Log in

View Full Version : Media Player Classic supports Matroska v2, *.mp4, *.mov, and MPEG-4 Timed Text


Pages : 1 2 3 4 5 [6] 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

icosium
22nd November 2005, 03:20
Hi all,
I have been experementing with Matroska container recently [RV10 (x264 was not convincining) + AACplus v2], trying to squeeze as much data in the audio stream. I wanted to use AACplus v2 instead of Real Audio AACplus (anybody tried this?). Anyway, after creating the aac file with Besweet + winamp encoders (v2@48Kbps) from a 2 Ch AC3, and play it in MPC, I get the following details in the "properties" window:

Audio: AAC 24000Hz mono 48Kbps [Audio]

Now, I dont know if it is just the way the internal MPC AAC decoder sees it (although it is based on FAADv2, which is supposed to support AAC SBR and PS), but I am worried that I am not getting the quality I want.

The same situation is repeated when I mux the Real Video stream with AAC stream, in a MKV container. Any ideas/similar problems?

Cheers,
PS: I also tried the latest release posted on the first page.

bond
22nd November 2005, 12:42
Hi all,
I have been experementing with Matroska container recently [RV10 (x264 was not convincining) + AACplus v2], trying to squeeze as much data in the audio stream. I wanted to use AACplus v2 instead of Real Audio AACplus (anybody tried this?). Anyway, after creating the aac file with Besweet + winamp encoders (v2@48Kbps) from a 2 Ch AC3, and play it in MPC, I get the following details in the "properties" window:

Audio: AAC 24000Hz mono 48Kbps [Audio]

Now, I dont know if it is just the way the internal MPC AAC decoder sees it (although it is based on FAADv2, which is supposed to support AAC SBR and PS), but I am worried that I am not getting the quality I want.

The same situation is repeated when I mux the Real Video stream with AAC stream, in a MKV container. Any ideas/similar problems?

Cheers,
PS: I also tried the latest release posted on the first page.first of all the name "aacplus" is the name of one implementation of the aac standard, the one from coding technologies, and not the name of the format (like "divx" is only one implementation of the asp standard and not the name of the format)

about your problem:
i guess you didnt set the flag to signal aac+sbr when muxing to mkv?
afaik faad2 always upmixes to the double frequency and channelnumber with small frequencies and mono as the aac specs do not require you to signal sbr

now the thing is that when you dont signal it mpc connects the graph, the aac decoder doesnt get the info that its sbr,ps and doesnt display the info, still it will upmix as it sees that its a low frequency and mono

breez
24th November 2005, 18:01
I recently experienced some stuttering on playback with ffdshow using AviSynth processing (http://forum.doom9.org/showthread.php?t=103179). I noticed Zoomplayer didn't exhibit it. Tried 6.4.8.4 and everything was smooth again! So there seems to be something up with the combination ffdshow+AviSynth+MPC 6.4.8.5 or newer.

edit: Something to do with AVI and Matroska splitters. Check the thread.

LeMoi
25th November 2005, 22:53
Any changelog for MPC 6.4.8.7 ?

Egh
26th November 2005, 00:53
yeah, would be nice to know what was changed in the last release of mpc.

(and vsfilter 2.37 as well :P)

rotflol
26th November 2005, 09:45
Well, 'Send "Now Playing" information to mIRC through GTSdll' is one of the new things. :)

Elias
26th November 2005, 13:20
Any changelog for MPC 6.4.8.7 ?I'll check with Gabest and see if he can tell me anything.

futurex
26th November 2005, 13:45
Hi elias, do you know where i can give suggestions for media player classic?

something i would really like is to have an option "dock to edges of the screen", as well as having the window not resizing each time a new video (with a different resolution) is played in the playlist. thanks :)

JoeBG
26th November 2005, 15:54
How can I manage it, that MPC directly starts with the UTF-8 subtitles so that I donīt have to switch :)

I mean this: Subtitle: UTF-8 [GPAC Streaming Text Handler (ger) (plain text)]

Egh
26th November 2005, 17:05
Well, 'Send "Now Playing" information to mIRC through GTSdll' is one of the new things. :)

but i didn't notice any difference with or without that option so far.

Liisachan
27th November 2005, 04:46
MPC 6.4.8.7:
In the VMR9 Renderless mode with "Use texture surfaces and render video in 3D" (Options | Playback | Output), if I happen to select "Bilinear (PS 2.0)" for resizing, the result is abnormally CPU-intensive and video frames are dropped radically when resized (for instance 200%)

- This doesn't happen with 6.4.8.6. New phenomenon in 6.4.8.7.
- The result is okay if I use simple "Bilinear" or if I don't use "Use texture surfaces and render video in 3D"

Can anyone reproduce this?

