Author: Derek Paquette
Date: 23:52:32 04/18/04
Go up one level in this thread
On April 19, 2004 at 02:38:27, Uri Blass wrote: >On April 19, 2004 at 02:07:10, Kurt Utzinger wrote: > >>On April 18, 2004 at 23:58:17, Larry Scott wrote: >> >>> >>> A few years ago the famed Eygptologist Dr. Robert Huebner successfully drew a >>>match against Deep Fritz6 Running on a Dual pent 800. The match was rather dull, >>>Huebner seemed to effortlessly draw the monster fritz. could this be repeated >>>today? >> >> >> No computer program will be able to beat a GM if the latter's >> aim is only to draw (provided a very boring playing style is used). >> Kurt > >By the same logic no GM is going to beat another GM in case that the aim of the >second GM is only to draw(provided a very boring style is used). > >If it is not the case then writing the right book for the program is enough to >win the game(you only need to guess the moves of the losing GM correctly and >prepare the right book and remember that in matches of machines against humans >the machines do not use the commercial book). > >Uri Another problem is that programs in man vs machine or organized tournaments with machines and Gm's the program always resigns too early, it resigns when its 'reasonable' to assume that it has lost, but infact not lost the program doesn't mind playing for another 45 moves, but it seems its done out of 'courtesy' It seems the program resigns once its 'over the hill' but no one has ever had to fight for the win, the GM might blunder in the next 45 moves, or get into time trouble, they don't use the computers best strengths to the advantage
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.