Author: Gian-Carlo Pascutto
Date: 04:20:16 05/06/02
Go up one level in this thread
On May 06, 2002 at 07:15:41, Vincent Diepeveen wrote: >On May 05, 2002 at 21:56:17, James Robertson wrote: > >>On May 05, 2002 at 19:25:07, stuart taylor wrote: >> >>>I feel that if a 1995 program can beat a top program of today in even one game, >>>that means that there are aspects which that older program knows better than the >>>new one, or why else would it win? >>>S.Taylor >> >>You can test this theory by playing a program against itself. =) >> >>James > >that is not a good test, when the version numbers are close >to each other. He was really talking about identical versions, i.e. no difference. Although they're identical, in a selfplay match they won't always draw. That obviously doesn't mean the version that won has more chess understanding than the other - they're identical! -- GCP
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.