Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: I insist: CSTAL is great but nobody seems to care... (Sinatra)

Author: Enrique Irazoqui

Date: 07:00:36 05/07/98

Go up one level in this thread


On May 06, 1998 at 20:39:42, Thorsten Czub wrote:

>On May 06, 1998 at 15:35:43, Enrique Irazoqui wrote:
>
>>But I bet he likes winning games. A hybrid of CSTAL and a fast searcher
>>might be a nice way to eat your cake and have it, no? When CST reaches
>>+2 or +3 without material advantage it might be a good moment to stop
>>speculating and start counting all these humble beans Chris likes so
>>much? :)
>
>IMO it would be enough to let genius5 compute in 2-player-mode in
>background.
>If genius says  > +1 pawn than knock off tal-function.
>(or even better: play with genius5 engine in background [oops - cheating
>!!!])

Genius will never have a +1 without material advantage, or at least no
more than once in a blue moon. Often I have seen CST evaluating at +3,
while the other programs thought it was between 0 and -0.XX. In these
cases, CST had a great position, one that any human player would pick.
For instance, in the game against Fritz 5 that I posted the other day.
It might be the right moment to switch to materialist and fast, well
before Genius shows a +1.

Besides, for doing what you suggest you need Genius and 2 computers. CST
should be self sufficient, no? Build up with Tal functions and then drop
them?

Enrique



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.