Is it still just i686 optimized?
Or are we targeting P4 optimizations _specifically_ now?
And how is Athlon (which has a _signficant_ share) performance?
Furthermore, are "aggressive" SSE optimizations being used?
Or just the scheduling? It's not changing instructions to SSE,
correct?
It's a _critical_consideration_ to those of us who need
accuracy/precession over performance. I don't mind on the Athlon,
because the FPU gets leveraged _regardless_ of SSE instructions.
But several SSE instructions in the P4's native SSE pipes are well known
for not even being precise (different results with each run of the
_same_ data), let alone most are _not_ accurate ("lossy math" with
short-circuited logic).
--- RELEASE NOTE COMPARISION (FC2 v. FC3) ---
>From (FC3 Release Notes):
http://download.fedora.redhat.com/pub/fedora/linux/core/3/i386/os/RELEASE-NOTES-en.html
"Fedora Core 3 is optimized for Pentium 4 CPUs, but also supports
earlier CPUs (such as Pentium, Pentium Pro, Pentium II, Pentium III, and
including AMD and VIA variants). This approach has been taken because
Pentium-class optimizations actually result in reduced performance for
non-Pentium-class processors, and Pentium 4 scheduling is sufficiently
different (while making up the bulk of today's processors) to warrant
this change."
As compared to (FC2 Release Notes):
http://download.fedora.redhat.com/pub/fedora/linux/core/2/i386/os/RELEASE-NOTES-en.html
"Fedora Core 2 is optimized for Pentium PRO (and later) CPUs, but also
supports Pentium-class CPUs. This approach has been taken because
Pentium-class optimizations actually result in reduced performance for
non-Pentium-class processors."
--
Bryan J. Smith b.j.smith@xxxxxxxx
------------------------------------------------------------------
"Communities don't have rights. Only individuals in the community
have rights. ... That idea of community rights is firmly rooted
in the 'Communist Manifesto.'" -- Michael Badnarik
