Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Dutch teletext: "Fritz plays in dutch championship"

Author: Djordje Vidanovic

Date: 08:48:38 04/07/00

Go up one level in this thread


On April 07, 2000 at 06:47:31, Jeff Nouveau wrote:

>On April 07, 2000 at 06:25:05, Djordje Vidanovic wrote:
>
>>On April 07, 2000 at 05:27:09, Jeff Nouveau wrote:
>>
>><it's a game of time and stress management and SOME chess knowledge> (emphasis
>>mine)
>>
>>You must be joking.  Try Kasparov, Anand or Kramnik at any time odds, say 5
>>minutes for you and 1 for them.  And them reconsider your idea about SOME chess
>>knowledge.
>>
>>
>>***  Djordje
>
>I'm sure you understood what I meant :o)
>Of course, chess knowledge is of first importance : I just wanted to express the
>fact that, for OTB play, the importance of time and stress management grows.
>
>I don't have any kind of chance against people over 2200 ELO, because the
>difference in chess ability is so huge that the time advantage doesn't make any
>difference.
>
>But when the relative strengths get closer, time and stress grow in influence.
>
>Did I reconsider enough ?
>
>Jeff


:-)) Thanks for this interesting exchange.

I am not sure about your having reconsidered the matter sufficiently :-)  I
believe that when you pit a chess program against a professional chess player,
the amount of knowledge is going to decide the match.  Naturally, the human will
be at a small disadvantage due to memory limitations, stress, etc., but the
program will be at an even greater disadvantage due to its inherent limitations
regarding learning, pattern recognition, etc.  Therefore it will be the
_chessic_ ability that decides the outcome, in the long run, of course.  Limited
statistical samples may show otherwise, but a large number of games will reveal
a simple pattern:  humans learn _very_ successfully and they learn the patterns
which is what matters and determines the strength of _any_ strong chess player.

Consequently, as you may have expected from my little treatise, I believe that
the strongest programs of today cannot yet cross the 2500 ELO mark.  They are
very strong at blitz, possibly around 2700, bordering on 2600 at rapid play, but
at traditional tournament tc's they are "only" around 2500, even on the fastest
of modern HW.

Thanks again.

***  Djordje




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.