I don't think my graphic card is new enough anyway to support PS 2.0 (I'm not sure) but in that case simple bilinear is supposed to be used automatically (the tooltip says so), so there shouldn't be any difference between "Bilinear" and "Bilinear (PS 2.0)"
On the other hand, if my card does support PS 2.0, then the above should mean, what was working with 6.4.8.6 is not working now with 6.4.8.7. That's what I'm guessing.

Egh
27th November 2005, 05:31
@ Liisachan: so what card do you have?

"- The result is okay if I use simple "Bilinear" or if I don't use "Use texture surfaces and render video in 3D""

if you don't use 3D then shaders are not used as well, naturally. And fancy 3D rotations and stuff are also unavailable then.

Just tested on my card, which for sure doesn't have PS 2.0 whatsoever. Couldn't detect any serious difference in speed between plain surfaces mode and 3d mode Bilinear PS 2.0.

BTW, have any test been done comparing cpu load in different those modes?

celtic_druid
27th November 2005, 06:34
I can set it to Bilinear (PS 2.0), then zoom, scroll and rotate and it still plays fine. I am running my own compile, but I don't think that the source has changed since the 6.4.8.7 release. ATI 9800 Pro. Bicubic (PS2.0) works fine to.

Liisachan
27th November 2005, 07:29
Then maybe it's my hw-side problem.
It's Quadro FX 500 (supporting PS 2.0a, according to Google). I'm clueless when it comes to HW, but the phenomenon is reproducable and very obvious for me.

Sample .csv by ffdshow: 6487 & PS don't mix well:

CPU load for the first 16 frames: VMR 9 Renderless / 3D / 200% / RGB32
--MPC 6.4.8.7----- --MPC6.4.8.6---
Simple Bil. PS 2.0 Simple Bil PS 2.0
0% 7% 7% 10%
100% 100% 100% 67%
1% 100% 1% 0%
34% 100% 0% 1%
0% 100% 51% 34%
1% 100% 0% 67%
0% 100% 34% 0%
34% 17% 100% 34%
67% 100% 26% 0%
34% 75% 50% 1%
0% 86% 34% 34%
1% 100% 0% 67%
50% 100% 34% 0%
34% 0% 0% 34%
34% 0% 1% 34%
34% 100% 0% 0%

JoeBG
27th November 2005, 18:43
How can I manage it, that MPC directly starts with the UTF-8 subtitles so that I donīt have to switch :)

I mean this: Subtitle: UTF-8 [GPAC Streaming Text Handler (ger) (plain text)]

Sorry for forcing this. But is there any chance for help regarding this topic?

Elias
27th November 2005, 19:27
Changelog for MPC 6.4.8.7 added in the first post!

SeeMoreDigital
27th November 2005, 19:50
Changelog for MPC 6.4.8.7 added in the first post!How about adding this link (https://sourceforge.net/project/showfiles.php?group_id=82303&package_id=84358) somewhere on your opening thread page?


Cheers

Stebs
27th November 2005, 20:19
Then maybe it's my hw-side problem.
It's Quadro FX 500 (supporting PS 2.0a, according to Google). I'm clueless when it comes to HWYour Card should be ok, it's based on the NV34GL Chip which is similar to a GeforceFX 5200, it's not really suited for the latest Games (too slow) but it can do PS 2.0 and VS 2.0 (also 2.0a, with is a bit "better").
Do you have the latest official Drivers installed for it? (81.67):
http://www.nvidia.com/page/partner_certified_drivers.html

Maybe also try installing latest Directx, ffdshow etc.
Here with a Geforce 6800GT, there was no difference between simple bilinear and PS2.0 bilinear.

PS. How did you benchmark such exact numbers for every frame? Some function in MPC or ffdshow I missed?

Yong
27th November 2005, 20:37
This problem is very easy to reproduce if use MPC(6.4.8.7, 6.4.8.6) internal MP4 splitter with x264/mp4 when do seeking
(its happend when i do seeking at same/random location several times... :rolleyes: ), Gabest MP4 dshow splitter have the same problem too :(
it cause libavcodec.dll "assertion failed!".

ive tried to do the same thing with Halli media splitter, it didnt cause the crash.

