[FFmpeg-devel] [RFC] H.264/SQV3 separation: h264data.h
Diego Biurrun
diego
Mon Dec 1 11:32:25 CET 2008
On Mon, Dec 01, 2008 at 11:25:59AM +0100, Diego Biurrun wrote:
> On Sat, Nov 29, 2008 at 05:28:27PM +0100, Michael Niedermayer wrote:
> > On Fri, Nov 28, 2008 at 09:46:19PM +0100, Diego Biurrun wrote:
> > > On Fri, Nov 28, 2008 at 07:14:57PM +0100, Michael Niedermayer wrote:
> > > > On Fri, Nov 28, 2008 at 05:48:50PM +0100, Diego Biurrun wrote:
> > > > > Will this need to be benchmarked at some point?
> > > >
> > > > id say it cant hurt to benchmark it before commit
> > >
> > > Hmmmmm
> > >
> > > I ran the following benchmarks on my K6-III with two different samples:
> > >
> > > http://samples.mplayerhq.hu/V-codecs/h264/cathedral-beta2-400extra-crop-avc.mp4
> > > http://samples.mplayerhq.hu/V-codecs/h264/Aladin.mpg
> > >
> > > I used the following MPlayer command lines to benchmark:
> > >
> > > mplayer -quiet -frames 1000 -benchmark -nosound -vo xmga
> > > mplayer -quiet -frames 1000 -benchmark -nosound -vo null
> > >
> > > I made five runs for each sample and vo and threw away the best and
> > > worst times. The results are attached.
> > >
> > > Unfortunately there seems to be a slight slowdown with the cathedral
> > > sample, with Aladin it's a bit inconclusive.
> > >
> > > Now I don't know how to continue from here. Were my benchmarking
> > > procedures sensible? (BTW, is there a good way to benchmark with ffmpeg
> > > directly?)
> >
> > time ffmpeg ... >&/dev/null
>
> Sure, but what options would I use?
>
> > > Are these intermediate results worth anything at all or
> > > should the benchmarks be done once the split is complete to be
> > > significant?
> > >
> > > I remember that WMV2 decoding times improved after the split from
> > > msmpeg4. Please point me in the right direction.
> >
> > iam not sure, one thing that may or may not help is figuring out which
> > table it is that causes the slowdown. I dont really belive it is
> > a specific table, rather gcc doing something randomly silly.
>
> I ran the test on my K6-III with gcc 4.1, but I could reproduce similar
> numbers on my G4 with gcc 4.3. So all in all this might actually be for
> real.
>
> I redid the benchmarks with the ugly but complete split I have in my
> local tree:
>
> http://www1.mplayerhq.hu/~diego/split.diff
>
> The numbers are attached. It still appears to be slower, but not that
> much.
>
> Since intermediate numbers do not appear to paint too accurate a picture
> my suggestion is to go ahead with the split and start comparing when it
> is finished. Michael?
.. and here are the numbers I forgot to attach as usual ..
Diego
-------------- next part --------------
mplayer-vanilla -quiet -frames 1000 -benchmark -nosound -vo null Aladin.mpg
BENCHMARKs: VC: 11.757s VO: 0.021s A: 0.000s Sys: 0.294s = 12.072s
BENCHMARK%: VC: 97.3898% VO: 0.1746% A: 0.0000% Sys: 2.4356% = 100.0000%
BENCHMARKs: VC: 11.744s VO: 0.021s A: 0.000s Sys: 0.284s = 12.048s
BENCHMARK%: VC: 97.4714% VO: 0.1756% A: 0.0000% Sys: 2.3530% = 100.0000%
BENCHMARKs: VC: 11.714s VO: 0.021s A: 0.000s Sys: 0.295s = 12.030s
BENCHMARK%: VC: 97.3718% VO: 0.1744% A: 0.0000% Sys: 2.4538% = 100.0000%
mplayer-vanilla -quiet -frames 1000 -benchmark -nosound -vo xmga Aladin.mpg
BENCHMARKs: VC: 11.606s VO: 1.580s A: 0.000s Sys: 0.285s = 13.471s
BENCHMARK%: VC: 86.1564% VO: 11.7275% A: 0.0000% Sys: 2.1161% = 100.0000%
