[FFmpeg-devel] [PATCH] h264 parallelized

Michael Niedermayer michaelni
Fri Jun 29 12:27:02 CEST 2007


Hi

On Fri, Jun 29, 2007 at 10:57:50AM +0200, Andreas ?man wrote:
> Hi
> 
> Michael Niedermayer wrote:
> > Hi
> > 
> > ive rerun the benchmark and the values are below to make it more clear
> > 
> > svn  : 0m0.737s, 0m0.735s, 0m0.742s, 0m0.746s, 0m0.746s
> > patch: 0m0.756s, 0m0.763s, 0m0.757s, 0m0.761s, 0m0.752s
> > 
> > testing done with:
> > time ffmpeg/ffmpeg-threadslices -i test.h264 -f null /dev/null 2>/dev/null
> > 
> 
> Uww..  bad figures indeed.
> I am a bit confused why there is such a big difference since the patch
> does not poke around at all in the speed critical paths of the code.
> It must be some kind of cache issues or inlining changes..
> 
> On what CPU is this?

model name	: AMD Duron(tm) Processor
stepping	: 1
cpu MHz		: 807.391
cache size	: 64 KB

ive tried another file (Aladin.mpg 995 frames 352x240, the other file
was 538 frames 160x128)
svn  : 0m10.828s, 0m10.777s, 0m10.848s, 0m10.799s, 0m10.742s avg:10.799
patch: 0m10.770s, 0m10.777s, 0m10.831s, 0m10.918s, 0m10.778s avg:10.815

ill do more tests

[...]
-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

I am the wisest man alive, for I know one thing, and that is that I know
nothing. -- Socrates
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: <http://lists.mplayerhq.hu/pipermail/ffmpeg-devel/attachments/20070629/4c12a543/attachment.pgp>



More information about the ffmpeg-devel mailing list