SeeMoreDigital
27th November 2005, 20:50
Hi Yong,

I don't have this problem. I can seek using the slide-bar and using pre-defined chapters ;)


Cheers

Yong
27th November 2005, 21:23
Hi Yong,

I don't have this problem. I can seek using the slide-bar and using pre-defined chapters ;)


Cheers
I see...
may be my computer too slow so it cause trobules :D
(Pentium4 2.4GHZ, Win XPSP2, 512 ddr266 sdram and Radeon9600XT 128MB with latest display driver.)
Im just tried to play MPC with VMR9 Renderless + render video in 3D + Bicubic A=-0.75(PS2.0) + x264(640x480 30fps) + Halli media splitter with fullscreen...
After serveral second ffdshow crashed with "assertion failed!" message... :rolleyes:
(because my computer cant render video clips realtime at fullscreen mode with VMR7/9 Renderless)

SeeMoreDigital
27th November 2005, 21:46
I see...
may be my computer too slow so it cause trobules :D
(Pentium4 2.4GHZ, Win XPSP2, 512 ddr266 sdram and Radeon9600XT 128MB with latest display driver.)
Im just tried to play MPC with VMR9 Renderless + render video in 3D + Bicubic A=-0.75(PS2.0) + x264(640x480 30fps) + Halli media splitter with fullscreen...
After serveral second ffdshow crashed with "assertion failed!" message... :rolleyes:
(because my computer cant render video clips realtime at fullscreen mode with VMR7/9 Renderless)Well I only have a 2.8GHz PC

Anyway, why not give this x264 720x576 with 2Ch AAC-LC with chapters in MP4.7z (http://homepage.ntlworld.com/seemoredigital/Temp_Test_Files/x264_720x576_with_2Ch_AAC-LC_with_chapters_in_MP4.7z) sample a try?


Cheers

Yong
27th November 2005, 22:02
Well I only have a 2.8GHz PC

Anyway, why not give this x264 720x576 with 2Ch AAC-LC with chapters in MP4.7z (http://homepage.ntlworld.com/seemoredigital/Temp_Test_Files/x264_720x576_with_2Ch_AAC-LC_with_chapters_in_MP4.7z) sample a try?


Cheers
Oh, so wired, only the video clip u provided can play smoothly with those setting i mentioned above :D(CPU usage below 75%)
I wonder why only my own video clip play at jerky frame rate... :rolleyes:

Elias
27th November 2005, 22:05
Oh, so wired, only the video clip u provided can play smoothly with those setting i mentioned above :D(CPU usage below 75%)
I wonder why only my own video clip play at jerky frame rate... :rolleyes:Well, if you made it via MP4Box, you need to add the accurate FPS for AVC, or else MP4Box will by default make it 25 FPS. That might be the problem.

Yong
27th November 2005, 22:12
Well, if you made it via MP4Box, you need to add the accurate FPS for AVC, or else MP4Box will by default make it 25 FPS. That might be the problem.
Hmm, no, all my video clips(x264) were encoded with cli x264 encoder with mp4 output, no muxing with mp4box or mp4creator, so its video stream only.

Peuj
28th November 2005, 11:01
MPC 6.4.8.7:
In the VMR9 Renderless mode with "Use texture surfaces and render video in 3D" (Options | Playback | Output), if I happen to select "Bilinear (PS 2.0)" for resizing, the result is abnormally CPU-intensive and video frames are dropped radically when resized (for instance 200%)

- This doesn't happen with 6.4.8.6. New phenomenon in 6.4.8.7.
- The result is okay if I use simple "Bilinear" or if I don't use "Use texture surfaces and render video in 3D"

Can anyone reproduce this?

I don't think my graphic card is new enough anyway to support PS 2.0 (I'm not sure) but in that case simple bilinear is supposed to be used automatically (the tooltip says so), so there shouldn't be any difference between "Bilinear" and "Bilinear (PS 2.0)"
On the other hand, if my card does support PS 2.0, then the above should mean, what was working with 6.4.8.6 is not working now with 6.4.8.7. That's what I'm guessing.


Hi Liisachan,

I reproduce exactly the same problem with the 6.4.8.7
I don't thave the problem too with 6.4.8.6 and if I switch from Bilinear (PS 2.0) to simple Bilinear the problem is fixed.

And I have the latest version of DirectX, ffdshow (and even when I don't use ffdshow I can reproduce the pb), the latest driver...

Liisachan
28th November 2005, 11:54
@Stebs: Thanks for your info & advice :) My DirectX is the latest one, and the driver is quite new too (not the latest), but I'll update/reinstall them anyway just in case.
About the 'benchmark' you can just use ffdshow (Video decoder config -> Info & Debug | OSD | Save to)

@Peuj: The changelog says "3) Shaders reworked a bit." so I'm guessing Gabest changed something a little, which caused some negative side effects for specific cases.

