Author: James C. Logan
Date: 13:33:37 05/17/05
Go up one level in this thread
>No engine that was playing without an opening book would play 1.a3. I would >suspect that all top engines would either play 1.e4, 1.d4 or 1.Nf3, depending on >the time control. > >I would think that NO opening book would be much better than a book in which >every possible opening move had an equal priority. > >jm Does anyone think that a computer can amass in minutes all the knowledge about opening theory that people have collected over years. Humans can remember what we read, and thus should have an advantage against a "no book" program. Maybe not too far into the opening though. Would it make sense to let the computer play through a few scores of games for each opening using no book, with learning on, and then pitting it against a human? I imagine this to be equal to the program creating it's own book. Maybe it's too much work for someone to hand feed the computer, especially if eventually called upon to make a refutation to a new move made by the program. I am not a programmer, just an observer, so I hope my comments are not too off target.
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.