Author: Mike Hood
Date: 06:10:04 07/01/03
Go up one level in this thread
On July 01, 2003 at 06:45:09, Thorsten Czub wrote: >On July 01, 2003 at 06:03:07, Christophe Drieu wrote: > >>On July 01, 2003 at 05:41:31, Graham Laight wrote: >> >>>When a GM is contemplating a move, he doesn't say to himself, "Hmmmmm. I would >>>give the resulting position a score of 1.723". >>> >>>Such an evaluation is nonsense anyway. There should properly be only 3 >>>evaluations: >>> >>>1. Winning position >>> >>>2. Drawing position >>> >>>3. Losing position >>> >>>It would be nice if a program could work as follows: >>> >>>"nb5. This position contains a possible bishop trap". >>> >>>"nd5. This puts more pressure on the opponent's king" >>> >>>"Opponent classification: bishop trap success rate = 25%" >>> >>>"Opponent classification: king attack success rate = 15%" >>> >>>"Choice = nb5". >>> >>>-g >>Rebel 12 (DOS) can give you some informations like that. >Chess System Tal (dos) is doing this. Yes... a few years ago Chris Whittington made some very interesting suggestions along the lines of the original post in this thread. To paraphrase his words from memory: "A move should be chosen by static evaluation of the current position based on positional considerations. Maybe a short search of three or four plies is necessary to avoid obvious blunders, but the evaluation should not be based on a depth search". Chris's ideas sounded very exciting when I first read them, and I have to admit that I was disappointed that the program based on these ideas, Chess System Tal, didn't perform as well as the deep searching bean counters. Maybe more research (or rather experimentation) should be done along this path.
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.