[Ffmpeg-devel] Fixed vs. Floating Point AAC
Wed Mar 8 23:33:04 CET 2006
On Wed, Mar 08, 2006 at 01:14:54PM -0800, Mike Melanson wrote:
> Rich Felker wrote:
> >Including the 'low precision' mode in tremor (which makes no audible
> >difference, i.e. you won't be able to distinguish them with
> >doubleblind testing)?? IIRC it's between 1.5 and 2x as fast as the
> >normal mode in tremor.
> >BTW the testing I cited was on athlon.
> Can you outline your profiling process? 'top'? Something a little
> more accurate, perhaps?
Oded tested them on IRC a few weeks ago. He was comparing both to
ffvorbis and his own vorbis decoder implementation. IIRC he used
either mplayer -benchmark or time oggdec ... (I forget which).
More information about the ffmpeg-devel