Author: Vincent Diepeveen
Date: 17:01:43 11/11/01
Go up one level in this thread
On November 11, 2001 at 09:48:44, K. Burcham wrote: > > >that sounds good uri, but that is what i always wonder. >when i analyse a position or a complete game with one of my programs, >how can we know the best gm move. just because a gm made a decision to >choose a certain move, are we to assume that choice is the best move >because it is a gm move. and then if a game is gm vs gm, if it is a loss >then what. since it is a loss to we assume the winning gm made correct moves. >after spending so much time anaylizing kasparov positions against deep blue, >it seems kasparov had several positions he played into that he did not make the >right choices. >so in gm games maybe we can say each move may be a gm level move but not always >the best move. of course with this level of play being so high, it is difficult >for most to decide between a "bad" or "good" gm move. but we do have our >programs. Nothing gets as well analysed like matches kasparov-kramnik. you're probably talking about the Seirawan and Nunn type commented games, where definitely not bad GMs, with undoubtful strategic deep insight in the game and with a chessprogram at their laptop hitting 8 ply or something, use that as the 'tactical verification' of their strategic proposed move. In short kind of worlds worst advanced possible player getting used to write analysis on the fly :) >kburcham
This page took 0.01 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.