Author: Peter Berger
Date: 08:07:37 10/07/02
Go up one level in this thread
On October 07, 2002 at 06:45:59, Rolf Tueschen wrote: In the last CSS magazine there was an interesting article by Walter Eigenmann concerning engine matches starting from positions that contain more or less complicated strategical moments, usually taken from famous games often covered in books for learners. I remember one game Fritz7-Junior(?) where Fritz7 played a competely classical minority attack in a logical and instructive way. I remember I did a similar experiment in 1998 and at that time no program seemed to understand this motive at all. Another field of complicated strategy where computers perform very well these days is passive defense, which is often very difficult even for the greatest human masters, who blunder very often or lose patience. Regarding ..Bf8 in the Kramnik game - I am not so sure that this move is a serious blunder btw - it also has some points , as it allows to attack a knight on d4 with e5 without allowing it to f5 with a double threat on e7 and g7. I think that many human GMs lose their games against Kramnik in a very similar way - in fact Kramnik is well known for playing and winning this exact variation against human super GMs. The decision of the Fritz team to let the program play the QGA against Kramnik is at least dubious IMHO. When it is about "unfair" knowledge used by the computer ( e.g. opening book) it should be mentioned that it is extremely likely that Kramnik had the position after Bf8 on his board in preparation matches btw - just look at the time he used for his moves .. 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.