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

[lovelove]

The screenshot looks strange, you should do that in the "Mouse" column ("Mouse Windowed"/"Mouse Fullscreen" columns in latest versions).
sorry, I had overlooked that. solved. :thanks:

[clsid]

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...)Can you commit a patch for it?

[sneaker_ger]

Why is it impossible to block LAV filters? I want to switch between Haali and LAV for Matroska files, but blocking LAV in the external filter list doesn't actually result in blocking.

Only way seems to be disabling mkv in the LAV Filter settings.

[blizeH]

Hi guys,

I have an Acer Revo 3610 and try to run all different MKV files (etc!) using it.

I find I very often have to go Options -> Output and select different ones (EVR, Haali, Uncompressed) to get it working smoothly but depending on what MKV it is, I always need to try 5 or 6 different settings before it runs smoothly with no stuttering or audio lag.

I'd suggest maybe the Reco simply isn't up to it, but every video will run smoothly with the right setting. I was wondering if there's a good MKV plugin that'll run nicely with slower processors? One I can install and not have to worry about it again?

Thanks!:)

[JanWillem32]

Can you commit a patch for it?For the vanilla EVR component, it's impossible, as we can't control its merit list. For EVR Sync and EVR CP I can set it up in the renderer_fixes branch by changing 2 lines and adding comments.
The code in the trunk will be a problem. I can make the same changes, but for the code in the trunk that will be the fourth item I would make a specific exception for. I wanted to integrate fixes for problems with NV12, YUY2, and some types of RGB format conversions with the internal codecs first (back in March). As that would clash with the work in the ffmpegupdate branch, I only included those fixes in renderer_fixes, hoping that the ffmpegupdate branch would bring equal or better solutions compared to my patch.
I've been wanting to change the mixer parts for months already, but the format conversion section for the internal codecs is not compatible with my edited code for the renderers' mixer section in the current state, and I don't know when it will be.

[betaking]

to all dev:Chinese (Simplified) mplayerc.sc.rc update again
http://hotfile.com/dl/126558220/af082c7/mplayerc.sc.rar.html

[roytam1]

@Underground78, ticket #1644 and #1647 need your attention.
As you aren't on IRC, I shout it here. ;-)

[Mistar Muffin]

Is there any way to make MPC-HC control the system volume rather than it's own independent volume? On my HTPC, Arcsoft Total Media Theatre and Windows Media Center (and many others) adjust the volume system wide. Example:

When I am in WMC watching Live TV, if I turn down the volume to 50%, then later switch to MPC-HC which says the volume is at 100% I have to click the system volume slider and move the system volume from 50% to increase the volume. I essentially have to manage two different volume controls. I use a wireless keyboard with my HTPC, and if I hit the mute hotkey in any part of the OS, including other media players, the system volume is muted. If it stays muted and I open MPC-HC, hitting the mute hotkey when MPC-HC does nothing to unmute the system volume. It just toggles the internal app mute. I have to change focus away from MPC-HC in order to adjust the system volume using my keyboard hotkeys. I've looked for a setting that would allow this but there does not seem to be one.

Here is a nearly 2 year old feature request that appears to have been incorrectly marked a dupe. I cannot see any connect between the initial feature request, comments, and eventual closing of the ticket.

http://sourceforge.net/apps/trac/mpc-hc/ticket/180

I don't know how to go about getting this ticket re-opened as I believe it should be.

Thanks for creating the best media player out there.

-Muffin

[clsid]

Change the audio renderer in MPC options to WaveOut and it will use the system volume mapping. Otherwise it will use its own.

[Mistar Muffin]

Change the audio renderer in MPC options to WaveOut and it will use the system volume mapping. Otherwise it will use its own.

This does not have any effect. Clicking Mute did not mute the system volume, nor did adjusting the slider change the system volume. They still operated independently.

Thanks.

[clsid]

Well, that used to work in the past. Maybe it only works on XP.

[Mistar Muffin]

Well, that used to work in the past. Maybe it only works on XP.

Fair enough. That said, it still seems like a perfectly valid feature request. Should I open another ticket or try to get this one reopened:

http://sourceforge.net/apps/trac/mpc-hc/ticket/180

Thanks.

