Author: Peter Kappler
Date: 16:12:34 10/20/99
Go up one level in this thread
On October 20, 1999 at 16:31:42, Joshua Lee wrote: >What positionally don't chessprograms understand yet? Lots. Give most micros a position devoid of tactics, requiring slow maneuvering and repositioning of pieces, and they will play at or below 2200 ELO. The trick for a human is to get this kind of position, and then accurately convert it, which almost always requires some type of tactical breakthrough. Here's a few things I think all programs have trouble understanding (unless there is an immediate tactical consequence): pawn-sacrifices to gain time in the opening pawn-sacrifices to open a file or secure a square for attacking purposes pawn-sacrifices to improve piece activity outside passers in certain endgames attacks based on pawn-storms bad bishops I could probably think of others - these are the first that come to mind. Different programs handle these problems with varying success, I think. I know that Bob does a lot of things in Crafty to try to *avoid* closed positions altogether, and I think he has spent a lot of time on the outside passer problem. Probably the most famous example of how badly a strong program can botch positional play is game 6 of the first Kasparov - Deep Blue match. --Peter
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.