Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: The Litmus Test

Author: Dann Corbit

Date: 11:15:28 06/05/03

Go up one level in this thread


On June 05, 2003 at 10:22:16, Bob Durrett wrote:

>
>When the top GMs cannot win against a chess engine even with an unlimited number
>of takebacks, then you will have a good chess engine.  Not until.

Then the engine will have and must have solved the game of chess.  If the number
of takebacks is unlimited, then that is (mathematically) the only possibility.
Therefore, we will probably never have a "good" chess engine, by your definition
of good.

>There is a demand for a good chess engine.  It would be used as an analysis
>tool, not as a playing partner.

You will have a long wait.

My definition of "a good chess engine" is different than yours.  For instance,
Golem is an engine I can beat, so it isn't very strong.  Yet it plays pretty
well and doesn't crash or make wrong claims.  That's a good engine.  Ruffian is
an incredibly powerful engine that can tussle with even the professional
programs and it is free.  That's a good engine.  Chess Tiger is a very strong
engine that can easily be used in conjunction with the Covetka stuff for the CAP
project.  That's a good engine.  CM9000 costs 'lunch money', is very strong, and
has a lot of features.  That's a good engine.

Come to think of it, they are pretty much all good engines.  And I have about
250 of them.




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.