>
> Hi all,
> I never get around to optimizing the bit i'm working on, so i optimized
> quantize_xrpow instead of finalising the mdct thing.
>
> This is a speedup for quantize_xrpow that is sort of taken from my old
> quantize() speedup [see lame3.11 or previous].
> This will only speed up quantize_xrpow if you aren't on x86 using the
> asm. (the x86 asm loop still beats this). It's about 25% faster than before.
>
Thanks Mike, I just added this. it should be in 3.16beta.
(Gabriel - I will try to get 3.16 on the web site this weekend. As you
saw on the web site, it has some new tunings of the bit allocation
from Jan Rafaj which help testsignal2 & testsignal4. On testsignal2,
Jan claims LAME is now a little better than the newest Xing coder,
but FhG still has the edge)
Mark
--
MP3 ENCODER mailing list ( http://geek.rcc.se/mp3encoder/ )
