different particular decoders, (him ffdshow and me Arcsoft audio decoder). Looks like there's something with our MPC-HC installs. Just posting here so we can continue the discussion in a relevant thread.

[SamuriHL]

Yea, try my custom build and let me know if that does or doesn't help. It's 3098 so it should be "the latest" as of a few hours ago. Remember it has NO internal filters at all since I don't ever use them. It SHOULDN'T fix anything for you at all related to this problem, but, I figured it couldn't hurt.

[Andy o]

I'm trying wiping out everything Arcsoft from my system first, will try your build later, thanks.

[SamuriHL]

No problem. I'll check on your progress in the morning. I'm going to bed soon. I hope you're able to figure it out.

[pankov]

Samuri,
I've tried your custom build and sadly it doesn't fix the problem.
:(
On the bright side I've reset my ffdshow audio decoder settings (just renamed the key in the registry) and the problem disappeared.
Now I'll have to hunt it down to a particular setting but at least now I have a direction.

Nevertheless, I still think there is a bug in MPC-HC, because it shouldn't be possible a single decoder/filter to cause such a behavior in a player.

[Andy o]

Hmm I wiped my Arcsoft stuff from my PC, even ran a reg cleaner, and still the problem's there. Tried Samuri's build and the xvidvideo build, installing and running from the folder (from the .zip), and still no go. Also took video totally out of the equation, making a DTS-HD .mka audio-only, and MPC-HC still doesn't close properly the file before moving on to the next. I'm gonna install the stuff on my laptop see if it happens here too.

[pankov]

Andy,
in this case which audio decoder do you use or you are bit-streaming with LAV?

[Andy o]

I'm not bitstreaming anything right now. I'm decoding the DTS-HD with the Arcsoft decoder, but I'm having this bug with it. I haven't fixed the DD+ bug I mentioned at the LAV thread either.

I just tried on my laptop, and even not having LAV installed, it still happens with the MPC splitter and the Arcsoft decoder.

[Andy o]

Just found something weird. I've stripped my setup to the basics. Using freshly downloaded MPC-HC with default settings, MPC codecs and mkv splitter. Only Arcsoft decoder added for DTS-HD MA. ffdshow is uninstalled. The bug keeps happening, but only happens the first time. If I avoid it by pausing or stopping before opening the next file, it works OK, and after that, I don't have to stop or pause, it just works. Anyone has any ideas?


EDIT:
hmm it seems it still keeps something open, cause it doesn't work with WASAPI exclusive via ReClock, but works seemingly fine when not using exclusive mode.

[SamuriHL]

I don't know what to tell you guys. The build I made rules out internal filters interfering. Very odd.

[Andy o]

Man this thing is really bugging me. I'll try with a third PC which I never even turn on anymore.

[Ger]

I have a faint memory of similar symptoms, and I think it may have been related to DirectShowSpy.dll in that case, but I don't remember any details. Maybe it prevented the filters from unloading properly, causing similar symptoms to what you guys are seeing.

[pankov]

Andy, Samuri,
I've narrowed my problem to 15 different values in the registry for ffdshow audio decoder.
Sadly I'm too tired after a 12 hours work day so I'll do my comparison/test tomorrow and I'll tell you which is the option that breaks it for ffdshow.
I hope this will point to the real problem in MPC-HC and let's hope someone will be able to fix it.

[pankov]

Well ... I couldn't spleep knowing a solution can be so close so I did it tonight.
The registry value that is causing my problems is ""addToROT"=dword:00000001". If the value is 0 everything is OK. If it's 1 I get multiple files playing at the same time which can sometimes lead to a crash.
Now can somebody tell me what this option does and why is it making MPC-HC behave so bad?

[SamuriHL]

I saw your ffdshow thread post. I know absolutely nothing about that setting at all.

[G_M_C]

Well ... I couldn't spleep knowing a solution can be so close so I did it tonight.
The registry value that is causing my problems is ""addToROT"=dword:00000001". If the value is 0 everything is OK. If it's 1 I get multiple files playing at the same time which can sometimes lead to a crash.
Now can somebody tell me what this option does and why is it making MPC-HC behave so bad?

Google lead me to this: http://msdn.microsoft.com/en-us/library/dd390650%28v=vs.85%29.aspx

As far as i understand addToROT should always be returned to 0 when a graph is terminated.

[Ger]

Probably the same thing:
http://msdn.microsoft.com/en-us/library/dd390650(v=vs.85).aspx

EDIT: I was too slow.

[Aleksoid1978]

i have encodec one file with Xmedia Recode and try to watch with MPC-HC but there is no VIDEO & AUDIO using a fresh installed MPC-HC but adding CoreAVC i'm able to watch the movie...anyone else having problem with this movie and MPC-HC

here is the sample movie : http://www23.zippyshare.com/v/51283375/file.html

here the mediainfo :

Test on this build - http://aleksoid.tosei.ru/Test/mpc-hc.exe.mov_BaseLine.7z

[Andy o]

That's quite a find pankov, great!

I have also solved my problem just now, but it didn't have to do with ffdshow. I didn't even have it installed, and there's no ffdshow registry keys. I was using the TMT3 decoder from its installation folder. If I only isolate the relevant files in another folder, and register the filter there, my problem is gone. Not necessary to uninstall TMT3 either, I assume it still works. Finally.

[SamuriHL]

That's probably why I'm not seeing the issue, Andy. I have a separate dir that I register a mix of the TMT3 ax file with the dll's from TMT5.

[Andy o]

Still having the other minor problem with DD+ where seeking causes channel corruption. The front left stays the same, the FR goes to C, the C goes to FR, and the rest are silent. If I don't seek, the whole sample runs OK. I'll worry more though, when an actual legitimate DD+7.1 movie is found!

BTW, the TMT3 DD+ dll is the same file as in TMT5 .186, could you check if your latest TMT5 has a newer version please?

[Andy o]

by the way, why use the dlls of TMT5? I know the DTS decoder is newer, but are there any documented differences?

[SamuriHL]

They haven't changed it in a very long time. It's only used for HD DVD as far as they're concerned, so, think about when that was last touched. :)

[SamuriHL]

Don't know. I'm thinking it's more bug free. :)

[Andy o]

So, the fact that downsampling is being done with newer decoders (TMT3 .190) doesn't depend on the dlls, but on the asaudiohd.ax?

[SamuriHL]

Yup. That's correct.

[ikarad]

@developpers: Anybody works upon sub renderer or not at this moment? Can we expect a sub renderer without bugs (because there are bugs with bluray subs) soon or not?

[nevcairiel]

There are no MPC-HC devs, so no-one is working on anything. The only thing thats happening occasionally are random bug fixes, but no-one is working on any major projects like that.

Personally, i would be willing on helping to work on a new sub renderer, but since i only do things when they are done right, that needs someone thats working on the renderers side of things, defining a new interface to get rendered subs onto the video frame.

Since i don't expect that to happen, i'll just do other things. =)

[takenori]

my display stat become blurry all of sudden.
did a clean install, reset everything to default, no dice.
http://img852.imageshack.us/img852/4060/desktop2011051123022976.jpg (http://img852.imageshack.us/i/desktop2011051123022976.jpg/)

Did reset everything on my amd graphic driver to factory state, just in case,
still no dice

any thought?

[namaiki]

Maybe check if your graphics driver has anti-aliasing or if ati that Morphological Anti-aliasing thing forced on.

[takenori]

Maybe check if your graphics driver has anti-aliasing or if ati that Morphological Anti-aliasing thing forced on.

yes, I did a factory reset on my catalyst. the morphological antialiasing turned off by default, and theres no option to disable antialiasing, just 3 option to use application setting, enhance, or override application setting.
And the blur is still there

[adam777]

Hello all,
Just a quick question - is there an option in MPC HC for reading a stream via pipeline?
Specifically, sending a stream from a different program to VLC's STDIN?
The other program also has an option of UDP streaming, but I could not get it to work either.
Thanks, Adam.

[nightfly]

Nvidia's 3d Vision player looks a lot like MPC.

link (http://www.nvidia.com/object/3d-vision-video-player-1.6.9-driver.html)

link to screen shot (http://3dvision-blog.com/nvidia-has-released-3d-vision-video-player-v1-5-2/)

Forgive me if this has been discussed before. But if not, it seems like they offer up the code/sdk for their 3D player to devs. Perhaps something MPC-HC could "plug-in" for 3D support?

[Thunderbolt8]

does anyone know how I can use dscaler5 ivtc mod with mph-hc (32) on win7 (64) ? I cannot add it exclude it from DEP in system settings.

[LigH]

@ nightfly:

Even MPC looks a lot like WMP 6.4 from times of Windows 98 still (therefore the name: "Media Player Classic"): It is just a very simple window with generic widgets around a renderer. There is no design to be robbed. ;)

[HoP]

i think one of them should be remove
my build: xhmikosr
MPC-HC_r3097_x64_MSVC2010

http://www.pixelz.fr/2/4/3/663d169f73375c164278f2d7c9fc1.png

[Ger]

Press the default button for "MPEG Media file", like it says here (http://sourceforge.net/apps/trac/mpc-hc/changeset/3088).

[WonderCsabo]

Hi Developers and sansnom05!

Did you manage to make any progress on ticket 1087 (http://sourceforge.net/apps/trac/mpc-hc/ticket/1087)?

[8GPMPY]

@8GPMPY: Desktop composition can be a bit of a strain on the GPU and memory. Also, EVR CP and VMR-9 (renderless) are not light renderers. Even on default settings, everything becomes a problem when displaying subtitles on low end GPUs. (Or even newer ones with animated subtitles.)
I'll be patching a lot of things in due time, but I don't have a perfect solution for you at the moment. Just try to relieve the strain from CPU and GPU by disabling some system items and hope for the best.
I'm getting crazy! :eek:
I had a spare Geforce 9400GT and tested it. On h.264 material it was great, but unfortunately it only has VP2 and is not able to provide DXVA on VC-1. [1]
Then I swapped it with a HD4350, which has UVD 2.2 instead of plain UVD of the HD3200 IGP [2].
But this shows the same lags as the HD3200.

Here is a MediaInfo of a problematic MP4 with 25 fps:
http://pastebin.com/9Wk7zFvc

And here a VC-1 with 59.940 fps:
http://pastebin.com/bFp4erQZ

Once again: on Windows XP both files play fine.

[1] http://en.wikipedia.org/wiki/Nvidia_PureVideo
[2] http://en.wikipedia.org/wiki/Unified_Video_Decoder

[oddball]

Any ideas what would prevent me from going fullscreen once I hit ctrl-enter or setting the hotkey to just enter? I can do thr alt-enter option but the normal fullscreen does nothing. Build 3104 btw but I am pretty sure this happened before with an earlier build.

[psychoirfan]

with evr/evr custom
http://i990.photobucket.com/albums/af27/pisipsycho/th_evr.jpg (http://s990.photobucket.com/albums/af27/pisipsycho/?action=view&current=evr.jpg)

with vmr9/vmr7/haali renderer..(others)
http://i990.photobucket.com/albums/af27/pisipsycho/th_vmr9.jpg (http://s990.photobucket.com/albums/af27/pisipsycho/?action=view&current=vmr9.jpg)

Colors are terrible and there is a strong denoiser like fft3dgpu with evr modes, cause of that i can't use dxva. Of course, all shaders off :).
The problem exist in windows media player too.
I think it's ati's driver problem, but i'm not sure. I cleaned ati driver and installed only driver(without ccc, avivo ...). It didn't work.

an avisynth script like
directshowsource("something.mkv")
has no problem.

win7 x86
ati mobility hd5730m
ccc 11.5

[Mangix]

that's due to dynamic contrast being enabled in ATi's CCC and DXVA. disable one of those and the first will look like the bottom.

[psychoirfan]

I installed again ccc. And closed all of video settings in panel. This time worked. http://forum.darkhardware.com/images/smilies/shot.gif I tried that before. I have no idea what's changed this time. Thanks for your help

[Aleksoid1978]

I little mistake - rev. 3107 fix rev.3105, not a 3015 :)

[CruNcher]

Could somebody take a look @ this and try to maybe reproduce ? http://forum.doom9.org/showpost.php?p=1501021&postcount=559

[pirlouy]

@CruNcher: can you stop posting big images ? Use thumbnails or links please.

And please, change your signature, use a tiny one. I'm sure you know a lot more things than me concerning multimedia, but think about others: big images use CPU, RAM, bandwith; and your signature bothers at least me when I want to read threads...

[Lincoln Burrows]

How can we report a bug in MPC-HC?

I have reported this before but no one have fixed yet.

If someone can look at this:

- When we change the language from english to another one (portuguese for example) - MPC-HC settings - we are not able to select audio streams anymore. This is only happening when we change the language to another one which is not english.

English (working):

http://i.imgur.com/PfUPz.png

Portuguese (not working):

http://i.imgur.com/gLRxm.png

[v0lt]

This is only happening when we change the language to another one which is not english.
This problem has three translations: Armenian, Portuguese (Brasil), and Slovak.

[Mercury_22]

Is it possible to make MPC-HC use an external "icon.ico" file like it does with the toolbar cause I'll really like (a simpler) way to use a different "icon.ico" for the x64 (or x86) version ?

P.S. And before you ask :)
e.g. to avoid the confusion between the x86 and x64 (formats / file association) on systems with both versions or
for people who like to use another ico (e.g. I really prefer the one from BE mod)

https://sourceforge.net/apps/trac/mpc-hc/ticket/1333

[JanWillem32]

The main file icon for .exe files is hard-coded in the executable header as a series of bitmaps. As for as I know, it's not possible to make it link to something else, as I've not seen interfaces to manipulate Windows Explorer to load external icons for .exe files.
The icons and other image-related resources can be changed for usage in the program itself without hard-coding them, though.

[LigH]

I believe, more exactly it is called a "resource"; they are not "in the header", but they are in specific file segments, separated from the code.

Resources in executables can be changed by a "resource editor" tool, but with some important restrictions. Especially their size and location must not change.

[madshi]

Especially their size and location must not change.
Actually they may. See freeware "Resource Hacker".

[6205]

IMHO those new 16x16 pix icons are too simple. Also even if i have desktop setup in 32-bit colors i see only their uglier 4-bit color versions.

Anyway... i would sugest to remove entire icon cache "mpciconlib.dll" and use generic icons only, available in main "mpc-hc.exe" file.
But even those need some cleanup (view picture), because they contain Aero style icons, which needs to be replaced. http://oi51.tinypic.com/2l9mjj7.jpg

MPC-HC was always lighweight and imo there no need to create new icon for every single new filetype which is supported. It's waste of time and resources.
I suggest to have 5 basic icons only, inside of mpc-hc.exe file:

1./ Main application icon
2./ Generic video file icon
3./ Generic audio file icon
4./ Generic playlist icon
5./ Geric optical disc icon(for CD audio, DVD or Blu-Ray discs)

Very good icons are in K-Lite Codec Pack version of MPC-HC. You should definitelly consider to use them.
http://8bf.deviantart.com/art/Media-Player-Classic-FINAL-132910148


Also good loking are these icons from DeviantArt.com

http://davidm147.deviantart.com/art/MPC-HC-Icon-Library-Glowing-208826509

http://sheppii.deviantart.com/art/xpAlto-MPC-HC-filetypes-203581644?q=boost%3Apopular%20mpc-hc%20icons&qo=4

[kumi]

The "Full Floating Point Processing" setting appears to be incompatible with ffdshow's "High quality YV12 to RGB conversion" setting. When both are checked on my system, video is either corrupted or repeats a few frames over and over (audio still works fine, though.)

Changing to "Half Floating Point Processing", or disabling "Full" fixes the problem. Or, disabling HQ conversion in ffdshow fixes it.

Software environment:
Windows XP SP3
MPC-HC r3112 (xvidvideo.ru)
ffdshow r3853 ICL12 (xvidvideo.ru)
EVR CP

[JanWillem32]

@kumi: That's well known on a lot of hardware. It's one of the reasons I've set RGB types as to be preferred dead last in the merits list.
For non-interlaced 8-bit 4:2:0 video these formats are preferred: NV12, YV12, I420/IYUV (in that order).
For interlaced 8-bit 4:2:0 video only NV12 is preferred.
For 8-bit 4:2:2 video these formats are preferred: YUY2, UYVY, YVYU (in that order).
The entire merits list counts 31 items, even very rare ones, mostly just to sort all Y'CbCr formats above the RGB formats.
For completeness: when using ffdshow tryouts, it's usually best to let the mixer do a lot of the conversion work. In the output tab of the video decoder, use "Set pixel aspect ratio in output media type", "Set interlace flag in output media type" and "Allow output format changes during playback".

[Dstruct]

Still no Color Management and Full/Half Floating Point Processing possible with version 1.5.2.3114! Options are grayed out. Working fine in 1.5.1.2903.

(Nvidia GeForce 7600GS, Windows XP SP3)

[JanWillem32]

The detection method in the trunk build is too strict. I've lowered the requirements for my builds months ago. Because the parts I write for the rendering engine are risky (to say the very least, as I usually break just as many features as I fix), it may take some time before integrating fixes like that.

[rsd78]

Hi,

I know this topic has been brought up, but I didn't see any new information recently so I thought I would ask.

Any plans or what is the status of fixing DXVA for Intel based platforms (e.g. Sandy Bridge)?

Thanks

[mr.duck]

I made program icons to replace the '321' clapperboard one.


Preview:

http://i.imgur.com/YqgkQ.png


These two icons are 256x256px. They are high quality. About the best I can do.

There is a bevelled version and a gloss version. I prefer the bevelled one, but gloss seems more popular these days.

What do you think?

[ranpha]

I made program icons to replace the '321' clapperboard one.


Preview:

http://i.imgur.com/YqgkQ.png


These two icons are 256x256px. They are high quality. About the best I can do.

There is a bevelled version and a gloss version. I prefer the bevelled one, but gloss seems more popular these days.

What do you think?

Can you made it so that it has a bright background? Else your icon set will be just out of place in Windows Explorer just like DivX Player/PowerDVD icon sets are.

Oh BTW, did you test the black icons on Windows 7 with Aero disabled?

[madshi]

Looks pretty good to me! However, there's still the problem of making such kind of icons look nice in just 16x16 pixels (and 32x32 and 48x48) for the Explorer.

[Midzuki]

http://i.imgur.com/YqgkQ.png

These two icons are 256x256px. They are high quality. About the best I can do.

There is a bevelled version and a gloss version. I prefer the bevelled one, but gloss seems more popular these days.

What do you think?

The same as madshi. :)

BTW, and just out of curiosity,
what's the name of that evil font? :D

[Hera]

D3DFS + PGUP/PGDOWN = PREV x2 / NEXT x2 chapter
PGUP/PGDOWN = PREV x1 / NEXT x1 chapter

Seems like you have the same function call twice somewhere in D3D FS mode.

EDIT: this seems relatively new development.

[v0lt]

These two icons are 256x256px. They are high quality. About the best I can do.

There is a bevelled version and a gloss version. I prefer the bevelled one, but gloss seems more popular these days.

What do you think?
here highlighted the middle finger?
:confused:

[mr.duck]

Can you made it so that it has a bright background? Else your icon set will be just out of place in Windows Explorer just like DivX Player/PowerDVD icon sets are.

Oh BTW, did you test the black icons on Windows 7 with Aero disabled?

I tried different backgrounds and with Aero disabled. Looks good :).


Looks pretty good to me! However, there's still the problem of making such kind of icons look nice in just 16x16 pixels (and 32x32 and 48x48) for the Explorer.
Thanks :). Yep it's a bit of a squeeze especially at 16x16! But did you see my icon library for MPC-HC (http://forum.doom9.org/showpost.php?p=1500873&postcount=44)? That goes down to 16x16. I know I can do it, just that I have only done the 256px icon so far for the program icon.


what's the name of that evil font? :D
Evil?? I thought it was a bit feminine. It's called Complete in Him (http://www.dafont.com/complete-in-him.font).


here highlighted the middle finger?
:confused:
Middle finger to the other bloated players ;)

Of course the logo concept is taken from the default startup screen logo thing. You know? Like the splash screen. It's just called the logo in the options panel.

[v0lt]

- When we change the language from english to another one (portuguese for example) - MPC-HC settings - we are not able to select audio streams anymore. This is only happening when we change the language to another one which is not english.
Check out this version - deleted (incorrect build)

[JanWillem32]

Check out this version - mpc-hc_3122_fix_translation.7z (http://www.mediafire.com/?acviijn6ytramcw)While we're at that subject, I've noticed that some languages don't have the normal tabs for the subtitles section in the options menu. Is that a similar issue? I was looking forward to editing this part of the menu. I can take a further look at it to see which languages are affected, of course.

[ranpha]

I tried different backgrounds and with Aero disabled. Looks good :).



Why don't you show us the icon in action in a default (white) WIndows Explorer window? Preferably in 'detail file list' mode. This icon set that is similar to DivX Player icons doesn't really inspire confidence in me.

[v0lt]

While we're at that subject, I've noticed that some languages don't have the normal tabs for the subtitles section in the options menu. Is that a similar issue?
I also noticed a problem with the list of subtitles and shaders on my build. The reason is not clear.

[Toddler Naruto]

Hi, I have a question, any chance of you adding an option/feature to minimize MPC-HC to the system tray?

[namaiki]

Hi, I have a question, any chance of you adding an option/feature to minimize MPC-HC to the system tray?

View-> Options-> Player-> Tray icon.

[Toddler Naruto]

View-> Options-> Player-> Tray icon.

Thanks namaiki :), I dunno why I didn't bother trying that out first lol, problem solved :D.

[Hera]

Speaking from the SSE build 1.5.2.3122,

The CPU drain imposed by EVR and EVRCP is quite high for me.
20%-30% CPU utilization to play a 720p MP4 (from YouTube).
More demanding video (and, in most cases, codecs) may lead to stuttering.
Compare that with WMPs 4%-12%.

I do recall being able to watch video with Atom underclocked 50%, this was a long time ago and with an old install of W7 (not I have more vanilla W7 SP1) and older NV drivers.
So it may be overhead from that W7 service which gives higher priority to audio or newer NV drivers? Meh.

Also, latest JanWilliam32's builds seems to get stuck on loading (infinite loop somewhere?) when GPU video scaling options are selected.

[nevcairiel]

WMP probably uses DXVA to off-load the decoding to the GPU, and your MPC configuration doesn't, at least thats what it sounds like.

[Hera]

WMP probably uses DXVA to off-load the decoding to the GPU, and your MPC configuration doesn't, at least thats what it sounds like.

No, I am not that thick.
MPC just has massive overhead compared to WMP regardless of whether it is EVR or EVR:CP.
I have no idea what this overhead is composed of and neither do I really care. It is just... bad.

[nevcairiel]

EVR is the same default EVR that WMP uses, the overhead in that case is not related to the renderer used.

[tetsuo55]

hera, do your samples have subtitles as that could explain it.
also have you tried resetting the registry settings?

[ceb]

Is it possible to view the thumbnail of .mkv videos? I don't use any other software besides MPC-HC.

[Hera]

hera, do your samples have subtitles as that could explain it.
also have you tried resetting the registry settings?

No. This was tested on an MP4 file ripped from YouTube.

Yes.

[Mercury_22]

MPC-HC's internal MPEG PS/TS/PVA splitter doesn't detect ALL the audio streams in a m2ts file
e.g. Sample (http://www.multiupload.com/MV7RNMR1LW) it detects only 7 out of 8 audio (it doesn't detect the last one DTS Express )
General
ID : 0 (0x0)
Complete name : D:\TEST_2_VIDEO_STREAMS.m2ts
Format : BDAV
Format/Info : Blu-ray Video
File size : 100 MiB
Duration : 31s 957ms
Overall bit rate : 26.2 Mbps
Maximum Overall bit rate : 48.0 Mbps

Video #1
ID : 4113 (0x1011)
Menu ID : 1 (0x1)
Format : AVC
Format/Info : Advanced Video Codec
Format profile : High@L4.1
Format settings, CABAC : Yes
Format settings, ReFrames : 3 frames
Codec ID : 27
Duration : 32s 32ms
Width : 1 920 pixels
Height : 1 080 pixels
Display aspect ratio : 16:9
Frame rate : 23.976 fps
Color space : YUV
Chroma subsampling : 4:2:0
Bit depth : 8 bits
Scan type : Progressive

Video #2
ID : 6912 (0x1B00)
Menu ID : 1 (0x1)
Format : AVC
Format/Info : Advanced Video Codec
Format profile : High@L3.2
Format settings, CABAC : Yes
Format settings, ReFrames : 2 frames
Format settings, GOP : M=3, N=24
Codec ID : 27
Duration : 2s 669ms
Width : 720 pixels
Height : 480 pixels
Display aspect ratio : 16:9
Frame rate : 23.976 fps
Standard : NTSC
Color space : YUV
Chroma subsampling : 4:2:0
Bit depth : 8 bits
Scan type : Progressive

Audio #1
ID : 4352 (0x1100)
Menu ID : 1 (0x1)
Format : DTS
Format/Info : Digital Theater Systems
Format profile : MA / Core
Muxing mode : Stream extension
Codec ID : 134
Duration : 31s 147ms
Bit rate mode : Variable
Bit rate : 1 576 Kbps / 1 510 Kbps
Channel(s) : 6 channels
Channel positions : Front: L C R, Side: L R, LFE
Sampling rate : 48.0 KHz
Bit depth : 24 bits
Compression mode : Lossless / Lossy

Audio #2
ID : 4353 (0x1101)
Menu ID : 1 (0x1)
Format : AC-3
Format/Info : Audio Coding 3
Format profile : Dolby Digital
Mode extension : CM (complete main)
Codec ID : 129
Duration : 31s 392ms
Bit rate mode : Constant
Bit rate : 320 Kbps
Channel(s) : 2 channels
Channel positions : Front: L R
Sampling rate : 48.0 KHz
Bit depth : 16 bits
Compression mode : Lossy
Stream size : 1.20 MiB (1%)

Audio #3
ID : 4354 (0x1102)
Menu ID : 1 (0x1)
Format : AC-3
Format/Info : Audio Coding 3
Mode extension : CM (complete main)
Codec ID : 129
Duration : 31s 168ms
Bit rate mode : Constant
Bit rate : 640 Kbps
Channel(s) : 6 channels
Channel positions : Front: L C R, Side: L R, LFE
Sampling rate : 48.0 KHz
Bit depth : 16 bits
Compression mode : Lossy
Stream size : 2.38 MiB (2%)

Audio #4
ID : 4355 (0x1103)
Menu ID : 1 (0x1)
Format : AC-3
Format/Info : Audio Coding 3
Mode extension : CM (complete main)
Codec ID : 129
Duration : 31s 168ms
Bit rate mode : Constant
Bit rate : 640 Kbps
Channel(s) : 6 channels
Channel positions : Front: L C R, Side: L R, LFE
Sampling rate : 48.0 KHz
Bit depth : 16 bits
Compression mode : Lossy
Stream size : 2.38 MiB (2%)

Audio #5
ID : 4356 (0x1104)
Menu ID : 1 (0x1)
Format : AC-3
Format/Info : Audio Coding 3
Mode extension : CM (complete main)
Codec ID : 129
Duration : 31s 168ms
Bit rate mode : Constant
Bit rate : 640 Kbps
Channel(s) : 6 channels
Channel positions : Front: L C R, Side: L R, LFE
Sampling rate : 48.0 KHz
Bit depth : 16 bits
Compression mode : Lossy
Stream size : 2.38 MiB (2%)

Audio #6
ID : 4357 (0x1105)
Menu ID : 1 (0x1)
Format : AC-3
Format/Info : Audio Coding 3
Mode extension : CM (complete main)
Codec ID : 129
Duration : 31s 168ms
Bit rate mode : Constant
Bit rate : 640 Kbps
Channel(s) : 6 channels
Channel positions : Front: L C R, Side: L R, LFE
Sampling rate : 48.0 KHz
Bit depth : 16 bits
Compression mode : Lossy
Stream size : 2.38 MiB (2%)

Audio #7
ID : 4358 (0x1106)
Menu ID : 1 (0x1)
Format : AC-3
Format/Info : Audio Coding 3
Mode extension : CM (complete main)
Codec ID : 129
Duration : 31s 168ms
Bit rate mode : Constant
Bit rate : 640 Kbps
Channel(s) : 6 channels
Channel positions : Front: L C R, Side: L R, LFE
Sampling rate : 48.0 KHz
Bit depth : 16 bits
Compression mode : Lossy
Stream size : 2.38 MiB (2%)

Audio #8
ID : 6656 (0x1A00)
Menu ID : 1 (0x1)
Format : DTS
Format/Info : Digital Theater Systems
Format profile : Express
Codec ID : 162
Duration : 85ms
Bit rate mode : Constant
Bit rate : 192 Kbps
Channel(s) : 2 channels
Channel positions : Front: L R
Sampling rate : 48.0 KHz
Bit depth : 16 bits
Compression mode : Lossy
Stream size : 1.99 KiB (0%)

[JanWillem32]

DTS Express has DTS HD-like data blocks. Just as with the full data stream with the two regular DTS HD formats, splitting and decoding it is a known issue.
Too bad these formats are so obscure. They are quite common, but there's still no open source decoder for them.

[mr.duck]

Finally, they are done. Getting the bevel to work at 16x16 was quite tricky, but I think I pulled it off.

http://i.imgur.com/ylsfy.png

[mark0077]

tyvm for the icons, they look gorgeous :) im very happy with them.

[Hera]

netbook 1080p MP4 from YouTube,

WMP CPU Usage,
http://i1221.photobucket.com/albums/dd467/QueenHera/WMPKicksAssYo-netbook.png

MPC CPU Usage,
http://i1221.photobucket.com/albums/dd467/QueenHera/WTFMPCYo-netbook.png

Notice that MPC consistently uses more CPU power than WMP.

notebook 720p MP4 from YouTube,

WMP CPU Usage,
http://i1221.photobucket.com/albums/dd467/QueenHera/WMPKicksAssYo-notebook.png

MPC CPU Usage,
http://i1221.photobucket.com/albums/dd467/QueenHera/MPCWutthehell-notebook.png
sry messed up the screenshot a bit...

Notice that MPC consistently uses more CPU power than WMP.

In all trials the video was played for a few minutes before taking the screenshot.

This is MPC + MP4 + EVR vs. WMP + MP4.
See signature for hardware details.
EDIT: Default Codecs, no filters applied on top.

[CruNcher]

@Hera
Default Internal Codecs vs Microsoft DXVA2 Codecs or same standard Vista/7 Microsoft Decoder for both ? and did you also compared it with Aero off ? also don't forget MPC-HC uses a whole internal color and vsync rendering engine you would need to turn that off also and compare :)
You also should take a look @ the newer experimental builds from JanWillem32.

[namaiki]

netbook 1080p MP4 from YouTube

You could try MPC-HC's EVR renderer instead of EVR-Custom Pres if you aren't already.

[Hera]

EDIT5:
Whats up with the experimental builds not having an all in-depth information overlay (Stats)?

@Hera
Default Internal Codecs vs Microsoft DXVA2 Codecs or same standard Vista/7 Microsoft Decoder for both ? and did you also compared it with Aero off ? also don't forget MPC-HC uses a whole internal color and vsync rendering engine you would need to turn that off also and compare :)
You also should take a look @ the newer experimental builds from JanWillem32.

Default MPC vs. Default WMP setup. MPC - EVR.
Obviously DXVA was utilized on both systems.

The comparison is WMP Aero ON vs MPC Aero ON.
Aero stays on as WMP runs just fine with it on.

V-Sync and color settings are NOT configurable using EVR. They are only configurable using EVR:CP. When using EVR:CP, I have V-Sync Disabled and do not have any post-processing enabled. Using EVR:CP, V-Sync is a fully useless CPU hog. Alternatives include using Aero or having "Wait for Flushes" turned on.
EDIT3: Might as well say that EVR:CP uses at least as much CPU as EVR, not less.

I already mentioned that the experimental builds get fully stuck when opening a file when using PS2.0 scaling options.
EDIT: Will try EVR now.
EDIT2: No improvement on notebook with experimental builds.
EDIT3: nope.avi

You could try MPC-HC's EVR renderer instead of EVR-Custom Pres if you aren't already.

The comparison is with EVR and *NOT* with EVR:CP.

EVR was used for fair comparison purposes only.

I use EVR:CP for daily purposes.

[Xizer]

Can someone tell me why MPC-HC's internal EVR filters crash the player when opening MPEG-TS recordings from one HDTV channel but works fine on MPEG-TS recordings from other channels? madVR and Haali work fine on these problem recordings.

Here are sample .ts files for you to try yourself and error information.

