[Ffmpeg-devel] [PATCH] another vorbis optimization
Luca Barbato
lu_zero
Thu Aug 10 20:26:30 CEST 2006
Loren Merritt wrote:
>>
>> what about moving bias somewhere else (like in the vorbis_context) and
>> spare a branch?
>
> I guess it's cleaner that way. The difference in speed is not measurable
> (perfectly predictable branches are cheap).
ok
>
> One branch (perfectly predictable) vs saving 224 integer additions when
> non-simd. Well, if you value simplicity of the code over speed in this
> less-common case (and I guess I do too, else I would've mmxed this
> loop), then I can remove it.
>
we could check if what's better for each arch.
>
> No, the commented-out version is faster here. But the question that
> matters is: Which is faster on cpus that don't have sse nor 3dnow? And I
> can't benchmark that.
I'll benchmark on ppc/ppc64 now, there is a missing bit for vorbis.h I
think.
lu
--
Luca Barbato
Gentoo/linux Gentoo/PPC
http://dev.gentoo.org/~lu_zero
More information about the ffmpeg-devel
mailing list