[MPlayer-dev-eng] non-free and undistributable files in tremor/

Hervé W. H.O.W.aka.V+mplayer at gmail.com
Thu Aug 3 02:08:41 CEST 2006


On 02/08/06, Diego Biurrun <diego at biurrun.de> wrote:
> On Wed, Aug 02, 2006 at 11:44:15AM +0200, Michael Niedermayer wrote:
> >
> > On Wed, Aug 02, 2006 at 11:21:16AM +0200, Diego Biurrun wrote:
> > >
> > > I sat down and did some benchmarks on my K6-III 500, Debian stable
> > > system with all services except getty instances stopped, running Linux
> > > 2.4.32 and MPlayer r19262.  I used the following two samples to test,
> > > the first is 10 minutes long, the second 70 minutes
> > >
> > > ftp://ftp.openbsd.org/pub/OpenBSD/songs/song37.ogg
> > > http://jdolan.dyndns.org/jaydolan/choons/Jay%20Dolan%20-%20Live%20(6-17-04).ogg
> > >
> > > I compared the internal tremor decoder with the FFmpeg Vorbis decoder
> > > and internal tremor with --enable-tremor-low.  Below are the results.
> > >
> > > It seems like ffvorbis still has some ground to cover.  Now if somebody
> > > could sit down and optimize it...
> >
> > could you (or someone else) do some more finegrained benchmarks
> > (START/STOP_TIMER around mdct, bitparsing, ...) or gprof?
> >
> > btw, which gcc version did you use? if it was 2.95, maybe a test with
> > latest gcc would be interresting too ...
>
> Sorry, I forgot to mention the gcc version, silly me.  It was 3.3.5 from
> Debian stable.
>
> Diego

What was configure's output for the line: "Checking for mm3dnow.h"
If it wasn't yes, copying the file mm3dnow.h from a gcc 4.x system (or
tarball or...) could cause better benchmark results for the lavc
decoder. Might be interesting.

-V



More information about the MPlayer-dev-eng mailing list