Login
Username:

Password:

Remember me



Lost Password?

Register now!

Sections

Who's Online
144 user(s) are online (81 user(s) are browsing Forums)

Members: 1
Guests: 143

skynet, more...

Headlines

 
  Register To Post  

« 1 ... 12 13 14 (15) 16 17 18 ... 30 »
Re: Mplayer Altivec test version for G4 and PA6T-1682M
Just can't stay away
Just can't stay away


See User information
@LiveForIt

Quote:
I think I downloaded, LibSDL from OS4Depot.net.
I switched to the one from https://code.google.com/p/os4sdl/ because Roman told me the OS4depot.net version is old and too buggy.

If you are using the os4depot.net SDL and ao_sdl works in your MPlayer versions it's obviously the other way round

Quote:
I did not change the size of the buffer
In the original MPlayer sources

#if defined(__MINGW32__) || defined(__CYGWIN__)
    #define SAMPLESIZE 2048
...
is
#if defined(__MINGW32__) || defined(__CYGWIN__) || defined(__AMIGAOS4__)
    #define SAMPLESIZE 2048
...


Edited by joerg on 2014/4/18 23:23:54
Go to top
Re: Mplayer Altivec test version for G4 and PA6T-1682M
Home away from home
Home away from home


See User information
@K-L

On my Sam440 Flex 800Mhz (Radeon 9250)

Quote:
mplayer -nosound -quiet -benchmark Superman.flv


This time i started all 3 MPlayer exe (and the video) from RAM Disk:

Beta 3 of MPlayer from Joerg with p96_pip:
BENCHMARKs: VC: 50.637s VO: 4.663s A: 0.000s Sys: 0.641s = 55.941s
BENCHMARK%: VC: 90.5198% VO: 8.3352% A: 0.0000% Sys: 1.1450% = 100.0000%

Beta 4 of MPlayer from Joerg with p96_pip:
BENCHMARKs: VC: 50.223s VO: 4.664s A: 0.000s Sys: 0.636s = 55.524s
BENCHMARK%: VC: 90.4542% VO: 8.4008% A: 0.0000% Sys: 1.1451% = 100.0000%

Beta 9 of MUI MPlayer from kas1e with p96_pip:
BENCHMARKs: VC: 50.796s VO: 4.625s A: 0.000s Sys: 0.634s = 56.055s
BENCHMARK%: VC: 90.6183% VO: 8.2501% A: 0.0000% Sys: 1.1316% = 100.0000%




Edited by samo79 on 2014/4/19 2:02:19
Go to top
Re: Mplayer Altivec test version for G4 and PA6T-1682M
Just can't stay away
Just can't stay away


See User information
@samo79

Quote:
On my Sam440 Flex 800Mhz (Radeon 9250)

Beta 3 of MPlayer from Joerg with p96_pip:
BENCHMARKs: VC: 36.848s VO: 4.853s A: 0.000s Sys: 0.755s = 42.456s
BENCHMARK%: VC: 86.7918% VO: 11.4308% A: 0.0000% Sys: 1.7775% = 100.0000%

Beta 9 of MUI MPlayer from kas1e with p96_pip:
BENCHMARKs: VC: 51.120s VO: 12.343s A: 0.000s Sys: 0.500s = 63.963s
BENCHMARK%: VC: 79.9214% VO: 19.2973% A: 0.0000% Sys: 0.7812% = 100.0000%
Would be nice, but that's impossible. There may be small differences because of the different MPlayer versions and compiler options, but the VC speeds can't be that different because of it. I'm only working on the VO, nothing else. Looks like you are using something like "lavdopts fast=1:skiploopfilter=all" in conf/mplayer.conf when using my version but not when using Roman's.

