Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Best Algorithm Variation

Author: Ed Trice

Date: 12:37:39 04/27/04

Go up one level in this thread


If your evaluation function is just material, you will search very deeply, but
the quality of the moves in the p.v. will be less than desireable. You can fold
in bonuses and merit in a pre-evaluator to help with move ordering, but from the
sounds of it, you want to test your program without intermediate goals encoded.

Large hash tables will be of help, but the history hueristic and move sorting
really work synergistically with good leaf node evaluation routines.



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.