Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: more examples for search-based stupidity

Author: Hristo

Date: 04:13:42 06/14/98

Go up one level in this thread


Uri, there is more than one way to look into a position!
You have to forget what you have learned over the years, just for a
moment, and try to imagine what would be the best(most natural) way to
teach a computer what is the meaning of the "chess" game...
hmmmm ... the program depends very heavily on a move-generator. In some
positions the M-Generator outputs more than 17 000 000 moves per
second(on PII 266 128M).But these moves are not used to create a tree of
possible positions.
The program(so far) doesn't need any hash tables!
In any case it is too early to say how well will this program play!
I do not have a game to show you! Once(if) I do everybody here will know
and everybody will be able to try it for free!


best regards.
Hristo

>I do not understand it because every chess program(it can be a
>slower search) and every human depends on search
>(humans know better than programs what to search).
>
>you cannot see in the games thorsten posted that hxg5 is a mistake
>without a search and if you have an evaluation function that say it
>without search your program
>can do mistakes in sacrificing when the opponent takes the material
>and the initiative for it is not enough.
>>Uri



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.