Author: Mogens Larsen
Date: 08:40:47 02/08/01
Go up one level in this thread
On February 07, 2001 at 21:56:50, James T. Walker wrote: >Hello Mogens, Hey Jim, >I believe you are right. When using auto232 games I think the book settings are >automatically set to "Optimal". The problem is this was one of the "manual" >games I played and did not make sure the book settings were already set to >"Optimal" before the game started. Which means I'll have to go back and replay >this game. If you publish the games I'll remember to take a look :-). >It does not change the fact that the book has a blunder in it >though. All books that has a core based on a number of (GM) games will contain blunders. The frequency will be low for obvious blunders and higher for refuted moves or variations. Besides, the book must also be able to exploit blunders or refuted moves made by the opponent. That can be partially eliminated by using a lower bound for the frequency, ie. only moves played at least three times. But not completely. That requires thorough move annotation, which is a time consuming task. >The "normal" setting should give more variety but not blunders. Thanks >for reminding me. Actually, I think it's possible to use normal and avoid major blunders. You just have to make sure that "Tournament Book" is enabled. But generally optimal is best. The number of possible errors in engine-engine matches are enormous. Regards, Mogens
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.