This sample .ts from FUSE crashes the player:

http://www.mediafire.com/?fbz51q9e6z90ehb

This sample .ts from TLC works fine:

http://www.mediafire.com/?aas6hgcaqa46q99

Samples are brief (fair use). Here's the error I get when playing the FUSE clip when my renderer is set to EVR Custom:

Problem signature:
Problem Event Name: APPCRASH
Application Name: mpc-hc.exe
Application Version: 1.5.2.3104
Application Timestamp: 4dca8d91
Fault Module Name: StackHash_647f
Fault Module Version: 6.1.7601.17514
Fault Module Timestamp: 4ce7ba58
Exception Code: c0000374
Exception Offset: 000ce653
OS Version: 6.1.7601.2.1.0.256.1
Locale ID: 1033
Additional Information 1: 647f
Additional Information 2: 647f62bebcc30009c0e45c6a9a0316cf
Additional Information 3: 2d6c
Additional Information 4: 2d6cc0a288f04eba030c32886eeacbec

I've noticed when comparing the two files with MediaInfo the Fuse one has these additional lines in its information that the TLC one lacks:

Format settings, GOP : M=3, N=24
Scan type : Interlaced
Scan order : Top Field First

System setup:
Windows 7 x64
MPC-HC x86 1.5.2.3104
AMD Radeon HD 6970

[namaiki]

Using EVR:CP, V-Sync is a fully useless CPU hog. Alternatives include using Aero or having "Wait for Flushes" turned on.

EVR CP's VSync prevents stuttering where the screens refresh rate is a multiple of the video's frame rate, which I experienced even with the alternatives.

That said, stuttering was rarely experienced in the EVR renderers compared with Haali's Renderer.

[Hera]

EDIT: Experimental Build offers much better FFPP performance.

EVR CP's VSync prevents stuttering where the screens refresh rate is a multiple of the video's frame rate, which I experienced even with the alternatives.

That said, stuttering was rarely experienced in the EVR renderers compared with Haali's Renderer.

Never had these issues. I haven't turned on V-Sync for EVR:CP in, like, 3? years. First happened on a C2D2.4Ghz and 8600MGT, stuttering when there was a lot of things going on. I though the computer couldn't handle the quality or something, then I turned off V-Sync. Magically, all the issues went away.

I never ever had any tearing with V-Sync disabled.

V-Sync = Trash that maxes out the CPU sometimes doing nothing productive. *This may be false on XP, but not for me - I had to wait for flushes to fix tearing*

Haali Renderer = The only way to enjoy HD content on a 2004 desktop PC. No V-Sync though as far as I can tell.

[Dstruct]

Version 1.5.2.3145


Sparkle GeForce GTS 450, Windows XP SP3


-> changing "Floating Point Processing" settings in Renderer Settings needs MPC to be restarted


Expected: change "Floating Point Processing" setting on the fly without the need to restart the plaer

[nevcairiel]

What you expect and what really is does not have to be the same. :p
not every setting can be changed during playback

[JanWillem32]

I've broken that function quite a few times during development (although it's not broken in my current developmental alpha build). It's tricky to get a good re-initialization of the allocator-presenter parts during playback when switching these kinds of things.

The Vsync functions are a bit difficult to handle. The easiest way to avoid tearing and stuttering on most video cards is by using the exclusive mode and only enabling the Vsync option in the video card's 3D settings panel. More advanced methods require a constant audio resampler to sync to a specific frame rate. I've been studying methods for those, and the current sync methods in EVR Sync, EVR CP and VMR-9 r.. Windowed mode poses the problem of not knowing when a frame is truly transferred to the backbuffer, as the Windows desktop decides that. There are a few ways of guessing it, but there's no universal Vsync solution for all systems.

[mr.duck]

tyvm for the icons, they look gorgeous :) im very happy with them.

Thx :)


The general consensus seems to be that everyone likes them.

Can my icons be used to updated the MPC-HC '321' logo?

[ceb]

Is it possible to view the thumbnail of .mkv videos? I don't use any other software besides MPC-HC.
Anyone? :thanks:

[JanWillem32]

A basic web search for "thumbnail view mkv" reveals that there are two programs that can function as an explorer plugin to generate thumbnails in folder view. As MPC-HC doesn't install anything but the settings in "HKEY_CURRENT_USER\Software\Gabest\Media Player Classic\" and the basic file associations, other system-wide functions, such as generating thumbnails, can only be handled by system-wide splitters, codecs and other plugins.
To be more specific on this item: one or several .DLL files serving as a plugin for the Windows explorer must be capable of splitting a .MKV file, decode a frame, resize it, and output that picture.

[SamuriHL]

Divx has that ability but it often crashes explorer in the process.

[nevcairiel]

On Windows 7 at least, thats a WMP component doing it, but me personally, i haven't got it to work with MKVs yet..

[clsid]

To get thumbnails:

1) registry tweak:
Windows Registry Editor Version 5.00

[HKEY_CLASSES_ROOT\.flv\ShellEx\{BB2E617C-0920-11D1-9A0B-00C04FC2D6C1}]
@="{c5a40261-cd64-4ccf-84cb-c394da41d590}"


2) Install a splitter and video decoder

3) On Windows 7 you need to force ffdshow as preferred H.264 decoder. Thumbnails won't work with the MS decoder.

Limitation: the first frame of the video is used as thumbnail. This can't be changed as it is simply how the MS shell extension works for file types that it does not support natively.

[ramicio]

It's only going to get proper support if MS adds it. Chance of MS ever adapting to something that's open source, zero.

[nevcairiel]

3) On Windows 7 you need to force ffdshow as preferred H.264 decoder. Thumbnails won't work with the MS decoder.


Ah, maybe thats what i forgot. WMP needs special preferred decoder settings.

[ceb]

I thought it were possible with only MPC-HC, guess I was wrong. :D
Thanks everyone for the replies!

[Hera]

Also the experimental build does get stuck after watching a few files in a row.

[v0lt]

[3144] jonasno
Possible fix to a hang bug and some useful notes to other devs.
:mad:
New bug: No change of the picture while seeking to pause. Few freezes when trying to rewind.

[Aleksoid1978]

:mad:
New bug: No change of the picture while seeking to pause. Few freezes when trying to rewind.

Fix. :)

[v0lt]

Fix. :)
Thank you! :thanks:
MPC-HomeCinema 1.5.2.3151 (x86) (http://www.mediafire.com/?xfgh33iws56bh)
- Fix : rev. 3144 broken seek while on pause;

[Mixer73]

The Vsync functions are a bit difficult to handle. The easiest way to avoid tearing and stuttering on most video cards is by using the exclusive mode and only enabling the Vsync option in the video card's 3D settings panel. More advanced methods require a constant audio resampler to sync to a specific frame rate. I've been studying methods for those, and the current sync methods in EVR Sync, EVR CP and VMR-9 r.. Windowed mode poses the problem of not knowing when a frame is truly transferred to the backbuffer, as the Windows desktop decides that. There are a few ways of guessing it, but there's no universal Vsync solution for all systems.

I think the problem I have now is that people have tried so hard to fix things that builds of MPC-HC over the past year will no longer play the files I would like to watch all day without dropping constant frames.

I play simple MPEG2 50hz files on a 60hz monitor and I have used MPC-HC since build 900 something, although recently I switched to PotPlayer because it plays my files fine, even with the same decoder and renderer selected. I do no need to use exclusive mode with PotP. I guess there's lots of funky stuff that MPC-HC does better with subs and so on but that's not an issue for me.

[jos99]

Divx has that ability but it often crashes explorer in the process.
Shark codec pack does it.

[rsd78]

Hi,

I know this topic has been brought up, but I didn't see any new information recently so I thought I would ask.

Any plans or what is the status of fixing DXVA for Intel based platforms (e.g. Sandy Bridge)?

Thanks

Anyone able to provide any information? Thanks!

[hayan]

broken seek (keep Hold down the "← or →" button) on flv or realmedia

[Skittelor]

little progress
http://img812.imageshack.us/img812/347/fileassociations3.th.png (http://img812.imageshack.us/img812/347/fileassociations3.png)
now controls are not connected with the settings. a lot of work.
Test-drive this build (http://cid-d561cc7d122f3f32.office.live.com/browse.aspx/Shared/MPCHC%5E_SWSCALER%5E_MOD) featuring some new internal software decoder options.
More info in the readme. Also see ticket 719 (http://sourceforge.net/apps/trac/mpc-hc/ticket/719).

http://public.sn2.livefilestore.com/y1p0zCeMT3iqjbz83hAB4ctETpL_fM5vaUJt1gG6NT7lowch9VgTX8vj1X5mNUtmFzzXNADVagu0ayI2Hbag1lvUg/whatsnew.png?psid=1

Are these changes going to make it to the trunk? They're great facelifts. It would be great if they did.

@v0lt and @bobdynlan: It'd be nice if you guys gave us updates once in a while :)

[bigzaj]

I'm having issues with keyboard shortcuts. For example I set pause to Ctrl-P but it does not respond at all. The cursor comes up but it does not pause. It was working fine until a few days ago. My system did not change but it still won't work. Also I have it set to open files in full screen, when in full screen I cannot right click or anything to get a menu, all I can do is exit (which the keyboard shortcut works for)

[namaiki]

I'm having issues with keyboard shortcuts. For example I set pause to Ctrl-P but it does not respond at all. The cursor comes up but it does not pause. It was working fine until a few days ago. My system did not change but it still won't work. Also I have it set to open files in full screen, when in full screen I cannot right click or anything to get a menu, all I can do is exit (which the keyboard shortcut works for)

Try reassign 'Toggle Pixel Shaders' which is also set for Ctrl+P.

Try disable D3D Fullscreen mode if it is enabled. (View-> Options-> Output)

[PlazzTT]

I have a HDTV capture which is 1440x1080. The DAR is 16:9, so the display width and height is 1920x1080.

When I use File->Save Image, is it possible to get the screenshot in 1920x1080 instead of 1440x1080 (without having to resize it in Photoshop later). The Snapshot option in VLC creates a 1920x1080 image.

I suppose it's up for debate which one it should be, just wondering if there's an option in MPC-HC.

[JanWillem32]

@bigzaj: Since r2964 or something all commands in D3D Fullscreen are like if you tap two or three times. It's annoying for me too, but I haven't exactly found the source of that bug yet. Menus in exclusive mode will have to be made by someone familiar with D3D GUI items. It would be great to have some more controls in addition to only the seekbar.

@PlazzTT: The "Save Image" function only saves a partially rendered item from the mixer. To get a full render, right-click in the main window, set "Full Screen", wait for the mouse cursor to disappear and use the "Print Screen" button on the keyboard. That will capture a fully rendered picture. Depending on your rendering settings, it's in a better quality usually, too. To only enable aspect ratio correction, without fully resizing a picture, use "View", "Video Frame", "Normal Size" first.
You will have to paste the resulting item from the clipboard in an image editor. Most editors will feature an automatic clipping function for if there's any black bars.
If you are into color accurate processing of the images: disable the color management in MPC-HC, take the screenshot and assign a proper HDTV, NTSC, PAL/SECAM color profile on the saved image.
Improving the image saving functions is on my development wish list, but it's not priority (see my thread for reference). I did improve the speed a bit with the 10-, 16- and 32-bit color surface modes when saving images. I also have a screenspace image saving function template for during development, but that one won't convert images to 8-bit, and most people don't know what to do with .PFM or .DDS image files.

[jos99]

Finally, they are done. Getting the bevel to work at 16x16 was quite tricky, but I think I pulled it off.

http://i.imgur.com/ylsfy.png
Nice icons:thanks:. Slight problem, I am getting the large icons OK, but most of my files are thumbnailed and for these I still get the old style MPC-HC icon in the bottom right corner, any advice how to fix? Also, the icon for the actual exe/shortcut is still old-style also.

[mr.duck]

Nice icons:thanks:. Slight problem, I am getting the large icons OK, but most of my files are thumbnailed and for these I still get the old style MPC-HC icon in the bottom right corner, any advice how to fix? Also, the icon for the actual exe/shortcut is still old-style also.

I was hoping they looked good enough to be adopted into the project, but so far nothing has happened.

I'm not sure on the procedure. Who gets to decide what icon to use?

[BatKnight]

I was hoping they looked good enough to be adopted into the project, but so far nothing has happened.

I'm not sure on the procedure. Who gets to decide what icon to use?
tetsuo55 is the MPC-HC Project Manager

Bat

[Underground78]

I was hoping they looked good enough to be adopted into the project, but so far nothing has happened.

I'm not sure on the procedure. Who gets to decide what icon to use?

You should probably visit irc://irc.freenode.com/#mpc-hc-dev.

[diizzy]

Hmm...
I'm sitting here with with two computers using Intel HD Graphics (ie built in graphics in first gen Core i3/i5 CPUs) with a rather interesting and somewhat annoying issue. About 95% of the time using EVR CP, EVR Sync or VMR-9 w/ 3D Surfaces MPC-HC tends to lag/stutter heavily when going full screen using a resolution above 1080p like 1920x1200 (this never occurs on 1080p or lower resolution). Sometimes it just works flawless which is very odd and a bit annoying since its not consistent. Resolution of the clip doesn't seem to matter and window mode always works just fine which makes me believe that it isn't a performance issue caused by video card or CPU. Both computers are connected with AC adaptor so they don't run on battery which may cause performance issues/throttling. Anyone else experiencing this? This occurs on two laptops (i3 and i5) running Win 7 SP1 and I've tried several different versions of video drivers without resolving the issue and both are running the newest driver available by Intel. Any suggestions on how to track this is issue down?

[CruNcher]

Performance issue with the realtime upscaling it seems did you tried all of the resizer options already ?

[tetsuo55]

I was hoping they looked good enough to be adopted into the project, but so far nothing has happened.

I'm not sure on the procedure. Who gets to decide what icon to use?I'll send an email to the contributers this weekend. Replacing the program icon is something that has to be done carefully.

[adam777]

Hi all,
I found out what seems to be an MPC-HC bug.
First, I select madVR as the video renderer in the output page.
Next, If I select madVR as a preferred external filter, internal subtitle renderer will not work with it (madVR loads fine).
When not selecting it in the external filters list or selecting it as a blocked (!) filter, it loads and plays subtitles fine.
I'm pretty convinced it's something rather new as I'm quite sure it was selected as a preferred filter for me for quite some time.
Adam.

[JanWillem32]

Loading MadVR and Haali renderers trough the internal renderer allocation unit is preferred. I can imagine that things become more complicated if you try to set these as an external filter. Not all interfaces might work.

Edit, just a development question:
Does anyone still use the "Regular offscreen plain surface" or "2D surfaces" options with VMR-9 r. (the two options are actually the same thing in the current implementation)? These are legacy options from VMR-7 r. to give VMR-9 r. very basic subtitle, OSD and stats screen support, without any of the more advanced rendering options (scalers, shaders, vsync options, et cetera). It's very dated and inefficient by now, as the unused items from the regular rendering path are still completely loaded. There hasn't been any maintenance on this rendering path anymore for years now.

[adam777]

Loading MadVR and Haali renderers trough the internal renderer allocation unit is preferred. I can imagine that things become more complicated if you try to set these as an external filter. Not all interfaces might work.

OK, :thanks:

[diizzy]

@ CruNcher
Doesn't make sense since the same clip runs flawless to what seems randomly, kinda hard to debug. :/

[mr.duck]

I'll send an email to the contributers this weekend. Replacing the program icon is something that has to be done carefully.

Ok, thanks. I'll keep an eye on this thread. Also, I have made generic file type icons in various sizes with the same logo if that's needed/useful.

[oddball]

Hey guys. I have another issue. I like to use auto switcher to match the fps to the refresh rate. However when playing back transport streams that are 29.97fps it initially switches to 60Hz on my HDTV (I have auto switch set to 59Hz NTSC but that's beside the point). But then it kinda detects the telecine after a moment and decimates the stream to 23.976fps. However the HDTV is now running at 60Hz and thus is not synced to the video fps.

Is there any way around this?

[nevcairiel]

No the switch only works on the info from the source filter, not anything the decoder does to the stream

[pawlol83]

Hello,
I have a problem with video rendering mode MadVR. Despite the choice of rendering mode I am still not able to call its instance. This "error" probably occurred on the version of MPC-Home Cinema 3110 (x86). In another player I have no problem running the video rendering mode MadVR. I tried to restore the default settings by deleting the settings file, deleting the entry from the registry, but it did not give satisfactory results. Using the built-in codecs solves the problem. If you use the ffdshow filter (set to download) I can not run rendering mode MadVR. Please refer to the problem.

FFdshow settings attached to this post.

Greetings

[cyberbeing]

There appears to be a bug in the MPC-HC MKV splitter (tested rev 3158) when used with CoreAVC 2.5.5 and VSFilter.dll, which breaks VSFilter's fps counter and causes slowdowns rendering complex subtitles. Enable VSFilter's statistics and you'll see what I mean. The VSFilter's framerate counter goes crazy, fluctuating between 0fps and 10000fps, when it should read something like 23.976fps.

MPC-HC MKV splitter, FFDShow, VSFilter = No bug.
Haali Media Splitter, CoreAVC, VSFilter = No bug.

This is a rather non-issue for me, since I always use Haali, but it should probably be fixed since it very well may effect the internal subtitle renderer as well with that same combo. In the meantime, people should be wary about using the MPC-HC internal MKV splitter with subtitles and CoreAVC.

[oddball]

Playing back the Big Buck Bunny clip 1 '1080p@24.0, 4:2:2 chroma, 70 Mbps CBR (with no zero stuffing) in a 73 Mbps Transport Stream' using either the internal MPEG2 decoder, LAV CUVID decoder or PowerDVD decoder results in garbled output. Using ffdshow as the decoder results in correct playback.

Can someone else test? I have tried different splitters and renderers and it makes no difference. Only ffdshow LIBMPEG2 seems to display it correctly.

EDIT: In the LAV CUVID thread they say it's the 4:2:2 chroma that is the issue here. Also that there should be a check for it since build 3123. Well I am using 3180 Madhi's build and it does not seem to work. I am not talking about DXVA decoding as obviously the hardware does not support it. But it does not decode it in software correctly either. Only ffdshow seems to. Not that I am bothered if MPC-HC stays broken since ffdshow works perfectly fine. Just thought I would report it as a bug ;)

[Underground78]

Just thought I would report it as a bug ;)

Yes, you should probably open a ticket on Trac.

[hayan]

How to disable the time tooltip?

When moving the mouse fast from up to down/down to up over the seek bar in "stopped" status

Time tooltip: I'm alive :p

[Mercury_22]

Can we please add an icon_x64 cause since there already are logo.1_x64 logo.2_x64 logo.3_x64 it will be in the logic of things to have a icon_x64 too! not to mention that (I know I've said this before but I haven't got an (official) answer) it will really, really, help people with both version on their system (e.g. to distinguish between file association (x64 or x86 MPC-HC) before start playing the file)
Also in the same logic used for the logos we can use the same icon + an "x64" or we can use a totally different icon (I prefer the one from BE mod)

What do you say (tetsuo :) ) ?

[Thunderbolt8]

I have two questions:


1. is there still someone working on VSfilter/directvobsub, is there a thread somewhere here?

2. regarding the title, would it be possible to add blu-ray subtitle support (PGS) for external files (.sup) which have the same name as the movie file (for mpc-hc and directvobsub/VSfilter)? (e.g. a mkv remux).
atm, it its necessary that these subs have to be muxed together with video and audio, otherwise they wont show. and it also seems that PGS subs are not supported at all by directvobsub atm?

[nevcairiel]

DirectVobSub supports PGS just fine, you just have to use a non-ancient version.

Also, why not just mux them into the mkv? :d

[SamuriHL]

Also, why not just mux them into the mkv? :d

Do I need to come over there and slap you? :D LOL! That's a HORRIBLE idea!!! :D

[nevcairiel]

Whats wrong with muxing subtitles inside the MKV? If you use one of those automatic tools, thats what they'll do too.

[Thunderbolt8]

because usually I prefer OCRed subs, the pgs ones would only be a measure of last resort when running into non OCR-friendly subs. but I guess I could do that, as long as they are not automatically enabled by default.

DirectVobSub supports PGS just fine, you just have to use a non-ancient version.thats strange. in which situation, only when they are muxed inside the .mkv or also outside? Im using DirectVobSub_2.40.3164_x86 and they dont work for me.

[SamuriHL]

Whats wrong with muxing subtitles inside the MKV? If you use one of those automatic tools, thats what they'll do too.

MUXING, ok, sorry, misread what you said. Man this cold needs to go away so I can stop taking cold medicine! :) Yes, muxing's good. Burning them into the video is not! :p

[Thunderbolt8]

ok, tried a bit more and indeed, both directvobsub and the mpc-hc filter play PGS subs outside of .m2ts files, but only if they are muxed into the .mkv - not as external files, whereas both play other externals subs like .srt and .ass fine.

so it would be nice if external PGS support could be added as well for both.

[Underground78]

How to disable the time tooltip?

When moving the mouse fast from up to down/down to up over the seek bar in "stopped" status

Time tooltip: I'm alive :p

You can still seek when playback is stopped and as the seekbar is enable, the tooltip shows up. I am not quite sure to understand why we can still seek when the playback is stopped ... What do you think ? I may add a special behavior not to show the tooltip in this case but I would rather follow the behavior of the seekbar. In personally not against changing the behavior of the toolbar if people think it would be better.

[Superb]

I hope to post a Hebrew translation for MPC-HC tomorrow. I'm almost done.

Something I've noticed: the whole "translate the rc file" way is ugly, hard, and annoying. Many (many!) strings are repeating themselves. Not all strings are in string tables (which are indeed easier to translate). There must be a better way.
Also, I wanted to ask: would it be possible to set the RTL window flag when using the Hebrew translation (and other future RTL languages)?

[mindbomb]

does anyone know if using full or half floating point processing effects chroma upsampling on nvidia hardware?

it seems to go hand in hand the use of it and chroma upsampling shaders, idk if that is a coincedance or if it's necessary.

[Underground78]

Something I've noticed: the whole "translate the rc file" way is ugly, hard, and annoying.

You are not supposed to translate the RC file but only the text file which contains only the strings (in the text folder, you can find mplayerc.rc.txt which is the base you should have used). It is not a big deal but for next time use the text file which will correspond to your translation).

[v0lt]

These strings are never used and duplicated. I can delete them all or half only.

IDS_AG_NAVIGATE "Navigate"
IDS_AG_OPEN_DISC "Open Disc"
IDS_AG_FILTERS "Filters"
IDS_AG_SUBTITLES "Subtitles"
IDS_MAINFRM_28 "Audio Language"
IDS_MAINFRM_29 "Subtitle Language"
IDS_AG_VIDEO_ANGLE "Video Angle"
IDS_AG_JUMP_TO "Jump To..."
IDS_AG_FAVORITES "Favorites"
IDS_AG_SHADERS "Shaders"
IDS_AG_VIDEO_FRAME "Video Frame"
IDS_AG_PANSCAN "PanScan"
IDS_AG_ASPECT_RATIO "Aspect Ratio"

IDS_SHADER_POPUP "Shaders"
IDS_FAVORITES_POPUP "F&avorites"
IDS_JUMPTO_POPUP "Jump To..."
IDS_VIDEOANGLE_POPUP "Video Angle"
IDS_SUBTITLELANGUAGE_POPUP "Subtitle Language"
IDS_AUDIOLANGUAGE_POPUP "Audio Language"
IDS_SUBTITLES_POPUP "Subtitles"
IDS_AUDIO_POPUP "Audio"
IDS_FILTERS_POPUP "Filters"
IDS_OPENCDROM_POPUP "O&pen Disc"
IDS_NAVIGATE_POPUP "Navigate"
IDS_VIDEOFRAME_POPUP "Video Frame"
IDS_PANSCAN_POPUP "Pan&&Scan"
IDS_ASPECTRATIO_POPUP "Aspect Ratio"
IDS_ZOOM_POPUP "Zoom"
IDS_RECENT_FILES_POPUP "Recent Files"

[JanWillem32]

@mindbomb: For nVidia hardware a bilinear filter is forced. For 8-bit working surfaces on ATi hardware a bilinear filter is used by choice (hardcoded in the rendering engine), 10- 16- and 32-bit working surfaces are unfiltered (nearest neighbor). For Intel hardware, I don't know exactly what the last generation and the one before that do. Intel IGPs became DirectX 10+ only recently, that's one of the reasons there are differences in the two generations.
To integrate a custom mixer so that the rendering engine can work with raw data from a DirectShow video decoder is no small effort. I don't have much experience with actual DirectShow stuff, and we could really use some extra developers to help with those parts. Unless a mixer is added, there's unfortunately no way to access the raw Y'CbCr structures directly from the decoder in the rendering engine.
Also note that the chroma up-sampling shaders won't work on interlaced content. (A modified version for the vertical 4:2:0 part could actually work by using the forced wave deinterlacer. That deinterlacer doesn't do more than writing the top and bottom field to one progressive frame, so it halves the field rate. It's a popular, but very bitrate inefficient way to transport near-progressive 1920×1080 frames inside interlaced 1920×540 fields.)

[Skittelor]

EDIT: In the LAV CUVID thread they say it's the 4:2:2 chroma that is the issue here. Also that there should be a check for it since build 3123. Well I am using 3180 Madhi's build...

Where do I get builds by madshi?

[v0lt]

MPC-HomeCinema 1.5.2.3193 (x86) (http://www.mediafire.com/?xfgh33iws56bh)
fixed bitrate calculation for DTS and DTS Hi-Res in DTSAC3Source
fixed bitrate calculation for DTS-core in BaseSplitter
Thanks to madshi.

[Superb]

I hope to post a Hebrew translation for MPC-HC tomorrow. I'm almost done.

Something I've noticed: the whole "translate the rc file" way is ugly, hard, and annoying. Many (many!) strings are repeating themselves. Not all strings are in string tables (which are indeed easier to translate). There must be a better way.
Also, I wanted to ask: would it be possible to set the RTL window flag when using the Hebrew translation (and other future RTL languages)?

Hebrew translation: http://www.sendspace.com/file/togbmg or http://www.mediafire.com/?5r6cfycfnol26d0

[Underground78]

Hebrew translation: http://www.sendspace.com/file/togbmg or http://www.mediafire.com/?5r6cfycfnol26d0

Thanks, I will have a look tonight.

[CruNcher]

Feature Request: http://forum.doom9.org/showpost.php?p=1506227&postcount=816

[ch1n4]

MPC-HomeCinema 1.5.2.3193 (x86) (http://www.mediafire.com/?xfgh33iws56bh)
fixed bitrate calculation for DTS and DTS Hi-Res in DTSAC3Source
fixed bitrate calculation for DTS-core in BaseSplitter
Thanks to madshi.

Is madshi releasing builds regularly or are these "special builds"?

[madshi]

I don't release MPC-HC builds at all (except for very special situations). 3193 was made by v0lt, not me. I just assisted a bit via PM.

[v0lt]

MPC-HomeCinema 1.5.2.3199 (x86) (http://www.mediafire.com/?xfgh33iws56bh)
fixed bitrate calculation for DTS in MpaDecFilter
fixed bitstream output for DTS with a bitrate less then 1536 kbit

@ch1n4
This is usual builds. Madshi helped. Many thanks to him.

[v0lt]

@Underground78
In r3177 there are tooltips on the buttons, in r3180 no longer exists.
IMHO, the problem in r3178.

[Underground78]

@Underground78
In r3177 there are tooltips on the buttons, in r3180 no longer exists.
IMHO, the problem in r3178.

I am currently improving the tooltip but I don't understand what you are talking about.

I have already fixed locally some cases where the tooltip was not hiding as it should. Also I have added back the auto-hide when the cursor stays still for a while.

[v0lt]

r3177: http://i.imgur.com/aIHJQ.png

[Underground78]

r3177: http://i.imgur.com/aIHJQ.png

I don't understand how I could have broke that ... I will have a look anyway. Edit : Problem found and fixed in r3201.

Also as nobody seems to agree on the tooltip position, I will add as soon as I can an option to disable it, enable it as a "standard" tooltip or enable it as it is now above the seekbar. :devil:

[Underground78]

Hebrew translation: http://www.sendspace.com/file/togbmg or http://www.mediafire.com/?5r6cfycfnol26d0

I am still working on it, will be done tomorrow. I have no idea about the RTL flag so I will need some help for that.

Edit : In fact here a first test build MPC-HC_Hebrew.7z (http://www.mediafire.com/?p7992oyb26x1po0), can you tell me if everything seems ok ?

[xiulet]

How to disable the time tooltip?

When moving the mouse fast from up to down/down to up over the seek bar in "stopped" status

Time tooltip: I'm alive :p

on full screen , when moving mouse on seekbar, the time tooltip is freezing on screen very often.

adéu.

[Superb]

I am still working on it, will be done tomorrow. I have no idea about the RTL flag so I will need some help for that.

Edit : In fact here a first test build MPC-HC_Hebrew.7z (http://www.mediafire.com/?p7992oyb26x1po0), can you tell me if everything seems ok ?Looking great. Thank you.
I've already noticed some small translation mistakes (hard to see when working only w/ the text file in front of you) and will provide a patch to fix them (hopefully this weekend) once the translation will get to the SVN repository.

I think RTL is a must here. Some of the strings in Hebrew are shown visually incorrectly when viewed in LTR.
I would check eMule's source code to see how they do it. They also use MFC and have some pretty generic function that turns the entire GUI to RTL. (specifically, one should look how the "RTLWindowsLayout=1" ini option is parsed there)

Thanks again. I really appreciate you taking the time to look this over.

[Underground78]

on full screen , when moving mouse on seekbar, the time tooltip is freezing on screen very often.

adéu.

Please test the build (http://www.mediafire.com/?p7992oyb26x1po0) I posted for Superb, it contains yet uncommitted changes for the tooltip that may fix your issue.

Looking great. Thank you.
I've already noticed some small translation mistakes (hard to see when working only w/ the text file in front of you) and will provide a patch to fix them (hopefully this weekend) once the translation will get to the SVN repository.

I think RTL is a must here. Some of the strings in Hebrew are shown visually incorrectly when viewed in LTR.
I would check eMule's source code to see how they do it. They also use MFC and have some pretty generic function that turns the entire GUI to RTL. (specifically, one should look how the "RTLWindowsLayout=1" ini option is parsed there)

Thanks again. I really appreciate you taking the time to look this over.

I will have a look tonight, thanks for the advice.

[v0lt]

Please test the build (http://www.mediafire.com/?p7992oyb26x1po0) I posted for Superb, it contains yet uncommitted changes for the tooltip that may fix your issue.
This build is better than 3201. Tooltips are very importunate in 3201.

[InTheLight]

I'm having some trouble with video_ts.ifo files in the playlist.

Right now I can't get MPC to advance to the next file in the playlist after a video_ts.ifo file has finished. Playback stops, but that's it. The /close command also isn't working.

Anybody know if this should be working or how I can make it work?

Thanks for any help.

Edit: Actually it doesn't look like this will be possible, and it makes sense, due to the nature of the *.ifo file(s). Oh well.

[Underground78]

This build is better than 3201. Tooltips are very importunate in 3201.

I think I have fixed all the issues: the tooltip now always hides when it should, it auto-hides after a while if you don't move the cursor and it does not show up when the cursor just quickly hovers the seekbar. I will try to commit those fixes tonight, next step will be to add an option to disable the tooltip completely and to choose if it must appears above or besides the seekbar.

Edit : committed as r3203.

[v0lt]

Thank you. Much better now.

[Underground78]

Looking great. Thank you.
I've already noticed some small translation mistakes (hard to see when working only w/ the text file in front of you) and will provide a patch to fix them (hopefully this weekend) once the translation will get to the SVN repository.

I think RTL is a must here. Some of the strings in Hebrew are shown visually incorrectly when viewed in LTR.
I would check eMule's source code to see how they do it. They also use MFC and have some pretty generic function that turns the entire GUI to RTL. (specifically, one should look how the "RTLWindowsLayout=1" ini option is parsed there)

Thanks again. I really appreciate you taking the time to look this over.

This new test build (http://www.mediafire.com/?x2r4fcpdbs31sab) should have the RTL flag but I cannot see any difference so I am not sure it works as it should ...

[Superb]

This new test build (http://www.mediafire.com/?x2r4fcpdbs31sab) should have the RTL flag but I cannot see any difference so I am not sure it works as it should ...Hmm, the GUI doesn't seem to be RTLed... So I guess the flag isn't applied correctly.

EDIT: after more testing, it seems the flag is applied to the right click menu! That's something. The rest of the UI is still LTR.

EDIT 2: Also, it is only applied right after changing to Hebrew. After restarting the player (where it is set to Hebrew), the right click menu becomes LTR again.

[Underground78]

Hmm, the GUI doesn't seem to be RTLed... So I guess the flag isn't applied correctly.

EDIT: after more testing, it seems the flag is applied to the right click menu! That's something. The rest of the UI is still LTR.

EDIT 2: Also, it is only applied right after changing to Hebrew. After restarting the player (where it is set to Hebrew), the right click menu becomes LTR again.

Do the dialogs (like the options for example) are RTLed ? Will try to do another test build tonight.

[Superb]

Do the dialogs (like the options for example) are RTLed ? Will try to do another test build tonight.Nope. They are not. Only the right click menu. :/

[Underground78]

Nope. They are not. Only the right click menu. :/

In this build (http://www.mediafire.com/?qb2p7fa6o906bry), I think everything is RTLed but the change is only totally applied after restarting the application.

