Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Contrast in playing strength.

Author: Amir Ban

Date: 07:19:05 09/01/98

Go up one level in this thread


On August 31, 1998 at 17:57:02, Mark Young wrote:

>Is computer Vs computer testing now useless in gauging a chess program’s
>strength playing humans? When Crafty gets killed playing Junior 5 by a wide
>margin. And Fritz 5 draws a match with Rebel 10 even when Rebel 10 has a 2x
>hardware advantage.

Sounds like you are saying that comp-comp testing show such great rating
differences that they can't possibly be realistic. This is not true, and you
should look less at one or two lopsided results like the above, and more at the
general picture as it is reflected in say the SSDF lists. Remember that Fritz 5
had some very one-sided results against some opponents, but the bottom line is
that it is 40-50 points ahead of the rest which is not hard to accept as
realistic. The list historically shows a gradual progress and a dependence of
ratings on CPU power which I think is quite easy to believe and to relate to the
actual playing power of programs.


>Is it time to abandon Computer Vs Computer testing all
>together? Or are we going to have two standards to judge chess programs? One
>chess program being the best playing other chess programs and one chess program
>being the best playing humans.

Playing against computers and playing against humans are different experiences.
No argument there. But to say that strength vs. computers and strength vs.
humans is unrelated, or even to say that there are some "weak" programs who are
actually stronger when it comes to humans is not supported in fact. Actually the
facts indicate the opposite.

It has always been the strongest programs who have also scored best against
humans. And vice versa: there is no example of a program that is not considered
to be at the top but that is scoring better against humans (than the top
programs). Someone mentioned CSTal on this thread as "proving" this point, which
strikes me really odd, since I've never heard of any achievement of this program
against a strong human.

Amir



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.