Jerry Feldman wrote:
> Actually, I did a number of (unofficial) benchmarks a while ago when I
> worked for partner engineering at HP. While most of the benchmarks ran
> well at 64-bit, one of my partner's tested their applications and found
> their 32-bit versions were significantly faster (I think this one was on
> HP-UX, but may have been Linux). Additionally, we found on the Digital
> Alpha that there were some applications that were slower in 64-bits. The
> Alpha was a full 64-bit chip with no native 32-bit mode. For the most
> part, the personal workstation as you point out, won't make too much of
> a different. Just remember that a 64-bit OS can address the full 4GB of
> the OP's system where a 32-bit OS must use PAE to access over 3GB. This
> restriction only comes into play if you have a very memory intensive
> application.
x86 is a very different story than the Alpha. x86_64 adds more registers,
which is already enough to boost program speed (fewer memory accesses
needed). x86_64 also means at least SSE and SSE2 are guaranteed to be
there, so you also benefit from those, whereas on 32-bit x86 only a few
libs have sse2 versions available.
Kevin Kofler
--
fedora-list mailing list
fedora-list@xxxxxxxxxx
To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-list
Guidelines: http://fedoraproject.org/wiki/Communicate/MailingListGuidelines
