Author: blass uri
Date: 09:50:24 02/20/00
Go up one level in this thread
On February 20, 2000 at 11:05:58, KarinsDad wrote: <snipped> >When you or anyone else analyzes a computer chess game, what do you really >discover? Do you discover the reason that a program made a mistake (the why) or >do you discover a mistake (the what) and hope that the programmer will be able >to piece together the cause and effect of the various algorithms which led to >that mistake (and possibly correct it for future games)? Sometimes it is possible to guess the why. I do not see only move but also evaluation and main line. If a program cannot see a simple tactics it is possible to guess the why. For example if you see that a program evaluate position as a mate against itself when there is a simple stalemate combination you can guess that the program have a problem in seeing some stalemates. If a program show after a move Rc2-c3 a main line of Qd3-b1 Rc3-c2 with evaluation of 0.00 you can guess that the program evaluates repetition too early and does not understand that Qb1-d3 is illegal after Rc3-c2 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.