[Underground78]

@Underground78, ticket #1644 and #1647 need your attention.
As you aren't on IRC, I shout it here. ;-)

I will try to take care of that but as I'm on holidays so I won't be here often for the next two weeks.

@betaking: please give us the updated txt file instead of the RC file.

[roytam1]

I will try to take care of that but as I'm on holidays so I won't be here often for the next two weeks.

@betaking: please give us the updated txt file instead of the RC file.

Thanks for updating. I'm afraid that you missed my newer custom_messages.iss in #1647 ;-)

[Underground78]

Thanks for updating. I'm afraid that you missed my newer custom_messages.iss in #1647 ;-)

Damn, you're right. :o

[Hera]

I did notice that MPC sound controls are bad.
For example,
1. Turn down the system volume to 1%
2. Open file in MPC w. D3DFS.
3. Use the function keys to up the volume to max
4. MPC:HC will (eventually) say that the volume is 100%
5. In reality, you will not be able to hear anything and system volume will still be 1%

[Aleksoid1978]

I did notice that MPC sound controls are bad.
For example,
1. Turn down the system volume to 1%
2. Open file in MPC w. D3DFS.
3. Use the function keys to up the volume to max
4. MPC:HC will (eventually) say that the volume is 100%
5. In reality, you will not be able to hear anything and system volume will still be 1%

MPC-HC didn't control System Volume.
MPC-HC has its own volume control, independent of the system.
It's not a bug.

[betaking]

I will try to take care of that but as I'm on holidays so I won't be here often for the next two weeks.

@betaking: please give us the updated txt file instead of the RC file.

OK!
http://hotfile.com/dl/126855273/2a70f84/mplayerc.sc.rc.zip.html

[Hera]

MPC-HC didn't control System Volume.
MPC-HC has its own volume control, independent of the system.
It's not a bug.

Then it is just bad design. :)

[xiulet]

hi all, my friends,

in fullscreen mode the last three or four versions , the keys( i use some with mouse) dont' work , in window mode are ok.

thanks :)

[v0lt]

in fullscreen mode the last three or four versions , the keys( i use some with mouse) dont' work , in window mode are ok.
This is a new "very useful" feature.

You can write a ticket.

[xiulet]

This is a new "very useful" feature.

ups,really, is a new feature? only is working in window mode , now?


:eek:

[khagaroth]

No, there are now separate settings for window and fullscreen mode. Recheck settings.

[xiulet]

No, there are now separate settings for window and fullscreen mode. Recheck settings.

ok thank you, i didn't see it because i have tu update catalan language :D

bye

[smok3]

is there a default shortkey for "loop playback this file"? or menu item?

[kasper93]

Hi, There is problem with timestamps in subresync. I don't know what causes the problem, but it not works like it should. Subtitles are presented properly every time but subresync show bad time. Screen bad (http://dl.dropbox.com/u/16282309/MPC-HC/bad.png), good (http://dl.dropbox.com/u/16282309/MPC-HC/good.png) It fail when subtitles are auto loaded, but after manual reload works good. With madVR even manually loaded subs don't work properly.
subtitle format:
{780}{924}{Y:b}FILM STANLEYA KUBRICKA
{925}{1075}{Y:b}MECHANICZNA POMARAŃCZA
{2409}{2457}/To ja...
{2458}{2530}/to znaczy Alex,|/oraz moi trzej druhowie,
{2533}{2623}/to znaczy Pete, Georgie i Dim.
{2631}{2675}/Siedzieliśmy w Barze Mlecznym Korova
{2677}{2778}/starając się wybrać rozsądny|/sposób spędzenia wieczoru.
{2816}{2893}/Bar Korova sprzedawał mleko z dodatkiem.(movie frame rate: 23.976 fps)

It looks like subresync module too soon load subs without checking FPS of movie and load it in 25fps instead of 23.976 fps, but I'm only guessing.

[JanWillem32]

The default is indeed 25.0 FPS. It's a task for the renderer to update it with the correct frame rate. The request call for a subtitle image takes a time parameter in 100 ns units.

[hdboy]

it's a minor problem, but I can't seem to reassign "display stats." mpc still uses ctrl+j no matter what I set.

