Author: Uri Blass
Date: 04:02:02 06/25/05
Go up one level in this thread
On June 25, 2005 at 06:04:06, Claude Le Page wrote: >It would be the case if each book move had been thoroughly analyzed by the >engine with the help of a team of GMs >It's far from being the case for all deviations :then you must use a database >and rely on statistics of the games >for a new move ,there are very few games , and their outcome does not reflect >always the value of the move: the Dilworth was banned for long because it's >first appearence was a loss ( against Alekhine!) >Experience shows ( don't ask me why : I don't Know )that up to move 10 engines >are rather weak , but after move 10 they are stronger than the GMs who make the >book >Therefore , limitation to 12 ( I already explained why 12 and not 10 , because >Ruy Lopez Open and KID main line )seems quite justified I do not believe that it has to do with the number of moves. if you use something automatic then it is more logical to use statistics. A line that is played by GM's very often is probably correct(otherwise someone could find that it is wrong by computer analysis). 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.