Author: Rolf Tueschen
Date: 02:43:00 07/16/02
Go up one level in this thread
On July 16, 2002 at 01:15:40, GuyHaworth wrote: > >Thanks, Rolf, for the implied compliment about my previous silence on 'matters >of debate' and about my statistical contribution to the CO7 workshop. > > >[ I personally don't think it would have been worth coming to Maastricht just to >hear my 15' on that. I was merely re-analysing the results of Heinz' big, >post-Karlsruhe/Ph.D./book, M.I.T. experiment on self-play ... to show that they >are more decisive than Heinz has yet said. ] > > >I have certainly confined myself in the past to technical/endgame matters. > >However, I do feel that computer-chess can 'add value' to human experience >rather than just displace it. In this case, that means that computers can add >to our enjoyment of the game, and that FIDE should recognise that, even >officially. > >Actually, FIDE have made positive official statements about computer chess in >the past. It is not impossible that they will do so in the future. > > >For the future, be assured that you will always be able to tell when I'm giving >technical advice and when I'm expressing an opinion. > > >Obviously, computers have an advantage in memory-access terms - in opening >books, endgame tables ... and in huge hash-tables of potential, evaluated >positions. Whether this is 'unfair' or not is debatable: no 'rules' are bieng >broken. > >In the interests of 'good play', one might level the playing=field a little, >allowing humans access to opening books and endgame tables. After all, Kasparov >would not have made that move-transposition which led to him losing the last >game of the 2nd match with Deep Blue. That would have led to a more >satisfactory outcome in my view: there is little satisfaction in a win gained >by an elementary opponent blunder. > > >g I have to thank you, all the best for your work. Of course computerchess has a lot to give to chessplayers. Let's keep the question of "rules being broken" open. When I say this I never want to say that it's a matter of justice already since there are no exact rules for machines and chess yet. Still there are reasons to believe that something is wrong with the databases. With the freedom of using computers in tournament chess would simply leave classical tournament chess. That is about GM having their own computer-like memory (eidetics). I don't need artificial memory to enjoy my chess. Chess is also the manifestation of the incomplete. This is the driving force in my hobby. I wouldn't enjoy chess where the handling of the features of say ChessBase products get a first-rank importance during play. This would completely pervert chess. Let me oppose your opinion on the blunder in the 6th game. Kasparov mainly supported classical chess with his mistake. Chess is not basically about memory - although you can't become super GM without eidetics, but chess is about the fight between psyches. Simply because in general almost equally educated and strong individuals play each other. So this is all about different psyches. In the 6th Kasparov was no longer "in himself" as Kasparov, that is the message. Reason the tension from complex irritations after the second game. In a way the 6th marked the end of friendly relations between human players and computerchess machines. Caused by the IBM/DB2 team. The engineers simply had (still have!) no understanding of the importance of psyche for chess. In vain I tried to explain the aspect from a scientifical angle. In experiments with human clients it's most important to consider the comfort of the client simply to exclude unwanted disturbances. It's a terrible influence on computerchess that the famous US team had that lack of smartness (sorry, but we must face the truth) and still superficially "won" the second match and thereby this famous prize too. Own goal in the end. If only these Hsu et al. would have had your understanding of the differences of levels of statements, the tournament situation for machines would be different too! Rolf Tueschen
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.