Author: blass uri
Date: 15:38:56 07/15/00
Go up one level in this thread
On July 15, 2000 at 18:09:05, Robert Hyatt wrote: >On July 15, 2000 at 13:52:39, blass uri wrote: > >>On July 15, 2000 at 13:16:45, Jeroen Noomen wrote: >> >>>On July 15, 2000 at 11:43:30, Daniel Chancey wrote: >>> >>>Hi, >>> >>>Junior made several mistakes: >>> >>>1. Closing the position >>>2. Not playing Bh6 to exchange the black bishop >>>3. The useless and time consuming bishop maneuver Bg5-e3-f4? >>>4. Castling long, right into the black attack >>>5. The lemon a2-a3??, which gives Black a nice target >>> >>>After 15 moves White was already desperate. You know, in these positions the >>>strongest programs simply look silly, using a 4,5 GHz processor or not! The >>>point is that calculating gets you nowhere in closed positions. Even a 5000 GHz >>>processor wil not prevent moves like a2-a3? Because this move will not lose >>>within the next 30 plies. >>> >>>It seems that programmers should have 2 versions of their program: >>> >>>1. A fast one to play other chess programs >>>2. A much slower one with a lot more chess knowledge, that can prevent >>> games like against Kramnik and Piket. >> >>I think that a better idea is to do only one slow program that can win other >>programs by king attacks. >> >>Chess system tal often wins other programs by king attacks but the programmer >>stopped to develop this program. >> >>You can download the games of the millenium tournament and see games that system >>tal wins by king attack >> >>see http://homepages.compuserve.de/hjsxy/page4.html >> >>Uri > > >Just don't conclude that it wins most of the games it plays... It did not win most of the games but it won some games and got 9.5 out of 15 that gave it first place with some other programs. 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.