Author: Ernst Walet
Date: 11:10:26 12/11/01
Go up one level in this thread
On December 10, 2001 at 10:24:40, Vincent Diepeveen wrote: >On December 10, 2001 at 10:21:08, Bas Hamstra wrote: > >>On December 10, 2001 at 10:16:25, Vincent Diepeveen wrote: >> >>>On December 10, 2001 at 08:05:42, Vincent Diepeveen wrote: >>> >>>>On December 10, 2001 at 06:02:46, Bas Hamstra wrote: >>>> >>>>>On December 10, 2001 at 04:13:05, Andrew Dados wrote: >>>>> >>>>>>On December 10, 2001 at 03:40:53, Russell Reagan wrote: >>>>>> >>>>>>>At http://www.ast.cam.ac.uk/~cmf/chess/theory.html it says: >>>>>>> >>>>>>>"Most top chess programs these days use bitboards." >>>>>>> >>>>>>>Is this true? >>>>>> >>>>>>No. The opposite is true. >>>>>> >>>>>>non bitboard programs: >>>>>>commercial: tiger, fritz, junior, diep, rebel, genius, nimzo >>>>>>amateur: most >>>>>>ferret non-bitboard, too. >>>>>> >>>>>>bitboard programs: afik none commercial. crafty, insomniac and few more. >>>>>> >>>>>>then to some extend all programs may use bitboards especially to carry around >>>>>>pawn structure info. >>>>>> >>>>>>-Andrew- >>>>> >>>>>Correction: Fritz uses bitboards. Non rotated. >>>>> >>>>>Bas. >>>> >>>>No it doesn't. >>> >>>Correction, i got email from Frans his official statement nowadays >>>is that he IS using bitboards. >> >>Not only that, but he even claims BB datastructure is not slower. Quite >>something from mr. efficiency, eh? >> >>Bas. > >I wonder why Fritz is slower on the K7 than on the P3. > >On my paper bitboards should be much faster on K7 as you can >do BSF/BSR to a 64 bits register without branch mispredictions, >and i don't see how to do that on the P3. > >Best regards, >Vincent By combining MMX registers, as he told me at Leiden. Ernst.
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.