Author: Robert Hyatt
Date: 10:34:22 05/18/01
Go up one level in this thread
On May 18, 2001 at 12:45:04, Olaf Jenkner wrote: >On May 18, 2001 at 09:38:18, Robert Hyatt wrote: > >>On May 18, 2001 at 08:51:55, Joshua Lee wrote: >> >>>What will the new chips do for chess? >> >> >>For bitboard programs, potentially a factor of 2x faster. For others, not >>much just for going to 64 bits. > >for 32-bit machines I know two ways of coding: > >1) >or eax,[m] >jnz @a >or edx,[m+4] >jnz @a >... >@a: > >2) >or eax,[m] >or edx,[m+4] >or eax,edx >jnz @a >... >@a: > >Does anybody know which is better? >It's difficult to guess because pipelining, pairing >and so on. > >Olaf The second is almost certainly better, as it has one less branch.
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.