[FFmpeg-devel] [RFC] H.264/SQV3 separation: h264data.h
Diego Biurrun
diego
Fri Nov 28 21:46:19 CET 2008
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:
> > Next up on the list is moving the tables in h264data.h to a .c file that
> > can be compiled and shared by the H.264 and SVQ3 decoder. This requires
> > adding an ff_ prefix to the names of the tables, removing the static
> > keyword and then moving h264data.h to h264data.c while replacing the
> > contents of h264data.h with appropriate extern declarations.
> >
> > I have done this in my local tree, but the patch is 174kB big (and
> > terribly boring), so I'm not sending it here. OK to proceed (in
> > several steps)?
>
> yes
>
> > 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?) 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.
Diego
-------------- next part --------------
mplayer-split -quiet -frames 1000 -benchmark -nosound -vo null cathedral-beta2-400extra-crop-avc.mp4
BENCHMARKs: VC: 27.329s VO: 0.021s A: 0.000s Sys: 0.261s = 27.611s
BENCHMARK%: VC: 98.9794% VO: 0.0766% A: 0.0000% Sys: 0.9440% = 100.0000%
BENCHMARKs: VC: 27.267s VO: 0.022s A: 0.000s Sys: 0.255s = 27.545s
BENCHMARK%: VC: 98.9935% VO: 0.0794% A: 0.0000% Sys: 0.9271% = 100.0000%
BENCHMARKs: VC: 27.590s VO: 0.021s A: 0.000s Sys: 0.261s = 27.873s
BENCHMARK%: VC: 98.9867% VO: 0.0764% A: 0.0000% Sys: 0.9369% = 100.0000%
mplayer-split -quiet -frames 1000 -benchmark -nosound cathedral-beta2-400extra-crop-avc.mp4
BENCHMARKs: VC: 26.697s VO: 4.159s A: 0.000s Sys: 0.236s = 31.092s
BENCHMARK%: VC: 85.8646% VO: 13.3769% A: 0.0000% Sys: 0.7585% = 100.0000%
BENCHMARKs: VC: 26.724s VO: 4.047s A: 0.000s Sys: 0.229s = 31.000s
BENCHMARK%: VC: 86.2068% VO: 13.0554% A: 0.0000% Sys: 0.7378% = 100.0000%
BENCHMARKs: VC: 26.827s VO: 4.051s A: 0.000s Sys: 0.236s = 31.114s
BENCHMARK%: VC: 86.2207% VO: 13.0196% A: 0.0000% Sys: 0.7596% = 100.0000%
mplayer-split -quiet -frames 1000 -benchmark -nosound -vo null Aladin.mpg
BENCHMARKs: VC: 12.223s VO: 0.021s A: 0.000s Sys: 0.300s = 12.545s
BENCHMARK%: VC: 97.4400% VO: 0.1681% A: 0.0000% Sys: 2.3919% = 100.0000%
BENCHMARKs: VC: 12.278s VO: 0.021s A: 0.000s Sys: 0.302s = 12.601s
BENCHMARK%: VC: 97.4393% VO: 0.1638% A: 0.0000% Sys: 2.3969% = 100.0000%
BENCHMARKs: VC: 12.220s VO: 0.021s A: 0.000s Sys: 0.293s = 12.534s
BENCHMARK%: VC: 97.4933% VO: 0.1706% A: 0.0000% Sys: 2.3361% = 100.0000%
mplayer-split -quiet -frames 1000 -benchmark -nosound Aladin.mpg
BENCHMARKs: VC: 12.253s VO: 1.559s A: 0.000s Sys: 0.314s = 14.125s