I don't really know if I can do better ... :/

Edit : You can have a look to the installer (http://www.mediafire.com/?b1olm4tt9545db1) also.

Edit 2 : Do you have a Sourceforge account ? We usually put the Sourceforge email address (*@users.sourceforge.net) of translators in Authors.txt.

[Superb]

In this build (http://www.mediafire.com/?qb2p7fa6o906bry), I think everything is RTLed but the change is only totally applied after restarting the application.

I don't really know if I can do better ... :/

Edit : You can have a look to the installer (http://www.mediafire.com/?b1olm4tt9545db1) also.

Edit 2 : Do you have a Sourceforge account ? We usually put the Sourceforge email address (*@users.sourceforge.net) of translators in Authors.txt.Hmm, I think there is too much RTL in some wrong places. Let me explain:
1. The main window looks like this when RTLed...
http://k.min.us/iOLXK.png
The green circles = good RTL. Should be RTL there.
Red circles = bad RTL. images are flipped. Positions wrong.

2. Settings window should be RTLed:
http://i.min.us/idgaHS.png

Oh, and I don't have a SF.net account and rather not have my e-mail posted in the AUTHORS file. Thanks.

EDIT: just tested the installer. It's great!

[Mercury_22]

OSD it's totally screwed up in the last version(s) when using EVR ! :
Entering or exiting full screen it triggers the last osd message (time or volume or...) and it won't go away unless some action involving osd it's done

[Underground78]

That's what I thought, too much RTL at same places and no RTL for the settings where it is needed ... Will try again ...

@Mercury_22 : I guess I know why ! Give me two minutes to commit something ...

[Mercury_22]

That's what I thought, too much RTL at same places and no RTL for the settings where it is needed ... Will try again ...

@Mercury_22 : I guess I know why ! Give me two minutes to commit something ...

It's working correctly now (Revision 3209):thanks:

[Underground78]

OSD it's totally screwed up in the last version(s) when using EVR ! :
Entering or exiting full screen it triggers the last osd message (time or volume or...) and it won't go away unless some action involving osd it's done

Can you try this build (http://www.mediafire.com/?d6f5c7m3djfvogx) please ?

Edit: Damn you have been too fast for me ! Great if it's fixed ! :)

[Mercury_22]

Can you try this build (http://www.mediafire.com/?d6f5c7m3djfvogx) please ?

Edit: Damn you have been too fast for me ! Great if it's fixed ! :)

You've said "Give me two minutes to commit something ..." so... :):p
Also there is a message missing : The [DXVA] in the status bar also when using EVR !
But I don't know if it was there before cause I usually use EVR-CP and in EVR-CP it's still there so...

[Underground78]

Also there is a message missing : The [DXVA] in the status bar also when using EVR !
But I don't know if it was there before cause I usually use EVR-CP and in EVR-CP it's still there so...

I think it has always been like this but I don't know why.

[Underground78]

Hmm, I think there is too much RTL in some wrong places. Let me explain:
1. The main window looks like this when RTLed...
http://k.min.us/iOLXK.png
The green circles = good RTL. Should be RTL there.
Red circles = bad RTL. images are flipped. Positions wrong.

I have fixed the problem with the toolbar, the seekbar and the video frame. Unfortunately, fixing the video frame has broken the OSD and I don't think I can have both correct at the same time ...

2. Settings window should be RTLed

I have found a way to have it RTLed but it breaks completely the layout ... I don't understand since the same way is used in eMule and it seems to work. :(

Edit: I will need to move on and stop working on getting the RTL layout to work for a while. If you think the main window is usable with this build (http://www.mediafire.com/?tndsrtprgjva2ej), I would like to commit the translation as a beta version. It will be easier for you to fix/update the translation and maybe someone can have an idea to fix the RTL layout in the options dialog. Of course, I will try again when I have time.

Tell me if you are ok with that ! Also I would like to be sure I haven't broken anything in the non-RTL translations so please report back if you think something is wrong with the English translation for example.

[kostik]

How can I load subtitles using MPC Decoder in fullscreen 3D?

[JanWillem32]

Switching subtitles is mapped to Y (next), U (previous) and W (on/off) keys by default. You can view and edit the key mapping in the View->Options->Keys tab. There are several other command options for subtitles there, too.

[Superb]

Edit: I will need to move on and stop working on getting the RTL layout to work for a while. If you think the main window is usable with this build (http://www.mediafire.com/?tndsrtprgjva2ej), I would like to commit the translation as a beta version. It will be easier for you to fix/update the translation and maybe someone can have an idea to fix the RTL layout in the options dialog. Of course, I will try again when I have time.

Tell me if you are ok with that ! Also I would like to be sure I haven't broken anything in the non-RTL translations so please report back if you think something is wrong with the English translation for example.Committing the current changes would be great. The build you posted is a wonderful starting point. We can improve a lot from there on.
Thanks again for the time and work!

On another note, once the translation will be committed... can someone explain to me how to update it via a text file (as opposed to the rc file)? I saw a couple of scripts and batch files. Dunno which is which. :)

[Underground78]

Committing the current changes would be great. The build you posted is a wonderful starting point. We can improve a lot from there on.
Thanks again for the time and work!

Committed as r3211. Thanks for your work ! :)

On another note, once the translation will be committed... can someone explain to me how to update it via a text file (as opposed to the rc file)? I saw a couple of scripts and batch files. Dunno which is which. :)

You should modify the text file and then use patch.bat. (You will need Perl.)

Also you should probably open a ticket on Trac so that we can keep track easily of your updates (see for example the ticket for Ukrainian translation (https://sourceforge.net/apps/trac/mpc-hc/ticket/1069)). When you want us to update the translation, just attach the text file to your ticket and we will commit the changes.

[bigzaj]

I'm having an issue on ONE computer with MPC-HC, when I try to fast forward or skip forward on AVI files it throws the audio out of sync by 2-3 seconds. Any thoughts on what could be causing this? Same files on different computer with MPC-HC configured the same dont have this issue.

[betaking]

I'm having an issue on ONE computer with MPC-HC, when I try to fast forward or skip forward on AVI files it throws the audio out of sync by 2-3 seconds. Any thoughts on what could be causing this? Same files on different computer with MPC-HC configured the same dont have this issue.

Known issues & bugs:

* Internal avisplitter crashes randomly for me.

[v0lt]

@devs
Does anyone know why it is difficult to make the function CDTSAC3Stream::GetMediaType? Why do a lot of types?

[Underground78]

I'm having an issue on ONE computer with MPC-HC, when I try to fast forward or skip forward on AVI files it throws the audio out of sync by 2-3 seconds. Any thoughts on what could be causing this? Same files on different computer with MPC-HC configured the same dont have this issue.

Have you checked the filters used to be sure they are exactly the same ?

[mikeggsm]

I think MPC's EVR Custom Presenter has a screenshot bug: the screenshots (File -> save image) often have the wrong aspect ratio depending on the video. It has been that way for many versions, still is as of 1.5.1.2903.

That's why I use EVR which always has the correct AR, so DXVA is unavailable as soon as there are subtitles.

[Underground78]

I think MPC's EVR Custom Presenter has a screenshot bug: the screenshots (File -> save image) often have the wrong aspect ratio depending on the video. It has been that way for many versions, still is as of 1.5.1.2903.

I think it is always like that with all supported renderers, the screenshot feature does not use the resized frame but the original one.

[mikeggsm]

I think it is always like that with all supported renderers, the screenshot feature does not use the resized frame but the original one.

Well no, since it works with plain EVR...

[nevcairiel]

Press Print-Screen, problem solved.

[mikeggsm]

Press Print-Screen, problem solved.

Printscreen cannot save the video at its original resolution, it saves the black bars added in fullscreen (so the screenshot is not adapted to viewing on a monitor of different AR, and I have 1980x1200 and 1980x1080).

I'll stick with EVR, I am having really inconsistent and frustrating results with EVR CP: right now if I download this video http://www.youtube.com/watch?v=y5vWhdq2LDI (Valkyrie Profile opening on PSX), in normal fullscreen viewing the whole video is zoomed a little so it is cropped on the borders compared to youtube's fullscreen, part of the image is just missing (eg at 1:44 the clouds above are cropped). EVR doesn't have that problem and never gives any problems on any version.

[namaiki]

Printscreen cannot save the video at its original resolution, it saves the black bars added in fullscreen (so the screenshot is not adapted to viewing on a monitor of different AR, and I have 1980x1200 and 1980x1080).


Plain EVR saves a resized image. It also saves black bars added in full screen - whatever is visible in the video area. (it is automated at least)
I suppose you would not be affected by that issue if all your videos are 1920x1080/1920x1200 and always in fullscreen.

Anyway, I cannot seem to reproduce the 'video is zoomed' issue: http://img189.imageshack.us/img189/797/sdnkjsdh.png (single monitor only, though)

Could you please post up a screenshot demonstrating how much zoom there is?

[mikeggsm]

Anyway, I cannot seem to reproduce the 'video is zoomed' issue: http://img189.imageshack.us/img189/797/sdnkjsdh.png (single monitor only, though)

Could you please post up a screenshot demonstrating how much zoom there is?

Thanks for trying, interesting that you don't have that problem, for me on a 1920x1080 monitor:

http://imageshack.us/photo/my-images/838/evrvsevrcp.jpg/

The down image is the correct one (with EVR), it's what I see when MPC is windowed and it's what youtube shows. But if I put EVR-CP in fullscreen I get the up image: it is cropped vertically (clouds/flowers are missing) but horizonally the same content is stretched a bit. I think the aspect ratio hasn't changed, EVR-CP has zoomed the image a bit, which then takes more space than the monitor can display so it ends up cropped.

I wish I could use EVR-CP...

[namaiki]

Could you check:
View-> Video Frame-> Touch window from inside?

[mikeggsm]

Wow, that was it, thanks a lot! It was on "Zoom 1".

Actually I had just found a way to solve the problem: delete HKEY_CURRENT_USER\Software\Gabest. But I had saved it as a .reg file, and every time I remerged it, the problem reappeared: I thought it was a conflict with settings of old builds, but no... (Underground78: yes, now I know why)

I used to have an aspect ratio problem with screenshots taken with EVR CP long ago (I posted about it here), but that has been solved apparently.

EDIT: no it has not been solved, argh! EVR-CP had solved the gray pillarbox problems I had with EVR, but this screenshot problem is really annoying and this time deleting the registry key doesn't solve it. Apparently it happens because EVR-CP saves the source pixels without taking into account the "display aspect ratio" embedded in the file. EVR doesn't have that problem since it doesn't save the source pixels, it saves whatever is on screen at that moment.

[Underground78]

Well the "Zoom 1" setting is saved in the registry, if you delete the key MPC-HC will use the default so no "Zoom 1".

[hajj_3]

isn't it about time we freezed the features and had a new stable build released, its been around 5 months. The official MPC-HC website still needs updating with the changelog, it still talks about vista, nothing about win7.

[betaking]

last mpc-hc svn 3238 internal video decoder can not play this video but 3237 play is ok!
General
ID : 0 (0x0)
Complete name : 02.MTS
Format : BDAV
Format/Info : Blu-ray Video
File size : 12.1 MiB
Duration : 11s 881ms
Overall bit rate : 8 515 Kbps
Maximum Overall bit rate : 18.0 Mbps

Video
ID : 4113 (0x1011)
Menu ID : 1 (0x1)
Format : AVC
Format/Info : Advanced Video Codec
Format profile : High@L4.0
Format settings, CABAC : Yes
Format settings, ReFrames : 2 frames
Format settings, GOP : M=1, N=26
Codec ID : 27
Duration : 11s 880ms
Bit rate mode : Variable
Bit rate : 7 911 Kbps
Maximum bit rate : 16.0 Mbps
Width : 1 440 pixels
Height : 1 080 pixels
Display aspect ratio : 16:9
Frame rate : 25.000 fps
Color space : YUV
Chroma subsampling : 4:2:0
Bit depth : 8 bits
Scan type : Interlaced
Scan order : Top Field First
Bits/(Pixel*Frame) : 0.203
Stream size : 11.2 MiB (93%)

Audio
ID : 4352 (0x1100)
Menu ID : 1 (0x1)
Format : AC-3
Format/Info : Audio Coding 3
Mode extension : CM (complete main)
Codec ID : 129
Duration : 11s 968ms
Bit rate mode : Constant
Bit rate : 256 Kbps
Channel(s) : 2 channels
Channel positions : Front: L R
Sampling rate : 48.0 KHz
Bit depth : 16 bits
Compression mode : Lossy
Delay relative to video : -80ms
Stream size : 374 KiB (3%)

Text
ID : 4608 (0x1200)
Menu ID : 1 (0x1)
Format : PGS
Codec ID : 144
Duration : 11s 375ms
Delay relative to video : -80ms

[Aleksoid1978]

last mpc-hc svn 3238 internal video decoder can not play this video but 3237 play is ok!
General
ID : 0 (0x0)
Complete name : 02.MTS
Format : BDAV
Format/Info : Blu-ray Video
File size : 12.1 MiB
Duration : 11s 881ms
Overall bit rate : 8 515 Kbps
Maximum Overall bit rate : 18.0 Mbps

Video
ID : 4113 (0x1011)
Menu ID : 1 (0x1)
Format : AVC
Format/Info : Advanced Video Codec
Format profile : High@L4.0
Format settings, CABAC : Yes
Format settings, ReFrames : 2 frames
Format settings, GOP : M=1, N=26
Codec ID : 27
Duration : 11s 880ms
Bit rate mode : Variable
Bit rate : 7 911 Kbps
Maximum bit rate : 16.0 Mbps
Width : 1 440 pixels
Height : 1 080 pixels
Display aspect ratio : 16:9
Frame rate : 25.000 fps
Color space : YUV
Chroma subsampling : 4:2:0
Bit depth : 8 bits
Scan type : Interlaced
Scan order : Top Field First
Bits/(Pixel*Frame) : 0.203
Stream size : 11.2 MiB (93%)

Audio
ID : 4352 (0x1100)
Menu ID : 1 (0x1)
Format : AC-3
Format/Info : Audio Coding 3
Mode extension : CM (complete main)
Codec ID : 129
Duration : 11s 968ms
Bit rate mode : Constant
Bit rate : 256 Kbps
Channel(s) : 2 channels
Channel positions : Front: L R
Sampling rate : 48.0 KHz
Bit depth : 16 bits
Compression mode : Lossy
Delay relative to video : -80ms
Stream size : 374 KiB (3%)

Text
ID : 4608 (0x1200)
Menu ID : 1 (0x1)
Format : PGS
Codec ID : 144
Duration : 11s 375ms
Delay relative to video : -80ms

Upload a short sample. Use TsMuxer to cut

[Hera]

isn't it about time we freezed the features and had a new stable build released, its been around 5 months. The official MPC-HC website still needs updating with the changelog, it still talks about vista, nothing about win7.

Is D3D keyboard issues fixed yet?

[Gleb Egorych]

Question about VSfilter as standalone filter. I use it along with Zoom Player and there is one interesting thing: if I load ZP from shortcut and start movie with subtitles then green arrow appears in system tray. If I start ZP from command line then the arrow does not appear. Is it indended behaviour? One more thing: after closing ZP this arrow doesn't disappear, need to move cursor over it to hide.

[kshycho]

How can I disable capture in Menu > Settings.
From the version 1.4 has been added to the menu by capture every program each time you check the TV card available. MPC longer starts.
In version 1.3.1249 does not have this problem.

[ney2x]

Question about VSfilter as standalone filter. I use it along with Zoom Player and there is one interesting thing: if I load ZP from shortcut and start movie with subtitles then green arrow appears in system tray. If I start ZP from command line then the arrow does not appear. Is it indended behaviour? One more thing: after closing ZP this arrow doesn't disappear, need to move cursor over it to hide.

If you got the VSFilter.dll from here (http://xhmikosr.1f0.de/index.php?folder=bXBjLWhjL21wYy1oY19hcHBzL3ZzZmlsdGVy) or here (http://www.xvidvideo.ru/) , then you will encounter problems like broken "vertical padding" and memory leak. But, if you get it here (http://www.mediafire.com/?wbp1fwdh2pigpr4) or here (http://www.cccp-project.net/beta/) then you are ok.

Please be aware that CCCP Project's revisions of DirectVobSub is fine (not broken and no memory leak and does not stay in tray icon upon closing any player).

[fastplayer]

Seeking leads to artifacts starting with rev3239. rev3237 is fine, so I assume it must be the changes in rev3238.
It happens regardless of container (AVI, MKV, MOV, MP4), renderer (madVR, EVR-CP), and splitter (internal, Haali).

[Gleb Egorych]

If you got the VSFilter.dll from here (http://xhmikosr.1f0.de/index.php?folder=bXBjLWhjL21wYy1oY19hcHBzL3ZzZmlsdGVy) or here (http://www.xvidvideo.ru/) , then you will encounter problems like broken "vertical padding" and memory leak. But, if you get it here (http://www.mediafire.com/?wbp1fwdh2pigpr4) or here (http://www.cccp-project.net/beta/) then you are ok.

Please be aware that CCCP Project's revisions of DirectVobSub is fine (not broken and no memory leak and does not stay in tray icon upon closing any player).
I do use CCCP Project's revision 2.40.2.3230 and I have the tray icon bug described above.

[JEEB]

If you got the VSFilter.dll from here (http://xhmikosr.1f0.de/index.php?folder=bXBjLWhjL21wYy1oY19hcHBzL3ZzZmlsdGVy) or here (http://www.xvidvideo.ru/) , then you will encounter problems like broken "vertical padding" and memory leak. But, if you get it here (http://www.mediafire.com/?wbp1fwdh2pigpr4) or here (http://www.cccp-project.net/beta/) then you are ok.

Please be aware that CCCP Project's revisions of DirectVobSub is fine (not broken and no memory leak and does not stay in tray icon upon closing any player).
Whether or not the tray icon stays depends on if the "using" application properly releases the filters or not. There might be extra bugs in a given filter that might leave the icon around, but it mostly is the fault of the using application (good example: CCCP's insurgent troubleshooting application's current version [is being rewritten by Movax the Blamed]).

As for the vertical padding thing: clicky (http://sourceforge.net/apps/trac/mpc-hc/ticket/119)

I have a separate branch for VSFilter in my directshow component building system from which you can get the current patches for reverting those commits for building a standalone VSFiler (clicky (https://github.com/jeeb/mpc-hc/commits/vsfilter) -- the top two commits).

[ney2x]

Whether or not the tray icon stays depends on if the "using" application properly releases the filters or not. There might be extra bugs in a given filter that might leave the icon around, but it mostly is the fault of the using application (good example: CCCP's insurgent troubleshooting application's current version [is being rewritten by Movax the Blamed]).

As for the vertical padding thing: clicky (http://sourceforge.net/apps/trac/mpc-hc/ticket/119)

I have a separate branch for VSFilter in my directshow component building system from which you can get the current patches for reverting those commits for building a standalone VSFiler (clicky (https://github.com/jeeb/mpc-hc/commits/vsfilter) -- the top two commits).

You might help clsid regarding VSFilter http://forum.doom9.org/showthread.php?p=1508278#post1508278

[Gleb Egorych]

Should VSfilter's tray icon appear always or only in some cases?

[Px]

Should VSfilter's tray icon appear always or only in some cases?
Depends on VSfilter settings, by default VSfilter loads only if corresponding subtitles for video is found, but user may turn always loading VSfilter

[Gleb Egorych]

Depends on VSfilter settings, by default VSfilter loads only if corresponding subtitles for video is found, but user may turn always loading VSfilter
I mean VSfilter is loaded but tray icon doesn't appear. I'm asking whether tray icon should appear every time VSfilter is loaded.

[fastplayer]

I mean VSfilter is loaded but tray icon doesn't appear.
FYI, that bug's been in VSFilter for years. Move the mouse over it and it disappears.

[fastplayer]

Seeking leads to artifacts starting with rev3239. rev3237 is fine, so I assume it must be the changes in rev3238.
It happens regardless of container (AVI, MKV, MOV, MP4), renderer (madVR, EVR-CP), and splitter (internal, Haali).
Fixed in rev3246 (http://mpc-hc.svn.sourceforge.net/viewvc/mpc-hc?revision=3246&view=revision).
Thanks, Aleksoid1978! :)

[v0lt]

@Underground78
Changeset 3205
Changed the cursor for the seek bar: the hand cursor will be used only when hovering the position marker.
Some users do not like it. They want back the hand cursor.

IMHO, this is a bad idea, when a different cursor for the seek bar and position marker.

[fastplayer]

^I agree. There should be only one cursor style for the entire seek bar.

[Qaq]

or an option in settings.

[Underground78]

@Underground78

Some users do not like it.

IMHO, this is a bad idea, when a different cursor for the seek bar and position marker.

^I agree. There should be only one cursor style for the entire seek bar.

Well I have changed it because some users (including Tetsuo55 and XhmikosR) asked me to do so because they thought they were no reason to use the hand cursor other than when you can drag something ...

You should probably discuss all that with them, I won't revert the change for now.

[v0lt]

All right. I will write my opinion here.

The Hand cursor (http://i.msdn.microsoft.com/dynimg/IC210310.png) indicates the place which can be clicked. Wherever you can click its use is not necessary. But in this case, it works very well.

I would put a usual cursor (http://i.msdn.microsoft.com/dynimg/IC210309.png) on the buttons. So it is clear that they can push.

Hand cursor is also not needed at video frame (it confuses users only).

[Underground78]

I don't really understand, if "The Hand cursor indicates the place which can be clicked" (which often seems to be the case), keeping it for the buttons and the video frame would make sense, wouldn't it ?

[v0lt]

I don't really understand, if "The Hand cursor indicates the place which can be clicked" (which often seems to be the case), keeping it for the buttons and the video frame would make sense, wouldn't it ?
Yes, but do not always need it.

For a video frame is better to remove the Hand cursor anyway.

[Underground78]

For a video frame is better to remove the Hand cursor anyway.

In fact I already have a patch for that but I was reluctant to commit it because I fear people would forget they can click on the video frame.

[v0lt]

I remember there was a suggestion that the Hand cursor occur at prompting the over a DVD menu item.

[Underground78]

I remember there was a suggestion that the Hand cursor occur at prompting the over a DVD menu item.

It would be the minimum but it was more thinking to the play/pause behavior when clicking on the video frame.

[v0lt]

Now this cursor is always, even if the Play and Pause is disabled.

But I am more worried about the current seekbar. If during playback, move the mouse on the slider, which moves quickly (for example, look at a clip), it would be <censored>.

[Underground78]

Now this cursor is always, even if the Play and Pause is disabled.

But I am more worried about the current seekbar. If during playback, move the mouse on the slider, which moves quickly (for example, look at a clip), it would be <censored>.

I think I don't understand what you are talking about.

[v0lt]

I think I don't understand what you are talking about.
translate: This is a very bad idea, when a different cursor for the seek bar and position marker. :devil:

or an option in settings.
No need to make excess options.

[fastplayer]

No need to make excess options.
Agreed. That would be pure overkill.

On another note: How about removing "error resilience" and "IDCT Algorithms" from the video decoder properties?
IMO the default settings of these options are fine for the vast majority of MPC-HC users.

[cengizhan]

the latest version 1.5.2.3251 on xvidvideo.ru is buggy. mouse wheel changes audio delay, left and right cursor keys doesnt jump. also it changed the corresponding key maps. resetting them doesnt help.

[Aleksoid1978]

the latest version 1.5.2.3251 on xvidvideo.ru is buggy. mouse wheel changes audio delay, left and right cursor keys doesnt jump. also it changed the corresponding key maps. resetting them doesnt help.

Try this build - http://aleksoid.tosei.ru/Dist/mpc_1.5.2.3252_x86.7z

For me it's work fine.

[cengizhan]

Try this build - http://aleksoid.tosei.ru/Dist/mpc_1.5.2.3252_x86.7z

For me it's work fine.

Didnt work at first. i think that other version corrupted my mpc-hc.ini file. i have deleted that and now it works.

[Dstruct]

Didnt work at first. i think that other version corrupted my mpc-hc.ini file. i have deleted that and now it works.

Same happened here. Had to reassign some keys with 3251. Some were still mapped correctly. Others not.

[Mercury_22]

:thanks: for Revision 3251 - Directory Listing
Modified Sun Jun 19 18:54:34 2011 UTC (19 hours, 28 minutes ago) by underground78

Fix: Use a different program id for x64 builds so that there is no conflict over file associations between x86 and x64 builds.

(Warning: You might need to reset file associations.)
This commit fixes ticket #846.
Finally :)

But without an "icon_x64" and the change of the "Play with MPC-HC" and "Add to MPC-HC Playlist" messages for the MPC-HC x64 version to "Play with MPC-HC64" and "Add to "MPC-HC64 Playlist" the "problem" it's not completely solved
IMHO the main exe "icon.ico" should be updated and an "icon_x64.ico" should be added

PS Most of the users (I know :) ) don't use/care about the "media icons", (mpciconlib.dll is almost 1/2 of MPC-HC.exe) so we should not bother with those

PS2 I prefer like many other users the simple yet very nice and adequate for a media / video player (hint :p) icon from the MPC-HC BE mod

[Underground78]

Same happened here. Had to reassign some keys with 3251. Some were still mapped correctly. Others not.

Strange, were you using both the x86 and x64 builds ?

@Mercury_22 : I will ask Tetsuo55 or XhmikosR about the icon, I think someone is redesigning them so we can probably ask him to draw a new icon.

[Dstruct]

Strange, were you using both the x86 and x64 builds ?

Only x86 from xvidvideo.ru

[TheElix]

Question to experts: where can I read about EVR CP's GPU control in more detail? When should I use each option in what circumstances? Etc. Also, when it is advisory to turn off desktop composition?

[Mercury_22]

...
@Mercury_22 : I will ask Tetsuo55 or XhmikosR about the icon, I think someone is redesigning them so we can probably ask him to draw a new icon.

What I'm saying is lets add some icon ( even the same, but preferable the one from BE mod if Alexins it's OK with that) as "icon_x64.ico" and then people can come with "proposals" for the new icons and once Tetsuo & CO :) decide which is better then make the (final) change
A first step it's really needed cause we've been waiting so long for this changes
Also when you talk with them please ask about adding the GUI patch from the BE mod to the SVN cause it's just adding some options to the GUI with no changes for those who don't want / like them, and since bobdynlan it's "M I A" there is no point to wait for him

PS If needed Here is BE mod (http://www.xvidvideo.ru/media-player-classic-home-cinema-x86-x64/media-player-classic-homecinema-x86-x64-1-5-2-3251.html) or direct link for mod BE patch 2976 (http://www.mediafire.com/?k819dx1mrcgk2ha)

[Underground78]

I will tell them about that (Tetsuo reads this thread so he might even answer you before I catch him on IRC).

[mark0077]

Guys, I have a filter priority issue I'm wondering can someone comment on. I have "Lav Splitter" set to "Prefer" in mpc-hc and expect it to be used as the "Preferred" splitter for all of the media types that it advertises. However Haali Splitter still gets loaded in instead of it. Shouldn't the fact that its "Preferred" in the external filters section of mpc-hc make it have higher preference over other registered filters on my machine? Maybe I misunderstood how it works.

[clsid]

Source filters can be forced for specific file extensions using settings in the Registry. Those take precedence over merits.

Here is a useful little tool that (amongst other things) can configure the preferred splitters:
http://fileforum.betanews.com/detail/KLite-Codec-Tweak-Tool/1183386542/1

[mark0077]

tyvm!

[cengizhan]

Didnt work at first. i think that other version corrupted my mpc-hc.ini file. i have deleted that and now it works.

it seems that there is a problem with recent versions. keymaps arent working correctly. i have assigned ctrl+right to jump forward (medium), but it skips to next key frame.

[namaiki]

it seems that there is a problem with recent versions. keymaps arent working correctly. i have assigned ctrl+right to jump forward (medium), but it skips to next key frame.

Did you disable "Fast seek(on keyframe)" in 'Tweaks'?

[tetsuo55]

@ Mercury_22 and Underground78

I agree that the x64 needs its own icon, for now the existing icon with x64 added on top should do (We're about to get shiny new icons)

I also think the file association should have some hint to distuinguish between 32 and 64 bit versions (this requires code changes)

The BE mod skinability options where still in an alpha or pre-beta state last i heard, and that would be the reason why its not yet included. Ofcourse i want to have them committed as soon as possible.

[cengizhan]

Did you disable "Fast seek(on keyframe)" in 'Tweaks'?

it doesn't matter whether fast seek is disabled or not. pressing ctrl+right should jump forward without pausing. but it pauses and shows next frame

this is my mappings : http://i53.tinypic.com/105d2l1.jpg

[rsd78]

Any updates as to whether Intel DXVA will be supported/fixed?

[Lincoln Burrows]

Is there a problem I should know regarding FLAC streams in Matroska files? Is MPC-HC capable of reading those streams?

[GrofLuigi]

Could please someone help me decode the key customizations I've made a long ago that are very important for me. The first number is obvious, but what about the others? In the recent builds, and also not so long ago, MPC-HC kept re-sorting the entries and adding (modifying?) some, so this might not be the 'real' ones that I've made.

I'll read the source if necessary (although I don't have a slightest clue about programming) just to get them back. That's how important they are to me. :)

[Commands2]
CommandMod0=890 3 0 "" 5 0 13
CommandMod1=891 b be "" 5 0 0
CommandMod2=892 b bc "" 5 0 0
CommandMod3=900 3 be "" 5 0 0
CommandMod4=899 3 bc "" 5 0 0
CommandMod5=904 b 27 "" 5 0 0
CommandMod6=903 b 25 "" 5 0 0

GL

[namaiki]

[Commands2]
CommandMod0=891 3 0 "" 5 0 13
CommandMod1=892 b be "" 5 0 0
CommandMod2=893 b bc "" 5 0 0
CommandMod3=901 3 be "" 5 0 0
CommandMod4=900 3 bc "" 5 0 0
CommandMod5=905 b 27 "" 5 0 0
CommandMod6=904 b 25 "" 5 0 0

In short, try the above.

edit: This may have been reverted.

[JohnLai]

Does UVD 2.2 support VC1 hardware acceleration? Cause I have a video file encoded with :
ID : 2
Format : VC-1
Format profile : MP@HL
Codec ID : WMV3
Codec ID/Info : Windows Media Video 9
Codec ID/Hint : WMV3
Description of the codec : Windows Media Video 9 - Professional
Duration : 1h 25mn
Bit rate mode : Variable
Bit rate : 4 000 Kbps
Width : 1 280 pixels
Height : 720 pixels
Display aspect ratio : 16:9
Frame rate : 29.970 fps
Bit depth : 8 bits
Scan type : Progressive
Compression mode : Lossy
Bits/(Pixel*Frame) : 0.145
Stream size : 2.40 GiB (95%)

and MPC didnt show any DXVA in use at all. Catalyst 11.6. Any confirmation on what went wrong?

[G_M_C]

Does UVD 2.2 support VC1 hardware acceleration?
[...]

Do you need DXVA to play the file fluently ?

And on the question if UVD2.2 would support VC-1 in DXVA: click (http://lmgtfy.com/?q=uvd2.2+vc1+dxva)

[JohnLai]

Well, I am using Pentium 4, so DXVA is very important. Main question is, why mpc can't use dxva with the specified VC-1 format above? It works fine with H.264 though.

[fastplayer]

^If it's WMV3, then it's not supported.

[JohnLai]

^If it's WMV3, then it's not supported.

O.o? Mind to elaborate more? Thanks in advance.

[fastplayer]

MPC-HC's DXVA decoder just can't accelerate WMV3. That's all.

[namaiki]

it doesn't matter whether fast seek is disabled or not. pressing ctrl+right should jump forward without pausing. but it pauses and shows next frame

this is my mappings : http://i53.tinypic.com/105d2l1.jpg
Did you check the mapping for "Framestep" and "Framestep back"? The default keys have recently changed.

[v0lt]

the latest version 1.5.2.3251 on xvidvideo.ru is buggy. mouse wheel changes audio delay, left and right cursor keys doesnt jump. also it changed the corresponding key maps. resetting them doesnt help.
I can fix it (I'll turn back IDs).

fixed in r3262.

MPC-HomeCinema v1.5.2.3262 (x86) (http://www.mediafire.com/?xfgh33iws56bh)

[cengizhan]

Did you check the mapping for "Framestep" and "Framestep back"? The default keys have recently changed.

Thanks. i thought that was all about jumping. the new framestep was also ctrl+right. removed it and fixed the problem.

i think it will be very helpfull if you make mapping list sortable by name, key etc. And the program should refuse and disable dublicate key mappings.

[clsid]

Good news. I have fixed the vertical padding functionality in VSFilter. Also, a patch has been committed that significantly improves performance of subtitles in VSFilter and MPC.

