[MPlayer-dev-eng] liba52 0.7.4

[Dominik 'Rathann' Mierzejewski]

On Saturday, 14 January 2006 at 19:23, Diego Biurrun wrote:
> On Sat, Jan 07, 2006 at 01:29:41PM +0100, Emanuele Giaquinta wrote:
> > On 1/7/06, Dominik 'Rathann' Mierzejewski <dominik at rangers.eu.org> wrote:
> > 
> > > Please do, and tell us what didn't get integrated (since you said "almost all").
> > 
> > Apart from the changes already integrated upstream, I've not ported the ones to
> > the non accelerated imdct in imdct.c (it has been totally rewritten).  I've
> > tested it succesfully with the accelerated resample/downmix/imdct on
> > ppc/altivec and x86/mmx+sse.  To test it create a new liba52 directory and do
> > the following:
> 
> I'd appreciate if somebody could look into this.  We have a bug in
> Bugzilla about failing AC3 samples that a newer version of liba52 should
> hopefully be able to play.  Furthermore an upgrade promises much
> improved performance.
> 
> Benchmarks comparing old and new version would be good.

My CPU is AMD Athlon Thunderbird 1GHz, MMX2, 3DNowEx, no SSE.
MPlayer compiled with -O3 -ffast-math -march=athlon (which is the default).

With 5:32 long 448kbit/s AC3 5.1 sample, the new liba52 is slower:

for i in `seq 1 5` ; do mplayer -channels 6 -benchmark -novideo -ao \
pcm:nowaveheader:file=/dev/null 51.ac3 ; done

$ benchavg.pl <mplayer-old-AC3-5.1.log
5 results processed
average  VC is: 0.000s
average  VO is: 0.000s
average  AO is: 7.294s
average SYS is: 0.268s
average tot is: 7.563s

$ benchavg.pl <mplayer-new-AC3-5.1.log
5 results processed
average  VC is: 0.000s
average  VO is: 0.000s
average  AO is: 7.740s
average SYS is: 0.291s
average tot is: 8.030s

This is about 6% slower.

With 23:31 long 192kbit/s AC3 2.0 sample, the new liba52 is slower, too:

for i in `seq 1 5` ; do mplayer -channels 2 -benchmark -novideo -ao \
pcm:nowaveheader:file=/dev/null 20.ac3 ; done

$ benchavg.pl <mplayer-old-AC3-2.0.log
5 results processed
average  VC is: 0.000s
average  VO is: 0.000s
average  AO is: 11.451s
average SYS is: 0.406s
average tot is: 11.857s

$ benchavg.pl <mplayer-new-AC3-2.0.log
5 results processed
average  VC is: 0.000s
average  VO is: 0.000s
average  AO is: 12.429s
average SYS is: 0.413s
average tot is: 12.842s

This is about 8% slower.

Maybe someone can provide samples for which the old liba52 fails and the
new one works.

Regards,
R.

-- 
MPlayer RPMs maintainer: http://rpm.greysector.net/mplayer/
"I am Grey. I stand between the candle and the star. We are Grey.
 We stand between the darkness ... and the light."
        -- Delenn in Grey Council in Babylon 5:"Babylon Squared"