David Cary Hart wrote: > On one hand, the FC3 packages are compiled with the "mtune=pentium4" > flag yet the kernel is distributed based on PentiumPro. Then it has the > additional overhead of the generic x86 code. Dave Jones replied: > You'll be hard pressed to find a real-world example where this makes a > noticable difference. Even microbenchmarks will likely be in the less > than 5% range. What about the 4G/4G split? Are we still where we were in April over the performance losses due to that? (In http://www.redhat.com/archives/fedora-test-list/2004-April/msg01888.html Alan Cox said that "its also about 10% slower on my laptop" and you replied: > It's somewhat strange that a few weeks ago when 4g/4g was temporarily > disabled, absolutely *no-one* said "Hey, it got 10% faster!" ) James. -- E-mail address: james | "Drums must never stop. Very bad if drums stop." @westexe.demon.co.uk | "Why? What will happen if the drums ever stop?" | "Bass solo."
