> >
> > How does everyone think about unifing float size to 32bit? It has merit to
> > x86, or other CPU?
>
> On x86, I found double to be not significantly slower than float with gcc.
> Given C's preference for doubles, I tend to code for double, even (especially)
> when perfomance is an issue.
>
> Monty
>
I am also tempted to make everything 64bit. But in my tests,
using all 32bit reals is about 4% faster. Right now lame uses
a mixture (inherited from the ISO code). Anyone can change this:
edit the definations of FLOAT and FLOAT8 in machine.h
Mark
--
MP3 ENCODER mailing list ( http://geek.rcc.se/mp3encoder/ )
