Author: Mogens Larsen
Date: 03:44:24 04/16/01
Go up one level in this thread
On April 16, 2001 at 05:33:47, Ed Schröder wrote: >What an utter nonsense is that. You and Enrique should know better than that. >It is like Christophe said, all arranged behind the curtains, no way back. >Discrimination of the worst kind. > >In no way it is proven that a multi-processor program is doing better than >a single-processor program. Just look at the latest Ausfess tournament Tiger >13.0 (not 14.0) topping above all the multi-processing entries: One argument for requiring that the programs be SMP capable is the publicity side of the contest. After Deep Blue and top performances by Deep Junior and Deep Fritz at GM tournaments, Rebel Century on a fast Athlon wouldn't sound very impressive in that context. Whether Rebel Century or the Tigers have any real chance against Deep Fritz on an eight cpu monster remains to be seen and I honestly doubt it. However, it would be fair if everyone was given the chance no matter how difficult it would be. Especially since they've abandoned ship on the rest of the SMP capable programs you mention as well, which doesn't imply an ambition to resolve the challenger question in an acceptable fashion IMO. >. Junior 4 x 500 Mhz >. Fritz 4 x 500 Mhz >. Ferret 4 x 400 Mhz >. Cilkchess 240 x Alpha at 250 Mhz >. P.ConNerS 186 x PII 450 Mhz >. Zugzwang 512 x Alpha at 300 Mhz Thanks for mentioning some of the important SMP capable programs. Are there others? Something called Dark Thought? Mogens.
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.