Author: Peter McKenzie
Date: 17:15:02 01/13/99
Go up one level in this thread
On January 13, 1999 at 04:24:30, Marc-Philippe HUGET wrote: >Hi everybody, > > In this moment, I test my program without books, I think it is a good meaning >in order to test quality of evaluation : pwan development, pieces development, >etc. Are you agree with me ? > For example, Crafty often begins with Knight Nf3 or Nf6 or Nc6. > > This idea is very strange, isn't it ? > > Opinion ? > > >Marc-Philippe Huget I think I've posted before on this topic but since you asked, here is my opinion: I think testing a program without an opening book is only of limited use at best. Why spend time on getting your program to play better in the early stages of the game, when you could be spending time on getting it to play better in the middle and later stages of the game? Sure, if your program wants to play the Ruy Lopez all by itself then thats cool but it may or not mean it will play the middlegame well. By letting the book take care of the first few moves, we can concentrate directly on the positions that really matter. Also, a decent book will have a good amount of variety while a deterministic program without a book may get stuck in a narrow range of positions. I haven't heard of any of the top chess programmers testing without opening books. cheers, Peter
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.