Author: Robert Hyatt
Date: 19:45:12 06/24/02
Go up one level in this thread
On June 24, 2002 at 04:12:29, Bernhard Bauer wrote: >On June 23, 2002 at 20:18:30, Sean Mintz wrote: > >>Its been rumored to use bitboards. If ported correctly, then yes it would run >>faster. Possibly significantly faster. > >Why should it run faster? >It's not clear, at least to me, why a smp program should run faster on >16 procs than on 8 procs. >It may double the nps but not necessarily find a good move earlier. >Please note that SMP is not a massive parallel system. >SMP is limited. >Kind regards >Bernhard I don't understand the "SMP is limited". If you mean "number of processors" then that is partially true, although I should point out that there are many 32-processor SMP systems around, starting at the mighty Cray T932 at the top of the food chain... As far as running faster, all I can say is mine definitely does run faster. Every processor will make the search time drop a bit...
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.