On Tuesday 28 December 2004 03:46 pm, Daniel Thor Kristjansson wrote: > On Tue, 28 Dec 2004, Isaac Richards wrote: > ]Won't be applied. Anything but the default compile settings (with the > ]exception of different arches) is unsupported. > > This patch removes the -mfpmath=sse performance recommendation, but > keeps the additional archs, such as the winchip, the updated C3 info, > and other Intel and AMD processors.
Have you profiled on each of those processors to see if setting the arch actually creates faster code? I've seen -march=athlon-xp create 50% slower code than -march=pentiumpro with a recent gcc..
Isaac _______________________________________________ mythtv-dev mailing list mythtv-dev@mythtv.org http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-dev
With this version of gcc
gcc (GCC) 3.4.2 20041017 (Red Hat 3.4.2-6.fc3)
I obtained similar results with pentiumpro and athlon-xp (though the athlon code was a bit smaller); but I got a big boost (from 18% of idle time to 25 %) when adding HAVE_3DNOW to MMX and also putting -m3dnow on the C flags (still with pentiumpro arch).
I have a athlon XP 2000+, and hauppage card, everything is encoded in RTJpeg.
Bye
Manu
pgphFfTrW8akS.pgp
Description: PGP signature
_______________________________________________ mythtv-dev mailing list mythtv-dev@mythtv.org http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-dev