I'm on Windows 2000. I'll test if I can reproduce the problem on my dual-boot Windows XP later.

EDIT: The problem was fixed after I updated the driver. Thank you very much Stebs :D Something must have been messed up.

Btw, celtic_druid's mplayerc2005.11.27.2kxp.7z doesn't start on Windows XP (it works excellently on 2k)

EDIT2: I thought it was fixed but I"m having the PS 2.0 problem again now (abnormal cpu load)... I'll try to re-install everything again to see if that will help.

Liisachan
3rd December 2005, 13:31
MPC 6.4.8.7: VMR9 Renderless + Bilinear (PS 2.0) = Abnormal CPU load for me.

- This doesn't happen on the previous versions. Only on 6.4.8.7. 6.4.8.6 is a-ok.
- This doesn't happen for VMR9 Windowed. Only Renderelss.
- This doesn't happen if I use Simple Biliniar (non PS) for resizing.
- This happens both on Win2k and XP.
- Both official 6.4.8.7 and celtic_druid's 6.4.8.7 have this problem.
- The problem is not fixed even after I:
(1) reinstalled directx_9c_oct05sdk_redist
(2) reinstalled forceware 81.67 and 81.85 (My card is Quadro FX 500)
(3) deleted HKEY_CURRENT_USER\Software\Gabest\Media Player Classic and redid MPC settings

Any suggestions...? :sad: :confused:

BTW, celtic_druid's mplayerc2005.11.27.2kxp.7z runs only on win2k for me. on windows xp, I get this error: "This application has failed to start because the application configuration is incorrect. Reinstalling the application may fix this problem."

Elias
3rd December 2005, 14:41
BTW, celtic_druid's mplayerc2005.11.27.2kxp.7z runs only on win2k for me. on windows xp, I get this error: "This application has failed to start because the application configuration is incorrect. Reinstalling the application may fix this problem."Celtic must be compiling it incorrectly if that's the case, or your Win2k might be b0rked. Bond has Win 2k, he can try to reproduce this error.

Egh
3rd December 2005, 22:25
BTW, celtic_druid's mplayerc2005.11.27.2kxp.7z runs only on win2k for me. on windows xp, I get this error: "This application has failed to start because the application configuration is incorrect. Reinstalling the application may fix this problem."

Nothing new here :) In his august release he managed somehow to b0rk XP theme controls in his build. I was unable to turn them on even with manifest file added to mpc folder.

He himself has only win2k so he doesn't test his builds on XP.

Liisachan
3rd December 2005, 22:31
imho MS is the one who is to blame here.
"Compatibility" is not in their dictionary.
Devs are supposed to have 98 me 2k xp 2003 etc etc to test one program? Baka na...

Liisachan
5th December 2005, 00:59
Dear Gabest:

This feature is cool but too powerful :D sending the info to every single channel you are in, almost flooding. How about sending the info only to the 'active' chanel (the window that gets focus), especially when you type /mpcvid manually in a specific chan?

Elias
5th December 2005, 01:26
Dear Gabest:

This feature is cool but too powerful :D sending the info to every single channel you are in, almost flooding. How about sending the info only to the 'active' chanel (the window that gets focus), especially when you type /mpcvid manually in a specific chan?http://www.mircscripts.org/comments.php?id=1736 try this.

Rash
5th December 2005, 02:27
This feature is cool but too powerful :D
Hahaha, I've never seen anyone saying a feature is "too" powerful. :D Funny.

Liisachan
5th December 2005, 04:04
@Elias
I like Gabest's own mpcinfo.dll for mIRC, as it's easily scriptable, like

/mpc --> me is playing... nopath($dll(mpcinfo.dll,file,...

Elias
5th December 2005, 06:17
@Elias
I like Gabest's own mpcinfo.dll for mIRC, as it's easily scriptable, like

/mpc --> me is playing... nopath($dll(mpcinfo.dll,file,...It has been using it all the time. If I remember correct, Gabest made mpcinfo.dll for this mIRC add-on from the beginning.

http://img228.imageshack.us/img228/9922/mediadisplayer1wz.th.png (http://img228.imageshack.us/my.php?image=mediadisplayer1wz.png)

Peuj
5th December 2005, 12:19
MPC 6.4.8.7: VMR9 Renderless + Bilinear (PS 2.0) = Abnormal CPU load for me.
...........
Any suggestions...? :sad: :confused:
Hi Liisachan,

a good news, this is the answer from Gabest about our problem:
"Yes, it's really bad, but I made the fixes already. There's also going to be a lot faster two pass bicubic resizer in the next version.


