Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: One viewpoint of correspondence chess

Author: Dana Turnmire

Date: 02:39:12 09/11/03

Go up one level in this thread


On September 11, 2003 at 05:19:33, Uwe Meißner wrote:

>On September 11, 2003 at 04:43:54, Dana Turnmire wrote:
>
>>From "The Mammouth Book of Chess" page 433.
>>
>>"...the human uses the computer to check over analysis he has done, and to reach
>>a verdict on random tactics.  It's a tool.  As computers get better and better,
>>such occurrances as this (finding mate in 9 etc.) will become more common, and
>>both players will be guided away from such perils by their machines, until
>>eventually there is no way out for one side or the other.  Since so many of the
>>moves were chosen by excluding moves for reasons other than human preference, it
>>will become impossible for a player to explain why he won a game, or for the
>>loser to understand why he lost.  They can perhaps point to the strategy they
>>employed, but it will really have been some random assisted tactical
>>implementations of the two players' strategies that have decided matters.
>>Looking at it in terms of the humans versus computers discussion, for a while
>>the game will have been drifting around inside the 20% of positions in which
>>humans are better than computers, or the 60% no-man's land where it isn't clear
>>who handles the position better.  However, should the game drift into the 20% of
>>positions that computers handle far better than humans, then that is the end of
>>the human involvement in the game.  The two computers are effectively battling
>>it out from then on."
>>
>>"Obviously, as the percentages get slanted more in favour of the computers, the
>>point at which it is the two computers locked in battle will become more
>>frequent, and occur earlier in the game."
>>
>>"Perhaps a ban on the use of computers in championship events could to some
>>extent be enforced by requiring players to be able to explain, if called upon by
>>the official bodies, how they happened to find any really strong
>>counter-intuitive moves.  This would be the equivalent of the drug test in
>>athletics."
>
>Interesting stuff, but the context is not quite clear to me. Are they speaking
>about correspondence chess or some sort of Advanced Chess a la Kasparow?
>I think good players will always be able to explain post mortem what did happen
>on the board; it's only a matter of time and work.
>Nevertheless there are some tricky endings (you can find in tablebases), that
>have 100-200 moves, where it seems impossible to explain the meaning of each
>single move. But also this is only a matter of time and motivation. In natural
>sciences there are much more complicated questions, which already have been
>solved even without computers.
>>
>Uwe

The Author FM Graham Burgess states  "One of the most perceptive commentaries on
the subject of computer-aided correspondence play was made by Peter Sowray (I
present his arguments here with very considerable embellishments of my own), a
strong over-the-board player who has also played a good deal of high-level
correspondence chess.  His view of the current situation is that the human uses
the computer to check over analysis he has done, and to reach a verdict on
random tactics."



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.