Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: 14 position test suite from Richter's

Author: Don Dailey

Date: 19:40:36 12/10/97

Go up one level in this thread


On December 10, 1997 at 15:25:29, Howard Exner wrote:

> I think this is true. The game of chess offers diverse possibilities.
> Karpov often likes to quote "there are many roads that lead to Rome"
> in drawing analogies to the many winning paths available in a
> chess game.

Yes!  I've noticed that often the computer chooses the efficient path
where the human chooses something with more beauty.  The computer
move often gets bereated because it is "ugly" and the human often
(unobjectively) concludes that the move was bad or weak.  And it
often turns out that there was nothing wrong after all with the
computers ugly choice.

>I'm beginning to believe that the only valid indicator of assessing
>computer chess strength is to rely on computer vs humans at 40/2.
>There are other indicators but none as accurate as the traditional way
>we rate ourselves, namely tournament play. Too bad there are so few
>human vs computer events out there.

I wish there were more too.  But even if there were, it's now becoming
more difficult to find worthy human opponents.  Anyone less than 2200
ELO is starting to become too weak to properly test against.  This is
not to say we don't get beat occasionally by the weaker player, but
realistically, for measuring chess strength, you need opponents that
are nearly the same chess strength as the test subject.

>Anyway, I felt I had to comment here since I agree with your above
>summary on test suites.

I appreciate the feedback, thanks for the comments!

I think the holy grail of computer chess developers is to find a fast
way to measure small improvements.  It may be possible to devise a
problem set that is reasonable accurate, but it will probably need to
have hundreds of positions consisting mostly of positional problems
but with some tactics thrown in.  Perhaps in a similar ratio to actual
games.  But these
are really hard to construct because the best move is often not as
clear cut as imagined.  Like you said, there are often lots of roads
leading to Rome!   Even tactical sets often turn out to be not so
straightforward, and routinely, alternate solutions are found.
This happened on the last couple of small sets posted on this group!

-- Don






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.