Author: Kurt Widmann
Date: 16:23:08 07/08/01
Go up one level in this thread
On July 08, 2001 at 18:41:30, Bertil Eklund wrote: >On July 08, 2001 at 17:48:37, Kurt Widmann wrote: > >>On July 08, 2001 at 16:28:38, Bertil Eklund wrote: >> >>>On July 08, 2001 at 14:20:26, Kurt Widmann wrote: >>> >>>>Why is the importance of opening theory so neglected by computerchess >>>>enthusiasts? I am seeing the use of booklines sometimes totally ingnored >>>>by posters of rated computer games. It is my opinion, that booklines are >>>>the most important part of a chessgame. The quality of "Book",or the "Guiding >>>>human hand" for chessprograms has a greate influence as to the outcome of >>>>a game. Every good chessplayer trys to reach a familiar position or get his >>>>oponent into a unfamiliare position not suitable for his/her style of play. >>>>The recent performance of Pocket Fritz against GM's shows a tippical >>>>exploitation of ones opponent book weakness. If Kramnik can not find a hole >>>>in the armor of Deep Fritz, it is safe to assume that he will not win >>>>the match. Therfore I am shure that Kramnik is studying DF,s book over >>>>and over again,knowing the importance of once oponents strength and >>>>weaknesses, including the all important openinglines. >>>>Why then are so few discussion on any forum about current opening theory, >>>>as to their valididy and or exploitations of them? >>>>Just curious, >>>>Kurt Widmann >>> >>>Hi! >>> >>>Kramnik are going to play (with white) Colle and/or queen-pawn openings in every >>>game where his positional understanding is superior to any program. >>>It is impossible to prepare against this for more then 2-4 moves. >>>Of course he doesn't play the latest sharp lines in sicilian defense. >>> >>>Bertil >> >>Well, I belief that Kramnik with colle b3 or c3 will not get more than >>a draw,because the remaining pawn structure is too simple for Fritzes >>25 ply or more depth search ability. His normal d4 lines are better >>than the colle. Against e4 Kramnik may have to take his chances with black >>playing the Rue Lopez. He may "unnerve" or "outbook" Fritz with the d5 line, >>even so, he never seem to play it. > >Maybee you are right but I have seen a lot of games with "Colle-structures" >where the human slowly can build up a king-attack. Of course his normal d4-lines >are (objectively) much better but also sharper and much more riscy against a >fast computer. > >Do you mean Marshall (d5)? It could be very dangerous to play for more then a >draw in this variation. I don't think it is a good choice to play normal >Ruy-Lopez variations against computers because it is to complicated and often >leads to open and complicated positions. I don't know much about the >Berlin-defence and if it is the right choice against computers. I have 434 >computer-games in this defence in my computer-base out of 94000 games played by >computers. > >Bertil > Yes I mean the marshall d5 line. My reason for suggesting d5 against Fritz is, that all lines are forcing, and well analysed. Deviations are nearly impossible, therefore Kramnik could prepare lines to exhaustion, eliminating any uncertainty created by a new bookline against his normal Berliner defence. I am shure that Kramnik does not need my advise,but Fritzes calculating power must not be underestimated. We just have to wait and see. Kurt Widmann
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.