Re: AMD64 64bit vs 32bit speed increase
On Fri, Feb 13, 2004 at 07:43:29PM -0500, C. Scott Ananian wrote:
> On Sat, 14 Feb 2004, Paul Brook wrote:
> > -msse2 -mfpmath=sse2. It's possible that oggenc also has custom sse based
> > routines which are not enabled when compiled for i386 (the default x86 debian
> > arch).
> I looked for these, but could not find any. I think the differences in
> floating-point code generation are most likely responsible.
This is another good reason for multiarch. making it possible to have different
compile arguments to support the different instruction sets, SSE, SSE2, 3DNow
I agree that, for the most part, these will not make a major change in overall
system performance but things like oggenc are a great example of the times you
would notice it.
I would be very happy if the next time to go to 'AutoRip' a CD I have to wait
for a much shorter time.