Does any of the other devs perhaps has some time to update the settings page of VSFilter with this suggestion:
http://mpc-hc.svn.sourceforge.net/viewvc/mpc-hc/trunk/src/filters/transform/VSFilter/DirectVobSubFilter.cpp?r1=3260&r2=3259&pathrev=3260

[Underground78]

Does any of the other devs perhaps has some time to update the settings page of VSFilter with this suggestion:
http://mpc-hc.svn.sourceforge.net/viewvc/mpc-hc/trunk/src/filters/transform/VSFilter/DirectVobSubFilter.cpp?r1=3260&r2=3259&pathrev=3260

Something like that (http://i.imgur.com/15KnH.png) ?

[clsid]

Yes, that looks good.

[GrofLuigi]

[Commands2]
CommandMod0=891 3 0 "" 5 0 13
CommandMod1=892 b be "" 5 0 0
CommandMod2=893 b bc "" 5 0 0
CommandMod3=901 3 be "" 5 0 0
CommandMod4=900 3 bc "" 5 0 0
CommandMod5=905 b 27 "" 5 0 0
CommandMod6=904 b 25 "" 5 0 0

In short, try the above.

edit: This may have been reverted.

Thank you, but I wanted an explanation on the meaning of the entries so I can figure out what was changed and why (I forgot why I made some of the entries, some were changed by MPC-HC itself maybe, and there might be errors).

Like, what does be, bc mean; what are the codes for modifier keys and where... I'll by trial and error if I have to, but I was hoping for an explanation.

GL

[v0lt]

@GrofLuigi
There was an error. It is fixed in v1.5.2.3262.
I do not recommend using versions 3247...3261.

[Underground78]

Yes, that looks good.

You can try this test build (http://www.mediafire.com/?5ef9tk7b01a22vi).

[clsid]

You can try this test build (http://www.mediafire.com/?5ef9tk7b01a22vi).The settings aren't saved correctly.
The buttons to increase/decrease buffer size should be inverted. Top one should increase instead of decrease.

[Underground78]

The settings aren't saved correctly.
The buttons to increase/decrease buffer size should be inverted. Top one should increase instead of decrease.

That was really stupid, damn copy/paste ... :o It should be fixed in this build (http://www.mediafire.com/?qad8mazqaonen31).

[clsid]

Buffer option works now. But animation option still broken.

[Underground78]

Buffer option works now. But animation option still broken.

Problem found, I was not reading the same key I was writing in the registry so the saved value was not reloaded ... I hope it is ok now : VSFilter_v3.7z (http://www.mediafire.com/?oukb4ur3rk0fxzy) !

[Midzuki]

Problem found, I was not reading the same key I was writing in the registry so the saved value was not reloaded ... I hope it is ok now : VSFilter_v3.7z (httpwww.mediafire.comoukb4ur3rk0fxzy) !

httpwww.mediafire.comoukb4ur3rk0fxzy cannot be OK :)

[namaiki]

http://www.mediafire.com/?oukb4ur3rk0fxzy

Regarding the "allow animation when buffering" checkbox-

In v3: Tick means animation is disabled when buffering, unticked means buffer with animation.

So, you could either reverse the function of the checkbox or reverse the displayed name/action of the checkbox. To disable buffering, user can change 'sub pictures to buffer' to '0'.

[Midzuki]

@namaiki: Thanx :thanks:

[Underground78]

Sorry for the link, I still don't understand how this happened ... :eek:

In v3: Tick means animation is disabled when buffering, unticked means buffer with animation.

So, you could either reverse the function of the checkbox or reverse the displayed name/action of the checkbox.

I have kept the same name than in MPC-HC and reversed the action of the checkbox: VSFilter_v4.7z (http://www.mediafire.com/?b892nk7946dl1o0).

To disable buffering, user can change 'sub pictures to buffer' to '0'.

Is that ok ? It was done like that in MPC-HC so I have kept it the same way.

[namaiki]

I have kept the same name than in MPC-HC and reversed the action of the checkbox: VSFilter_v4.7z (http://www.mediafire.com/?b892nk7946dl1o0).

Everything seems to work as it should. Awesome work! :D

Is that ok ? It was done like that in MPC-HC so I have kept it the same way.
I think it should be fine as long as animation is enabled by default. Also, I might recommend buffer set to 4 subpictures by default due to flickering subtitles on 3 when there is more than a single subtitle line on the screen.

[clsid]

Working OK now, so time to commit.
Changing buffer size to 4 should be fine. Please change it in MPC as well. The old default in VSFilter was 10, but I don't know if a larger buffer than 4 has much real benefit.

I did notice a crash when changing the settings during playback. Perhaps that can be solved by reading the setting into two vars, one for usage during playback, and one purely for the GUI. So no real-time update of the setting.

I did some basic benchmarks to test the performance of subs. These numbers were gathered with ProcessExplorer and from just one run per test, so take results with a grain of salt.
I measured the total CPU time used by MPC while playing a 720x480 H.264 clip with subtitles that have karaoke effects. Clip duration 1m46s.

Without subs
6.74 seconds

New VSFilter, Buffer = 0
12.38

VSFilter 2.39, Buffer = 0
13.09

New VSFilter, Buffer = 4
10.54

New VSFilter, Buffer = 4, Animation on
13.81

New VSFilter, Buffer = 10
11.01

New VSFilter, Buffer = 10, Animation on
13.76

VSFilter 2.39, Buffer = 10
10.78

Conclusions:
- Heavy subtitles can have a big performance impact. Up to 50% of the total CPU usage.
- Animation with buffering uses the most CPU time. Followed by Animation without buffers. Buffering without animation uses the least CPU time.
- Animation can use up to 80% more CPU time than no animation.
- New VSFilter is not much faster than the old one. I expected a bigger difference with the recently added performance tweaks. Perhaps the VSFilterMod additions have some negative effect on performance.

I might do another test with basic SRT subs later to measure performance in case of a non-animated sub.

[tetsuo55]

That is kind of disappointing.

please do the basic SRT test.

[JEEB]

Well, unfortunately VSFilter has been and will continue to be slow. No real news there. And I kind of expected there to be not much of real performance-up, although the "multithreaded VSFilter" guy is relatively enthusiastic, which is a good thing.

Anyways, if we want real speed-ups, someone needs to make fontconfig less awful on Windows so we can use libass instead :3

[clsid]

With a quickly manually created SRT I get results between 7,5 and 7,9 seconds. So those differences are small.
Btw, all the above tests were done with the standard EVR renderer.

EVR CP has a very nasty 'bug' that makes it consume CPU cycles even when paused or when playback has stopped. Hopefully that can be given some priority to fix.

I also tried with the internal sub renderer (and EVR CP instead of EVR), but I got weird results. Both SSA and SRT have same CPU time of 17.0s (buffer 4 + anim), no subs gave 14.4. The measurement method might be inaccurate.

[namaiki]

EVR CP has a very nasty 'bug' that makes it consume CPU cycles even when paused or when playback has stopped. Hopefully that can be given some priority to fix.

I believe it's because it's continually scanning the screen's refresh rate.


I did notice a crash when changing the settings during playback. Perhaps that can be solved by reading the setting into two vars, one for usage during playback, and one purely for the GUI. So no real-time update of the setting.
There is the 'Apply changes immediately' checkbox in the 'Misc' tab. I suppose that could be disabled by default. (please don't get rid of it)

[Underground78]

So everybody agrees with that:
I think it should be fine as long as animation is enabled by default. Also, I might recommend buffer set to 4 subpictures by default due to flickering subtitles on 3 when there is more than a single subtitle line on the screen.
There is the 'Apply changes immediately' checkbox in the 'Misc' tab. I suppose that could be disabled by default. (please don't get rid of it)

If everybody does, I will commit the changes and also update default settings for MPC-HC.

Edit : Here what MPC-HC does to choose if it enables or not animation when buffering :

bool bAllowAnimationWhenBuffering = true;
SYSTEM_INFO SysInfo;
GetSystemInfo(&SysInfo);
if (SysInfo.dwNumberOfProcessors < 3) {
bAllowAnimationWhenBuffering = false;
}

Maybe I should keep something like that ? Although I am not sure it is really useful to check the number of processors if VSFilter is not multithreaded ?

[fastplayer]

Assuming that the code is actually not multithreaded, this sure looks like a lazy man's performance categorization. It is pure nonsense IMO.

[Underground78]

Assuming that the code is actually not multithreaded, this sure looks like a lazy man's performance categorization. It is pure nonsense IMO.

Yeah but I need to be sure the code is not multithreaded at all.

[fastplayer]

Yeah but I need to be sure the code is not multithreaded at all.
Even if it was, it wouldn't be dependent on this setting at all. There's no variable that saves the number of cores and uses it subsequently throughout the code.
We had similar nonsense in ffdshow that Stargazer removed when he cleaned up the colorspace conversion code.

[Underground78]

Even if it was, it wouldn't be dependent on this setting at all. There's no variable that saves the number of cores and uses it subsequently throughout the code.
We had similar nonsense in ffdshow that Stargazer removed when he cleaned up the colorspace conversion code.

I am not sure to understand, it is just an hint for the user I guess so why would you need a variable to save the number of core ? However I am quite sure it was wrongly chosen and I will probably delete it.

[fastplayer]

I am not sure to understand, it is just an hint for the user I guess so why would you need a variable to save the number of core ? However I am quite sure it was wrongly chosen and I will probably delete it.
The code basically says that anything with less than 3 cores is not suitable of running sub animations which is total nonsense of course. Especially considering that the code itself is not multithreaded at all.

My suggestion:
bool bAllowAnimationWhenBuffering = true;
SYSTEM_INFO SysInfo;
GetSystemInfo(&SysInfo);
if (SysInfo.dwNumberOfProcessors < 3) {
bAllowAnimationWhenBuffering = false;
}

r.fSPCAllowAnimationWhenBuffering = !!pApp->GetProfileInt(IDS_R_SETTINGS, _T("SPCAllowAnimationWhenBuffering"), bAllowAnimationWhenBuffering);
replace with
r.fSPCAllowAnimationWhenBuffering = !!pApp->GetProfileInt(IDS_R_SETTINGS, _T("SPCAllowAnimationWhenBuffering"), true);

[clsid]

There is the 'Apply changes immediately' checkbox in the 'Misc' tab. I suppose that could be disabled by default. (please don't get rid of it)Disabling that option is not a good idea. It is useful for seeing effects of font and size changes.
The crash only happens with the two new settings. Just do not apply those changes directly.

The rationale behind the number of processors check is that a CPU with multiple cores is usually a reasonably fast one.

I believe it's because it's continually scanning the screen's refresh rate.That could be made conditional based on the playback state. Maybe the scan interval can be increased as well to lower CPU usage. Isn't the refresh rate typically a constant value? Except maybe at the start if rate changing is enabled.

[Underground78]

Disabling that option is not a good idea. It is useful for seeing effects of font and size changes.
The crash only happens with the two new settings. Just do not apply those changes directly.

I've changed the code so that those two settings are not applied immediately : VSFilter_v5.7z (http://www.mediafire.com/?77q365o7fq0td7o). I have also change the default to 4 and true but I can still change this.

The rationale behind the number of processors check is that a CPU with multiple cores is usually a reasonably fast one.

This is how I understood it at first that's why I am not sure what to do. Even if it is a test for lazy, we may not want to always enable this setting.

[clsid]

Working fine now.

The core detection code is only used for the default setting. It is only used if the value was not yet set in the Registry/INI.

[cyberbeing]

I measured the total CPU time used by MPC while playing a 720x480 H.264 clip with subtitles that have karaoke effects. Clip duration 1m46s.

CPU time alone is not a good way to judge VSFilter performance. The real question is if there is a gain in FPS with extremely heavy scripts, which are unable or barely playback in real-time on slower computers.

I did some testing, and in most cases gommorah's fixed string manipulation patch results in a 0-2% increase in avg framerate on difficult samples when VSFilter is CPU limited and prebuffering is disabled (r3255 vs r3262 from xvidvideo.ru). Some subtitle samples and CPUs likely benefit from the patch more than others, but at least this resolved one of many causes of VSFilter slowdowns. Since the code hasn't really changed much, it's likely the optimizations from VS2010 vs VS2005 results in most of the gain over Guliverkli2.

AMD X2 4800+ (939) @2.64Ghz on WinXP SP3

VSFilter r3262 (xvidvideo.ru VS2010)
KaraokeSample1: ~22 fps
KaraokeSample2: ~20.5 fps
TypesettingSample1: ~20.5 fps

VSFilter r3255 (xvidvideo.ru VS2010)
KaraokeSample1: ~21.75 fps
KaraokeSample2: ~20 fps
TypesettingSample1: ~20 fps

VSFilter r3242 (VS2010 WPO+PGO)
KaraokeSample1: ~22.5 fps
KaraokeSample2: ~22.75 fps
TypesettingSample1: ~21.25 fps

VSFilter 2.39.5.2 (Guliverkli2 VS2005)
KaraokeSample1: ~20.75 fps
KaraokeSample2: ~16.75 fps
TypesettingSample1: ~19.25 fps

Note: The above are rough measurements, averaged over a few runs on 1000-2000 frame clips. They are accurate enough to see which builds are faster/slower with the samples tested, but not accurate enough to determine exact performance gains/losses in the way I tested. The r3242 WPO+PGO build I just added for comparisons sake. I haven't taken the time to make a new build, but I suspect a r3262 WPO+PGO build would have a similar 0-2% gain with the patch like the xvidvideo.ru builds do.

[Underground78]

The core detection code is only used for the default setting. It is only used if the value was not yet set in the Registry/INI.

Yeah I know but what should I do ? Keep 4 and true as default settings for VSFilter ? Should I port them back to MPC-HC (getting rid of the core check) ?

I never use animation in subtitles so I don't really know what to do ...

[VipZ]

Yeah I know but what should I do ? Keep 4 and true as default settings for VSFilter ? Should I port them back to MPC-HC (getting rid of the core check) ?

I think the 3 core check seems kind of odd, as this eliminates Core2Duo's but includes AMD Phenom X3 which are typically clocked lower and slower clock for clock. Especially since its not multithreaded.

Maybe if a check is desired something like this,
WinXP = 3 cores (Mostly to keep it away from older P4's with HT)
Vista+Windows7 = 2 cores

[clsid]

I say commit with 4 as default in both VSFilter and MPC. Use true for the anim option. It can always be changed later if needed.

[neb1236]

Hi, since revision 3262 (fixed vertical padding) placement of Drawing is wrong (I don't know why and how).
Here is a self explanatory image, I used Aegisub to render it so we can see the anchor point, but I have tested on a video, and the result is the same:
http://img808.imageshack.us/img808/8107/examplerb.png

Thanks for reading.

[namaiki]

Hi, since revision 3262 (fixed vertical padding) placement of Drawing is wrong (I don't know why and how).
Here is a self explanatory image, I used Aegisub to render it so we can see the anchor point, but I have tested on a video, and the result is the same:
http://img808.imageshack.us/img808/8107/examplerb.png

Thanks for reading.
It would be much appreciated if you could upload your video/subtitle sample so that this issue can be reproduced.

[neb1236]

It would be much appreciated if you could upload your video/subtitle sample so that this issue can be reproduced.

Of course, here it is, under the form of a test case: http://www.mediafire.com/?6at2mvvzhb3xd97

[clsid]

With the old VSFilter 2.39 it also isn't perfect. Without padding the blue boxes are slightly larger, but the black boxes are positioned correctly. With padding enabled, the black boxes are stretched vertically. But padding is something not really suitable for subs that need specific positioning above the video.

With new VSFilter, the black boxes are positioned wrong even with padding disabled.

[neb1236]

Oh, sorry, forgot to mention the reference image, (I didn't made the square the same size), here is reference image, and the result with r3262, it's not a very good testcase:
http://img717.imageshack.us/img717/2954/resultu.png

[cyberbeing]

I can confirm what neb1236 is seeing when Vertical Padding is set to Original Height in r3262.

CCCP's r339/r1096 basevideofilter revert patches doesn't have the positioning bug seen in neb1236's r3262 image and fixes Vertical Padding in VSFilter. Identical results to Guliverkli2.
Edit: My mistake, it turns out I tested the wrong build which only had the r339/r1096 basevideofilter revert patches. The build with r339/r1096 basevideofilter revert patches and gommorah's patch did have the bug. It looks like CLSID fixed this in r3270.

CLSID's custom Vertical Padding Fix patch in r3261 does have the positioning bug seen in neb1236's r3262 image.

[clsid]

r3261 should have the same effect as reverting 339/1096. Some fresh eyes looking at that code would be helpful. I might have overlooked something.

[Eliminateur]

noone of the internal filters appear to be working i select all of them, ffmpeg and dxva and i always get a pin failed to render message.
no matter if i disable them in ffdshow or not
r3262 x64

[v0lt]

mpc-hc_3268_dts_ac3_test1.7z (http://www.mediafire.com/?wveo5je94w57y8g)
DTSAC3Source:
+ Added support DTSWAV (DTS-CD)
+ Added search DTS and AC3 headers (skip garbage in the beginning)
+ Improved support for AC3 and E-AC3

[JohnLai]

Next question, does full floating point processing work with ATI? I can't use it, if I do, it resulted in either black screen or flickering black and white. (in dxva mode)

[TheElix]

Next question, does full floating point processing work with ATI? I can't use it
I can. I have an HD6950.

I have a different problem which is quite terrible. I can't change files with PgUp/PgDn when I'm in D3DFS. When I try to do so everything's normal except one thing: video is frozen on the last frame of the previous video. It's as if the video card fails to see the need to show any changes! Sound is running normal but seekbar as well as mouse cursor aren't showing up. Catalyst 11.6, everything is set to use application settings. All video enhancers are off. I use EVR CP and tried every option in renederer settings. Tried different splitters as well. Tried uninstalling the player with its settings and with a fresh install it's the same thing.
:confused:

[JanWillem32]

I can. I have an HD6950.

I have a different problem which is quite terrible. I can't change files with PgUp/PgDn when I'm in D3DFS. When I try to do so everything's normal except one thing: video is frozen on the last frame of the previous video. It's as if the video card fails to see the need to show any changes! Sound is running normal but seekbar as well as mouse cursor aren't showing up. Catalyst 11.6, everything is set to use application settings. All video enhancers are off. I use EVR CP and tried every option in renederer settings. Tried different splitters as well. Tried uninstalling the player with its settings and with a fresh install it's the same thing.
:confused:It's an old problem of the renderer termination sequence in exclusive mode. It causes the old renderer instance to keep attached to an exclusive mode session on a display output, and it disallows new sessions to take possession of it. I re-wrote that part of the renderer fist in March, and I've made several optimizations for this part after that. Note that EVR is usually more sensitive to this problem than VMR, due to different unloading speeds. I've yet to integrate any optimizations I've made over time since January. There are some regressions in my version of the renderer core that inhibit me from requesting integration of the code I wrote. The most notable problem is the incompatibility with some the original VSync code.

[TheElix]

JanWillem32, are you a wizard (http://angrywhitedude.com/wp-content/uploads2/2010/05/are-you-wizard.jpg)? The problem is gone with your build! Why don't other development folks use your method to prevent this problem?
P.S. And I've almost made up my mind to delete the system.

[JanWillem32]

I'm not going to try to break up my re-written parts of the renderer for partial integration. I've re-written almost the complete rendering engine, and the allocator-presenter is re-written for more than a quarter of the total code. There's little resemblance in some parts of the rendering method. That causes incompatibility with a lot of the older code. The best I can do at the moment is to provide users with tester builds and interact in the thread I made. I use this method as a way to work towards offering the renewed version of the internal renderer sections as a complete replacement, rather than a patch of the original code. I most certainly know I have bend over backwards to ever get the three VSync items to work again as they used to. I'll just have to make it work, but it's very hard for me to understand the cohesion in the many 64-bit items it uses, with timescales in millisecond, 100 nanosecond, nanosecond, second and per second timing units. I would never have written it in this way. (I'm usually lazy in this regard and just activate the video card's own internal VSync, or let desktop composition handle the VSync in windowed mode, but I'm a lot more familiar with DirectX 10 and 11 than 9, anyway.)

[TheElix]

I see. Well, your method is the only one that makes internal autochange frequency work for me. That just makes one more reason for me to stick to your versions for now. By the way, does autofrequency in your build work for you with D3DFS?

[JanWillem32]

The exclusive mode frequency adjustment part is officially on the "to do" list, but if the display frequency is already set while the program is in windowed mode, the code I wrote will happily accept normal (whole number) and NTSC refresh rates (those are 1/1.001 adjusted) from the currently active state in windowed mode. If the exclusive mode is activated before this kind of frequency change command is used, that command is simply negated.
I didn't know this function worked at all in my builds. I haven't paid much attention to this part, I just tried to fix the critical things first. I'll have to add this function to the "to test if it works in all cases" list then...
If it doesn't really work in exclusive mode I can try to make a link to the allocator-presenter. It could then inherit the custom set frequency number directly from the user-set options and enable it on an output by itself, instead of inheriting the frequency from the windowed mode before it.

[v0lt]

@Underground78 and other developers

Then it was:
- Never
- Always
- While Playing

Now it is:
- Never
- Always
- While Playing
- While Playing Video
:confused:

Can do better than three lines?
Еxample:
- Never
- Always
- While Playing Video

[namaiki]

- While Playing
- While Playing Video[/I]
:confused:

Can do better than three lines?


While Playing <- This means MPC-HC is always on top.
While Playing Video <- This means MPC-HC is only on top if a video is open in MPC-HC and it is playing.

[v0lt]

While Playing <- This means MPC-HC is always on top.
While Playing Video <- This means MPC-HC is only on top if a video is open in MPC-HC and it is playing.

How it works I understand. But why do so difficult? Who needs these four modes? Three is enough.

[namaiki]

How it works I understand. But why do so difficult? Who needs these four modes? Three is enough.

Feature request on the tracker:
http://mpc-hc.svn.sourceforge.net/viewvc/mpc-hc?view=revision&revision=3247

It doesn't really hurt anybody to just keep it there, unless there is a better way to implement.

Personally, I do use the video only on top mode.

[Underground78]

How it works I understand. But why do so difficult? Who needs these four modes? Three is enough.

Some people might want to keep the player on top when playing audio files (for example if they want to see informations about the track) and not only for video files. I don't think having a 4 menu items really matters ...

[quasit]

Isn't native PCM support (MKV) gonna be fixed? I tried latest nightly and I still get same error as before, http://pastebin.com/WknqTs3P
Yes, it works using LAV splitter but then I can't change between subtitles (just enable/disable). I tried Haalii Media Splitter but then I get out of sync for PGS subtitles (MKV).

It says in the bug report that you should change audio render, it doesn't help. It also says some sound cards don't support PCM. So what supports it then? Some $200 sound card? Why can't a MPCHC splitter be added that supports PCM and doesn't have any subtitle issues like LAV and HMS?
http://sourceforge.net/apps/trac/mpc-hc/ticket/665

[e-t172]

It also says some sound cards don't support PCM.

Nonsense. PCM is the standard audio format for all computer applications. A soundcard which doesn't support PCM is just a very expensive paperweight.

[LigH]

In general yes, but ... ;)

PCM = "Pulse Code Modulation" means: raw, uncompressed samples. But samples may be represented in many different formats, and not all formats need to be supported by all soundcards.

There are integer samples (WAV 2CC: 0x0001), where 16 bit is the most common and usual resolution, but 24 bit or 32 bit samples are possible too (let's forget about 8 bit). Video DVDs might even have 20 bit samples in a rather strange interleaving (not supported by WAV).

Furthermore there are floating point samples (WAV 2CC: 0x0003) in IEEE single or double precision format (32 or 64 bit).

IIRC, the "Microsoft Wave Mapper" was the special ACM "codec" of VfW to convert sample formats between WAV storage and sound card hardware support on-the-fly. DirectShow will have a similar special filter.

[v0lt]

Some people might want to keep the player on top when playing audio files (for example if they want to see informations about the track) and not only for video files.
I have not seen this features in any audio player. There's enough options "never" and "always".
Videoplayer should have one additional option. The second option is excess.

Now we have a bug with option Options...->Player->Other->Always in top :(

[clsid]

I also suggest to associate Ctrl+A with "while playing video" instead of "always". That is more useful imho.

[Underground78]

I have not seen this features in any audio player. There's enough options "never" and "always".
Videoplayer should have one additional option. The second option is excess.

Why only one ?! I can't understand your point ... However I agree with clsid suggestion and we will probably change the shortcut if everybody agrees. We can also change the order, if people prefers the new option to come first.

Now we have a bug with option Options...->Player->Other->Always in top :(

That's true, thanks for reporting it. (I personally tend to think that three states checkboxes are often not really clear and I had never noticed this one ...)

I can do that to fix the bug:

http://i.imgur.com/wiUZg.png

[fastplayer]

(I personally tend to think that three states checkboxes are often not really clear and I had never noticed this one ...)
+1
They are even less clear when there's no tooltip or any description whatsoever...
The dropdown menu makes more sense, especially if there are more than 3 choices to be made of. :)

[clsid]

I would suggest to just remove it from the options page. A drop down looks ugly there. I also do not see any need to have the same option in two places. Having it in the View menu is the natural place.

[Underground78]

I would suggest to just remove it from the options page. A drop down looks ugly there. I also do not see any need to have the same option in two places. Having it in the View menu is the natural place.

I am not against doing that too.

[v0lt]

I can do that to fix the bug:
If everything remains as it is, it is necessary to correct the settings dialog. The order of options not need to change.

Isn't native PCM support (MKV) gonna be fixed?
fixed in r3271

MPC-HomeCinema 1.5.2.3271 (x86) (http://www.mediafire.com/?xfgh33iws56bh)

[Underground78]

If everything remains as it is, it is necessary to correct the settings dialog. The order of options not need to change.

I don't think I change it, do I ? Or maybe we are not talking about the same thing ?

[v0lt]

@Underground78
Do anything or I will return back three options. :devil:

[Underground78]

I will do something but let us discuss that before doing anything ... The combobox fix was already ready when I posted the screenshot but as some people seem to think there is no need to have the options twice, I will need to ask Tetsuo and/or XhmikosR opinion about that.

[boyumeow]

...

Can do better than three lines?
Еxample:
- Never
- Always
- While Playing Video

How about
- Never
- While Playing
- While Playing Video

[tetsuo55]

What about this

Dropdown:
* Never
* Always
* Only during playback

Checkbox:
[] Do not apply "always on top" for audio files.

We could change that checkbox to "always on top during playback"

[clsid]

3 choices plus checkbox is more complex.

I don't see the whole problem with having 4 choices instead of 3. It is just nitpicking.

[TheElix]

- While Playing
- While Playing Video

Isn't self-explanatory at all. I agree with those who want to remove/rename this option.

[clsid]

Not self-explanatory? How more obvious does it need to be? Well, it could be renamed to "While playing audio or video" for those who are mentally retarded.

[tetsuo55]

The options are confusing for 2 reasons i think

1. Since MPC-HC is a video player, "while playing" and "while playing video" are exactly the same thing.

2. The alternative option "while playing video" is technically aimed at the niche group of users that use mpc-hc as their audio player. They want to have it on top for video, but not audio which makes sense.

If we want to keep 4 options will will have to make extremely clear what each option means, or alternatively we can seperate out this special audio users option into a checkbox.

[v0lt]

test build - mpc-hc_3282_ontop_n_a_wpv.7z (http://www.mediafire.com/?z8aqq31w7so5156)
(On top: Never, Always, While Playing Video)

[tetsuo55]

After discussing with underground i came up with this suggestion:


[]On Top
-[]Only while File is playing
-[]Disable when audio-only file is playing

It should catch all 4 usecases.

[nevcairiel]

IMHO, MPC-HC is a video player, and this feature is just bloating up the options for no good reason. I would just deny it. Just because someone asks for something doesn't mean you have to do it, you know.

[robpdotcom]

I use MPC for playing audio; I have quite a few BluRay audio discs that I have ripped to mka, so it's useful for bitstreaming/decoding DTS-HD (are there any audio players that can use LAVaudio or ffdshow?). But I still agree with Nev.

[TheElix]

I use MPC for playing audio; I have quite a few BluRay audio discs that I have ripped to mka, so it's useful for bitstreaming/decoding DTS-HD (are there any audio players that can use LAVaudio or ffdshow?). But I still agree with Nev.What about foobar? It can decode DVD-A images, dunno about blu-ray. Also supports ASIO/WASAPI/kernel streaming.

[xiulet]

mpc-hc_3268_dts_ac3_test1.7z (http://www.mediafire.com/?wveo5je94w57y8g)
DTSAC3Source:
+ Added support DTSWAV (DTS-CD)
+ Added search DTS and AC3 headers (skip garbage in the beginning)
+ Improved support for AC3 and E-AC3

hi, i try to play The Dark Side Of The Moon (2003 Hibrid SACD) that is a dtswav burn and get the tipical noise on mph hc(3280), with vlc an foobar ( with DTS decoder component) i get multi-ch pcm on my a/v receptor. Are testing the support DTSWAV (DTS-CD) or are full implemented right now ?

Only my h/k dvd player(s/pdif optical) was showing dts 5.1 from my receptor, but now is broken :(.

I am using hdmi1.3a cable both( ati 5850--> h/k 255).
thanks, adéu. :cool:

[Aleksoid1978]

hi, i try to play The Dark Side Of The Moon (2003 Hibrid SACD) that is a dtswav burn and get the tipical noise on mph hc(3280), with vlc an foobar ( with DTS decoder component) i get multi-ch pcm on my a/v receptor. Are testing the support DTSWAV (DTS-CD) or are full implemented right now ?

Only my h/k dvd player(s/pdif optical) was showing dts 5.1 from my receptor, but now is broken :(.

I am using hdmi1.3a cable both( ati 5850--> h/k 255).
thanks, adéu. :cool:

Upload one dtswav for test.

[xiulet]

this one :DTSWAV_test (http://www.megaupload.com/?d=CZBQSV2K)

adéu :p

[v0lt]

this one :DTSWAV_test (http://www.megaupload.com/?d=CZBQSV2K)
mpc-hc_3268_dts_ac3_test1.7z (http://www.mediafire.com/?wveo5je94w57y8g) - plays
MPC-HomeCinema 1.5.2.3283 (http://www.mediafire.com/?xfgh33iws56bh) - does not play

"Added search DTS and AC3 headers" feature not implemented yet, because it may work on some video files.

[xiulet]

mpc-hc_3268_dts_ac3_test1.7z (http://www.mediafire.com/?wveo5je94w57y8g) - plays
MPC-HomeCinema 1.5.2.3283 (http://www.mediafire.com/?xfgh33iws56bh) - does not play

"Added search DTS and AC3 headers" feature not implemented yet, because it may work on some video files.

yes i see mpc-hc_3268_dts_ac3_test1.7z play this individual file, but when using the menu option AUDIO CD, to play entire cd wavdts file from my combo player, it get the noise again.

thanks, adéu.

[v0lt]

but when using the menu option AUDIO CD, to play entire cd wavdts file from my combo player, it get the noise again.
What MPC-HC is to blame? :confused:

[xiulet]

What MPC-HC is to blame? :confused:

noo, i try tu help with this , my english is not good , may be you understand bad my commentary :(

i can upload the full cd audio for testing but i thing this should be illegal , isen't it?

No problem , using vlc or foobar200 it works.

adéu.

[v0lt]

DTSAC3Source filter does not work with the disk, it works with files (.dts, .wav, .dtswav).

CDDAReader filter works with the disk. I have not seen it.

[xiulet]

DTSAC3Source filter does not work with the disk, it works with files (.dts, .wav, .dtswav).

CDDAReader filter works with the disk. I have not seen it.

ok, then i will use any of other two programs, thanks

adéu.

[boyumeow]

After discussing with underground i came up with this suggestion:


[]On Top
-[]Only while File is playing
-[]Disable when audio-only file is playing

It should catch all 4 usecases.

Just on my point of view, I agree with v0lt to have 3 option rather than 4.

I personally using MPC-HC for video and audio.
In my case, [Always] is not needed when I'm not even using MPC-HC to play any video (cause I only have 1 screen).
I listen to music using my ears rather than using my eyes (sorry).

Thanks.

[Underground78]

If we really want to have only 3 options, we probably should rename "While Playing Video" to "While Playing" and add a new checkbox in the "Tweaks" page which would be enable by default to disable on top while playing for audio-only files.

(I won't be able to work on that before I can fix my computer, the SATA chip seems to be out of order.)

[KornX]

ive put in ticket 1452...
(to fill my 24GB)
http://sourceforge.net/apps/trac/mpc-hc/ticket/1452
and i am willing to buy a beer or sth for the prob solver :)

KornX

[clsid]

If we really want to have only 3 options, we probably should rename "While Playing Video" to "While Playing" and add a new checkbox in the "Tweaks" page which would be enable by default to disable on top while playing for audio-only files.I still vote for 4 options. Putting checkboxes everywhere just sucks. Related options should be together as much as possible. You don't simplify things by making them complex.

Name the choices "While playing audio or video" and "While playing video". That should be clear for everyone.