Go to top
Re: Mplayer Altivec test version for G4 and PA6T-1682M
Home away from home
Home away from home


See User information
@joerg

Quote:
Maybe the buffer size change is important: http://www.sendspace.com/file/e4i0qh


No, with the audio activated i have the same crash

Quote:
Symbol info:
Instruction pointer 0x6C60E780 belongs to module "mplayer-joerg-beta4" (HUNK/Kickstart)

Stack trace:
module mplayer-joerg-beta4 at 0x6C60E780 (section 5 @ 0x7AA75C)
module mplayer-joerg-beta4 at 0x6C60E754 (section 5 @ 0x7AA730)
module mplayer-joerg-beta4 at 0x6C5F6B8C (section 5 @ 0x792B68)
module mplayer-joerg-beta4 at 0x6BE7EAF0 (section 5 @ 0x1AACC)
module mplayer-joerg-beta4 at 0x6BE79D00 (section 5 @ 0x15CDC)
module mplayer-joerg-beta4 at 0x6BE70CE8 (section 5 @ 0xCCC4)
module mplayer-joerg-beta4 at 0x6BE728E4 (section 5 @ 0xE8C0)
native kernel module newlib.library.kmod+0x000020ac
native kernel module newlib.library.kmod+0x00002d14
native kernel module newlib.library.kmod+0x00002ef0
_start()+0x170 (section 1 @ 0x16C)
native kernel module dos.library.kmod+0x00024cd0
native kernel module kernel+0x0003b4b0
native kernel module kernel+0x0003b530

Go to top
Re: Mplayer Altivec test version for G4 and PA6T-1682M
Home away from home
Home away from home


See User information
@joerg

Quote:
Would be nice, but that's impossible. There may be small differences because of the different MPlayer versions and compiler options, but the VC speeds can't be that different because of it. I'm only working on the VO, nothing else. Looks like you are using something like "lavdopts fast=1:skiploopfilter=all" in conf/mplayer.conf when using my version but not when using Roman's.


You right for sure, I am also surprised for such difference !
I searched now and it was because i copied your exe in the same directory of the LiveForIt archive, and in that config settings he use: skiploopfilter=all
so my test was affected by that ..

Edit: Infact now results are completely different, see the old post

Go to top
Re: Mplayer Altivec test version for G4 and PA6T-1682M
Just can't stay away
Just can't stay away


See User information
@samo79

Quote:
Edit: Infact now results are completely different, see the old post
Something is still wrong, 4.x seconds VO isn't possible with MUI MPlayer, it has to be > 10.

Go to top
Re: Mplayer Altivec test version for G4 and PA6T-1682M
Home away from home
Home away from home


See User information
@joerg

Mmm don't know why, will check ...

Go to top
Re: Mplayer Altivec test version for G4 and PA6T-1682M
Not too shy to talk
Not too shy to talk


See User information
SAM440EP-Flex 800MHz, R9250, p96_pip (Superman.flv)

mplayer-joerg-beta4

BENCHMARKs: VC: 50.594s VO: 4.699s A: 0.000s Sys: 0.583s = 55.876s
BENCHMARK%: VC: 90.5461% VO: 8.4099% A: 0.0000% Sys: 1.0440% = 100.0000%

mui-mplayer-beta9

BENCHMARKs: VC: 50.917s VO: 12.469s A: 0.000s Sys: 0.321s = 63.707s
BENCHMARK%: VC: 79.9231% VO: 19.5722% A: 0.0000% Sys: 0.5046% = 100.0000%

CD32/A500/A600/A600+Furia/A1200/A4000D+A2320+PiccoloSD64/Sam440 flex 800MHz RAM 1GB HD7750 128MB OS4.1 SBLive! ->
Go to top
Re: Mplayer Altivec test version for G4 and PA6T-1682M
Quite a regular
Quite a regular


See User information
Sam440EP ITX, 667mhz, Graphics M9.

Clip tested, superman.flv.

Everything in ram.

Joerg Beta 4

BENCHMARKs: VC: 55.843s VO: 5.143s A: 0.000s Sys: 0.633s = 61.619s
BENCHMARK%: VC: 90.6263% VO: 8.3472% A: 0.0000% Sys: 1.0265% = 100.0000%


Kas1e Beta 9

BENCHMARKs: VC: 57.911s VO: 13.490s A: 0.000s Sys: 0.335s = 71.736s
BENCHMARK%: VC: 80.7280% VO: 18.8052% A: 0.0000% Sys: 0.4668% = 100.0000%


Cheers

Go to top
Re: Mplayer Altivec test version for G4 and PA6T-1682M
Just can't stay away
Just can't stay away


See User information
@samo79

Quote:
No, with the audio activated i have the same crash

Quote:
Symbol info:
Instruction pointer 0x6C60E780 belongs to module "mplayer-joerg-beta4" (HUNK/Kickstart)

Stack trace:
module mplayer-joerg-beta4 at 0x6C60E780 (section 5 @ 0x7AA75C)
It's a bug in libSDL, it crashes in the SDL_SYS_CreateThread() function.