BENCHMARK%: VC: 86.7425% VO: 11.0377% A: 0.0000% Sys: 2.2199% = 100.0000%
BENCHMARKs: VC: 12.190s VO: 1.569s A: 0.000s Sys: 0.311s = 14.070s
BENCHMARK%: VC: 86.6324% VO: 11.1538% A: 0.0000% Sys: 2.2138% = 100.0000%
BENCHMARKs: VC: 12.215s VO: 1.567s A: 0.000s Sys: 0.315s = 14.097s
BENCHMARK%: VC: 86.6501% VO: 11.1156% A: 0.0000% Sys: 2.2343% = 100.0000%
-------------- next part --------------
mplayer-vanilla -quiet -frames 1000 -benchmark -nosound -vo null cathedral-beta2-400extra-crop-avc.mp4
BENCHMARKs: VC: 27.886s VO: 0.021s A: 0.000s Sys: 0.248s = 28.155s
BENCHMARK%: VC: 99.0450% VO: 0.0757% A: 0.0000% Sys: 0.8793% = 100.0000%
BENCHMARKs: VC: 27.822s VO: 0.021s A: 0.000s Sys: 0.251s = 28.095s
BENCHMARK%: VC: 99.0294% VO: 0.0764% A: 0.0000% Sys: 0.8942% = 100.0000%
BENCHMARKs: VC: 27.962s VO: 0.021s A: 0.000s Sys: 0.252s = 28.235s
BENCHMARK%: VC: 99.0329% VO: 0.0743% A: 0.0000% Sys: 0.8928% = 100.0000%
mplayer-vanilla -quiet -frames 1000 -benchmark -nosound cathedral-beta2-400extra-crop-avc.mp4
BENCHMARKs: VC: 27.427s VO: 4.061s A: 0.000s Sys: 0.230s = 31.719s
BENCHMARK%: VC: 86.4693% VO: 12.8047% A: 0.0000% Sys: 0.7260% = 100.0000%
BENCHMARKs: VC: 27.555s VO: 4.080s A: 0.000s Sys: 0.234s = 31.870s
BENCHMARK%: VC: 86.4616% VO: 12.8036% A: 0.0000% Sys: 0.7349% = 100.0000%
BENCHMARKs: VC: 27.581s VO: 4.032s A: 0.000s Sys: 0.233s = 31.846s
BENCHMARK%: VC: 86.6070% VO: 12.6622% A: 0.0000% Sys: 0.7308% = 100.0000%
mplayer-vanilla -quiet -frames 1000 -benchmark -nosound -vo null Aladin.mpg
BENCHMARKs: VC: 12.203s VO: 0.022s A: 0.000s Sys: 0.292s = 12.517s
BENCHMARK%: VC: 97.4926% VO: 0.1719% A: 0.0000% Sys: 2.3355% = 100.0000%
BENCHMARKs: VC: 12.256s VO: 0.020s A: 0.000s Sys: 0.293s = 12.569s
BENCHMARK%: VC: 97.5104% VO: 0.1602% A: 0.0000% Sys: 2.3295% = 100.0000%
BENCHMARKs: VC: 12.305s VO: 0.021s A: 0.000s Sys: 0.290s = 12.616s
BENCHMARK%: VC: 97.5286% VO: 0.1693% A: 0.0000% Sys: 2.3022% = 100.0000%
mplayer-vanilla -quiet -frames 1000 -benchmark -nosound Aladin.mpg
BENCHMARKs: VC: 12.232s VO: 1.562s A: 0.000s Sys: 0.305s = 14.099s
BENCHMARK%: VC: 86.7575% VO: 11.0805% A: 0.0000% Sys: 2.1620% = 100.0000%
BENCHMARKs: VC: 12.086s VO: 1.585s A: 0.000s Sys: 0.300s = 13.971s
BENCHMARK%: VC: 86.5054% VO: 11.3472% A: 0.0000% Sys: 2.1474% = 100.0000%
BENCHMARKs: VC: 12.240s VO: 1.603s A: 0.000s Sys: 0.294s = 14.136s
BENCHMARK%: VC: 86.5840% VO: 11.3363% A: 0.0000% Sys: 2.0797% = 100.0000%
More information about the ffmpeg-devel
mailing list