BENCHMARKs: VC: 11.588s VO: 1.553s A: 0.000s Sys: 0.302s = 13.443s
BENCHMARK%: VC: 86.1995% VO: 11.5558% A: 0.0000% Sys: 2.2447% = 100.0000%
BENCHMARKs: VC: 11.589s VO: 1.575s A: 0.000s Sys: 0.297s = 13.461s
BENCHMARK%: VC: 86.0885% VO: 11.7035% A: 0.0000% Sys: 2.2080% = 100.0000%
mplayer-vanilla -quiet -frames 1000 -benchmark -nosound -vo null cathedral-beta2-400extra-crop-avc.mp4
BENCHMARKs: VC: 26.279s VO: 0.021s A: 0.000s Sys: 0.255s = 26.556s
BENCHMARK%: VC: 98.9602% VO: 0.0803% A: 0.0000% Sys: 0.9595% = 100.0000%
BENCHMARKs: VC: 26.452s VO: 0.021s A: 0.000s Sys: 0.256s = 26.729s
BENCHMARK%: VC: 98.9633% VO: 0.0775% A: 0.0000% Sys: 0.9592% = 100.0000%
BENCHMARKs: VC: 26.309s VO: 0.021s A: 0.000s Sys: 0.250s = 26.579s
BENCHMARK%: VC: 98.9824% VO: 0.0778% A: 0.0000% Sys: 0.9398% = 100.0000%
mplayer-vanilla -quiet -frames 1000 -benchmark -nosound -vo xmga cathedral-beta2-400extra-crop-avc.mp4
BENCHMARKs: VC: 25.665s VO: 4.030s A: 0.000s Sys: 0.239s = 29.935s
BENCHMARK%: VC: 85.7381% VO: 13.4634% A: 0.0000% Sys: 0.7985% = 100.0000%
BENCHMARKs: VC: 25.702s VO: 4.022s A: 0.000s Sys: 0.235s = 29.960s
BENCHMARK%: VC: 85.7892% VO: 13.4253% A: 0.0000% Sys: 0.7855% = 100.0000%
BENCHMARKs: VC: 25.783s VO: 4.047s A: 0.000s Sys: 0.231s = 30.060s
BENCHMARK%: VC: 85.7702% VO: 13.4616% A: 0.0000% Sys: 0.7682% = 100.0000%
-------------- next part --------------
mplayer-split -quiet -frames 1000 -benchmark -nosound -vo null Aladin.mpg
BENCHMARKs: VC: 11.741s VO: 0.021s A: 0.000s Sys: 0.296s = 12.059s
BENCHMARK%: VC: 97.3671% VO: 0.1754% A: 0.0000% Sys: 2.4575% = 100.0000%
BENCHMARKs: VC: 11.731s VO: 0.021s A: 0.000s Sys: 0.296s = 12.047s
BENCHMARK%: VC: 97.3714% VO: 0.1755% A: 0.0000% Sys: 2.4531% = 100.0000%
BENCHMARKs: VC: 11.785s VO: 0.021s A: 0.000s Sys: 0.299s = 12.105s
BENCHMARK%: VC: 97.3561% VO: 0.1718% A: 0.0000% Sys: 2.4721% = 100.0000%
mplayer-split -quiet -frames 1000 -benchmark -nosound -vo xmga Aladin.mpg
BENCHMARKs: VC: 11.622s VO: 1.555s A: 0.000s Sys: 0.300s = 13.477s
BENCHMARK%: VC: 86.2354% VO: 11.5391% A: 0.0000% Sys: 2.2255% = 100.0000%
BENCHMARKs: VC: 11.661s VO: 1.578s A: 0.000s Sys: 0.297s = 13.537s
BENCHMARK%: VC: 86.1477% VO: 11.6575% A: 0.0000% Sys: 2.1947% = 100.0000%
BENCHMARKs: VC: 11.688s VO: 1.557s A: 0.000s Sys: 0.296s = 13.540s
BENCHMARK%: VC: 86.3178% VO: 11.4976% A: 0.0000% Sys: 2.1846% = 100.0000%
mplayer-split -quiet -frames 1000 -benchmark -nosound -vo null cathedral-beta2-400extra-crop-avc.mp4
BENCHMARKs: VC: 26.625s VO: 0.021s A: 0.000s Sys: 0.254s = 26.900s
BENCHMARK%: VC: 98.9769% VO: 0.0784% A: 0.0000% Sys: 0.9447% = 100.0000%
BENCHMARKs: VC: 26.573s VO: 0.021s A: 0.000s Sys: 0.253s = 26.847s
BENCHMARK%: VC: 98.9790% VO: 0.0798% A: 0.0000% Sys: 0.9412% = 100.0000%
BENCHMARKs: VC: 26.598s VO: 0.021s A: 0.000s Sys: 0.255s = 26.874s
BENCHMARK%: VC: 98.9704% VO: 0.0798% A: 0.0000% Sys: 0.9498% = 100.0000%
mplayer-split -quiet -frames 1000 -benchmark -nosound -vo xmga cathedral-beta2-400extra-crop-avc.mp4
BENCHMARKs: VC: 26.114s VO: 4.014s A: 0.000s Sys: 0.234s = 30.362s
BENCHMARK%: VC: 86.0090% VO: 13.2212% A: 0.0000% Sys: 0.7698% = 100.0000%
BENCHMARKs: VC: 26.195s VO: 4.036s A: 0.000s Sys: 0.232s = 30.463s
BENCHMARK%: VC: 85.9916% VO: 13.2481% A: 0.0000% Sys: 0.7603% = 100.0000%
BENCHMARKs: VC: 26.132s VO: 4.048s A: 0.000s Sys: 0.235s = 30.415s
BENCHMARK%: VC: 85.9174% VO: 13.3102% A: 0.0000% Sys: 0.7724% = 100.0000%
More information about the ffmpeg-devel
mailing list