Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: John McCarthy === AI as Sport

Author: Steve Coladonato

Date: 11:17:52 02/10/00

Go up one level in this thread


This article and the one by Stefan Meyer-Kahlen bring up some interesting
points.  It seems that chess programming is being solved in essentially the same
way by all programmers (generalization).  And it also seems that hardware
development is driving the increase in chess program strength, not the
algorithms in the program.  With this in mind, the "brute force" method of
solution seems to be the state of the art at this time.  And are not EGT's a
solution to the brute force method for endings?  Likewise, opening books are a
solution to the "brute force" method for the opening.  Given sufficient hardware
capabilities, these two areas could be solved with "brute force" programming.
Which leaves us the "middle" game as the current focus of algorithms.

I am not a chess programmer but I do understand that the current algorithms are
quite good at determining a "best" move for a given position.  Is the paradigm
that is used for these algorithms the way things should proceed?  Or, should the
programmers try to find other ways?

I am not attempting to criticize any of the work that has been accomplised so
far as it is rather amazing.  I am just wondering if the field has gotten into a
rut and `can't see the forest for the trees`?



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.