Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Forward with Opening Theory

Author: Thomas Mayer

Date: 10:14:29 05/17/05

Go up one level in this thread


Hi John,

> And if you take away randomness, it is very easy, given enough time, to find
> a way to beat ANY engine, because the engine will always reply the same way,
> given the same CPU and the same time control. Therefore, if such a match
> were to take place, it would be imperative that the human would have no time
> to prepare against the engine.

well, you can introduce also some randomness by the time usage. E.g. Quark takes
care of the opponent time - therefor to repeat the same game you must play
against it in exactly the same speed. This is nearly impossible. Nakamura at
playchess did throw Quark out of the book almost always with the same variation
and very early in the games, anyway it was not possible for him to repeat the
last games. Of course a position learner would do here even better.

Greets, Thomas

P.S.: Quark Anti-Human against Nakamura at playchess in bullet was one of the
most exciting match I have ever seen. And even in bullet the GM was capable of
getting 3 wins... Very impressive !



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.