[JarrettH]

Is there a proper way to setup mpc-hc so CUVID takes over DVDs? Whenever I've blocked the microsoft and mpc-hc ones, the mpc-hc non-dxva one takes priority or 'low merit' words appear beside it :thanks:

[nevcairiel]

LAV CUVID is not compatible with DVDs yet.

[TheElix]

I want to report a long-persisting bug. When you open MPC-HC in D3DFS mode any command you try to enter with a keyboard gets doubled. For example, if I press pause, it pauses and unpauses; if I press PgDn, it skips two chapters; if I press "s" to change subtitles, it skips one, and so on... The only way to fix this is to press the left mouse button on the screen as if making MPC-HC an active window. Please, fix it.

[Qaq]

I want to report a long-persisting bug. ...if I press pause, it pauses and unpauses
Same thing with me. I use JanWillem32's builds. Thanks for the mouse tip.

[TheElix]

Also, please make Direct3D+AutoChangeMonitorMode work for everyone if possible. It works for me only in JanWillem32's builds unless I jump to another file in which case resolution is reset to default and remains the same in the newly opened file. It only works if I open the file along with the player. As for others I hear it doesn't work for them at all. It's a shame because it's a really helpful function.

[boyumeow]

I still vote for 4 options. Putting checkboxes everywhere just sucks. Related options should be together as much as possible. You don't simplify things by making them complex.

Name the choices "While playing audio or video" and "While playing video". That should be clear for everyone.

I agree with clsid for not putting checkboxes everywhere.

In my case, "While playing audio and video" is as good as "Always", I don't see the need for always on top when we are not even using MPC-HC, that's why I vote for 3 options. While we could use the word "Always" for "While playing audio and video" only, unless I have misunderstood it.

Just to summarize, I agree with clsid: 4 options without extra checkboxes. Thanks.

p.s. I'm not rushing U, Underground78, just your PC :D.

[tetsuo55]

Agreed, lets do clsid's idea.

Always
Never
When playing video
When playing audio and video

[clsid]

I see there are attempts to update the FFmpeg code again. Maybe you guys should reconsider putting it in a DLL. That would save a lot of trouble getting it compatible with VS2010. Only compiling it with GCC will require much less custom code and the difference with ffdshow would be minimal.

[HeadlessCow]

I agree with clsid for not putting checkboxes everywhere.

In my case, "While playing audio and video" is as good as "Always", I don't see the need for always on top when we are not even using MPC-HC, that's why I vote for 3 options. While we could use the word "Always" for "While playing audio and video" only, unless I have misunderstood it.

Just to summarize, I agree with clsid: 4 options without extra checkboxes. Thanks.

p.s. I'm not rushing U, Underground78, just your PC :D.

The difference between "Always" and "While playing audio and video" is that if you pause it, it stops being always on top. I use that all the time because I'll have a video playing in the corner of my screen, but sometimes need to see a window that's overlapping the video area (generally the menubar of whatever app I'm using while I watch video).

[xiulet]

hi, when playing cd-audio the mpl-hc make a pause to find next song( or timeline), in audio-cd with no pauses between songs this is not good :)
there are a alternative CDDA Reader Filter to use? or will be possible to fix this? Foobar2000 plays well with no pauses.

thanks, adéu.

[v0lt]

I can do that to fix the bug:
http://i.imgur.com/wiUZg.png
I tried to translate. It looks badly on Windows 7.

[Underground78]

Do you think giving more space for "On top" text will fix the problem? (I can also use two lines, one for the label and one for the combobox.)

About r3299 (https://sourceforge.net/apps/trac/mpc-hc/changeset/3299/trunk), I agree with the fact using default font size is usually better but here it is not so great because the path can be very long and with the default size it will be always cut. (A real fix would be to use more than one line instead of cutting the message but I am not sure it is easy to do ...)

[v0lt]

About r3299 (https://sourceforge.net/apps/trac/mpc-hc/changeset/3299/trunk), I agree with the fact using default font size is usually better but here it is not so great because the path can be very long and with the default size it will be always cut. (A real fix would be to use more than one line instead of cutting the message but I am not sure it is easy to do ...)
Vice versa. It is now well. I used default OSD size = 10.

Do you think giving more space for "On top" text will fix the problem? (I can also use two lines, one for the label and one for the combobox.)
The problem is not only a width. Combobox looks bad in Windows 7 at this place.

About width:
eng - On top: while playing video
rus - Поверх других окон: при воспроизведении видео

[clsid]

Why not remove that option like I suggested before? It is unneeded since the option is already in the View menu.

[Hera]

I want to report a long-persisting bug. When you open MPC-HC in D3DFS mode any command you try to enter with a keyboard gets doubled. For example, if I press pause, it pauses and unpauses; if I press PgDn, it skips two chapters; if I press "s" to change subtitles, it skips one, and so on... The only way to fix this is to press the left mouse button on the screen as if making MPC-HC an active window. Please, fix it.

This is really annoying.

[Underground78]

If you don't want it to be lost again in this thread, you should probably open a new ticket in Trac for this bug report.

[TheElix]

Where is this, can you point me to where I might leave a ticket?

[JanWillem32]

http://sourceforge.net/apps/trac/mpc-hc/wiki/How_to_Report_Issues
That's a basic overview of the ticket system.

[boyumeow]

Why not remove that option like I suggested before? It is unneeded since the option is already in the View menu.

Agree, it might helps for translations that v0lt is encountering.

[Underground78]

This has been discussed a lot on IRC yesterday and we will probably rework the options page so that the combobox fits. Tetsuo and XhmikosR don't really like the idea of removing the option from the options dialog.

[clsid]

If they insist on having the option twice, then so be it.

[v0lt]

Changeset 3310
instead of AviSplitter: fixed r3305 (round-off error)
must be AviSplitter: fixed r3305 (arithmetic overflow)
sorry

please correct if possible.

[Underground78]

I changed the log message, it won't change in Trac but at least it will be correct in the svn log. (It is probably possible to resync Trac with the svn but I don't know if Sourceforge let the admins do that).

[v0lt]

Thanks.

[Masutin]

1) MPC exits the Full Screen mode when switching to another window. Can I prevent it?
2) A request for the Go To function: Converting the value of minutes that exceeds 60 to the corresponding values of hours and minutes, e.g. 0:99 to 1:39. BSPlayer does this.

[TheElix]

If you don't want it to be lost again in this thread, you should probably open a new ticket in Trac for this bug report.
It's already been reported by someone else, ticket #1493 (http://sourceforge.net/apps/trac/mpc-hc/ticket/1493).

[v0lt]

About "Go To frame" function. Does anyone really need a feature to manually set the fps?

[TheElix]

Can't really see why. Maybe if you want to compare same movie/different release with different frame rates on the same frame, but I usually use the 'eyes and framestep' method. Let's treasure the time of the developers and let them spend it on something more critical.

[v0lt]

About OSD.
I suggest the writing "Image saved" instead of "Image saved to: C: \ ... \ image.png".
The path is visible in the status bar.
People (who have requested this feature) is not against it

[Underground78]

At the beginning he asked for the path if I remember correctly. It can be useful when using it fullscreen I guess.

[clsid]

Edit: nevermind, fixed already in 3322

[XinHong]

Revision 3323 - Directory Listing
Modified Tue Jul 5 06:05:53 2011 UTC (9 minutes, 5 seconds ago) by aleksoid

Ticket #1493, thanks to judelaw.

Thanks Judelaw & aleksoid :)

[Aleksoid1978]

Thanks Judelaw & aleksoid :)

Need more testing - if can, test and report here result. :)

[Gleb Egorych]

A question about the standalone MPEG splitter: with the splitter Zoom Player always loads subtitle renderer on MPG files which breaks DXVA acceleration. Is it planned to be fixed?

[Aleksoid1978]

A question about the standalone MPEG splitter: with the splitter Zoom Player always loads subtitle renderer on MPG files which breaks DXVA acceleration. Is it planned to be fixed?

What fixed ??? Other player all fine with MPEGSplitter.

[tetsuo55]

A question about the standalone MPEG splitter: with the splitter Zoom Player always loads subtitle renderer on MPG files which breaks DXVA acceleration. Is it planned to be fixed?I think the player is supposed to blacklist the subtitle renderer for these codecs. The splitter is correct in its behaviour by demuxing the subtitles.

[Masutin]

1) I'll explain my previous request about the Go To box. It is purely for convenience of finding a particular position. Without specifying the hour, you just enter e.g. 0:99 and get 1:39.
Functions to go to the start and end position would be useful too.
2) If this forum is not dedicated entirely to development, can I avoid exiting the Full Screen mode when switching to another window?

[Gleb Egorych]

What fixed ??? Other player all fine with MPEGSplitter.

I think the player is supposed to blacklist the subtitle renderer for these codecs. The splitter is correct in its behaviour by demuxing the subtitles.
LAV Splitter does not expose subtitle pin on mpg files, Haali doesn't either.

[Aleksoid1978]

LAV Splitter does not expose subtitle pin on mpg files, Haali doesn't either.

But - MPG files can have subtitle ... or not ???

[nevcairiel]

They can, MPG files can have VOB subtitles.
And if they do, LAV Splitter will export a subtitle pin, but only then.

[tetsuo55]

has anyone seen the new hotfix amd driver?

this is supposedly the changelog:

Quote:
Catalyst 11.6b Hotfix resolves the following issues:

- Blue screen erorrs (BSOD) seen when connecting HDMI and DisplayPort based displays
- Screen flickering when the system is idle
- Mouse pointer corruption found in previous AMD Catalyst releases
- Fixes for various OpenGL issues
- VSync bug fixed
- Improved image quality

http://support.amd.com/us/kbarticles/Pages/AMDCatalyst116bHotfix.aspx

[SamuriHL]

I had only seen 11.6a which applied only to some new cards, so, thanks for posting that.

[tetsuo55]

I seem to be seeing DXVA corruption with this new driver.

Can anyone confirm?

[TheElix]

Try 11.7 Early Look version http://developer.amd.com/tools/gDEBugger/Pages/default.aspx I have no issues with them.

[tetsuo55]

The DXVA problem is a false positive.

@TheElix, the driver i posted is newer than the one in your link

[TheElix]

@tetsuo55, Yet I don't see any DXVA corruption. HD6970 here.

[Aleksoid1978]

A question about the standalone MPEG splitter: with the splitter Zoom Player always loads subtitle renderer on MPG files which breaks DXVA acceleration. Is it planned to be fixed?

Try rev. 3343

[JanWillem32]

has anyone seen the new hotfix amd driver?

this is supposedly the changelog:

Quote:
Catalyst 11.6b Hotfix resolves the following issues:

- Blue screen erorrs (BSOD) seen when connecting HDMI and DisplayPort based displays
- Screen flickering when the system is idle
- Mouse pointer corruption found in previous AMD Catalyst releases
- Fixes for various OpenGL issues
- VSync bug fixed
- Improved image quality

http://support.amd.com/us/kbarticles/Pages/AMDCatalyst116bHotfix.aspxFor all users: as usual, don't forget to set up settings again in the Catalyst Control Center after installing. EVR will use all those filters that are enabled by default in the video settings panel in any video. Enabling automatic deinterlacing, pulldown detection and using settings on internet video is fine, but the other options should be set up manually by the user, according to taste. Note that all video enhancement filters will inhibit the renderer from displaying reference pictures correctly.
The filters are also costly in performance. EVR uses a lot more resources typically than VMR-9, but that's only because of those extra filters.

Possibly related to the new driver, I have a black screen in VMR-9 w. and VMR-9 r. in mixer mode. VMR-9 r. without mixer mode has horrible color clipping. Can anyone else verify that?

[TheElix]

VMR-9 r. without mixer mode has horrible color clipping. Can anyone else verify that?I dunno about VMR-9 r., but I noticed some nasty feature in CCC drivers. I set Limited RGB pixel format in CCC for my display, but when I choose different resolution (or frame rate) pixel format is reverted to Full RGB. In other words, I have to choose pixel format for each resolution I use, otherwise I get clipping.

[Gleb Egorych]

Try rev. 3343
Thanks, it works. But I reviewed your patch, the splitter still always adds subtitle pin on TS files, which leads to the same effect as with MPG format. What's the purpose to add fake subtitle pin?

[Aleksoid1978]

Thanks, it works. But I reviewed your patch, the splitter still always adds subtitle pin on TS files, which leads to the same effect as with MPG format. What's the purpose to add fake subtitle pin?

For MPEG-TS need add fake subtitle pin, because - subtitle stream can be found not at start of file, and if no fake stream add ... we can't play subtitle.

[ikarad]

I don't know if developpers works on sub renderer but this problem still exist with the last version of mpc-hc. It's a problem with blu-ray subs.
https://sourceforge.net/apps/trac/mpc-hc/ticket/48

[Eliminateur]

internal DXVA filters don't seem to be working as expected, i checked VC1, H264 and VC1 for DXVA yet i open any H264 file i have(MKV, CCV1 reported) and ffdshow loads.
for some mp4 files, then it does load the DXVA filter.
i'm using r3262
even if i disable ffdshow h264 decoding, then it fails to rende the pin.

I don't want to enable ffdshow decoding ad it won't be able to use internal mpc-hc subtitler.
thius is the mediainfo of one file for example

Video
ID : 1
Format : AVC
Format/Info : Advanced Video Codec
Format profile : High@L5.0
Format settings, CABAC : Yes
Format settings, ReFrames : 8 frames
Format settings, GOP : M=1, N=24
Codec ID : V_MPEG4/ISO/AVC
Duration : 1mn 32s
Width : 1 920 pixels
Height : 1 080 pixels
Display aspect ratio : 16:9
Frame rate : 23.976 fps
Color space : YUV
Chroma subsampling : 4:2:0
Bit depth : 8 bits
Scan type : Progressive
Writing library : x264 core 98 r1649 c54c47d
Encoding settings : cabac=1 / ref=8 / deblock=1:0:0 / analyse=0x3:0x113 / me=umh / subme=9 / psy=1 / psy_rd=1.00:0.00 / mixed_ref=1 / me_range=24 / chroma_me=1 / trellis=1 / 8x8dct=1 / cqm=0 / deadzone=21,11 / fast_pskip=0 / chroma_qp_offset=-2 / threads=6 / sliced_threads=0 / nr=0 / decimate=1 / interlaced=0 / constrained_intra=0 / bframes=3 / b_pyramid=2 / b_adapt=1 / b_bias=0 / direct=1 / weightb=1 / weightp=2 / keyint=250 / keyint_min=25 / scenecut=40 / intra_refresh=0 / rc_lookahead=40 / rc=crf / mbtree=1 / crf=16.0 / qcomp=0.60 / qpmin=10 / qpmax=31 / qpstep=4 / ip_ratio=1.40 / aq=1:1.00
Language : English

[Underground78]

What is your graphic card ? It might fail because the streams you are trying to play are not compatible with hardware decoding. In that case, the decoder will reject the connexion so that it fallbacks to software decoding.

[Eliminateur]

AMD 5800 cat 11.4
i've tried ffdshow with dxva and they decode via dxva ok

[Underground78]

AMD 5800 cat 11.4
i've tried ffdshow with dxva and they decode via dxva ok

Have you tried disabling compatibility checks in DXVA decoder control panel ?

[clsid]

Skipping the level check (which should be safe) might do the trick.

[Eliminateur]

all level check are already skipped(was never set in 1st place)

[clsid]

Disable custom mediatype option in Haali. MPC might not recognize CCV1.

[Eliminateur]

pinpon, clsid hit the nail!, disabled custom mediatype and mpc decoder sprang to life, many thanks (a guide or something should be posted for this kind of stuff)

[Thunderbolt8]

is there any chance that someone could update mpc-hc's DXVA stuff that it also supports interlaced content?
at least from a logical perspective it would make more sense if those formats are supported which actually need more cpu/rendering power.

[TheElix]

Is it somehow possible to make black bars in letterboxed video grey? It is to prevent burn-in on plasma displays.

[JanWillem32]

I can add a function to change the background color, but I don't have a clue how to add a color selection object to the options menu. For black letterboxes encoded in a video you can use a modified letterbox or pillarbox pixel shader.
https://sourceforge.net/apps/trac/mpc-hc/ticket/1334

[TheElix]

Thanks. It works, but not in all cases. Sometimes there're videos with varying aspect ratio (The Dark Knight, for example). So a constant letterbox won't work. I wonder if there's a way of letterbox/pillarbox autodetection.
Oh, and it won't work with madVR, for obvious reasons.

[G_M_C]

I think you can relax somewhat. I've owned a plasma since april 2007, and I have never seen image retention or burn in.

I use my plasma as my normal day-to-day TV-screen. I have watched many movies, both thru MPC-HT and regular BD, and i have never used 'gray letterboxes'. Never seen anything to worry about.

What you can see is (what i call) 'overexcited' phosphors where undertitels have been printed. But i personally dont count that as image retention, cause the effect is gone after whatching something else for 5 or 10 minutes.

[JanWillem32]

An autodetecting letterbox/pillarbox function would cause masking on every black screen, large portions of night scenes and parts of images with any dark objects at a border. Don't forget that a lossy encoded letterbox/pillarbox really isn't completely black at all. The function would need a minimum detection level above black that it may mask. A function like this will have to be integrated into the rendering engine, as the multi-pass detection and execution parts can't be efficiently done with only independent pixel shaders.

[TheElix]

I'm not talking about this level of autodetection. Let it search automatically for 100% black portions of fixed aspect ratio rectangles (seen in 2.35:1 movies, 4:3 movies on 16:9 screen etc.). Screw those lossy encoded letterboxes/pillarboxes.

G_M_C
Some plasmas have in-built image renention protection. Pioneers have image orbiter. Samsung can make letterboxes/pillarboxes grey. And some panels are more resistant to IR than others. You are lucky.

[clsid]

Use ffdshow to resize to screen resolution. It has an option to specify the color of the padded parts of the video.

[TheElix]

Thanks for the tip... It works but at cost of greatly increased CPU consumption. And it's not contributing to the PQ either, I think. But look at me, whining by myself all day long. I guess I'll have to live with what options available ATM.

[Underground78]

I can add a function to change the background color, but I don't have a clue how to add a color selection object to the options menu. For black letterboxes encoded in a video you can use a modified letterbox or pillarbox pixel shader.
https://sourceforge.net/apps/trac/mpc-hc/ticket/1334

If you tell me what I need to change (I can probably figure this out myself but I am lazy when I can :rolleyes:) I can add the option (if we can find somewhere where it fits).

[TheElix]

I can add the option (if we can find somewhere where it fits).It'd fit nicely in Tweaks or Miscellaneous tabs)

[cengizhan]

1.5.2.3357 : Alt+1,2,3, doesnt work.

Update: closed and opened a few times and then after that it worked.

[Underground78]

It'd fit nicely in Tweaks or Miscellaneous tabs)

Yes, probably.

1.5.2.3357 : Alt+1,2,3, doesnt work.

It works for me.

[Momber]

Replace jump forward and backward buttons in Windows 7 taskbar by next and previous buttons as they are probably used more often
Are you completely out of your mind now?
The jump forward and backward buttons need to be restored immediately.

[Underground78]

Are you completely out of your mind now?
The jump forward and backward buttons need to be restored immediately.

Why ? All people I have talked with thought that being able to go to the next/previous item was a lot more useful.

[Aleksoid1978]

Are you completely out of your mind now?
The jump forward and backward buttons need to be restored immediately.

I agree - need restored jump forward and backward

[Underground78]

We have discussed that again on IRC today and we won't change that. It seems that a majority of people think it makes more sense to be able to jump to next/previous item.

[JarrettH]

I get this corruption on menus after updating to 3391 from 3262 playing back a DVD

http://farm7.static.flickr.com/6020/5938148633_0e04fbbfc7_o.jpg

Same thing happened when I tried to play Zodiac...

http://farm7.static.flickr.com/6007/5938723742_249687b756_o.jpg

What is "Line 21 Decoder 2" under Filters?

[Aleksoid1978]

I get this corruption on menus after updating to 3391 from 3262 playing back a DVD

http://farm7.static.flickr.com/6020/5938148633_0e04fbbfc7_o.jpg

Same thing happened when I tried to play Zodiac...

http://farm7.static.flickr.com/6007/5938723742_249687b756_o.jpg

What is "Line 21 Decoder 2" under Filters?

Temporary don't use Microsoft DVT-DVD Video Decoder, DVD playback corrupted only with it's decoder.

[JarrettH]

It doesn't happen during the movie so far. Thanks for the info :cool:

[Aleksoid1978]

I get this corruption on menus after updating to 3391 from 3262 playing back a DVD

http://farm7.static.flickr.com/6020/5938148633_0e04fbbfc7_o.jpg

Same thing happened when I tried to play Zodiac...

http://farm7.static.flickr.com/6007/5938723742_249687b756_o.jpg

What is "Line 21 Decoder 2" under Filters?

Try this build - http://aleksoid.tosei.ru/Test/mpc-hc.exe_DVD_test.7z

On DVD with "Microsoft DVT-DVD Video Decoder".

[fastplayer]

Thanks a lot for rev3396 (http://mpc-hc.svn.sourceforge.net/viewvc/mpc-hc?view=revision&revision=3396), Underground78!
Now I know for sure whether I have or have not changed a setting. :thanks:

[Underground78]

Thanks a lot for rev3396 (http://mpc-hc.svn.sourceforge.net/viewvc/mpc-hc?view=revision&revision=3396), Underground78!
Now I know for sure whether I have or have not changed a setting. :thanks:

You're welcome, I've just fixed the case where filters are selected or unselected using the context menu.

[LigH]

@ Aleksoid1978:

Please avoid fullquotes without purpose.

Either reduce quotes to the important part, or do not quote at all ... but try to avoid waste of space due to repeating everything.
__

@ JarrettH:

"Line 21" might contain "Closed Captions" in an NTSC signal, slightly outside the visible video. A quite outdated system of subtitling/teletext. See: http://en.wikipedia.org/wiki/Closed_captioning

[Mercury_22]

@alexins
Can you please update the BE Mod patch (the GUI part not the "test-drive mod") against latest SVN cause I'm unable to apply the old patch (mod BE patch 2976) due to too many conflicts :o

[sawg]

rev3357 can play DVD and MPEG2 with DXVA in EVR.
rev3362 can't play.
rev3394 has some error fixed, but DVD still can't play with DXVA(Internal Filters: MPEG2-DXVA/NVIDIA Video Decoder) in EVR.

http://www.rehost.org/images/1310793842.jpg

[oddball]

This is bugging me.

Forced subtitle flag is ignored if you set subtitles to off during playback, close player then restart or play another file. I would think the logical option would be for forced subs to ALWAYS kick in when you play and the option to turn them off is then yours. At present if you play a file with forced subs and forgot you turned it off for the last file then you end up having to pause, go back and enable them in order to see forced subs.

Unless I am missing some option where forced subs are always shown.

If not then can we have an option to always show forced subs on playback even if you set subtitles off during the last file played?

EDIT: I just checked a file where forced subs are embedded but it does not have the secondary 'player must always show forced subs' flag as no so I am guessing I will need to remux. However I still think the player should have this as an option even if the file has the no flag set.

EDIT2: I just tried remuxing a file with the forced subs track flagged as 'Yes' and MPC-HC still ignores it if the previous movie had subs disabled. I think this is either a bug or a sorely missed option.

[dukestravels07]

I have a weird problem.

Some of my mkv files are locking up during playback.

I usually test every mkv file by playing it and jumping forward every 30 mins until i see the end credits. This worked fine with the films I am now having issues with.

When I play the file it would freeze up about 45 mins in then restart from beginning.

Restarting the film resulted in buffering, which stopped at 5% and then just locked up.

I am using dxva on an hd3600 ati agp. Not the best, but it works for my needs.

Settings in mpc-hc are:

vmr 9 (renderless)

Any ideas?
Why does it buffer a file on a local hard drive?

[v0lt]

This worked fine with the films I am now having issues with.
Before it was normal? Which version has no problems?

[dukestravels07]

Before it was normal? Which version has no problems?

Sorry, I did not phrase that well. The "normal" part is my playing the file and jumping forward in 30min segments until I know it can get to the end credits...thus telling me the file works ok.

[betaking]

last mpc-hc svn have seek bug! if i change after playbak-every time-play next in the folder play some files seek to file end time mpc-hc is crash file name is [01F].mkv [02F].mkv or CD_A.avi and CD_B.avi !
i made sceen recoder! and i use Internal splitter and Internal video decoder!
http://hotfile.com/dl/124365080/e2b3458/Video_2011-07-18_124914.wmv.html
Faulting application mpc-hc.exe, version 1.5.2.3415, time stamp 0x4e22c6a4, faulting module mpc-hc.exe, version 1.5.2.3415, time stamp 0x4e22c6a4, exception code 0xc0000005, fault offset 0x0005c6f0, process id 0x1564, application start time 0x01cc450630f48fa0.

[Vern Dias]

Would it be possible to set up hot keys for custom AR's (image sizes) that I have defined in the Pan and Scan context menu? (Edit, new, define sizes and offsets?)

Or is there some other way to define custom AR's that I am missing ala Zoom Player? I need at least 10 preset image sizes, accessible from hot keys (but not the numeric keypad keys).

Thanks, Vern

[rahzel]

Can someone please explain dithering to me?
As I understand it, dithering is enabled if:
1) you enable "10-bit RGB output"
2) you enable "force 10-bit RGB input"
3) you enable FFPP/HFPP

Would 1 and 2 give the same result, just that forcing 10-bit input would force MPC to always output 10-bit RGB? If so, do you need to have both of these checked?
In order of best to worst, does it go 10-bit -> HFPP (16-bit FP?) -> FFPP (32-bit FP?) If you just want to use FFPP, do you just enable that or do you need 10-bit RGB enabled?

Lastly, what confuses me in the renderer wiki, is that it says that dithering can be used with these modes. I thought dithering was already being used with these modes?

Pardon my ignorance, but please help me understand these settings.

[fastplayer]

Can someone change the hand cursor to a normal cursor when hovering over combo boxes in the options dialog?

[v0lt]

Can someone change the hand cursor to a normal cursor when hovering over combo boxes in the options dialog?
This is a distinctive feature of the player. :D

[JanWillem32]

@rahzel: Dithering options are to become selectable by the user. The "force 10-bit RGB input" option will be removed. (For reference, see the builds in my thread.)
FFPP and HFPP only set the precision for internal processing surfaces for the rendering chain. The "10-bit RGB output" option is intended to enable 10-bit back buffers, and in exclusive mode 10-bit display output, too. The changes can be seen in the stats screen. Changing these options while rendering is active works in my builds, not in the reference build.

[rsd78]

Hi, I thought I'd ask again, but any chance the issue with the macroblocking and intel iGPUs/dxva will be fixed? Thanks

[rahzel]

@rahzel: Dithering options are to become selectable by the user. The "force 10-bit RGB input" option will be removed. (For reference, see the builds in my thread.)
FFPP and HFPP only set the precision for internal processing surfaces for the rendering chain. The "10-bit RGB output" option is intended to enable 10-bit back buffers, and in exclusive mode 10-bit display output, too. The changes can be seen in the stats screen. Changing these options while rendering is active works in my builds, not in the reference build.
So you mean currently, dithering is enabled with these modes, but later is going to be selectable by user in future builds?

Sorry, can you try to explain the 10-bit mode and FFPP/HFPP in layman's terms (a little easier to understand). I don't know a whole lot on this (but I'm sort of starting to understand! :)).

Thanks.

[khagaroth]

Updated Czech translation - Ticket 1505 (https://sourceforge.net/apps/trac/mpc-hc/ticket/1505). But can't reopen the ticket, I'm getting "Warning: No permission to change ticket fields. " (related Trac bug - http://trac.edgewall.org/ticket/8884), can someone take a look at that.

[JanWillem32]

So you mean currently, dithering is enabled with these modes, but later is going to be selectable by user in future builds?Indeed, with several modes.Sorry, can you try to explain the 10-bit mode and FFPP/HFPP in layman's terms (a little easier to understand). I don't know a whole lot on this (but I'm sort of starting to understand! :)).

Thanks.In EVR CP and VMR-9 r. there are 4 surface formats, the one used for the mixer and renderer stages depends on the renderer setting.
m_SurfaceType = m_bFullFloatingPointProcessing? D3DFMT_A32B32G32R32F : m_bHalfFloatingPointProcessing? D3DFMT_A16B16G16R16F : m_bHighColorResolution? D3DFMT_A2R10G10B10 : D3DFMT_X8R8G8B8; (actual code, but should not be too hard to read)
With every step of higher surface precision, the precision losses for every step of processing of the decoded input to display output are less than with X8R8G8B8.
At the end of rendering, the output is dithered or rounded to A2R10G10B10 or X8R8G8B8, depending on the display format and the selected option for dithering.
The display format can be A2R10G10B10 when both the D3D exclusive mode and 10-bit output options are activated, else it's always X8R8G8B8.
All of this can be seen in the stats screen.
As a side-note, A32B32G32R32F, A16B16G16R16F and A2R10G10B10 disable the forced bilinear filter on chroma in the video mixer with ATi and some Intel GPUs. Chroma up-sampling shaders are required if you don't want the chroma to be scaled by nearest neighbor in these cases.

[rahzel]

okay, so as of now, the best setup would be to set it to 10-bit RGB output with FFPP, and if you want bilinear chroma upsampling instead of nearest neighbor, you have to enable the yv12 chroma upsampling shader?

if i'm having choppy playback issues (which is probably a result of lack of processing power by the gpu?) should I start doing this until it plays smooth:
1. disable yv12 chroma upsampling
2. disable FFPP
3. disable 10-bit RGB output

Or does 10-bit RGB output have no affect on the GPU processing load and is more of a compatibility/capability of your display thing?

I have 4 computers (including my notebook) and only my HTPC that has the fastest GPU of the 4 can handle FFPP -- the others stutter badly. All of my computers are AMD based (780G/HD3200, HD6250M, 880G/HD4250 and HD 5570).

[JanWillem32]

The yv12 chroma upsampling shader is rather bad, it uses some variant of the NTSC analog matrix to perform RGB - Y'CbCr conversion, and then blurs with a box blur. I've written two with proper SD and HD matrices, one bilinear and one spline5. (See my pixel shader pack, I can also write some more things if requested. The spline5 type is split for the integer 8- and 10-bit surfaces, and the floating point 16- and 32-bit surfaces.)
The 10-bit option uses the same amount of memory and processing as the default 8-bit setting. The only valid reason I've not set it as a default is because of the chroma conversion. The 16- and 32-bit options take the same amount of processing, but double and quadruple the memory load respectively. The 10-bit display output also comes relatively for free, it just requires a compatible video card. Many video cards expose 10-bit display formats as valid for output, but actually setting the format makes the driver crash the player (not too serious, though). I don't expect you'll have much trouble with the actual 10-bit display output option, as even my previous X1800 XT could handle that.
For reducing memory load, you can switch FFPP for HFPP, or even to 10-bit. All of your GPUs can at least handle the bilinear chroma up-sampling shader without issues. For further reducing the processing load, you can also disable the video filters in the video section in CCC, don't forget to also disable the forced AA and AF filtering settings in the 3D section (it's called "application preference"). For adding some customized filtering stages, from mere video jokes to filters for daily use, I've written plenty of other pixel shaders, too. Most are pretty light, but there are some really heavy ones as well.

[rahzel]

thanks a lot.

[Hera]

MPC holds on to the file even after CTRL-C.
That is, I have to close MPC completely to edit the file or eject the thumbdrive / USB HD...

[Underground78]

Which filters do you use ?

[sawg]

http://www.rehost.org/images/1311182402.jpg
rev3357 can play DVD and MPEG2 with DXVA in EVR CP.
rev3362 can't play.
rev3394 has some error fixed, but DVD still can't play with DXVA(Internal Filters: MPEG2-DXVA/NVIDIA Video Decoder) in EVR CP.
=>
11.07.21
rev3432 and rev3436 can't play DVD and MPEG2 with DXVA in EVR CP.

[v0lt]

@sawg
I do not have this problem. But I do not use NVIDIA Video Decoder.

Why do you have two NVIDIA Video Decoder and two renderers in the filters?
Show me a graph that you have get. (GraphStudio (http://blog.monogram.sk/janos/tools/monogram-graphstudio/)->File->Conect to Remote Graph...)

[nuhkka]

how come on these newer builds when you seek a video, it won't take you to where you clicked but somewhere close to it.. like not "frame-accurate" accurate seeking or whatever it's called that iwas able to do in earlier versions (the one posted on sf main site)

is there a way to fix it?

[namaiki]

Check: View-> Options-> Tweaks.

[Hera]

Which filters do you use ?