:)

Reino
5th December 2005, 13:43
Maybe this is a stupid question, but somehow with earlier versions I could see the specific bitrate of a videofile in the properties (Shift+F10).
Now I see information like this: "XVID 640x352 25.00fps". Am I missing something here, because I've also tried previous version but now suddenly they too don't show me a bitrate value anymore but only fps. Is there some setting/tweak which I missed or something?
Or maybe Gabest would be so kind to make it display like this "XVID 640x352 25.00fps 1606kbps" for instance in the next MPC update? :)

Liisachan
5th December 2005, 13:59
@Peuj Oooh! So it's MPC-side, after all..... Good news, but funny why the other ppl can't reproduce that then.

@CoRoNe Ctrl+4

Reino
5th December 2005, 14:22
Not what I meaned exactly. I had it displayed in properties (Shift+F10) with some previous versions and now it's gone :s.

rotflol
7th December 2005, 02:12
Works for me (6.4.8.7).


What doesn't work, however, is that when I downloaded the X3 HD trailer from Apple's site and changed the extension to .mp4, there's no sound in MPC. Here's what I get:
http://img481.imageshack.us/img481/9797/mpcx8ob.jpg

Haali's Matroska splitter works fine (when I disable the built-in splitter in MPC).

Reino
7th December 2005, 10:31
With those QuickTime trailers I noticed this too. Somehow the integrated MP4 Splitter isn't capable of splitting AVC1 together with AAC, because the Elecard MP4 Demultiplexer (and I guess Haali as you say) have no problem rendering these trailers.

Elias
7th December 2005, 13:19
This aac/avc in mov issue is something the bento4 code does not yet support. Gabest won't fix this because he had lots of issues when he tried.

SeeMoreDigital
7th December 2005, 14:11
Quite honestly, I don't think changing the .MOV file extension to .MP4 is a sensible thing to do anyway.

If you want force MediaPlayer Classic to play .MOV files using a "direct-show" filter, you should really be visiting and changing these MPC options: -

http://img207.imageshack.us/img207/3095/mpcformats6aa.png

..........Not that it will cure the AAC audio issue either!

Or force YAMB to de-mux/re-mux the streams within the .MOV container to the .MP4 container!


Cheers

bond
7th December 2005, 19:22
What doesn't work, however, is that when I downloaded the X3 HD trailer from Apple's site and changed the extension to .mp4, there's no sound in MPC.you know, gabests splitter is for MP4 files, i repeat, MP4. not mov files
you cant create a mp4 file by renaming the extension of a mov file

Elias
7th December 2005, 21:26
you know, gabests splitter is for MP4 files, i repeat, MP4. not mov files
you cant create a mp4 file by renaming the extension of a mov fileAmen brother.

Stebs
7th December 2005, 23:29
you know, gabests splitter is for MP4 files, i repeat, MP4. not mov files
you cant create a mp4 file by renaming the extension of a mov fileSure, but under Windows, this is the only realistic way to watch HD AVC Movs, you just have to use Haali Media Splitter, rename them to mp4 and it works...
Performance of Windows Quicktime with HDTV Content is just ridiculously slow, sometimes only around 10 fps!
Playback with ffdshow (when renamed to mp4) plays the same file very smooth and the CPU is only at 60-80 %...
If I remember right, the Makers of Quicktime Alternative gave the tip to use Haali Media Splitter + rename to mp4 in order to watch HDTV AVC Movs.

I think a lot of people dont really care if renaming movs to mp4 is "right", it's simply just about beeing able to watch those damn files if you happen to get one of those (normally I try to avoid movs as much as possible).

It's ok that this "trick" still does not work with internal mp4 splitter, but nevertheless I think it would be great if it would work someday, even if renaming movs to mp4 is "evil" because it is not "the way it was meant to play".
Edit: Another way to play those movs with realistic speed is to use VLC.
Edit2: Ah, been to slow ;)
VLC is great, just its UI is awfull, I really like that of MPC...

Elias
7th December 2005, 23:31
Sure, but under Windows, this is the only realistic way to watch HD AVC MovsNo. Just use VLC.

bond
8th December 2005, 14:59
Sure, but under Windows, this is the only realistic way to watch HD AVC Movs, you just have to use Haali Media Splitter, rename them to mp4 and it works...nonsense!
you dont need to rename to .mp4 for getting haali to work. haali supports .mov files (with .mov extension) no need for renaming

Stebs
8th December 2005, 16:02
Hmm, but how is it possible then to let Quicktime Alternative play normal movs and let AVC movs be played by ffdshow (since they are too slow with Quicktime)? - Would be glad if this is somehow possible in MPC without the renaming trick...