Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: The law of diminishing returns

Author: Rolf Tueschen

Date: 04:12:19 07/14/02

Go up one level in this thread


On July 14, 2002 at 01:01:43, Dann Corbit wrote:

>I strongly expect that where there is no clear move, the computers will change
>their mind over and over in the search.
>
>Lots of opening positions with no pawn snatches within 20 ply, and perhaps 25
>chessmen on the board, the computer will jump all over the place.

That was what I said when I asked if Ed had done some research about that.
Apparently he did not. But to insinuate that the 'change' should speak or should
be correlative with 'strength', that is not substantiated at all. But I insist
that Ed had a fine idea to do some research at all! It's a good beginning. It's
years ago that I read something about another little experiment. The reason for
the general situation is the understandable security thinking of the different
teams.

Rolf Tueschen

>
>I also suspect that if you took those same positions that you analyzed and
>analyzed them at 24 hours with the 5 top professional programs you will
>definitely get different answers for some of them.
>
>If you could go deep enough (let's say 30 plies) even in these positions there
>is sure to be some deeply hidden tactical shot that wins at least one pawn.  But
>finding it is just too hard for programs right now.
>
>Of course, the pawn grab may *not* be the best move, but that is what the
>computer will probably come up with eventually.



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.