Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Rybka >3000 elo as author = IM

Author: Stuart Cracraft

Date: 14:07:26 12/16/05

Go up one level in this thread


On December 16, 2005 at 08:31:36, Steve B wrote:

>hi Adams
>
>well there were several Master players that tried their hands at chess
>programming including a World Championthese would incluse but not be limited to:
>
>Botvinnik-WorldChampion
>Hans Berliner-Correspondence World Champion
>Larry Kaufman-USCF Master
>
>none of them however acheived a very strong result although i imagine the
>hardware of the day had a great impace on the strength of the program
>
>perhaps with todays hardware their programs would be world class strength
>
>Botvinniks treatise on programming a computer to play chess:

Not to sure about that interpretation.

Botvinnik's result was garbage.

Berliner was the top theoretical dog for MANY years and a chief
exponent of knowledge. His Hitech did have very high results for the
time (2400 USCF). This is when Belle (2200-2300) was tops in the 80's
and just before Deep Thought hit. So Berliner did reach the top of
the computer chess world for a time, albeit with an insufficient
hardware implementation.

As for Larry and Dan (Dailey), their programs have always been solid
masters.

I think we're comparing apples and oranges.

Most of previous-years were not SuperGrandmaster strength and
beyond, whereas nowadays, they are (getting to be.)

Overall I would say that being an Expert or above and being a programmer
is conducive to a stronger program than being below an expert. But
there have been many, many exceptions.

It is too loose a correlation for it to be useful and should not
be used to detract or bolster Rybka's seemingly-clear result/progress.

Stuart



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.