Go to top
Re: Mplayer Altivec test version for G4 and PA6T-1682M
Home away from home
Home away from home


See User information
@joerg

Quote:
It's a bug in libSDL, it crashes in the SDL_SYS_CreateThread() function.


Ok undestand, Roman recently fixed an unrelated issue in SDL, if you have an idea of how to fix this one i think you can report your SDL problem to him

Go to top
Re: Mplayer Altivec test version for G4 and PA6T-1682M
Just can't stay away
Just can't stay away


See User information
sdl ao should be fixed, at least it doesn't crash any more: http://www.sendspace.com/file/zogbjy
No vo changes.


Edited by joerg on 2014/4/19 23:14:48
Go to top
Re: Mplayer Altivec test version for G4 and PA6T-1682M
Home away from home
Home away from home


See User information
@joerg

Yep no crash now, but sound seems completely broken :-/

Go to top
Re: Mplayer Altivec test version for G4 and PA6T-1682M
Quite a regular
Quite a regular


See User information
@samo79

Same here, mostly just get static.
Do hear the odd sound on some clips.
Only briefly though and totally distorted.

Cheers

Go to top
Re: Mplayer Altivec test version for G4 and PA6T-1682M
Home away from home
Home away from home


See User information
@TiredOfLife

Yep

@Joerg

As we are here still also another long standing bug in sound, this one was present also in the original version from afxgroup

In menu:
Audio Option --> Mute

Doesn't work

Go to top
Re: Mplayer Altivec test version for G4 and PA6T-1682M
Just can't stay away
Just can't stay away


See User information
@samo79

Quote:
@Joerg

As we are here still also another long standing bug in sound, this one was present also in the original version from afxgroup

In menu:
Audio Option --> Mute

Doesn't work
That and the "Stay on Top" menu can only work if the ARexx support is built in, but I didn't add it to my MPlayer version. I'll remove the menu items which depend on ARexx. My new vo_pip.c is based on the old vo_p96_pip.c I implemented for Andrea's MPlayer 1.0-pre8, I don't use any code from the code.google.com MPlayer, the only exception in beta5 was ao_sdl.c and that may be the reason it doesn't work correctly, I'll switch back to the original MPlayer version.

Go to top
Re: Mplayer Altivec test version for G4 and PA6T-1682M
Home away from home
Home away from home


See User information
@joerg

Quote:
That and the "Stay on Top" menu can only work if the ARexx support is built in, but I didn't add it to my MPlayer version.


Ok, so that means the "mute" problem in the old version from afxgroup was caused by a bug in his ARexx implementation ?

Quote:
I don't use any code from the code.google.com MPlayer


But do you plan to do so at some point or atleast commit/merge your changes in Google site ?
Current code in Google already have all the missing parts you might need and that parts were present in the old version from Andrea but not yet on yours (double click for fullscreen, drag&drop, ARexx etc..)

Quote:
the only exception in beta5 was ao_sdl.c and that may be the reason it doesn't work correctly, I'll switch back to the original MPlayer version.


Instead of SDL (that it's also broken now) do you think it will be possible to use the AHI code from the kas1e version ?



Go to top
Re: Mplayer Altivec test version for G4 and PA6T-1682M
Just can't stay away
Just can't stay away


See User information
@samo79

Quote:
Current code in Google already have all the missing parts you might need
Except for MPlayer itself
I wont reimplement my pip vo for the old MPlayer version there.

Quote:
Instead of SDL (that it's also broken now) do you think it will be possible to use the AHI code from the kas1e version ?
I can't use anything from the code.google.com MPlayer or any other old MPlayer port as long as they aren't updated to the to the current MPlayer version, using the rest of the old code with a current MPlayer version will very likely cause even more problems than using the code.google.com ao_sdl did.

Go to top
Re: Mplayer Altivec test version for G4 and PA6T-1682M
Home away from home
Home away from home


See User information
@joerg
Quote:

I switched to the one from https://code.google.com/p/os4sdl/ because Roman told me the OS4depot.net version is old and too buggy.

