Author: Bob Durrett
Date: 18:30:28 09/25/03
Over the last few years, I have been looking at new books as they come out and from time to time have seen descriptions of the mental processes used by strong players when they do "planning." My impression is that is a procedural matter. Procedures are subject to programming! [True?] For example: Christian Kongsted, in his book "How to Use Computers to Improve Your Chess," gave an example where Capablanca won a technically won endgame by first thinking about what must be accomplished in specific terms [keep the c pawn from advancing, control the fifth rank, and the like] and then figuring out where the pieces need to be placed to achieve those aims. Then he worked out [i.e. computed] the ways to get the pieces to go to where they needed to be. Similarly, many examples exist in the literature [especially endgames] where the game is approached in stages whereas one first achieves some immediate aim and then sequentially goes on to later stages. Intuitively, it seems to me that all of that could be programmed. Perhaps it would require something a little different from the inner workings of current-day chess engines, but the key point is that such step-by-step PROCEDURES should be programmable. Doesn't this seem reasonable? Bob D.
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.