[MPlayer-dev-eng] Dithering 16-bit output for libmad

[D Richard Felker III]

On Sun, Jan 30, 2005 at 02:13:46PM +0300, Vladimir Mosgalin wrote:
> Since there are only few integer operations, the performance drop is
> very small, less than 0.1% of total cpu usage. It is about 10-50 times
> less than performance penalty from using libmad instead of mp3lib or
> from using lavcresample.

I like this a lot in principle, but I'm concerned about performance
still. Which cpus did you test this on, and how did you do the
benchmarks? I would expect it might be a lot more of a problem on
older ones with slow multiply, and it might make more of a difference
when playing movies too (not just .mp3 files) if cache coherency is an
issue and the extra code hurts cache performance.

Rich