Author: blass uri
Date: 14:35:32 07/15/00
Go up one level in this thread
On July 15, 2000 at 16:45:16, Dann Corbit wrote: >On July 15, 2000 at 12:24:17, Mogens Larsen wrote: > >>On July 15, 2000 at 11:59:08, Michael de la Maza wrote: >> >>>Junior is humming along, not losing a single game, and forcing some of the best >>>players in the world to play perfectly in order to salvage draws. >>> >>>Then Kramnik illustrates how to punish Junior. >>> >>>_All_ of the GMs _instantly_ get it. Some of them apply the lessons better than >>>others, but all of them make Junior look silly and none of them lose. >>> >>>GMs are amazing. >>> >>>When the history of computer chess is written, it will be divided into the >>>pre-Kramnik and post-Kramnik era. >> >>If you had followed the recent tournaments with computer program participation, >>you would have noticed that Kramnik was far from being the first Grandmaster to >>utilize an anti-computer strategy succesfully. > >This is true, but mostly, the GM's do not try to blockade or other common >anti-computer strategies, but rather "play their normal gams." I know that adams tried to play anti-computer against Deep Junior. Adams played an opening when he won Junior at blitz but at tournament time control Junior played better and adams could not win. I guess that Junior avoided the mistakes that it does at fast time control and when both sides do not do mistakes it is a draw. Adams was sure that he was going to win because he won Deep Junior at blitz in this opening and it is known that humans are better at long time control but he was wrong. I guess that they now see that the best anti-computer strategy is to play for a king attack and not for endgame but they do not always get the positions that they want. Example:Anand played based on the assumption that the computer will do a mistake by c4. Fritz does this mistake but not Junior and he could not get nothing from the opening. > >I have exchanged email with GM's who have played high profile computer matches. >I can tell you that for sure some of them really don't have any idea what sort >of strategy is successful against computers. Did these GM's play in the tournaments that Junior played? > >A very high profile event like this one will (I think) get the message out. Now >that computers are proving to be real competition, I think it is an opportunity. > >It is an opportunity for the program vendors to get some really high exposure. >It is an opportunity for the GM's to learn how to beat them. I think five years >ago no GM's took the computers seriously -- indeed they did not need to. But >now, they need to rethink that position. Once they learn to play well against >computers, several things will happen. > >One thing that will happen is that the programmers will work to break these >strategies. They will be successful, and so the GM's will have to learn new >strategies. > >When I played crafty the first time (I think it was version 11.something or >12.something) I found I could block the pawns. I did not even know about >winboard, so I had to do it with an actual chessboard (I'm not very good at >visualizing). The latest versions of crafty will absolutely not let me do that. I think that it is better to do a program that knows to play closed positions and not a program that does not let the opponent to close the position. I know that Junior does not try to avoid closing the position but try to play the best move. 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.