If you are using the os4depot.net SDL and ao_sdl works in your MPlayer versions it's obviously the other way round


os4depot version of libsdl are old. Last one are on code.google and should be used 100%, there was some necessary bugs fixed. I use it with all my ports, and also for sdl output in muimplayer (together with SDL files from latest official mplayer sources) and everything works, and no crashes in terms of sound.

Quote:

I can't use anything from the code.google.com MPlayer or any other old MPlayer port as long as they aren't updated to the to the current MPlayer version, using the rest of the old code with a current MPlayer version will very likely cause even more problems than using the code.google.com ao_sdl did.


ahi_dev output will works without problems with any version (at least i tested with few). I can test and with latest one, but i assume that all make no sense to worry about sdl fixes, or about ahi addon. Just skip all those suggestions imho, as all what we need its your new vo_p96_pip.c shared somewhere (does not matter where and how), so i from my side can just implemnt it to muimplayer with no probs.

Or, if you want to check it with sound, just grab ahi_dev.c , it will works for sure with any mplayer version. At least, the fact that i use sdl sound output files (sdl_common.c and ao_sdl.c) , from mplayer sources of 04.04.2014, that mean the way of handling sound outputs are the same and didn't changed, so ahi_dev from google code can be used (basically, that ahi_dev is not "from google code", but its from morphos version of mplayer, liveforit just for some reassons submit it on google code).

@samo

Quote:

Apart the new p96 driver, do you think it will be possible to update to the new r37148 revision code (updated by Joerg) to be use directly into your current MUI port ?

Maybe it's possible that using a recent version some problems related to the corrupted video will be "auto-fixed"...


Yep, probably core can be updated later. But i feel demotivated now, because liveforit didn't share vo_comp, in which i was in interest, and because of what i share with him fixed cgx_wpa. Probably we need someone else (again and again) who can make vo_comp and 100% share code.

And btw, imho don't throw on joerg all possible bugs from everywhere, we lucky that he works on vo_p96pip.c improvements, and cleary you should't expect new mplayer with every possible addon in it. Later you will ask him to write gui ?:) Let's just wait when he done with p96 overlay drivers, and be in hope he will share that with us, by any of ways, and there will be no stop factors to share code.

And, sadly, we seems will not have vo_comp.c sources, so probably if someone want to see muimplayer with it, we need someone else who will make it and normally share what he do. Damn, the same and the same again and again. Probably 120 version of mplayer is better :)

@Joerg
Will you in interst to make vo_comp.c , but with 100% sharing of code ? (we can make some bounty, etc).


Edited by kas1e on 2014/4/20 16:21:27
Edited by kas1e on 2014/4/20 16:28:43
Edited by kas1e on 2014/4/20 16:30:30
Join us to improve dopus5!
AmigaOS4 on youtube
Go to top
Re: Mplayer Altivec test version for G4 and PA6T-1682M
Home away from home
Home away from home


See User information
@kas1e

Quote:
but i assume that all make no sense to worry about sdl fixes, or about ahi addon. Just skip all those suggestions imho, as all what we need its your new vo_p96_pip.c shared somewhere


Quote:
And btw, imho don't throw on joerg all possible bugs from everywhere, we lucky that he works on vo_p96pip.c improvements, and cleary you should't expect new mplayer with every possible addon in it. Later you will ask him to write gui ?:)


Yeah that's true, i ask him only considering the current situation, but i know that theorically we wouldn't have any need of an entire new (full) version from him but just his updated components..

2 versions (at max) would be more than enough imho, your separated MUI port AND the afxgroup one (that could be just replaced with a final merged based release containing the latest modifications of both forks of LiveForIt and Joerg)

This in an ideal world of course ..

Quote:
Yep, probably core can be updated later. But i feel demotivated now, because liveforit didn't share vo_comp, in which i was in interest, and because of what i share with him fixed cgx_wpa. Probably we need someone else (again and again) who can make vo_comp and 100% share code.


Quote:
Probably 120 version of mplayer is better :)


Argg hope not, It would be better to just convince them to share their source

Go to top

  Register To Post
« 1 ... 12 13 14 (15) 16 17 18 ... 30 »

 




Currently Active Users Viewing This Thread: 1 ( 0 members and 1 Anonymous Users )




Powered by XOOPS 2.0 © 2001-2023 The XOOPS Project