Vanilla + Haali Splitter

[Eliminateur]

As a side-note, A32B32G32R32F, A16B16G16R16F and A2R10G10B10 disable the forced bilinear filter on chroma in the video mixer with ATi and some Intel GPUs. Chroma up-sampling shaders are required if you don't want the chroma to be scaled by nearest neighbor in these cases.

so if i understand correctly:
independently of using D3DEX or 2D surface if we use FFPP+10bit output the chroma upsampler is nearest neighbour?
this only happens when using 10 bit processing?

shouldn't d3dex solve this, seeing as it enables undithered A2R10G10B10 display output?

also, i found a bug in D3DEX and chapter/file jump:
with/without gui support, the keyboard pageup/pagedown keys work ok but with my mouse "up/down" keys it fails, if i press the up key several times maybe it will go back one chapter and pause the video at random.
Keys are marked as "X1 up" "X2 down"
i'm not using any mouse software
rev 1.5.2.3423 x64

[rahzel]

Indeed, with several modes.In EVR CP and VMR-9 r. there are 4 surface formats, the one used for the mixer and renderer stages depends on the renderer setting.
m_SurfaceType = m_bFullFloatingPointProcessing? D3DFMT_A32B32G32R32F : m_bHalfFloatingPointProcessing? D3DFMT_A16B16G16R16F : m_bHighColorResolution? D3DFMT_A2R10G10B10 : D3DFMT_X8R8G8B8; (actual code, but should not be too hard to read)
With every step of higher surface precision, the precision losses for every step of processing of the decoded input to display output are less than with X8R8G8B8.
At the end of rendering, the output is dithered or rounded to A2R10G10B10 or X8R8G8B8, depending on the display format and the selected option for dithering.
The display format can be A2R10G10B10 when both the D3D exclusive mode and 10-bit output options are activated, else it's always X8R8G8B8.
All of this can be seen in the stats screen.
As a side-note, A32B32G32R32F, A16B16G16R16F and A2R10G10B10 disable the forced bilinear filter on chroma in the video mixer with ATi and some Intel GPUs. Chroma up-sampling shaders are required if you don't want the chroma to be scaled by nearest neighbor in these cases.
I have another question. On the current build, if I don't have D3DEX checked, I understand that I'm not truly getting 10-bit output, but from my eyes, looking at a grayscale ramp, the gradients are definitely smoother. Is it the dithering that's enabled with 10-bit output that's smoothening out the gradients?
I don't use D3DEX because I can't seem to exit fullscreen... how are you supposed to do so?

[sawg]

@sawg
I do not have this problem. But I do not use NVIDIA Video Decoder.

Why do you have two NVIDIA Video Decoder and two renderers in the filters?
Show me a graph that you have get. (GraphStudio (http://blog.monogram.sk/janos/tools/monogram-graphstudio/)->File->Conect to Remote Graph...)
Please forgive me if I make any errors; my English is a little weak.

rev3357 and rev3432, I can't use Internal Filters: MPEG2-DXVA too.

grf file:
http://www.mediafire.com/?dke4boknfl7bzwt
http://www.megaupload.com/?d=VMQ4BJOP

mpc3357_mpeg2ts.grf
http://www.rehost.org/images/1311219382.jpg
mpc3357_dvd.grf
http://www.rehost.org/images/1311219381.jpg
mpc3436_mpeg2ts.grf
http://www.rehost.org/images/1311219384.jpg
mpc3436_dvd.grf
http://www.rehost.org/images/1311219383.jpg

[Underground78]

Check: View-> Options-> Tweaks.

More precisely "Fast seek (on keyframe)", note that this option has been disabled by default in latest build.

Vanilla + Haali Splitter

Have you tried using the internal splitter and see what happens ?

[Aleksoid1978]

Please forgive me if I make any errors; my English is a little weak.


fix in rev.3442 (https://sourceforge.net/apps/trac/mpc-hc/changeset/3442/trunk)

[e-t172]

Please forgive me if I make any errors; my English is a little weak.fix in rev.3442 (https://sourceforge.net/apps/trac/mpc-hc/changeset/3442/trunk)

You made his english better? :p

[JanWillem32]

so if i understand correctly:
independently of using D3DEX or 2D surface if we use FFPP+10bit output the chroma upsampler is nearest neighbour?
this only happens when using 10 bit processing?

shouldn't d3dex solve this, seeing as it enables undithered A2R10G10B10 display output?The only parameter required for the ATi driver to skip the bilinear filter is that the surface format for the mixer is anything but X8R8G8B8. In this case the 10-, 16- and 32-bit formats will cause the driver to skip the bilinear filter. (Not that I particularly care, I prefer to set my own set of chroma up-sampling methods, anyway. Outperforming the bilinear kernel isn't hard to do: on 4:2:0 it only blurs every corner pixel of the 2×2 block by adding up the pixels next to it. Its pixel weights are 1/16 diagonal + 3/16 horizontal + 3/16 vertical + 9/16 of the nearest neighbor.)I have another question. On the current build, if I don't have D3DEX checked, I understand that I'm not truly getting 10-bit output, but from my eyes, looking at a grayscale ramp, the gradients are definitely smoother. Is it the dithering that's enabled with 10-bit output that's smoothening out the gradients?
I don't use D3DEX because I can't seem to exit fullscreen... how are you supposed to do so?In my builds you can set dithering to level 0, to disable it. That will cause the input values to round to the display format. A completely pure gradient can be made by enabling the "color gradients" shader as "post-resize" (does need a dummy video running at a normal video rate in the background). A nice color set to produce banding with red, green and blue is in my case:
#define ColorTopLeft float4(0, .015625, 0, 1)
#define ColorTopRight float4(0, 0, .015625, 1)
#define ColorBottomLeft float4(.015625, 0, 0, 1)
#define ColorBottomRight float4(0, 0, 0, 1)
It should show 8-bit banding very nicely, but you may set any value in the interval [0, 1] for the target float4(red, green, blue, 1).
You can simply switch the options for dithering during that test to compare to compare the amount of banding against the ditherer's noise. There's also the "Dithering Test" option that will show what the dither filter is doing.
The exclusive mode can be closed by default by Ctrl+C, or exited by Alt+X (Alt+F4 like in all regular windows applications works, too). The key commands can be edited in the "View", "Options", "Keys" menu.

For indicating what the renderer is doing, see the stats screen (default key Ctrl+J) during playback. It will indicate the formats used by the player, timing statistics, and more information.

[v0lt]

MPC-HomeCinema 1.5.2.3445 (x86) (http://www.mediafire.com/?xfgh33iws56bh)
Fix : DVD playback with Nvidia PureVideo Decoder.
corrected time precision (for display).
fixed Russian translation

[sawg]

MPC-HomeCinema 1.5.2.3445 (x86) (http://www.mediafire.com/?xfgh33iws56bh)
Fix : DVD playback with Nvidia PureVideo Decoder.
corrected time precision (for display).
fixed Russian translation
Thanks for your help.
rev3445 fix MPEG2 files and DVD with Nvidia PureVideo Decoder.

But Internal Filters: MPEG2 Video (DXVA) still can't use.
Please tell me what I should do.

[v0lt]

But Internal Filters: MPEG2 Video (DXVA) still can't use.
Please tell me what I should do.
It always has been. Use a built-in software decoder.

[clsid]

So what exactly is the purpose of the recent MPEG2 changes?

[Tessier]

I have a problem with loading external subtitles. It's ok when the movie is stored on a local HDD.
However, I also have a router with an USB HDD connected to it and shared using SMB (the router uses OpenWRT). Whenever I open a movie from this remote HDD, no subtitle is loaded. Other players (e.g. Potplayer) loads the subtitle correctly and all files are correctly accessible from windows.
I have found a similar problem (old...), but no solution:
http://www.cccp-project.net/forums/index.php?topic=2600.15

[Underground78]

Is this the same issue: https://sourceforge.net/apps/trac/mpc-hc/ticket/610 ?

[fairchild]

Could someone be so kind as to give me an answer to these question I've been having. I've tried searching the net but haven't found an answer. When running DXVA (through an external ffdshow's DXVA decoder which uses NV12) and applying the YV12 Chroma Upsampling shader, do I need to enable this shader as a pre-resize shader or post-resize shader or both?

I read that this shader is still needed when running an external dxva decoder on ATI hardware as it still doesn't do proper chroma upsampling. If I run the internal MPC-HC DXVA decoder then the YV12 chroma upsampling is handled correctly thus not needing to use that shader?

Thanks!

[Silent Rain]

Guys, how about the fix ticket #1533 (https://sourceforge.net/apps/trac/mpc-hc/ticket/1533)?
Is it possible?

[Wile-E-Coyote]

So what exactly is the purpose of the recent MPEG2 changes?
I'd like to know as well.

[sawg]

So what exactly is the purpose of the recent MPEG2 changes?
rev3362
Add : Support DVD Menu marker with DXVA in EVR Custom and EVR Sync renderer, big thanks to judelaw.
rev3363
Change : Improved cursor behavior in DVD mode.
rev3369
Change : improve OSD message at File/DVD open.
Fix : Zoom at open DVD + DXVA decoder.
rev3378
Add : Support DVD Menu marker with DXVA in EVR.
rev3380
Fix the "Prev Angle (DVD)" hotkey and add an OSD message when changing the angle using the hotkeys or the "Navigate" menu.
Patch partially by judelaw.
This commit fixes ticket #1509.
rev3394
Change : improve normal Zoom at open DVD + DXVA decoder.
Fix : Corrupt DVD playback with Microsoft DVT-DVD Video decoder after rev.3362 - Disable "Line 21 Decoder 2".
rev3407
Disable DVD subtitle mixing in EVR-CP and EVR-Sync for Microsoft DTV-DVD Video Decoder
rev3436
Fix : Disable multiple output pin from internal MPEG2 software decoder, it's broken normal MPG playback.
Add : Missing class member in Mpeg2DecFilter, it't improve the debugging.
Change : Remove submenu in "Navigate" if no multiple stream present.
rev3442
Fix : DVD playback with Nvidia PureVideo Decoder.


DVD?

[Underground78]

Guys, how about the fix ticket #1533 (https://sourceforge.net/apps/trac/mpc-hc/ticket/1533)?
Is it possible?

I never saw this behavior but I will try again tonight.

[clsid]

rev3362
Add : Support DVD Menu marker with DXVA in EVR Custom and EVR Sync renderer, big thanks to judelaw.
rev3363
Change : Improved cursor behavior in DVD mode.
rev3369
Change : improve OSD message at File/DVD open.
Fix : Zoom at open DVD + DXVA decoder.
rev3378
Add : Support DVD Menu marker with DXVA in EVR.
rev3380
Fix the "Prev Angle (DVD)" hotkey and add an OSD message when changing the angle using the hotkeys or the "Navigate" menu.
Patch partially by judelaw.
This commit fixes ticket #1509.
rev3394
Change : improve normal Zoom at open DVD + DXVA decoder.
Fix : Corrupt DVD playback with Microsoft DVT-DVD Video decoder after rev.3362 - Disable "Line 21 Decoder 2".
rev3407
Disable DVD subtitle mixing in EVR-CP and EVR-Sync for Microsoft DTV-DVD Video Decoder
rev3436
Fix : Disable multiple output pin from internal MPEG2 software decoder, it's broken normal MPG playback.
Add : Missing class member in Mpeg2DecFilter, it't improve the debugging.
Change : Remove submenu in "Navigate" if no multiple stream present.
rev3442
Fix : DVD playback with Nvidia PureVideo Decoder.That is the SVN changelog. I read that daily. It does not explain things in detail. It seems that DVD menu effects were fixed when using DXVA, but that broke many things for which workarounds have been implemented. For example is blocking Line21 correct when internal subs are disabled? Does DVD playback still work properly with the non-EVR renderers?

[Silent Rain]

I never saw this behavior but I will try again tonight.

Great!

Specification of my PC:
CPU: Intel Core 2 Quad Q6600 @ 2.40 GHz
GPU: NVIDIA GeForce 8600 GT
RAM: 4 GB
OS: Windows 7 x64 (SP1)

[Underground78]

I have tested the samples you linked in the ticket and I cannot reproduce your problem using the internal decoder and splitter (I also tested LAVSplitter and it works too). Have you tried the latest nightly (http://goo.gl/8NqfJ) ?

[Silent Rain]

I have tested the samples you linked in the ticket and I cannot reproduce your problem using the internal decoder and splitter (I also tested LAVSplitter and it works too). Have you tried the latest nightly (http://goo.gl/8NqfJ) ?

Yes, I've tried.
On Splash PRO, PotPlayer, The KMPlayer, SPlayer and other video players all works like lightning! :eek:
Problems with rewinding only in MPC-HC :mad:

[Underground78]

Maybe it's because I am not using Windows 7. I will ask if v0lt and Aleksoid can reproduce.

[JanWillem32]

I just tested it. The trunk build with default codecs and splitters enabled did freeze twice on me (after a whole lot of tries). It basically pauses, until I jumped to some other part again. I couldn't get it to freeze while testing VMR-9 r., with LAVSplitter, or with modified versions of VMR-9 r. or EVR CP. (I did test those a bit less than during the first test run, though.)

[Underground78]

I use VMR-9 so it might explain why I can't reproduce.

[Silent Rain]

Underground78, bring your PC configuration, please.
And show me your stats while jumping on the timeline (Ctrl + J). With the default settings (with internal decoders and splitters).

[Underground78]

Underground78, bring your PC configuration, please.
And show me your stats while jumping on the timeline (Ctrl + J). With the default settings (with internal decoders and splitters).

CPU: AMD Athlon X2 6000+ @ 3.00 GHz
GPU: ATI HD2600XT
RAM: 2 GB
OS: Windows XP (SP3)

I can't show you the stats as the renderer doesn't support it.

[6205]

1./ I have updated to new stable build 1.5.2.3456 and i am missing shaders. Specially "Sharpen Complex" shader, which is very usefull for some anime in blurry, low resolution. How to get it back?


2./ I see that new icons are uglier than before. Why is that? You could at least fix them, because i see only 4-bit version of icons, which are very primitive and ugly.

[Underground78]

About 1/, use "Select pre/post-resize Pixel Shaders". About 2/ have you tried doing the association process again ?

[6205]

About 2/ have you tried doing the association process again ?

Yes i have tried that many times, including delete icon cachce in Windows 7. Even if my desktop is in 32 colors, small icons in 16x16 size are displayed in their ugliest 4-bit version.

[Ayukawa]

I am using MPC-HC 64-bit on a laptop with switchable graphics. Win 7 64-bit. Ati HD 6770M and Intel HD3000 chipsets. Ati driver 8.830.6.1000 dated 3/15/11. All came with the laptop. Having an issue with DXVA. Have tried cleared settings.

With MPC-HC 1.5.0.2827: h264 will play with DXVA on the Ati chipset with excellent video quality. VC1 and mpgs also play with DXVA, but there are some graphics artifacts with mpgs.

With the intel chipset, I get horrible marcoblocking, making it unusable. Did not try VC1 or mpgs because I mostly play h264 videos.

With MPC-HC 1.5.2.3456: h264 refuse to play on the Ati chipset. It says "Opening" then "Stop". Does not crash. The length of the video is shown in the bar, but moving the slider to any other position will not result in video play. VC1 and mpg play fine with DXVA. In fact, the mpg now play without the previous video artifacts.

With the intel chipset, the h264 macroblocking is gone, but I still get video artifacts, mainly in the form fuzziness around hardcoded subtitles and logos. The video is better if not using DXVA. 1.5.0's video quality is the same as the nondxva version. VC1 defaults to FFmpeg decoding. Mpgs play fine with DXVA.

I briefly tried 1.5.1 from sourceforge, and the behavior is the same as 1.5.2.

I would like to use 1.5.2 and DXVA with h264 videos, but am not sure what the problems is. I'm sorry if this is a known issue. Please direct me in the right direction if it is.

[6205]

About 1/, use "Select pre/post-resize Pixel Shaders

Okay thanks.
It seems like Ctrl+P makes picture noticeable sharper.

Ctrl+Alt+P seems to do nothing...

[Underground78]

Ctrl+Alt+P seems to do nothing...

I think it's normal if you only select the shader for pre-resize.

About the icons I don't really know but I don't think I see this behavior (I'm still using XP).

[6205]

About the icons I don't really know but I don't think I see this behavior (I'm still using XP).

OK... meantime i will not use default icon cache until it's resolved.

Here is picure of 4-bit versus 32-bit icons.
http://forum.doom9.org/showpost.php?p=1498967&postcount=17239

I asume than on 32-bit desktop i should get 32-bit version, but not in this case...

[Underground78]

For me the icons look like that: http://i.imgur.com/mJWvA.png, seem to be the 16-bit version.

[JaLooNz]

Sorry if this has been asked before, before the latest stable version it is possible to use the video frame to drag the application around. However, the latest stable seems to have disabled the drag, any events simply pause/play the video. Is there anyway to get back the previous behavior? :thanks:

[Underground78]

Sorry if this has been asked before, before the latest stable version it is possible to use the video frame to drag the application around. However, the latest stable seems to have disabled the drag, any events simply pause/play the video. Is there anyway to get back the previous behavior? :thanks:

It's a known problem, read ticket #1427 (https://sourceforge.net/apps/trac/mpc-hc/ticket/1427) for more information and more specifically this comment (https://sourceforge.net/apps/trac/mpc-hc/ticket/1427#comment:20) for a possible workaround.

[Toddler Naruto]

Hi, question, is it possible to use the association icons from an older build with a newer one?

[JanWillem32]

Sure, as long as the internal numbers for the icon list hasn't changed (very rarely happens). Even if that happens, nothing will crash because of it. The icon file is only read by the Windows Explorer.

[Toddler Naruto]

Sure, as long as the internal numbers for the icon list hasn't changed (very rarely happens). Even if that happens, nothing will crash because of it. The icon file is only read by the Windows Explorer.

Thanks, that's good to know, hope it's as simple as installing the older build on a different hard drive, deleting the icons for the new build, and then moving the icons from the old build folder to the new build folder.

[JanWillem32]

You only have to unpack and copy-replace the newer mpciconlib.dll, no extra installation required (except maybe re-association from the "Options", "Formats" menu, but that's all).

[Toddler Naruto]

You only have to unpack and copy-replace the newer mpciconlib.dll, no extra installation required (except maybe re-association from the "Options", "Formats" menu, but that's all).

Thanks alot for the help :), what you said worked for me :cool:.

[6205]

Thanks alot for the help :), what you said worked for me :cool:.

I do NOT recommend to do that. Icons will not fit to correct file type. Many things have changed since revision 3088 - last revision with old icons. Since then MPC-HC supports many new formats and new icons has been added to icon library, therefore old mpciconlib.dll file will not work properly. Also ordering is different.

For some file types will be icons missins, for some will MPC-HC display wrong icon. For example(when you see linked image) for MKV file will MPC-HC show D2V icon (no. 22) from old icon library. http://oi54.tinypic.com/ly91j.jpg

Anyway, if you wanna keep at least some old icons, i have deleted entire mpciconlib.dll and made afterward re-association process. This way i have all icons the same for all file types which are now taken from mpc-hc.exe file directly.

[Mercury_22]

@ Mercury_22 and Underground78

I agree that the x64 needs its own icon, for now the existing icon with x64 added on top should do (We're about to get shiny new icons)

I also think the file association should have some hint to distuinguish between 32 and 64 bit versions (this requires code changes)

The BE mod skinability options where still in an alpha or pre-beta state last i heard, and that would be the reason why its not yet included. Ofcourse i want to have them committed as soon as possible.

Any news about the x64 exe's new icon ?
I've "made" :) as you propose a new icon with the "x64 edition" for the x64 version that we can use until something better comes up
icon_x64.ico & MPC_HomeCinema_1.5.3.3482_x64 with the new icon (http://www.multiupload.com/SYYW8WPEOR)
http://i51.tinypic.com/dlrp1j.png http://i54.tinypic.com/qz55vm.png

P.S. I've included a new icon for the x86 version too

Also the GUI part of the BE mod it's working without any problem for quite some time now

[v0lt]

@Mercury_22
A new icon is bad. Nothing to see.

[Mercury_22]

@Mercury_22
A new icon is bad. Nothing to see.
Why a new icon for the x64 version "is bad"?
As you can see many other people (everybody else) seems to think that "is good" cause as I've said many, many, times we need to easily distinguish between the two version (file format association ) on systems with both x86 and x64 version

[v0lt]

Why a new icon for the x64 version "is bad"?
А small icon (16x16) like a heap of pixels.
Middle icon (48x48) looks bad too.

[Mercury_22]

А small icon (16x16) like a heap of pixels.
Middle icon (48x48) looks bad too.

OH you mean the icon is bad ? :) not the idea for a new icon for x64 MPC-HC ?
If that is what you mean then like I've said this is so we can add the code for a separate icon_x64.ico "until something better comes up "
Fell free to come up with something better but this was the closest icon to the current one (so the user still recognize the well known MPC-HC "brand" :) ) but different, that I could come up with

[6205]

Why a new icon for the x64 version "is bad"?
As you can see many other people (everybody else) seems to think that "is good" cause as I've said many, many, times we need to easily distinguish between the two version (file format association ) on systems with both x86 and x64 version

Uhm, no. This is nonsense. A icon represents a branding. Give me a example of different branding of 32 vs 64bit application somewhere else. Both version should use the same icons. End of discussion. You may have your own opinion, but it's wrong :)

Ability to distinguish between the two versions on systems with both versions should be done differently. As an example should MPC-HC developers take Microsoft approach. Has 64 bit Internet Explorer or Windows Media Player different icons?

[Mercury_22]

Uhm, no. This is nonsense. A icon represents a branding. Give me a example of different branding of 32 vs 64bit application somewhere else. Both version should use the same icons. End of discussion. You may have your own opinion, but it's wrong :)

Ability to distinguish between the two versions on systems with both versions should be done differently. As an example should MPC-HC developers take Microsoft approach. Has 64 bit Internet Explorer or Windows Media Player different icons?

So you're happy you can't distinguish between the two version of Internet Explorer or Windows Media Player ? and you think that approach is making things easier for the user ? Really ?
In your logic the mpc-hc64.exe should also be renamed to mpc-hc.exe too NO?

[Underground78]

We plan to make the x86 and x64 builds totally independent at least settings-wise and for the context menu entries (which means that there will be MPC-HC and MPC-HC x64). I don't know what is planned for icons but I understand your point.

[Px]

My 5 cents on 32 vs 64 icons: they must be generally same, but have small difference, so user can understand that this is one application and easily distinguish different bitness versions

[v0lt]

I forgot why you need two versions of the player at the same time? Do not have enough x86 version?

[pankov]

Guys,
I have a problem with VSFilter and more precisely with it's ability to delay the subtitles and even more precisely the way ZoomPlayer controls it in real time to change the delay value. I guess ZoomPlayer uses some kind of interface to do it but I'm not sure.
The latest version that I've found on xvidvideo.ru to be working is 1.5.2.3251 and the next one there (1.5.2.3314) is broken. Sadly there are ~60 versions in between but I can't find any of them so I can't narrow it down more.
Also I apologize for not being able to explain the problem more technically but I hope you'll be able to understand.

[betaking]

can not compiled last mpc-hc svn
1>DSUtil.cpp : warning C4819: 该文件包含不能在当前代码页(936)中表示的字符。请将该文件保存为 Unicode 格式以防止数据丢失
FontInstaller.cpp
1>DSUtil.cpp(1955): warning C4819: 该文件包含不能在当前代码页(936)中表示的字符。请将该文件保存为 Unicode 格式以防止数据丢失
1>DSUtil.cpp(2412): error C2001: 常量中有换行符
1>DSUtil.cpp(2413): error C2001: 常量中有换行符
1>DSUtil.cpp(2412): fatal error C1057: 宏扩展中遇到意外的文件结束
GolombBuffer.cpp
H264Nalu.cpp
HdmvClipInfo.cpp
MediaTypeEx.cpp
MediaTypes.cpp
NullRenderers.cpp
text.cpp
vd.cpp
vd_asm.cpp
1>已完成生成项目“C:\mpc-hc\src\DSUtil\DSUtil.vcxproj”(build 个目标)的操作 - 失败。

[betaking]

{&DXVA_Intel_H264_ClearVideo, _T("H.264 bitstream decoder, ClearVideo?)},// Intel ClearVideo H264 bitstream decoder
{&DXVA_Intel_VC1_ClearVideo, _T("VC-1 bitstream decoder, ClearVideo?)},// Intel ClearVideo VC-1 bitstream decoder
change to
{&DXVA_Intel_H264_ClearVideo, _T("H.264 bitstream decoder, ClearVideo")},// Intel ClearVideo H264 bitstream decoder
{&DXVA_Intel_VC1_ClearVideo, _T("VC-1 bitstream decoder, ClearVideo")},// Intel ClearVideo VC-1 bitstream decode
compiled is ok!

[JanWillem32]

It's probably the locational ANSI-Unicode conversion that's giving you problems. I'll correct it. Thanks for noticing.

[betaking]

It's probably the locational ANSI-Unicode conversion that's giving you problems. I'll correct it. Thanks for noticing.

Thanks, and UDPReader.cpp have this problem too! so i conver UDPReader.cpp to Unicode too!

[JanWillem32]

Interesting, can you test if saving UDPReader.cpp as UTF-8 does solve your problem? It can be done with the normal notepad, with "Save as". The developers can later on take a look at some other files that include the somewhat code page sensitive characters for conversion and prevent problems like this from happening again.

[betaking]

Interesting, can you test if saving UDPReader.cpp as UTF-8 does solve your problem? It can be done with the normal notepad, with "Save as". The developers can later on take a look at some other files that include the somewhat code page sensitive characters for conversion and prevent problems like this from happening again.

test if saving UDPReader.cpp as UTF-8 does solve your problem? It can be done with the normal notepad, with "Save as".
test not solve my problem!
saving UDPReader.cpp as Unicode solve my problem!

[Underground78]

I forgot why you need two versions of the player at the same time? Do not have enough x86 version?

Some people want to use the x64 version but some DS filters still only exist on x86 I guess.

Guys,
I have a problem with VSFilter and more precisely whit it's ability to delay the subtitles and even more precisely the way ZoomPlayer controls in real time it to change the delay value. I guess ZoomPlayer uses some kind of interface to do it but I'm not sure.
The latest version that I've found on xvidvideo.ru to be working is 1.5.2.3251 and the next one there (1.5.2.3314) is broken. Sadly there are ~60 versions in between but I can't find any of them so I can't narrow it down more.
Also I apologize for not being able to explain the problem more technically but I hope you'll be able to understand.

If you don't want your issue to be lost, you will need to report it on Trac.

Between does someone know if some graphic cards support DXVA for video resolution higher than 1080p ?

[betaking]

Some people want to use the x64 version but some DS filters still only exist on x86 I guess.



If you don't want your issue to be lost, you will need to report it on Trac.

Between does someone know if some graphic cards support DXVA for video resolution higher than 1080p ?

GeForce 410M, GT 520 GF119

VP5

D

April 2011

Introduced 4k x 2k video decoding

[GTPVHD]

http://www.anandtech.com/show/4380/discrete-htpc-gpus-shootout/11

The GT 520 is the first (and only GPU as of now) to support the VDPAU Feature Set D.

We asked NVIDIA about the changes in the new VDPAU feature set and what it meant for Windows users. They indicated that the new VPU was a faster version, also capable of decoding 4K x 2K videos. This means that the existing dual stream acceleration for 1080p videos has now been bumped up to quad stream acceleration.

3488 needs to be changed to allow 4k x 2k DXVA decoding on GT 520 and future Nvidia Kepler GPUs(GeForce 600 series) with the new Feature Set D hardware decoder.

[G_M_C]

@ GTPVHD & betaking
Anadtecht's article also mentioned that the 520 doesnt have a way of inputting 4K resolutions (over HDMI). The connection is 1.4, but Nv hasnt implemented the 4k part of the specification.

So despite (techically)beeing able to decode a 4K video, there is no way to get it done in reality (anyway as far as i understand Anandtech's article :p)

[nevcairiel]

You can decode the 4k x 2k just fine, the problem it talks about is actual output of anything higher then 1080p over HDMI. But of course, you can still down-scale the decoded result. :)

[hoborg]

Hi.
I am using MPC-HC only as a player, all internall filters are disabled.
For decoding i use LAV splitter + audio decoder + FFDshow for video.

The problem i have is that MPC-HC freeze randomply when opening .MPG files if is use libmpeg2 as a decoder in ffdshow.

There is no such issue if i open MPG directly in graphstudio. So can this be problem of MPC-HC player?

[v0lt]

It's probably the locational ANSI-Unicode conversion that's giving you problems. I'll correct it. Thanks for noticing.
_T("©") -> _T('\xA9')
_T("™") -> _T('\x99')
In this way we can solve the problem?
(I can not test right now)

[JanWillem32]

The trade mark sign is officially located at U+2122.
Using numerical references to characters should work, but there's also a problem in UDPReader.cpp with the common character é (U+00E9). ANSI encoding of characters common to the code page I'm using might not work for other code pages. The solution of converting to UTF-8 would be rather painless. Unfortunately, it doesn't work, only UTF-16LE does. The problem with that encoding is that it takes twice the space for each file, and mostly stores zeroes for the characters contained in the [0, 127] interval.
I can understand the compiler having problems with the code page mess, but the fact that it fails on UTF-8 needs some investigating. We might need to change a compiler setting or something to make it work.

[Underground78]

3488 needs to be changed to allow 4k x 2k DXVA decoding on GT 520 and future Nvidia Kepler GPUs(GeForce 600 series) with the new Feature Set D hardware decoder.

Yes, that's why I asked.

Hi.
I am using MPC-HC only as a player, all internall filters are disabled.
For decoding i use LAV splitter + audio decoder + FFDshow for video.

The problem i have is that MPC-HC freeze randomply when opening .MPG files if is use libmpeg2 as a decoder in ffdshow.

There is no such issue if i open MPG directly in graphstudio. So can this be problem of MPC-HC player?

It can be, when you say randomly, how often is it ?

[hoborg]

It can be, when you say randomly, how often is it ?

Well, often. Lets say it is 1:1 - one time it is working, second not.
It happend with all MPEG including VOBs.

I tested the same on slow computer (atom) and by some reason i was not able to reproduce it, but opening the video is much slower.
Here i have i5 2400S and opening is nearly instant - can be the CPU speed source of the issue? Maybe it need some delay?

BTW, i try to set MPC-HC internal MPEG splitter instead of LAV splitter and crash no longer happend.
So it must be something with LAV spliter + MPC-HC combination.

[VipZ]

The problem i have is that MPC-HC freeze randomply when opening .MPG files if is use libmpeg2 as a decoder in ffdshow.


I found libmpeg2 to be very unstable, soon as I switched to libav, all issues disappeared.

[hoborg]

I found libmpeg2 to be very unstable, soon as I switched to libav, all issues disappeared.

Well, some DVDs are not working correctly (navigation) with libav, no problem with libmpeg2.

[Px]

I forgot why you need two versions of the player at the same time? Do not have enough x86 version?
Well, I use three players :D, bsplayer as main, to view video on TV, mpc x32 to view video on main display, and mpc x64 when "something is wrong" :rolleyes:

[v0lt]

3488 needs to be changed to allow 4k x 2k DXVA decoding on GT 520 and future Nvidia Kepler GPUs(GeForce 600 series) with the new Feature Set D hardware decoder.
Has anyone tested it?

There are few other data.
http://en.wikipedia.org/wiki/Nvidia_PureVideo
The fifth generation of PureVideo HD, introduced with the Geforce GT 520, has significantly improved performance when decoding H.264 (and, from an already much higher level, other codecs), it is also capable of decoding 4K x 2K videos.

The fifth generation PureVideo HD is sometimes called "PureVideo HD 5" or VP5, although this is not an official Nvidia designation. This generation of PureVideo HD corresponds to Nvidia VDPAU Feature set D which due to limitations in all current VDPAU drivers does not support decoding higher resolutions than 2048x2048.
Therefore need the real tests.

[GTPVHD]

Ignore the wiki, it's written by a clueless person. If Anandtech, which is always in contact with Nvidia says it supports it, it supports it. The documentation cited by the wiki article is outdated.

Anyone can edit Wikipedia and post unverified and false info. Anandtech's reviewer for the article confirmed it with Nvidia the company themselves.

[nevcairiel]

The Hardware supports it, but i don't think the software (the driver) properly exposes it through DXVA yet.

[VipZ]

Well, some DVDs are not working correctly (navigation) with libav, no problem with libmpeg2.

True, I don't play DVD's with navigation so didn't come across this potential bug.

[Rectal Prolapse]

I've just noticed that this ticket has been closed:

http://sourceforge.net/apps/trac/mpc-hc/ticket/71

Congratulations on fixing this long-standing issue that has been a part of mpc-hc for 2 years! Kudos to v0lt for the fix - you will be remembered always.

[Toddler Naruto]

I do NOT recommend to do that. Icons will not fit to correct file type. Many things have changed since revision 3088 - last revision with old icons. Since then MPC-HC supports many new formats and new icons has been added to icon library, therefore old mpciconlib.dll file will not work properly. Also ordering is different.

For some file types will be icons missins, for some will MPC-HC display wrong icon. For example(when you see linked image) for MKV file will MPC-HC show D2V icon (no. 22) from old icon library. http://oi54.tinypic.com/ly91j.jpg

Anyway, if you wanna keep at least some old icons, i have deleted entire mpciconlib.dll and made afterward re-association process. This way i have all icons the same for all file types which are now taken from mpc-hc.exe file directly.

