Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Khalifman and Gelfand on computer

Author: ujecrh

Date: 13:22:40 05/22/00

Go up one level in this thread


On May 20, 2000 at 13:16:29, blass uri wrote:

>
>I believe that the main reason that the programmers of Fritz and Junior did not
>care about adding knowledge in the opening(like not getting out with the queen)
>is the fact that they know from experience that they do not play without book.
>

I completely agree with this.

A good example is the notion of tempo in the opening (Nimzovitsh nicely explains
this in "My system"). This is really not difficult to write some code so that
the chess program knows how to gain (or avoid a loss of) tempo in the opening
(without tactical reasons for it of course) but I know commercial programs that
do not have this kind of evaluation. They simply, for instance, move pieces
again and again if, as far as their evaluation is concerned, this is the best
move.

If a programmer wanted to play a better opening without any book then he would
simply add this kind of knowledge to his engine. Not difficult but also not
necessary as long as book takes care of it.

Ujecrh



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.