Author: Bruce Moreland
Date: 11:28:44 05/13/98
Go up one level in this thread
On May 12, 1998 at 14:30:31, Danniel Corbit wrote: >I am a programmer, but I admit I know very little about computer chess >right now. I am trying to learn. At any rate, for any GM, we can get >the PGN of every game they have every played. If we analyze every move >to great depth and store the value of each position, I think that would >have great value. Surely, they have tendencies towards certain openings >and positions? My argument regards your use of the word "great". You can figure out what they play, but the odds of a successful ambush somewhere out in the game tree are slight. The tools for devising these ambushes are poor. A computer, even thinking for a week, is at a disadvantage in the highly *strategic* opening portion of the game. And the odds of finding a cook in the more tactical middlegame phase are very slight. When I was a beginner, I read a chess book, and was all hopped up and ready to go against my usual patzer opponent, I was dead sure he'd fall into an opening trap I'd prepared for him. Of course he varied five moves prior to the trap. The chess game tree is *big*. bruce
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.