Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: The Limits of Positional Knowledge

Author: kurt

Date: 21:49:06 11/11/99

Go up one level in this thread


Why can positional knowledge not be obtained by adopting current
opening lines as some programs do? To reach GM rating a good
opening book knowledge is an absolute necessity.
I belive,programmers should prepare book.Depth search ability
has advanced far enough.Awareness of positional weaknesses etc
is now part of most programs at least par with human players.
Long range planning like in correspondence games should be achievable
with a narrow but deep book.This should overcome human ability to
deploy pieces creating a "JUDO" style like unbalance of forces which is
difficult to dedect with weighted ply search.To reach supremacy the learning
function together with an up to date book data may be the biggest factor.
Programmers should team up with reliable data collectors to prepare a
book best suitable to provide guidance into a manageable position. Marketing
such book updates may prove to be quite provitable.




This page took 0.02 seconds to execute

Last modified: Thu, 07 Jul 11 08:48:38 -0700

Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.