Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Rebel GM challenge shows: it's not aggressive enough to win a game

Author: Paulo Soares

Date: 11:56:05 09/06/99

Go up one level in this thread


On September 06, 1999 at 11:04:58, Robert Hyatt wrote:

>On September 06, 1999 at 06:50:00, Paulo Soares wrote:
>
>>On September 06, 1999 at 01:14:49, Jouni Uski wrote:
>>
>>>After 5(?) games no wins so far. Bad, bad. Probably an aggressive program like
>>>Fritz, Chessmaster, Mchess have already won couple of games...
>>>
>>>Jouni
>>
>>I believe that any program will have problems against GMs,
>>mainly in the opening.
>>
>>Paulo
>
>
>In the opening, computers _are_ GM players, because they are using GM games.
>I think that computers get shredded in endgames, because they know so little
>compared to GM players that know so much there...


The problem that I see in the opening is that a GM can prepare theoretical
novelties, and even if these novelties are not good for played against another
GM, programs will already be in difficulties, because I find very difficult
that a program gets to overcome a line prepared by a GM.
The option of playing Gambit Benko against Rebel was of GM Hoffman and,
according to him, 15 ...Rb7 were a theoretical novelty, in the case an
improvement in Gambit Benko's line. The move 16. Qc1 it was not good,
and I think any program could do a bad move in this position.

Paulo



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.