Hi, I already deleted the new mpciconlib.dll file, then I moved/copied the old mpciconlib.dll file to where the new version used to be, and finally I re-associated the icons.

Thanks for the help, though I did not need it, the old icons are working fine for me.

[ceb]

I have an ATI 5770 and use EVR CP, is the "YV12 Chroma Upsampling" shader still needed?
What is the difference between pre-resize and post-resize pixel shaders? I tried both but I hardly see any difference, post-resize seems slightly better (I use Sharpen Complex 2 and YV12 Chroma Upsampling). Which one should I use?

[fairchild]

I have an ATI 5770 and use EVR CP, is the "YV12 Chroma Upsampling" shader still needed?
What is the difference between pre-resize and post-resize pixel shaders? I tried both but I hardly see any difference, post-resize seems slightly better (I use Sharpen Complex 2 and YV12 Chroma Upsampling). Which one should I use?

I've asked this several times and never have gotten an answer. From what I've read if you use say ffdshow video decoder and set it to output RGB32 and have High Quality YV12 to RGB Conversion and Dithering checked, then the chroma upsampling bug won't affect you.

For sharpening, you want to generally use the post-resize pixel shader. For YV12 chroma upsampling I don't know if you should use post or pre-resize pixel shaders.

[leeperry]

I've asked this several times and never have gotten an answer.
how about comparing it yourself using this test pattern (http://www.mediafire.com/?7cjlfuid35u1hbe).

[fairchild]

how about comparing it yourself using this test pattern (http://www.mediafire.com/?7cjlfuid35u1hbe).

I've run that clip and I can't really notice much of a difference between different renderers and decoder combos. Is there a particular letter that I should be focussing on or something? I understand that if chroma isn't upsampling correctly then the red letters can blur into the black?

[leeperry]

I've run that clip and I can't really notice much of a difference between different renderers and decoder combos. Is there a particular letter that I should be focussing on or something? I understand that if chroma isn't upsampling correctly then the red letters can blur into the black?
You need to magnify a lot, the point resizing of the old ATi drivers would make big squares...but you don't want the chroma to be too blurry either IMHO. You could compare to what mVR does, personally my fav chroma upsampler in mVR is Softcubic80...100 is too blurry.

[JohnLai]

Problem Event Name: APPCRASH
Application Name: mpc-hc64.exe
Application Version: 1.5.3.3497
Application Timestamp: 4e31d056
Fault Module Name: mpc-hc64.exe
Fault Module Version: 1.5.3.3497
Fault Module Timestamp: 4e31d056
Exception Code: c0000005
Exception Offset: 00000000006583aa
OS Version: 6.1.7601.2.1.0.256.1
Locale ID: 1033
Additional Information 1: 151d
Additional Information 2: 151d52bb7e837a966a10eaa9d53f8929
Additional Information 3: 53c2
Additional Information 4: 53c2c7d603524dfaff1d0d80dcf60c81

Crashes still happen.....when exiting from D3D mode. If I played any media without D3D mode, no crash.

[v0lt]

@JohnLai
A long time is it? Which version is right?

[JohnLai]

If you want the exact version of non-crash, I dont know. However, built 3267 was fine.

Ever since modification 3286 onwards, kept getting stackhash error and I stopped to test at version 3329. Only recently, I started to test again and mpc keeps getting APPCRASH after exiting D3D mode since 3485.

[pancserzso]

Some mp4 files trigger a hard restart on my computer using DXVA!

1. first play, AMD graphics driver stops working if I continue working after this, but without graphic activities, then everything works ok.
2. if I start one of those files AGAIN, when MPC-HC opens then 2nd time, enlarges the window and in a second my system restarts!

I am using these versions:
x264 build 1995
MPC-HC build 1.5.2.3423
EVR Custom default, only using internal filters
AMD Radeon 5650M, Catalyst 11.3
Win 7 64

Here is one of the files which gives me this "instant restart":
http://ifile.it/2tawgez

Can you tell me what could be wrong? Could you test the mp4 movie on your system?

[JanWillem32]

Confirmed BSOD because of driver failure. This sample is 3840×2160 and that's why it doesn't work with DXVA. (I don't know why it should crash, though.)

[Aleksoid1978]

Some mp4 files trigger a hard restart on my computer using DXVA!

1. first play, AMD graphics driver stops working if I continue working after this, but without graphic activities, then everything works ok.
2. if I start one of those files AGAIN, when MPC-HC opens then 2nd time, enlarges the window and in a second my system restarts!

I am using these versions:
x264 build 1995
MPC-HC build 1.5.2.3423
EVR Custom default, only using internal filters
AMD Radeon 5650M, Catalyst 11.3
Win 7 64

Here is one of the files which gives me this "instant restart":
http://ifile.it/2tawgez

Can you tell me what could be wrong? Could you test the mp4 movie on your system?

Get latest version from www.xvidvideo.ru
after 3488 DXVA is off for more then FullHD resolution.

[ceb]

I've asked this several times and never have gotten an answer. From what I've read if you use say ffdshow video decoder and set it to output RGB32 and have High Quality YV12 to RGB Conversion and Dithering checked, then the chroma upsampling bug won't affect you.

For sharpening, you want to generally use the post-resize pixel shader. For YV12 chroma upsampling I don't know if you should use post or pre-resize pixel shaders.
Yeah post-resize seems slightly better, to me it looks like it has less artifacts (I hope this is the correct term).
Thanks for your answer. :)
how about comparing it yourself using this test pattern (http://www.mediafire.com/?7cjlfuid35u1hbe).
I tried it but don't see any difference (perhaps because I'm using a 'recent' video card and latest drivers [11.7]), so I guess I'm fine without it. :thanks:

[Underground78]

If you want the exact version of non-crash, I dont know. However, built 3267 was fine.

Ever since modification 3286 onwards, kept getting stackhash error and I stopped to test at version 3329. Only recently, I started to test again and mpc keeps getting APPCRASH after exiting D3D mode since 3485.

You mean a build between rev 3267 and 3286 added the bug ?

[JohnLai]

Correct, except errors are different after exiting D3D mode. Stackhash (3286-3329,stopped test) and Appcrash (3485 - ?). I am using xhmikosr's builds for testing.

[Underground78]

Correct, except errors are different after exiting D3D mode. Stackhash (3286-3329,stopped test) and Appcrash (3485 - ?). I am using xhmikosr's builds for testing.

Can you try narrowing down the range of revisions where the bug was introduced using the builds from here http://www.xvidvideo.ru/media-player-classic-home-cinema-x86-x64/ ?

[Silent Rain]

The video freezes while jumping on the timeline!
Tested on LAV Video Decoder (http://files.1f0.de/lavf/LAVFilters-video3.zip) & MPC-HC 1.5.3.3497 (http://xhmikosr.1f0.de/index.php?folder=bXBjLWhj). Connected to the player all three .ax-files (http://jpegshare.net/a5/32/a53226be9609ecee3fd45ff2011dd6e0.png.html) in external filters (http://postimage.org/image/8p3z7db8/). All the internal decoders and splitters disabled in the settings (http://postimage.org/image/8pfjysro/) of MPC-HC.


Bleeding Through - Orange County Blonde And Blue (Shock Session).mp4: (http://www.mediafire.com/?imhgm8wobdzc2so)

http://jpegshare.net/thumbs/b4/72/b4721f8c91a9ea7eedad4f93bb1da5bf.jpg (http://jpegshare.net/b4/72/b4721f8c91a9ea7eedad4f93bb1da5bf.jpg.html) http://jpegshare.net/thumbs/30/e8/30e8b1e4952b2b7205c4f9c1b117dcac.jpg (http://jpegshare.net/30/e8/30e8b1e4952b2b7205c4f9c1b117dcac.jpg.html) http://jpegshare.net/thumbs/5e/72/5e720798a2285998b5b59fa169cf169f.jpg (http://jpegshare.net/5e/72/5e720798a2285998b5b59fa169cf169f.jpg.html)


Bleeding Through - Death Anxiety.mkv: (http://www.mediafire.com/?tfbjtlic22ycuma)

http://jpegshare.net/thumbs/f6/f9/f6f90f065b893d3b9856987b28a5a5d9.jpg (http://jpegshare.net/f6/f9/f6f90f065b893d3b9856987b28a5a5d9.jpg.html) http://jpegshare.net/thumbs/4e/21/4e21c07c91e5d7ab617ba15492ae94dd.jpg (http://jpegshare.net/4e/21/4e21c07c91e5d7ab617ba15492ae94dd.jpg.html) http://jpegshare.net/thumbs/1f/64/1f64725768a1b0de889ddad3f324a262.jpg (http://jpegshare.net/1f/64/1f64725768a1b0de889ddad3f324a262.jpg.html)


Especially well seen these freezing, if you turn off DXVA (http://postimage.org/image/2q5bgnosk/) and Fast seek (on keyframe) (http://postimage.org/image/2q5d471ac/).

[JohnLai]

Can you try narrowing down the range of revisions where the bug was introduced using the builds from here http://www.xvidvideo.ru/media-player-classic-home-cinema-x86-x64/ ?

Which bug? Stackhash or Appcrash?

[Underground78]

Let's start with the first one to appear so the "Stackhash".

[JohnLai]

Strange, xvidvideo's builds didn't have stackhash error, instead BEX64 error came up. Below are the list:
3268 = no problem
3337 = BEX64
3338 = BEX64
3357 = BEX64
3363 = BEX64
3391 = APPCRASH
3403 = APPCRASH
3423 = APPCRASH
3432 = APPCRASH
3436 = BEX64
3446 = APPCRASH
3462 = BEX64
3471 = BEX64
3488 = BEX64
3491 = APPCRASH
3509 = BEX64

[Underground78]

Hmm, I should have tell you, there is a bug in the website that prevent you from seeing some builds ... :/ Put "Количество строк:" on 25 if you want to see the rev between 3268 and 3337.

Also does this happen with the x86 build too ?

[JohnLai]

Anyway..... I only tested 3 x86 builds from xvidvideo's builds....on 64bit win7 SP1.
3268 = perfectly fine.
3337 = BEX
3509 = BEX -->no '64' wording

Well...another results from xhmikosr's builds 64bit:
3280 = no problem
3285 = no problem
3291 = no problem
3297 = no problem
3322 = no problem
3324 = BEX64
3326 = BEX64

So...somewhere beween 3322 and 3324...something messed up, no idea what but changelog is :
Revision 3324 - Directory Listing
Modified Tue Jul 5 11:38:59 2011 UTC (3 weeks, 4 days ago) by aleksoid

Fix : x64 build crash if in Options try to use "WinLIRC" or "uICE" and press "Cancel".

Revision 3323 - Directory Listing
Modified Tue Jul 5 06:05:53 2011 UTC (3 weeks, 4 days ago) by aleksoid

Ticket #1493, thanks to judelaw.

Revision 3322 - Directory Listing
Modified Tue Jul 5 00:16:05 2011 UTC (3 weeks, 4 days ago) by aleksoid

Fix : MPEGSplitter, breaks standalone filter.

[Underground78]

Can you try this test build (http://www.mediafire.com/?7rx0pabsqq0e780) which has r3323 reverted and tell me if it helps ? Also try this debug build (http://www.mediafire.com/?i8myk3ril0l31en) and tell me if you get some useful informations when it crashes.

[JohnLai]

The results:
Test build = BEX
Debug build = no problem, I think? (Since no errors come out when exiting D3D)

[Underground78]

It seems the problem was not introduced where you think it was ... Are you really sure the problem never happens with r3322 from xhmikosr's website ?

[JohnLai]

Yup, I double checked again with 3322 and 3324 (64bit and 32bit) from xhmiskosr.

[Ayukawa]

Concenring the problem of DXVA not working on a laptop with ATI/Intel switchable graphics mentioned on the previous page, the last version that works on xvidvideo.ru is 1.5.1.2925.x64. After that the video will not play on the ATI chipset. Thanks.

[Underground78]

Yup, I double checked again with 3322 and 3324 (64bit and 32bit) from xhmiskosr.

It seems extremely doubtful that rev 3324 introduced the problem so if it's not 3323, I'm running out of ideas.

Here what Tetsuo told me on IRC tonight:
<tetsuo--> Underground78: that crash johnylai has
<tetsuo--> Underground78: its caused by his: driver, virusscanner, filter or trojan/virus
<tetsuo--> Underground78: tell him to update his videocard drivers and virusscanner, if that doesnt help tell him to only use internal filters
<tetsuo--> i cannot log in here, so i cant tell him

Concenring the problem of DXVA not working on a laptop with ATI/Intel switchable graphics mentioned on the previous page, the last version that works on xvidvideo.ru is 1.5.1.2925.x64. After that the video will not play on the ATI chipset. Thanks.

What is the first non working version you have found on xvidvideo.ru ?

[nimfe]

Anyone noticed that using MPC-HC and Logitech Gaming Software (older or 8.0.127) with 'Bicupic A=-0.60 PS2.0)' or 'Bicupic A=-0.75 PS2.0)' resizer option just result garbage output? Curiously other options dont..

In non current Logitech Software you could just disable MPC-HC but in newer one it wont help. Now I know logitech software is garbage as general but there is no real way use gaming keyboards like g150/g19 without it.

Perhaps option to disable keyboard lcd interface would be in order? Ofcoz it does not make sense that only those 2 shaders fail but i leave that to coders. :)

Non zoomed: http://img8.imageshack.us/img8/9933/36nonzoom.png
Fullscreen (any other zoom% does same though): http://img836.imageshack.us/img836/5599/34fullscreengrab.png

Mpc version: 1.5.2.3456 64bit (and yes other versions dont work either), EVR-sync.
Other stuff: W7-64, NVidia Gtx460 with 280.19, Asus M4A79T with Phenom2 x4 955, G510 keyboard and uptodate drivers.

[rahzel]

I have an ATI 5770 and use EVR CP, is the "YV12 Chroma Upsampling" shader still needed?
What is the difference between pre-resize and post-resize pixel shaders? I tried both but I hardly see any difference, post-resize seems slightly better (I use Sharpen Complex 2 and YV12 Chroma Upsampling). Which one should I use?

I've asked this several times and never have gotten an answer. From what I've read if you use say ffdshow video decoder and set it to output RGB32 and have High Quality YV12 to RGB Conversion and Dithering checked, then the chroma upsampling bug won't affect you.

For sharpening, you want to generally use the post-resize pixel shader. For YV12 chroma upsampling I don't know if you should use post or pre-resize pixel shaders.
Using leeperry's provided test pattern, I just compared MPC's YV12 chroma upscaler shader and ffdshow's high quality YV12 -> RGB conversion. For me, putting the YV12 shader in pre-resizer provided better results -- comparable to ffdshow. There are other alternatives to chroma upsampling (madVR, custom shaders), but considering you have an ATI card, if you enable full/half floating point processing or 10-bit RGB output, you need to enable that shader (or some sort of chroma upsampler), otherwise MPC will use nearest neighbor and it looks very bad (lots of aliasing and blocking). You really have to zoom in pretty good to see the difference, though.

[Ayukawa]

What is the first non working version you have found on xvidvideo.ru ?
It's 1.5.1.2928.x64. The official 1.5.2.3456 and the latest revision I tried last night also don't work. Thanks.

[JohnLai]

Driver = Ati Catalyst 11.6b and 11.8 preview, HD5450
Antivirus = Comodo Internet Security
I only use internal transform filter H264/AVC (DXVA) decoding for H.264 (I disabled all source and transform filters except H.264 DXVA) , ffdshow 3949 64bit, haali media splitter. Later I will test to use only internal filters.

[betaking]

Driver = Ati Catalyst 11.6b and 11.8 preview, HD5450
Antivirus = Comodo Internet Security
I only use internal transform filter H264/AVC (DXVA) decoding for H.264 (I disabled all source and transform filters except H.264 DXVA) , ffdshow 3949 64bit, haali media splitter. Later I will test to use only internal filters.

Ati Catalyst 11.6b and 11.8 preview is not good driver for ati 5xxx!

[ryrynz]

With all the BE builds on Xvideo.ru that I've tried jumping is broken. The standard builds available from Xvideo work perfectly. The jumping problems seem to only occur with the small and medium jumps. Can someone pass that along?

[JohnLai]

Ati Catalyst 11.6b and 11.8 preview is not good driver for ati 5xxx!

:confused: Thus, enlighten me on which version of Catalyst that is a good driver for 5000 series and I shall retest again once I have the time.


Btw, Underground78, why your debug build didnt crash when I tested it?

[iron2000]

With all the BE builds on Xvideo.ru that I've tried jumping is broken. The standard builds available from Xvideo work perfectly. The jumping problems seem to only occur with the small and medium jumps. Can someone pass that along?

I too experienced jumping problems with the BE builds.
Instead of jumping forward, it jumps back to the time when the jump is initiated.

Haven't been using the standard build for a long time now.

[ryrynz]

I too experienced jumping problems with the BE builds.
Instead of jumping forward, it jumps back to the time when the jump is initiated.

Haven't been using the standard build for a long time now.

Yeah that's exactly it, it doesn't happen all the time though.
I'd also rather use the BE builds (The original is just plain ugly) why that mod hasn't been incorporated into the standard builds I don't know, but first things first the jumping needs to be fixed.

[ryrynz]

:confused: Thus, enlighten me on which version of Catalyst that is a good driver for 5000 series and I shall retest again once I have the time.


Btw, Underground78, why your debug build didnt crash when I tested it?

Use 11.7

[JohnLai]

No luck, same result with 11.7.

After some testing with 3514:
If display the statistics (ctrl + j) while exiting = BEX64 StackHash_1dc2 = not always, sometimes only
Not sure what condition, sometimes gets APPCRASH
If OSD is displayed (such as fast forward etc) while exiting = BEX64 StackHash_26e5

Ignore above, still 3514:
Only 3 random errors occured
BEX64 StackHash_26e5 (always)
BEX64 StackHash_1dc2 (rare)
APPCRASH (often)

[Underground78]

Btw, Underground78, why your debug build didnt crash when I tested it?

No idea and that makes difficult to find the bug ...

It's 1.5.1.2928.x64. The official 1.5.2.3456 and the latest revision I tried last night also don't work. Thanks.

Ok so that's r2928 which breaks it, I'm quite sure of that but I'm not sure how to fix it. This rev disables DXVA for some Intel chips like Sandy Bridge as it made MPC-HC crash. I will ask Casimir about that.

[ceb]

Using leeperry's provided test pattern, I just compared MPC's YV12 chroma upscaler shader and ffdshow's high quality YV12 -> RGB conversion. For me, putting the YV12 shader in pre-resizer provided better results -- comparable to ffdshow. There are other alternatives to chroma upsampling (madVR, custom shaders), but considering you have an ATI card, if you enable full/half floating point processing or 10-bit RGB output, you need to enable that shader (or some sort of chroma upsampler), otherwise MPC will use nearest neighbor and it looks very bad (lots of aliasing and blocking). You really have to zoom in pretty good to see the difference, though.
I have not enabled full/half floating point processing or 10-bit RGB output, I don't even know what they do. :p
You mentioned "nearest neighbor", is it connected to the output resizer options? I have it at bicubic A=-1.00.
Anyway since you have to zoom in a lot to see the difference I should be fine without it.

[misha0209]

hi all,
if this was discussed, probably was but i do not have the time to look into 894 pages.
so, my question is regarding EVR-CP and EVR-SYNC.
can someone explain to me which one i should be using? what is the difference?
in EVR-CP i get no dropped frames. in EVR-SYNC i get dropped frames each time i change my position in video timeline. is EVR-SYNC experimental? is it recommended over EVR-CP, for a 60Hz screen?
thanks!

[Mercury_22]

Can someone please add a "Key" / keyboard shortcut for "After Playback"-> "Once"-> "Shutdown" cause it defeats the purpose of having this commands since we can't use them from a Remote control
I'm not "familiar" with MPC-HC's code but it doesn't seem that much work to me and the benefits are well worth the "Cost" :) :confused:

P.S. See Ticket 1155 (http://sourceforge.net/apps/trac/mpc-hc/ticket/1155)

[Hera]

Anyone get,
D3DFS + No Audio Codec = No ability to interact with MPC:HC

Without D3DFS, a popup informs you about some gibberish with audio codec. File had Vorbris audio.

[rahzel]

I have not enabled full/half floating point processing or 10-bit RGB output, I don't even know what they do. :p
You mentioned "nearest neighbor", is it connected to the output resizer options? I have it at bicubic A=-1.00.
Anyway since you have to zoom in a lot to see the difference I should be fine without it.
to me, the biggest reason to enable 10-bit output/ffpp/hfpp is so that MPC can enable dithering, which helps with banding. if you look at a smooth grayscale ramp, you may notice banding, in which one of the aforementioned settings can help.

banding is introduced because most video cards only output RGB 444 0-255 while video content is usually YCbCr420 16-235, so both the luma and chroma have to be upsampled. dithering really smooths out gradients. this is easier to notice than fixing chroma upsampling, but other than looking at a grayscale ramp, again, the difference isn't that noticeable when watching a video.

there are a few things that need to be scaled when playing videos on a PC. the scaler in the MPC output settings is for resizing videos to your display AFAIK.

[pankov]

Guys,
I have a problem with VSFilter and more precisely with it's ability to delay the subtitles and even more precisely the way ZoomPlayer controls it in real time to change the delay value. I guess ZoomPlayer uses some kind of interface to do it but I'm not sure.
The latest version that I've found on xvidvideo.ru to be working is 1.5.2.3251 and the next one there (1.5.2.3314) is broken. Sadly there are ~60 versions in between but I can't find any of them so I can't narrow it down more.
Also I apologize for not being able to explain the problem more technically but I hope you'll be able to understand.

I've added a ticket to tracker
http://sourceforge.net/apps/trac/mpc-hc/ticket/1585
and I hope it will be easy to fix it.

[ryrynz]

Pankov if you head here

http://www.xvidvideo.ru/media-player-classic-home-cinema-x86-x64/stranitsa-2.html

and also click on page 1 at the bottom of the page you can access more builds in the range you specified.

[tetsuo55]

No luck, same result with 11.7.

After some testing with 3514:
If display the statistics (ctrl + j) while exiting = BEX64 StackHash_1dc2 = not always, sometimes only
Not sure what condition, sometimes gets APPCRASH
If OSD is displayed (such as fast forward etc) while exiting = BEX64 StackHash_26e5

Ignore above, still 3514:
Only 3 random errors occured
BEX64 StackHash_26e5 (always)
BEX64 StackHash_1dc2 (rare)
APPCRASH (often)Hi, I discussed this further with some others and checked on MSDN.
The advice they gave was(in order of what to try):

1. Update all your drivers to latest
2. Run anti-spyware, anti-virus on the whole system
3. Format and reinstall windows (making sure your computer does not get infected over LAN/WAN, connect it to the network after installing virusscanner/firewall)
4. Replace the PSU
5. Replace the RAM

Good luck.

[tetsuo55]

Wouldn't it be awesome if windows and mpc-hc where THIS stable: http://www.fastcompany.com/magazine/06/writestuff.html?page=0%2C0

[Uxorious]

It seems MPC-HC is having problems with 7.1 AAC content.
(I've had the problem forever, but currently I'm on v1.5.3.3474 as shipped in K-Lite Mega Codec Pack 7.5.4).

If I play a 7.1.m4a file in MPC-HC, I get this constant scratchy sound (I can hear what it is, but it's distorted).
Take the same file and convert it to a WAV (with Nero), and MPC-HC plays it fine.

Anybody else with this problem, or is it likely to be something on my end?

[ryrynz]

That's an amazing read Tetsuo, highly recommend it to everyone.

[JohnLai]

Dear tetsuo55, I just formatted my pc around 1 week ago and the problem still remains same. I already make sure all of my drivers are up to date. Done checking for virus stuff by uninstalling comodo and scans using Avira today, nothing detected.

RAM tested using Memtest86, no problem.
I got two PSU : Silverstone ST50F and Gigabyte GE-N400-C2 = same result.

Those errors only occur when exiting from D3D mode. If D3D mode is not used, it exited fine.
So, I am entirely unsure why errors happens with ATI card after version 3322.

Just in case, I tested with my Nvidia 8400GS using 280.19 drivers and it works wonder, no error at all.

[Underground78]

If it doesn't happen with all computers, it will make the problem really difficult to understand ... :(

It seems MPC-HC is having problems with 7.1 AAC content.
(I've had the problem forever, but currently I'm on v1.5.3.3474 as shipped in K-Lite Mega Codec Pack 7.5.4).

If I play a 7.1.m4a file in MPC-HC, I get this constant scratchy sound (I can hear what it is, but it's distorted).
Take the same file and convert it to a WAV (with Nero), and MPC-HC plays it fine.

Anybody else with this problem, or is it likely to be something on my end?

You should probably open a ticket on Trac and upload a small 7.1 AAC sample.

[pankov]

Pankov if you head here

http://www.xvidvideo.ru/media-player-classic-home-cinema-x86-x64/stranitsa-2.html

and also click on page 1 at the bottom of the page you can access more builds in the range you specified.

Thanks, ryrynz.
I'm sure I did look for versions on different pages but I guess either I've somehow missed them or they simply were not present the other day. Nonetheless I managed to narrow it down between 1.5.2.3262 which is OK and 1.5.2.3268 which is broken.
Looking at the changelog I'm pretty sure it's commit 3264 that broke it
Add two new options in VSFilter GUI like it is done in MPC-HC:
- "Sub pictures to buffer" (default to 4)
- "Allow animation when buffering" (enable by default).

[Underground78]

I don't think this commit really broke something, it just triggered an existing bug. Can you try playing a bit with the options ? I would guess that if you set "Sub pictures to buffer" to 0, I will work again.

[pankov]

Underground78,
I've tried that but it didn't help.
I've also tried unchecking the "Allow animation when buffering" checkbox but again no success.
:(

[tetsuo55]

Dear tetsuo55, I just formatted my pc around 1 week ago and the problem still remains same. I already make sure all of my drivers are up to date. Done checking for virus stuff by uninstalling comodo and scans using Avira today, nothing detected.

RAM tested using Memtest86, no problem.
I got two PSU : Silverstone ST50F and Gigabyte GE-N400-C2 = same result.

Those errors only occur when exiting from D3D mode. If D3D mode is not used, it exited fine.
So, I am entirely unsure why errors happens with ATI card after version 3322.

Just in case, I tested with my Nvidia 8400GS using 280.19 drivers and it works wonder, no error at all.Many of use ATI, i cannot reproduce the probleon my ATI.

Can you try this build? http://www.mediafire.com/download.php?9ap6i7kizkwj0a7

[Underground78]

Underground78,
I've tried that but it didn't help.
I've also tried unchecking the "Allow animation when buffering" checkbox but again no success.
:(

Ok, now I think I get it ... I forgot that the interface was used by others players which don't update it each time it's modified. I should have kept the legacy functions, I will try fixing that today.

[pankov]

Underground78,
thank you very much!
:)

[JohnLai]

Many of use ATI, i cannot reproduce the probleon my ATI.

Can you try this build? http://www.mediafire.com/download.php?9ap6i7kizkwj0a7

Still crashed with StackHash_0a9e. Apparently users Qaq and TheElix also encountered the same problem.
http://forum.doom9.org/showthread.php?t=161047&page=8

[Uxorious]

If it doesn't happen with all computers, it will make the problem really difficult to understand ... :(

You should probably open a ticket on Trac and upload a small 7.1 AAC sample.

Well it happens on all computers I have tried it with.

Just hard to believe I'm the only one throwing 7.1 AAC content at it, but I guess I'll file a Trac issue and see.

Thanks.

[Underground78]

Well it happens on all computers I have tried it with.

This part was not for you, sorry for the misunderstanding.

Just hard to believe I'm the only one throwing 7.1 AAC content at it, but I guess I'll file a Trac issue and see.

Bugs reported here are often lost after a few days that's why we ask everybody to open tickets.

[Uxorious]

Well it happens on all computers I have tried it with.

Just hard to believe I'm the only one throwing 7.1 AAC content at it, but I guess I'll file a Trac issue and see.

Thanks.

Actually I just spotted something.
When I create an AAC file with 7.1 content and play it with MPC-HC, now if I hover my mouse over the FFA icon I see this:
ffdshow audio decoder:
Input: AAC, 6.1, 48000 Hz, 39 kbps (libfaad2)
Output: PCM, 6.1, 48000 Hz, 16-bit integer

I'm guessing it might be a libfaad2 bug...

[Underground78]

@pankov: See the Trac ticket you opened, I have put a test build for you there.

@Uxorious: It might be, I suppose you were using the internal filter at first ?

[v0lt]

Ticket #1606. Do not change the order of the audio tracks. (http://sourceforge.net/apps/trac/mpc-hc/ticket/1606)
What do you think about this?

[sneaker_ger]

Resorting tracks is a bad idea IMHO, because then the order would be different to other programs. People often take a short glimpse at a file with MPC-HC prior to remux/recode to find out the order of the tracks.

[Ayukawa]

Sorry for the newb question, but I'm new to using MPC-HC and using DXVA. Is there any advantage to using DXVA over software decoding if your CPU can handle the decoding? Both in terms of video quality and power consumption?

[Barlow]

Unless you're comparing to a broken decoder video output from the decoder is the same, so quality of the DXVA decoder is not better. However DXVA restricts which video renderer you can use. You won't be able to use madVR, which uses high quality scaling.
Power consumption should be better when using DXVA, as a special purpose piece of hardware will always be more efficient than a general purpose cpu.

[VipZ]

Sorry for the newb question, but I'm new to using MPC-HC and using DXVA. Is there any advantage to using DXVA over software decoding if your CPU can handle the decoding? Both in terms of video quality and power consumption?

I think its quite common that people think DXVA is more efficient, in every test I have run CPU is a fair bit more efficient.

Current total system:
Idle: 325w
CPU (RGB32 HQ): 350-355w avg
CPU (NV12): 355-360w avg
DXVA: 360-365w avg

CPU decoding does fluctuate more than DXVA depending on scene.

In the case of ATI HD5850 paired with a Core i7 860 or i7 2600k, the CPU is more efficient in pretty much every test I have run.

Any recent quad core can play 720p and even some 1080p in mostly idle states.

So DXVA makes for worse efficiency and more complex playback IMO

[SamuriHL]

Can I throw a wrinkle in that discussion? :D LAV CUVID offloads the decoding to the GPU and supports hardware deinterlacing. For a lot of people that's ideal because it allows you to use any renderer you want including madVR. And is great for older CPU's that may not have the power to handle 1080p decoding plus the other tasks necessary for post processing, etc. Just something to consider. :)

[G_M_C]

Can I throw a wrinkle in that discussion? :D LAV CUVID offloads the decoding to the GPU and supports hardware deinterlacing. For a lot of people that's ideal because it allows you to use any renderer you want including madVR. And is great for older CPU's that may not have the power to handle 1080p decoding plus the other tasks necessary for post processing, etc. Just something to consider. :)

You forget to mention that that only goes for people using a Nvidia GPU....

[VipZ]

Can I throw a wrinkle in that discussion? :D LAV CUVID offloads the decoding to the GPU and supports hardware deinterlacing. For a lot of people that's ideal because it allows you to use any renderer you want including madVR. And is great for older CPU's that may not have the power to handle 1080p decoding plus the other tasks necessary for post processing, etc. Just something to consider. :)

Do you any tests for how efficient LAV CUVID is? I haven't had a Nvidia graphics card for ages, but am interested to try this one day.

You forget to mention that that only goes for people using a Nvidia GPU....

Hehe, where is my LAV StreamVID :)

[SamuriHL]

You forget to mention that that only goes for people using a Nvidia GPU....

Yes, quite true.

[SamuriHL]

Do you any tests for how efficient LAV CUVID is? I haven't had a Nvidia graphics card for ages, but am interested to try this one day.


I don't. I have anecdotal evidence from my own personal experience. I have a very ancient AMD X2 4200+ machine that can't really do CPU decoding all that well. Certainly not for madVR. I dropped that 450 in place, loaded up LAV CUVID, and am now able to use madVR no problem on that machine. With full deinterlacing which is awesome.


Hehe, where is my LAV StreamVID :)

GLWT. Not going to happen. :) You get Cyberlink Video Decoder in HAM mode. Being both AMD and nVidia in my house, I've been through this already. ;)

[Hera]

EDIT: Does the power argument hold ground for a case of low-end GPU vs. high-end CPU?

Can I throw a wrinkle in that discussion? :D LAV CUVID offloads the decoding to the GPU and supports hardware deinterlacing. For a lot of people that's ideal because it allows you to use any renderer you want including madVR. And is great for older CPU's that may not have the power to handle 1080p decoding plus the other tasks necessary for post processing, etc. Just something to consider. :)

LAV CUVID seems to require more CPU power than the internal MPC h264 codec.
Tested on 1080p h264 source with animated subtitles using Tester Build of MPC using exclusive fullscreen mode and HFPP.
Using LAV CUVID (regardless of settings) caused frame dropping during the subtitle animation. Buffered frames went down to 1, graph drew a montain range and video stuttered.
Using default codec caused no such thing. Buffered frames remained at 3 and video did not stutter.

