Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Deep Blue Match game 2 of match 2

Author: Robert Hyatt

Date: 15:19:35 06/20/01

Go up one level in this thread


On June 20, 2001 at 14:26:20, Vincent Diepeveen wrote:

>On June 20, 2001 at 12:21:04, Geo Disher wrote:
>
>>Game two had a couple of key moves that were questioned by Kasparov after the
>>match.  Could someone tell me what those moves were?  I am thinking of just
>>letting my computer evaluate those positions for a few months if they will run
>>that long to see if the deep blue moves ever show up as preferred moves.  I
>>understand that even running the programs that long will not come close to the
>>number of moves DB went through but still I am interested in the result.  I
>>think others have done this in the past but maybe not with the newer software.
>
>Kasparov questionned Be4 move.
>
>This distracted audience from the poor level of Kasparov.
>
>Be4 is simple explainable. note deep blue searched around 11 ply there,
>diep version 1997 played Be4 too, because diep had high penalty for
>opposite bishops in those days. Note that Qb6 later appeared to be
>easily winning for white after some extensive analysis of Seirawan
>and later i only repeated those. The opposite bishops ending there
>is simply won for white.
>
>It's understandable that in 1997 more as 1 program didn't play Qb6 but
>preferred the first few plies Be4 there. Note that also in the Qb6
>line there are some great sacraficial lines which are from computerviewpoint
>not so relevant, but those also draw for black (sacraficing a6 and e4).
>
>Crafty in those days played Be4 too up to a certain depth.


Not any version I had.  It wanted to take the pawn until ply=21 which took a
long time to reach.  At depth 21 the score dropped enough that Be4 looked good
enough to play...  but that was more than 24 hours of computing.



>
>Basically all programs which had a big penalty for opposite bishops.
>
>Initially DIEP likes Be4 now too, but already quickly it gets way bigger
>search depths as Deep Blue ever got, and then Qb6 fails higher and higher
>and in the end it gets played.
>
>That kasparov questionned this move shows 2 things
>  a) he knows nothing from computerchess
>  b) he distracted the attention of the audience from his bad level
>     in this game.
>
>     Kasparov has made not only weak opening moves in this game, not
>     only missed a draw at the moment he resigned, but he also made
>     some major strategical errors. Seirawan explains these major
>     errors by saying: "kasparov never plays ruy-lopez with black,
>     i do, so i would play it way better as kasparov here".
>
>Best regards,
>Vincent



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.