Author: Christopher R. Dorr
Date: 12:29:26 06/30/99
Go up one level in this thread
Unless you *know* that it's a great move (by having played it against the computer before, and having experienced good results), I think it is a mind-bogglingly bad idea to use a decent TN (theoretical novelty) against a computer. This seems to be a bad choice by Sokolov, for several reasons 1. The psychological edge of having dropped a TN on somebody is lost. A GM pays attention to a TN...a computer simply looks at the position like any other. 2. In well known lines, the human has significant experience with the long-term ideas inherent in the position. This is a crucial edge for a human vs. a computer. In a TN situation, definitionally, the human will not have had experience playing out the positions generated, at least not in tournament play. Again, this strongly favors the computer, which doesn't care about such things at all. 3. Against a human GM, a TN often earns you a time advantage. Sometimes as much as 15-30 minutes, in slow tournaments. Against a comp, it does no such thing. But given the unfamiliarity of the ensuing positions (due to point 2), the human will have to spend extra time sorting the newness of the position out. the comp just goes along it's merry way, playing at it's normal time usage. 4. As soon as it's been played, it's surprise value is lost. Given that points 1 and 2 above indicate that you the human gets no benefit from having played it, it seems that you have blown a decent innovation for nothing. I'd be interested in hearing why he played it. I'm very surprised that an experienced GM like Sokolov did this against a computer. Chris On June 30, 1999 at 03:22:01, Brett Clark wrote: >From the Chessbase website: > >"Fritz saved the day for the computers with a brilliant win against the Ivan >Sokolov. The 2620 GM played a novelty which he had prepared together with Ulf >Andersson and Jan Timman for the latter’s match against Garry Kasparov last >year. Unfortunately Fritz completely refuted the idea with some masterful >tactical strokes which stunned Sokolov and forced him to resign in 22 moves." > >Someone help me out here. I'm a 1300 patzer, so I don't know if this is true or >not. I recall reading some posts that Sokolov just made a bad mistake, and >Fritz really didn't do anything worthy of praise. I'd like to hear some >opinions on the Chessbase spin doctor's version of the win.
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.