EDIT: But, free and works with non-EVR renderers is a good thing
EDIT2: Error resistant too while MPC's codec many times corrupts with Haali Splitter.... >.> PLEASE FIX THIS.

[Eliminateur]

i see there's a commit to support exporting settings, yet i fail to find a button or menu option to do such thing, where should i be looking?

Also, since hi10p content is starting to spread, is there a way for you to add a checkbox so that hi10p content bypasses DXVA?, just like the "bypass for SD content" a "bypass for not compliant"?
or the level check does that already?

[Underground78]

i see there's a commit to support exporting settings, yet i fail to find a button or menu option to do such thing, where should i be looking?

In the "Misc" page of the options dialog.

Also, since hi10p content is starting to spread, is there a way for you to add a checkbox so that hi10p content bypasses DXVA?, just like the "bypass for SD content" a "bypass for not compliant"?
or the level check does that already?

Hmm I thought it was already the case but maybe that was for 4:4:4.

Edit: Looks like 10-bit samples I have just fall-back to software decoding.

[G_M_C]

I don't. I have anecdotal evidence from my own personal experience. I have a very ancient AMD X2 4200+ machine that can't really do CPU decoding all that well. Certainly not for madVR. I dropped that 450 in place, loaded up LAV CUVID, and am now able to use madVR no problem on that machine. With full deinterlacing which is awesome.
[...]

That can be, but in JanWillems 'expirimental renderer' thread i read that no current Nv card does 10bit output. If that's true, Ati is the way to go imho. I've got a HD5770, and can do all the things you use MadVR for too. But when i go with Ati's driver-deinterlacing etc. i can also use JanWillems full floating point chroma upsampling and 10bit out.

All-in-all i'd say thare are tings going for and against both big Video-card brands. And the bottom line is that if you buy a current mid-to-high range card, either brand will do fine.

[nevcairiel]

10bit output is seriously overrated. You won't see a difference in the image between 10bit and 8bit+dithering, because like 99.9% of all screens internally work with 8bit or even less.

The point is, you want madVR + GPU deinterlacing, then NVIDIA is currently the only way to go. EVR will never be a replacement for madVR, no matter how long JanWillem works on it - the basic architecture is just limited.
Anyway, what good is 10-bit Output if EVR doesn't even support 10-bit Input? :)

[bozek]

madFlac doesnt work with x64 MPC-HC, right?

[betaking]

madFlac doesnt work with x64 MPC-HC, right?

x64 MPC-HC only use x64 filters! and madflac have no x64 verson! you can use lav filters x64!

[rahzel]

madFlac doesnt work with x64 MPC-HC, right?
just use the 32-bit version.

[Eliminateur]

In the "Misc" page of the options dialog.



Hmm I thought it was already the case but maybe that was for 4:4:4.

Edit: Looks like 10-bit samples I have just fall-back to software decoding.

feh turns out i xvidvideo hadn't compiled that verison, i downloaded 3522 and confused myself reading backwards in time, just downloaded 3838 and it's there.

on the 10bit output, yeah, even with skip level check it skips the renderer(i need to open a case in ffdshow because ffd dxva still load even on 10bit with all checks in place, at least ont he pre-10bit rev i have)

[clsid]

Latest ffdshow properly falls back to software decoding for 10bit stuff.

[JohnLai]

Underground 78, ignore the PM I sent earlier, I think I might found the issue, key and command stuff...........

First, for 'exit' command, set 'modifier' to none and 'key' to nothing. Set the 'mouse' to 'Left double click'.

Second, make sure mouse setting for 'fullscreen' command is set to 'none'.

Now, play any video file with D3D mode enabled and left double click to exit. The crash will happen.

[Underground78]

I can reproduce now but only with the non-debug build which makes debuging a bit tricky ...

[Ayukawa]

Ok so that's r2928 which breaks it, I'm quite sure of that but I'm not sure how to fix it. This rev disables DXVA for some Intel chips like Sandy Bridge as it made MPC-HC crash. I will ask Casimir about that.
Any updates/thoughts on this, Underground78? I've been trying out 1.5.0.2827 with DXVA on a Sandy Bridge machine, and I've not had any crashes. Thanks.

[Fadeout]

Latest ffdshow properly falls back to software decoding for 10bit stuff.

When will it go right into MPC-HC build?

[Underground78]

Any updates/thoughts on this, Underground78? I've been trying out 1.5.0.2827 with DXVA on a Sandy Bridge machine, and I've not had any crashes. Thanks.

No news from Casimir so far but I think the latest build which doesn't include this fix is r2825.

When will it go right into MPC-HC build?

It's already done. If you think it's not, you will have to give me a sample because I can't reproduce with any 4:4:4 / 10-bit sample I have.

[ForceX]

I think he meant 10 bit H.264 support in MPC-HC's internal H.264 decoder.

[Underground78]

I think he meant 10 bit H.264 support in MPC-HC's internal H.264 decoder.

Oh maybe, it's planed, it will normally be added the "ffmpegupdate" branch.

[tetsuo55]

Underground 78, ignore the PM I sent earlier, I think I might found the issue, key and command stuff...........

First, for 'exit' command, set 'modifier' to none and 'key' to nothing. Set the 'mouse' to 'Left double click'.

Second, make sure mouse setting for 'fullscreen' command is set to 'none'.

Now, play any video file with D3D mode enabled and left double click to exit. The crash will happen.Impressive detective work! Looks like you found a way to make 2 commands overlap resulting in a crash.

[Aleksoid1978]

Upload 10 bit H.264 sample for test anybody.

[v0lt]

2JanWillem32
Changeset 3564
-"23.976Hz (24P)"
-"29.97Hz (NTSC)"
+"24/1.001Hz (NTSC24)"
+"30/1.001Hz (NTSC30)"
WTF?

Why not?
"23.976Hz (NTSC24)"
"29.97Hz (NTSC30)"

Ordinary people also use the player.

[JanWillem32]

After converting about a hundred of those truncated numbers, I probably was pretty anal about converting them all, including some strings. (It's been a few months since I changed those.)
I do like how uniform it looks now, with the reference to NTSC and the actual slowdown value. But it's indeed a bit more technical.
Will "ordinary people" that can't read x/1.001Hz actually use the display frame rate changer? (That's what those strings are for.) It's a rather advanced feature, I believe.

[v0lt]

After converting about a hundred of those truncated numbers, I probably was pretty anal about converting them all, including some strings.
This is plain text which read eyes or not? Is it parsed?

[Underground78]

@JanWilliem32: I think too that you should keep the truncated numbers from strings.

@v0lt: Don't worry, those changes are still in a branch, not in the trunk.

[CruNcher]

Upload 10 bit H.264 sample for test anybody.

http://forum.doom9.org/showpost.php?p=1491609&postcount=185

[pgase]

Video Renderer error...
Audio Ok

.mkv

Media Type 0:
--------------------------
Video: RealVideo 4 720x536 25.00fps

AM_MEDIA_TYPE:
majortype: MEDIATYPE_Video {73646976-0000-0010-8000-00AA00389B71}
subtype: Unknown GUID Name {30345652-0000-0010-8000-00AA00389B71}
formattype: FORMAT_VideoInfo2 {F72A76A0-EB0A-11D0-ACE4-0000C0CC16BA}
bFixedSizeSamples: 1
bTemporalCompression: 0
lSampleSize: 1
cbFormat: 146

VIDEOINFOHEADER:
rcSource: (0,0)-(0,0)
rcTarget: (0,0)-(0,0)
dwBitRate: 0
dwBitErrorRate: 0
AvgTimePerFrame: 400000

VIDEOINFOHEADER2:
dwInterlaceFlags: 0x00000000
dwCopyProtectFlags: 0x00000000
dwPictAspectRatioX: 90
dwPictAspectRatioY: 67
dwControlFlags: 0x00000000
dwReserved2: 0x00000000

BITMAPINFOHEADER:
biSize: 40
biWidth: 720
biHeight: 536
biPlanes: 0
biBitCount: 0
biCompression: RV40
biSizeImage: 0
biXPelsPerMeter: 0
biYPelsPerMeter: 0
biClrUsed: 0
biClrImportant: 0

pbFormat:
0000: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
0010: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
0020: 00 00 00 00 00 00 00 00 80 1a 06 00 00 00 00 00 ........€.......
0030: 00 00 00 00 00 00 00 00 5a 00 00 00 43 00 00 00 ........Z...C...
0040: 00 00 00 00 00 00 00 00 28 00 00 00 d0 02 00 00 ........(...Ð...
0050: 18 02 00 00 00 00 00 00 52 56 34 30 00 00 00 00 ........RV40....
0060: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
0070:|00 00 00 22 56 49 44 4f 52 56 34 30 02 d0 02 18 ..."VIDORV40.Ð..
0080: 00 0c 00 00 00 00 00 19 00 00 01 48 10 20 40 00 ...........H. @.
0090: 80 00 €.


Media Type 1:
--------------------------
Video: RealVideo 4 720x536 25.00fps

AM_MEDIA_TYPE:
majortype: MEDIATYPE_Video {73646976-0000-0010-8000-00AA00389B71}
subtype: Unknown GUID Name {30345652-0000-0010-8000-00AA00389B71}
formattype: FORMAT_VideoInfo {05589F80-C356-11CE-BF01-00AA0055595A}
bFixedSizeSamples: 1
bTemporalCompression: 0
lSampleSize: 1
cbFormat: 122

VIDEOINFOHEADER:
rcSource: (0,0)-(0,0)
rcTarget: (0,0)-(0,0)
dwBitRate: 0
dwBitErrorRate: 0
AvgTimePerFrame: 400000

BITMAPINFOHEADER:
biSize: 40
biWidth: 720
biHeight: 536
biPlanes: 0
biBitCount: 0
biCompression: RV40
biSizeImage: 0
biXPelsPerMeter: 0
biYPelsPerMeter: 0
biClrUsed: 0
biClrImportant: 0

pbFormat:
0000: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
0010: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
0020: 00 00 00 00 00 00 00 00 80 1a 06 00 00 00 00 00 ........€.......
0030: 28 00 00 00 d0 02 00 00 18 02 00 00 00 00 00 00 (...Ð...........
0040: 52 56 34 30 00 00 00 00 00 00 00 00 00 00 00 00 RV40............
0050: 00 00 00 00 00 00 00 00|00 00 00 22 56 49 44 4f ..........."VIDO
0060: 52 56 34 30 02 d0 02 18 00 0c 00 00 00 00 00 19 RV40.Ð..........
0070: 00 00 01 48 10 20 40 00 80 00 ...H. @.€.

[JanWillem32]

@JanWilliem32: I think too that you should keep the truncated numbers from stringsThat's okay by me, I'll replace them. Notations of 23.(976023) Hz and 29.(970029) Hz seem mathematically correct and readable. http://en.wikipedia.org/wiki/Repeating_decimal

I remember why I put it in there again. Someone asked why the stats screen showed some different numbers than the expected exact 23.976 Hz:
-average frame rate fluctuating a bit near the expected frame rate
-reference frame rate as 23.97604 Hz
-detected frame rate as 23.97602 Hz when locked
-detected screen refresh rate as 23.97602 Hz

The actual targets are by specification 24/1.001, due to an audio engineering problem from the 1953 color TV standard. (I won't complain about archaic video standards this time...)
Average frame rates are subject to changes in synchronization throughout the renderer chain and different clock properties of the various components. You would need to link up everything to an atomic clock to get a very stable clock for rendering.
The reference frame rate is often noted in units of 100 ns, so it's often rounded.
The detection parts can correct a reference frame rate a bit, if it's near 24/1.001, 24, 25, 30/1.001 30, 48/1.001, 48, 50, 60/1.001 or 60 Hz.

[madshi]

I'd suggest using "23.976" and "29.970" because that's what is usually used by everyone.

Google hit table:
23.976: 124000000 hits
23.976023: 112000 hits
24/1.001: 20500 hits

Personally, I prefer "24/1.001" (and that's also what eac3to outputs, btw), but I understand that "23.976" is what everybody knows and understands.

[Underground78]

I'd suggest using "23.976" and "29.970" because that's what is usually used by everyone.

I agree with that, some users might be puzzled by the unusual bracket notation (although it's mathematically correct).

[v0lt]

If need to calculate, it is preferable to use fractions. In this case, it is not necessary.

[bur]

When you remove the menu and window bar (by pressing Ctrl+0) you can't move the window anymore. Maybe you could make it so, that the window can be dragged&dropped by clicking and holding the video? I'm not sure that's possible in Windows though.

Another related idea: An option so that after a time of inactivity MPC automatically removes all elements like seek bar, menu, etc and shows them again when the cursor is moved inside the window.

[Underground78]

When you remove the menu and window bar (by pressing Ctrl+0) you can't move the window anymore. Maybe you could make it so, that the window can be dragged&dropped by clicking and holding the video? I'm not sure that's possible in Windows though.

We are currently working on fixing that.

[v0lt]

Changeset 3587
Add the ability to use different mouse button bindings in windowed and fullscreen modes. Patch mostly by taemun.
Why? Why do so difficult? :confused:

[JarrettH]

Discovered a bug I think!

- Core i3 2100 (Intel HD Graphics 2000)
- ASUS P8H67 EVO
- 4GB DDR3
- Realtek integrated audio

Basically, VMR9 and plain EVR work, but when I switch to EVR-CP I get audio and the window resizes as if it were playing back video (both normal), but no video displays. Any clue as to why this is?

Another problem I'm encountering is using plain EVR with DVDs I can't disable subtitles. Something is rendering them and I don't know what it is (I can enable original DVD subtitles under navigate, but they overlap). I unticked the auto-load subtitles option and it still happens. There's no vobsub or ffdshow in the background. Is there a filter I should be blocking?

I may just blame both problems on the Intel graphics. :p

[ForceX]

Update your graphics driver to the latest, disable 10 bit RGB input and output option, disable any Floating Point Processing option from the Renderer Settings menu.

[v0lt]

Another problem I'm encountering is using plain EVR with DVDs I can't disable subtitles. Something is rendering them and I don't know what it is (I can enable original DVD subtitles under navigate, but they overlap).
Which DVD is this happening? Take a screenshot of this subtitles.

What video decoder do you use? Take a screenshot of the settings decoder.

[Underground78]

Why? Why do so difficult? :confused:

I don't see how this is difficult. People who don't need this feature can just assign the same binding to both windowed and fullscreen modes and really that's not difficult.

Speaking of key bindings, does someone use the ASCII type (I think it's of no use and broken anyway but I want to be sure) and/or need to know the numeric ID ?

[v0lt]

People who don't need this feature can just assign the same binding to both windowed and fullscreen modes and really that's not difficult.
Why set up the same thing twice?
For keys will be the same? :eek:

[Underground78]

Why set up the same thing twice?
For keys will be the same? :eek:

Because it's the only way to let people use different key bindings for windowed and fullscreen modes... Setting up the same button twice is not a big deal as most important mouse bindings are already set up by default. One thing we could do is to put by default something like "Use windowed binding" for fullscreen bindings but I'm not sure it's worth it.

[v0lt]

Because it's the only way to let people use different key bindings for windowed and fullscreen modes...
New feature gives this possibility. But who really needs it? About how many percent?

[bur]

When you remove the menu and window bar (by pressing Ctrl+0) you can't move the window anymore. Maybe you could make it so, that the window can be dragged&dropped by clicking and holding the video? I'm not sure that's possible in Windows though.

We are currently working on fixing that.I also found a way to handle it in the meantime. When you allow the status bar to be shown you can drag&drop the window using it.

[betaking]

New feature gives this possibility. But who really needs it? About how many percent?

You do not need, does not mean that other users do not need!
PS:I know you do not like adding some new features, such as the last time I mentioned channel switching. However, MPC-HC is not a person you develop, if the whole idea as you do. As to what open source is not it!
PPS:to Underground78, big thanks for you!
Revision 3548 - Directory Listing
Modified Tue Aug 2 16:05:47 2011 UTC (4 days, 18 hours ago) by underground78Make all message boxes translatable.
This commit fixes ticket #1610.
This change makes me a better translation of these messages out to allow more Chinese users some problems to find a good solution!

[v0lt]

You do not need, does not mean that other users do not need!
With such a worldview player to replace the browser, text editor and explorer. Because someone needs it. :(

[Underground78]

With such a worldview player to replace the browser, text editor and explorer. Because someone needs it. :(

Don't exaggerate... I know most people (like me) probably won't be using this feature but those people also probably won't change the mouse bindings as default settings will be fine for them. But as I told you if it really puzzles you, I'm not against adding a new default behavior for the fullscreen bindings which would be "default to windowed binding" so that users don't have to set up the binding twice.

@betaking: When you have an updated translation ready, don't forget to send us the updated translation file so that we can commit it. :)

[betaking]

Don't exaggerate... I know most people (like me) probably won't be using this feature but those people also probably won't change the mouse bindings as default settings will be fine for them. But as I told you if it really puzzles you, I'm not against adding a new default behavior for the fullscreen bindings which would be "default to windowed binding" so that users don't have to set up the binding twice.

@betaking: When you have an updated translation ready, don't forget to send us the updated translation file so that we can commit it. :)

OK! can you make this word to translatable?
PPageAccelTbl.cpp
m_list.InsertColumn(COL_RMCMD, _T("RemoteCmd"), LVCFMT_LEFT, 80);
to
m_list.InsertColumn(COL_RMCMD, ResStr(IDS_AG_REMOTECMD), LVCFMT_LEFT, 80); :)

[JanWillem32]

If we really want to translate all strings, I think the translators will have will have quite a handful on the ErrBox messages I already made (I'm going to add a lot more, too): http://sourceforge.net/apps/trac/mpc-hc/browser/branches/renderer_fixes/src/filters/renderer/VideoRenderers/DX9RenderingEngine.cpp .

[Underground78]

OK! can you make this word to translatable?
PPageAccelTbl.cpp
m_list.InsertColumn(COL_RMCMD, _T("RemoteCmd"), LVCFMT_LEFT, 80);
to
m_list.InsertColumn(COL_RMCMD, ResStr(IDS_AG_REMOTECMD), LVCFMT_LEFT, 80); :)

It's not really high priority but it will probably be done.

If we really want to translate all strings, I think the translators will have will have quite a handful on the ErrBox messages I already made (I'm going to add a lot more, too): http://sourceforge.net/apps/trac/mpc-hc/browser/branches/renderer_fixes/src/filters/renderer/VideoRenderers/DX9RenderingEngine.cpp .

We usually don't translate things until they get into the trunk I think.

[betaking]

It's not really high priority but it will probably be done.



We usually don't translate things until they get into the trunk I think.
@Underground78:last Chinese (Simplified) mplayerc.sc.rc
http://hotfile.com/dl/126125207/7558fc9/mplayerc.sc.rar.html

[JarrettH]

Which DVD is this happening? Take a screenshot of this subtitles.

What video decoder do you use? Take a screenshot of the settings decoder.

Actually I just investigated using an ffdshow setup to decode DVDs instead, it only happened using the Microsoft DTV decoder :thanks:

[Underground78]

@Underground78:last Chinese (Simplified) mplayerc.sc.rc
http://hotfile.com/dl/126125207/7558fc9/mplayerc.sc.rar.html

Thanks, I have created a new ticket on Trac (https://sourceforge.net/apps/trac/mpc-hc/ticket/1643) asking for the previous translators' approval before commiting. :)

[JohnLai]

So...any chance of my double click as command for exiting from D3D mode to be fixed?

[Underground78]

So...any chance of my double click as command for exiting from D3D mode to be fixed?

Exiting the whole application right ? It must be a race condition or something like that but I'm still not totally sure. I will try to have a look before the end of this week.

[Jtacdf]

Hi, the latest rev. 3592 seems to break some function of vsfilter on ass subs.
http://i197.photobucket.com/albums/aa315/Jtacdf/MPC-HC%20samples/Wrong.jpg
Last known working on rev. 3557.
http://i197.photobucket.com/albums/aa315/Jtacdf/MPC-HC%20samples/Right.jpg
Forgotten to link a sample.
http://www.mediafire.com/file/6b0fgpbvet7bbfe/%5BDoki%5D%20Suzumiya%20Haruhi%20no%20Yuuutsu%20%282009%29%20-%2008%20%281920x1080%20h264%20BD%20FLAC%29%20%5B79EC4B68%5D%20Sample.7z

[JanWillem32]

That looks like the /p drawing mode. Can you demux the ASS or SSA text subtitle file and look inside if it's indeed that?

edit: Verified, the sample indeed has colored rectangles using the /p drawing mode.

[Jtacdf]

I'm not sure. Opening up the ass in aegisub, the code goes something like "{\p1\c&H6C44D7&\pos(1107,1008)}m 0 0 l 233 0 233 750 0 750{\p0}" for that specific sample in the screenshot.

[clsid]

That bug is still there in 3605.

[Underground78]

This bug was introduced somewhere between r3557 and r3577.

Can you please try this build with r3600 reverted: http://www.mediafire.com/file/8dr5satajdzzqbp/mpc-hc.exe.7z ?

Edit: It seems to fix the problem, can you open a ticket on Trac please ?

[clsid]

The first part of this change is the cause:
http://mpc-hc.svn.sourceforge.net/viewvc/mpc-hc/trunk/src/Subtitles/RTS.cpp?r1=3560&r2=3559&pathrev=3560

[Underground78]

I reopened ticket #1563 (https://sourceforge.net/apps/trac/mpc-hc/ticket/1563) as r3560 was a fix for it.

[clsid]

The sample file from that ticket does not crash with LAV Splitter after reverting the patch. However, I also see no subtitles.

[Underground78]

Without reverting r3560 and using the internal splitter, I see two subtitles tracks and one does have subs. (Obviously, when r3560 is reverting, MPC-HC just crashes.)

[clsid]

LAV Splitter sees only one subtitle track. MediaInfo sees only one as well.

[betaking]

@Underground78:last Chinese (Simplified) mplayerc.sc.rc update again
http://hotfile.com/dl/126189202/ef394e6/mplayerc.sc.rc.txt.html
fix Compile error!

[Aleksoid1978]

Without reverting r3560 and using the internal splitter, I see two subtitles tracks and one does have subs. (Obviously, when r3560 is reverting, MPC-HC just crashes.)

MP4Splitter for some subtitle stream - add another, as plain text.

[CruNcher]

Hi could somebody work on that the Graph gets fully out of memory when the video is closed so that if you register a new external filter the same time you used a older one @ playback and close the video the new one gets utilized instead of the old instance (overwrite the same file and register while MPC-HC is open), so you don't need to restart MPC-HC anymore just for the new filter to become active :(

[Underground78]

@Underground78:last Chinese (Simplified) mplayerc.sc.rc update again
http://hotfile.com/dl/126189202/ef394e6/mplayerc.sc.rc.txt.html
fix Compile error!

Commited as r3609 ! :thanks:

[betaking]

Commited as r3609 ! :thanks:

To:Underground78,My friend zhfi be a new patch and a function for Play next in the folde to optimize this feature patch! If you are interested I will send you to see!:p

[Underground78]

To:Underground78,My friend zhfi be a new patch and a function for Play next in the folde to optimize this feature patch! If you are interested I will send you to see!:p

Please open a ticket on Trac. :)

[betaking]

Please open a ticket on Trac. :)

done!
https://sourceforge.net/apps/trac/mpc-hc/ticket/1646

[betaking]

If your folder contains audio and video files, in accordance with the practice before the check Play next in the folder, then finished playing the audio file will automatically play video files, and so was the entire folder where the files are played one time! But to apply this patch only after the file under the file name and extension automatically recognize the player! For example your file name contains 01, then play the end of this file will automatically play the file contains 02 instead of playing back the file 01 file!for example 01.MKV 02.MKV CD1.AVI CD2.AVI

[JanWillem32]

You mean that the file sorting method for when adding a folder to the playlist, should sort by extension first, and then by file name?

[betaking]

You mean that the file sorting method for when adding a folder to the playlist, should sort by extension first, and then by file name?

Bingo! you are right!:)

[betaking]

can not compiled last mpc-hc svn MPCVideoDec.ax for standalone filters,but compiled mpc-hc.exe is ok!
正在创建库 C:\mpc-hc\bin10\Filters_x86\MPCVideoDec.lib 和对象 C:\mpc-hc\bin10\Filters_x86\MPCVideoDec.exp
1>libavcodec.lib(log.o) : warning LNK4217: 本地定义的符号 __iob 在函数 _av_log_default_callback 中导入
1>libavcodec.lib(interf_dec.o) : warning LNK4049: 已导入本地定义的符号 __iob
1>libavcodec.lib(sp_dec.o) : warning LNK4049: 已导入本地定义的符号 __iob
1>libmingwex.a(wcrtomb.o) : warning LNK4217: 本地定义的符号 ___mb_cur_max 在函数 _wcrtomb 中导入
1>libmingwex.a(mbrtowc.o) : warning LNK4049: 已导入本地定义的符号 ___mb_cur_max
1>MPCVideoDecFilter.obj : error LNK2019: 无法解析的外部符号 "unsigned int __cdecl GetAdapter(struct IDirect3D9 *,struct HWND__ *)" (?GetAdapter@@YAIPAUIDirect3D9@@PAUHWND__@@@Z)，该符号在函数 "protected: void __thiscall CMPCVideoDecFilter::detectVideoCard(struct HWND__ *)" (?DetectVideoCard@CMPCVideoDecFilter@@IAEXPAUHWND__@@@Z) 中被引用
1>C:\mpc-hc\bin10\Filters_x86\MPCVideoDec.ax:fatal error LNK1120:1 个无法解析的外部命令
1>已完成生成项目“C:\mpc-hc\src\filters\transform\MPCVideoDec\MPCVideoDec.vcxproj”(build 个目标)的操作 - 失败。

生成失败。

已用时间 00:00:39.04

[Underground78]

I know what the problem is, I will try to fix that as soon as possible.

[betaking]

I know what the problem is, I will try to fix that as soon as possible.

thanks!

[Underground78]

thanks!

Fixed at r3622 (http://sourceforge.net/apps/trac/mpc-hc/changeset/3622/trunk).

[GTPVHD]

Microsoft has released a security update for Visual Studio 2010 SP1. All MPC-HC devs should install this update.

http://www.microsoft.com/download/en/details.aspx?displaylang=en&id=27049

http://support.microsoft.com/kb/2565057

[betaking]

Microsoft has released a security update for Visual Studio 2010 SP1. All MPC-HC devs should install this update.

http://www.microsoft.com/download/en/details.aspx?displaylang=en&id=27049

http://support.microsoft.com/kb/2565057

after install this update compiled last mpc-hc svn no problem!:)

[v0lt]

Microsoft has released a security update for Visual Studio 2010 SP1.
Would again like last time? :mad:
Now everything works without problems. Should not rush.

[CruNcher]

Does someone work on a DXVA Mpeg-2 Studio Profile (4:2:2) fallback to the Software Decoder for the internal MPC-HC Decoder ?

[Underground78]

Does someone work on a DXVA Mpeg-2 Studio Profile (4:2:2) fallback to the Software Decoder for the internal MPC-HC Decoder ?

Hmm, currently we only check (s->chroma_format < 2), I'm not sure if it's sufficient.

[Aleksoid1978]

Does someone work on a DXVA Mpeg-2 Studio Profile (4:2:2) fallback to the Software Decoder for the internal MPC-HC Decoder ?

Give a sample.

[v0lt]

Give a sample.
maybe it is. (http://forum.doom9.org/showpost.php?p=1491609&postcount=185)

[Aleksoid1978]

Does someone work on a DXVA Mpeg-2 Studio Profile (4:2:2) fallback to the Software Decoder for the internal MPC-HC Decoder ?

It's already done in rev. 3123 (http://mpc-hc.svn.sourceforge.net/viewvc/mpc-hc?view=revision&revision=3123)

[betaking]

to all dev:
compiled last mpc-hc svn standalone filters and mpc-hc.exe give me some warning C4018 !
for example
1>DXVADecoderVC1.cpp(214): warning C4018: “<”: 有符号/无符号不匹配
1>DXVADecoderVC1.cpp(215): warning C4018: “>=”: 有符号/无符号不匹配
1>DXVADecoderVC1.cpp(225): warning C4018: “>=”: 有符号/无符号不匹配
Able to compile successfully, but this is normal does it?
and use reshack open mpc-hc.exe-String Table-2582-1033 give me a RichEdit line insertion error.

[Underground78]

to all dev:
compiled last mpc-hc svn standalone filters and mpc-hc.exe give me some warning C4018 !
for example
1>DXVADecoderVC1.cpp(214): warning C4018: “<”: 有符号/无符号不匹配
1>DXVADecoderVC1.cpp(215): warning C4018: “>=”: 有符号/无符号不匹配
1>DXVADecoderVC1.cpp(225): warning C4018: “>=”: 有符号/无符号不匹配
Able to compile successfully, but this is normal does it?

That's "normal", those warnings were previously disabled, we have re-enabled them so that they can be eventually fixed.

and use reshack open mpc-hc.exe-String Table-2582-1033 give me a RichEdit line insertion error.

No idea about that, why would you need to use ResHack ?

Edit: @CruNcher: I can confirm, it seems that it already fall-backs to software decoding for 4:2:2 MPEG2.

[CruNcher]

It's already done in rev. 3123 (http://mpc-hc.svn.sourceforge.net/viewvc/mpc-hc?view=revision&revision=3123)

Huh right it seems i get garbage even without any DXVA @ all :( ? Build 1.5.3.3627

http://img824.imageshack.us/img824/6174/garbagestudio.png

[Underground78]

I would guess that the internal MPEG2 decoder based on libmpeg2 doesn't support 4:2:2.

[CruNcher]

I could swear it worked maybe it was another build though not from xhmikosr or it was decoded with ffdshow that should do it to so im a little surprised same for dscaler

I mean falling back to the internal decoder makes no sense if it doesn't support it then it's better to block the connection and fallback into the chain search.

Strange ffdshow seems to also fail now so only dscaler works currently.

Dscaler works (libmpeg2) YUY2

http://img215.imageshack.us/img215/9661/dscalerworks.png

though i would be for a fallback solution so MPC-HCs internal Decoder as long as DXVA is needed and if 4:2:2 content is tried then fallback to the chain and let the user define his prefered 4:2:2 Decoder (Dscaler,Mainconcept,Nero,Bitcontrol,ffdshow ect) :)

PS: Seems to be a bug in ffdshow YV12 needs to be disabled if the interlace flag is on else it corrupts

http://img171.imageshack.us/img171/2259/ffdshowworks.png


I pinpointed the problem in ffdshow to be the interlace flag in combination with YV12 selected (enabled it causes problems for Studio Profile input)
The output gets NV12 now and correctly rendered (seems Nvidia related) so YV12 should not be used when Studio Profile gets in and interlace flag is enabled instead NV12, this was tested with VMR9 Mixer mode and only happens though with YUV Mixing disabled with Yuv Mixing enabled it works with the default setup.

[betaking]

to all dev:last Chinese (Simplified) mplayerc.sc.rc
http://hotfile.com/dl/126489476/d115ecd/mplayerc.sc.rc.rar.html

[lovelove]

Is there a way to prevent MPC-HC from pausing the video upon single mouse left-click (i.e. left-mouse click should not trigger any action)?

[TheElix]

Is there a way to prevent MPC-HC from pausing the video upon single mouse left-click (i.e. left-mouse click should not trigger any action)?
http://rghost.ru/17604401/thumb.png (http://rghost.ru/17604401.view)

[Underground78]

The screenshot looks strange, you should do that in the "Mouse" column ("Mouse Windowed"/"Mouse Fullscreen" columns in latest versions).

[clsid]

Can AYUV input for the EVR CP renderer be disabled? It doesn't work correctly, giving either awful performance or borked video. See LAV Filters topic.

[Underground78]

Can AYUV input for the EVR CP renderer be disabled? It doesn't work correctly, giving either awful performance or borked video. See LAV Filters topic.

You should probably create a ticket and put JanWillem (janwillem32) in Cc.

[v0lt]

I would guess that the internal MPEG2 decoder based on libmpeg2 doesn't support 4:2:2.
It does not, because based on the old and much modified libmpeg2.

[JanWillem32]

Can AYUV input for the EVR CP renderer be disabled? It doesn't work correctly, giving either awful performance or borked video. See LAV Filters topic.When there's no support for a AYUV surface format on the video card, software emulation is enabled by the EVR mixer. Vanilla EVR does the same thing as EVR CP and EVR Sync in this regard. (And it's indeed very slow.)
It only gets borked because the software emulation mixer can't handle anything but X8R8G8B8 output frames.
As AYUV input isn't accepted natively by any video card at the moment, I don't have any problem with disabling it as a valid format until a proper custom mixer is written for the renderer. (As usual, C++ developers are very welcome to help write a custom mixer. The one I tried to write only made garbage images out of every incoming frame...)