Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Question about Evaluation Functions

Author: Matt White

Date: 14:44:18 03/09/02

Go up one level in this thread


On March 06, 2002 at 11:54:44, William H Rogers wrote:

Is this a-la Turing's approach? I like the idea of using the branching factor to
determine, at least to an extent, what the mobility of pieces is. Has anyone
tried using this type of information in a move-ordering function rather than in
the eval itself? It seems to me to be more logical to use non-material features
in ordering moves, and strictly material features in evaluating. This would
probably allow deeper searching, no?

Thanks for your responses, and anticipating further response,
Matt

>What I used to do was to count all of the legal moves when generating moves and
>add its total to the eval function. As you switch sides, you also count the
>enemys moves also and add its total to the eval. As there are only about 40 or
>so legal moves per side it does not offset the score very much, except when you
>severely limit your opponents ability to move. It also can help keep you from
>getting into a position where you have very little moves on you next turn.
>My suggestion is to play with it.
>Bill



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.