[ageback]

I'm the one who's maintaining the Simplified Chinese language cooperate with betaking.
Here's the latest sc txt file.
http://hotfile.com/dl/127037433/d174e41/mplayerc.sc.rc.zip.html

Anyone who has SVN authentication please update it.

[kasper93]

The default is indeed 25.0 FPS. It's a task for the renderer to update it with the correct frame rate. The request call for a subtitle image takes a time parameter in 100 ns units.

Thanks for explanation. So it should be fixed because now autoloaded subtitles are always load with default frame rate to subresync. But it's really minor bug and it can wait.

[Mistar Muffin]

I did notice that MPC sound controls are bad.
For example,
1. Turn down the system volume to 1%
2. Open file in MPC w. D3DFS.
3. Use the function keys to up the volume to max
4. MPC:HC will (eventually) say that the volume is 100%
5. In reality, you will not be able to hear anything and system volume will still be 1%

MPC-HC didn't control System Volume.
MPC-HC has its own volume control, independent of the system.
It's not a bug.

Then it is just bad design. :)

I know it's not a bug, but at the minimum I believe the functionality should be offered. A simple checkbox in the options for those that want it? This would allow users who are accustomed to the current behavior to remain unaffected while significantly helping those of us with HTPCs that are looking for unified volume control.

I'm really hoping someone with the ability can re-open this ticket:

http://sourceforge.net/apps/trac/mpc-hc/ticket/180

It's a feature request for this volume control and it is 20 months old. It was incorrectly closed about 2 months ago. I and other users would like it to still be a possibility.

Thank you.

[v0lt]

deleted
(bad translation)

The player must not change global system settings. There are special tools for this.

[Mistar Muffin]

These user to configure the global volume control using special software. Why set up 10 times the same thing in different software, because you can do it once?

I'm not sure what you are asking.

[Hera]

Just stop capturing the volume control keys, those keys are meant for system volume - thats the whole fix.
Just like on WMP on W7 - WMP has its own sound controls but using the FN keys to control volume controls system volume and doesn't adjust WMP volume.

[betaking]

compiled last mpc-hc give me warning C4800 !
1>MainFrm.cpp(7420): warning C4800: “int”: 将值强制为布尔值“true”或“false”(性能警告)
1>MainFrm.cpp(7420): warning C4800: “int”: 将值强制为布尔值“true”或“false”(性能警告)

[JanWillem32]

That's a generic warning for when a type is implicitly converted to "bool" (without explicit reinterpret, static or dynamic casting). There are a lot of those warnings when compiling MPC-HC. It's mostly a performance warning.
The 1-bit boolean type can be 1 (true) or 0 (false). However, it's always stored at pointer size (32 bits on x86 or 64 bits on x64), unless when packing multiple boolean bits into another type. Packing booleans is often bad for performance, so we rarely do that.
To optimize this kind of code, we evaluate if the code path can be changed to take the original "int" for usage as a binary buffer, or if it is used in any math, store it in a "size_t, "ptrdiff_t", "__int64", "unsigned __int64" or "double".

This isn't the best example of performance degradations caused by type-casting. This part of MainFrm.cpp isn't used often, and isn't performance critical.
The floating-point math in the video+subtitle geometry and subtitle generator uses "float" for storage everywhere, but the processor's floating point control word is set to the default double precision. This causes many useless conversions between "float" and "double" in these performance critical functions. The default floating point control word is fine, the problem is storing items in a "float" type. Correctly changing every "float" to "double" is a lot of work to adapt the code, but I've already seen good performance gains in the (alpha status) code I've written.

edit: I was wondering what temporal video renderer functions users would like to see in MPC-HC.
Previously, because of limitations in the renderer queue and VSync functionality, I couldn't add temporal functions. That changed earlier this week. I've already written deinterlacers and a basic frame interpolation function (that actively uses the current frame rate, internal renderer timings and the output refresh rate). Does anyone have a nice idea for a new function, or information on variants of frame interpolation and deinterlacing?

[v0lt]

compiled last mpc-hc give me warning C4800 !
1>MainFrm.cpp(7420): warning C4800: “int”: 将值强制为布尔值“true”或“false”(性能警告)
1>MainFrm.cpp(7420): warning C4800: “int”: 将值强制为布尔值“true”或“false”(性能警告)
fixed in r3681

[bur]

Could you add an auto-complete feature to the file open dialog?

