Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Is It really so, that D.Fritz was better later because of changes?

Author: Yatheen Manicka

Date: 19:10:14 10/26/02

Go up one level in this thread


I believe that the Deep Fritz team was allowed to work on the Opening Book (and
they did) in the course of the match. By choosing openings that avoided an early
exchange of queens, they could keep the match tactically complicated. In
addition, they could have simply instructed Deep Fritz to avoid even queen
exchanges in the middlegame by just increasing the material value of Fritz's own
queen. This is similar to what Kurt Utzinger did with ChessMaster. This change
can be made without needing to alter the program code itself (which was
prohibited).
My point is: Deep Fritz has demontsrated how to play effective anti-human chess
in its last 4 games. Even if Kramnik had not bludered in game 5 and fought for a
draw in game 6, Deep Fritz did not come close to losing in the last 4 games.
Hence, its anti-GM strategy was in fact effective. Deep Junior can simply
emulate this approach from game 1 of its match against Kasparov, effectively
preventing any human wins.
Ed



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.