Author: Volker Böhm
Date: 01:32:15 02/28/06
Go up one level in this thread
Hi, there is no real reason to not implement learning. But learing is not important at all for a chess engine thus it will be implemented if you don´t have other things to do. There are too many chess positions to be able to learn on a relevant part of them (you just have to implement a large book with radom selection of moves). Thus learing does only solve the following points: 1. Somebody trying to find a draw or a win by throwing the engine out of book very early (example 1. a3) and try similar moves as long as he needs to find a way to win. 2. Contests with engine vs. engine with a large amount of games between the same engines using a small book. Greetings Volker
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.