[nijiko]

General
ID : 31337 (0x7A69)
Complete name : I:\***.ts
Format : MPEG-TS
File size : 11.4 GiB
Duration : 1h 54mn
Overall bit rate : 14.4 Mbps

Video
ID : 256 (0x100)
Menu ID : 1024 (0x400)
Format : MPEG Video
Format version : Version 2
Format profile : Main@High
Format settings, BVOP : Yes
Format settings, Matrix : Custom
Format settings, GOP : M=3, N=15
Codec ID : 2
Duration : 1h 54mn
Bit rate mode : Variable
Bit rate : 13.5 Mbps
Maximum bit rate : 24.0 Mbps
Width : 1 440 pixels
Height : 1 080 pixels
Display aspect ratio : 16:9
Frame rate : 29.970 fps
Color space : YUV
Chroma subsampling : 4:2:0
Bit depth : 8 bits
Scan type : Interlaced
Scan order : Top Field First
Compression mode : Lossy
Bits/(Pixel*Frame) : 0.290
Stream size : 10.8 GiB (94%)

Audio
ID : 272 (0x110)
Menu ID : 1024 (0x400)
Format : AAC
Format/Info : Advanced Audio Codec
Format version : Version 2
Format profile : LC
Muxing mode : ADTS
Codec ID : 15
Duration : 1h 54mn
Bit rate mode : Constant
Bit rate : 140 Kbps
Channel(s) : 2 channels
Channel positions : Front: L R
Sampling rate : 48.0 KHz
Compression mode : Lossy
Delay relative to video : -94ms
Stream size : 115 MiB (1%)


MPEG Spliter 1.5.*.* can not play this file well.
AAC is broken for playing.
But MPEG Spliter 1.0.0.5 in Gulik project has no problem.

[nijiko]

BTW, all MPC series players, such as potplayer, kmp, also have the same problem.
Non-MPC series, such as VLC, splash, is fine.

Haali MPEG-TS can play well before 1:40:00,
after that will be mute.

[Aleksoid1978]

General
ID : 31337 (0x7A69)
Complete name : I:\***.ts
Format : MPEG-TS
File size : 11.4 GiB
Duration : 1h 54mn
Overall bit rate : 14.4 Mbps

Video
ID : 256 (0x100)
Menu ID : 1024 (0x400)
Format : MPEG Video
Format version : Version 2
Format profile : Main@High
Format settings, BVOP : Yes
Format settings, Matrix : Custom
Format settings, GOP : M=3, N=15
Codec ID : 2
Duration : 1h 54mn
Bit rate mode : Variable
Bit rate : 13.5 Mbps
Maximum bit rate : 24.0 Mbps
Width : 1 440 pixels
Height : 1 080 pixels
Display aspect ratio : 16:9
Frame rate : 29.970 fps
Color space : YUV
Chroma subsampling : 4:2:0
Bit depth : 8 bits
Scan type : Interlaced
Scan order : Top Field First
Compression mode : Lossy
Bits/(Pixel*Frame) : 0.290
Stream size : 10.8 GiB (94%)

Audio
ID : 272 (0x110)
Menu ID : 1024 (0x400)
Format : AAC
Format/Info : Advanced Audio Codec
Format version : Version 2
Format profile : LC
Muxing mode : ADTS
Codec ID : 15
Duration : 1h 54mn
Bit rate mode : Constant
Bit rate : 140 Kbps
Channel(s) : 2 channels
Channel positions : Front: L R
Sampling rate : 48.0 KHz
Compression mode : Lossy
Delay relative to video : -94ms
Stream size : 115 MiB (1%)


MPEG Spliter 1.5.*.* can not play this file well.
AAC is broken for playing.
But MPEG Spliter 1.0.0.5 in Gulik project has no problem.

Upload a sample for test.

[pirlouy]

I've tried adding ffdshow audio Processor as an external filter; I've chosen "Prefer", BUT it does NOT appear when I launch videos. Someone to confirm ?
FWIW, I use recent MPC-HC and ffdshow builds.

[nijiko]

Upload a sample for test.

It's a copyrighted clip...
And I think some Japanese TV signal videos are the same format (1440*1080i interlaced TS), and the key is "Muxing mode : ADTS"

[roytam1]

It's a copyrighted clip...
And I think some Japanese TV signal videos are the same format (1440*1080i interlaced TS), and the key is "Muxing mode : ADTS"

You may send the sample privately.
Without a sample, it is impossible to debug.

[Underground78]

Also cut the file and upload the smallest sample that can be used to reproduce the problem.

[robpdotcom]

Question:

Under view > Pan&Scan, there are presets (Scale to 16:9 TV, Zoom to Widescreen, etc) and an option to edit custom settings. Is there any way to assign a hot key to these, or to toggle between them?

[ramicio]

You can assign a hot key to toggle through the presets, but only one way.

[STaRGaZeR]

Hi there, I'm here to report an OLD issue. Recently I've upgraded from rev25xx to the last one. Among other things, I noticed 2 nasty bugs. The key here is that both "Remember last window position" and "Remember last window size" are checked. Now imagine this situation: the player is configured so that the window is on the top left corner of the screen, snapped to the edge of the desktop, with a size of 50% that of the screen. Now the bugs:

- In rev25xx if the video dimensions changed, the window changed its vertical dimension to accomodate the new video. Now it doesn't change at all, so we have huge black bars. You can reproduce it enabling and disabling ffdshow's resizing filter for example, or by opening a 1920x1080 video and then a 1920x800 one.
- In rev25xx the window size was correctly remembered when closing the player in fullscreen. To reproduce, do this: resize the window something you like. Close it and reopen, it'll be there as it should. Now play something, and enter fullscreen. Without leaving fullscreen, right click and select exit so the player closes. Open the player again: now the window is as big as the screen. This is obviously not wanted at all. In rev25xx, the window will go back to the previous size as it should.

I have narrowed it down to between rev2616 and rev2646 (no builds available in the middle). It probably was revisions 2636-2638. Any possibility of reverting or fixing this?

[upyzl]

Hi there, I'm sorry I have a issue with it...

My notebook is T1500 + 2GB RAM + MHD 2400 + Win7 x86

when I use 1.5.1 & 1.5.2 (include 1.5.2.3456 release) & 1.5.3, playing 720p would be very sloooow (or very hard)... No matter what splitter or decoder or renderer I use...
But, 1.5.0.2827 can be normal, quite OK. (Note: both are DXVA off, in fact both's DXVA are slow)

I want to up to date for new features... so, what's the problem on earth?
(Note: ALL are OK on my desktop)

[STaRGaZeR]

Hi there, I'm here to report an OLD issue. Recently I've upgraded from rev25xx to the last one. Among other things, I noticed 2 nasty bugs. The key here is that both "Remember last window position" and "Remember last window size" are checked. Now imagine this situation: the player is configured so that the window is on the top left corner of the screen, snapped to the edge of the desktop, with a size of 50% that of the screen. Now the bugs:

- In rev25xx if the video dimensions changed, the window changed its vertical dimension to accomodate the new video. Now it doesn't change at all, so we have huge black bars. You can reproduce it enabling and disabling ffdshow's resizing filter for example, or by opening a 1920x1080 video and then a 1920x800 one.
- In rev25xx the window size was correctly remembered when closing the player in fullscreen. To reproduce, do this: resize the window something you like. Close it and reopen, it'll be there as it should. Now play something, and enter fullscreen. Without leaving fullscreen, right click and select exit so the player closes. Open the player again: now the window is as big as the screen. This is obviously not wanted at all. In rev25xx, the window will go back to the previous size as it should.

I have narrowed it down to between rev2616 and rev2646 (no builds available in the middle). It probably was revisions 2636-2638. Any possibility of reverting or fixing this?

Okay, it seems that the problem is caused by r2637. Why was autozoom disabled when "Remember last window size" is enabled? They're perfectly compatible with each other. Disabling it is not a solution for tickets #293 and #678, but a workaround (no more option, no more problem, instead of fixing it).

#293 is unrelated to this, and #678 is a problem with the option not being properly saved in the ini.