Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: A New Topic: How Much Testing to get rid of the ChessFactor?

Author: Thomas Mayer

Date: 07:06:06 01/26/06

Go up one level in this thread


Hi Rolf,

well, first you must declare what chess strength IS. IMO this isn't clear. Of
course we have something, most call it strength, measured in Elo by sucess in
games.
You know, if they could, many engines would laugh about humans. Think about
Rybka & Fritz talking with each other: "Hey Fritz, have you seen this stupid
human ? He didn't do that bad, but then he missed a mate in 10 and did even
lose... They are sooooo weak in easy positions, unbelieveable that they even
have a chance from time to time" - Fritz Answering: "And that Karjakin against
Topalov, he even had a better position but then overseeing some simple tactical
shoots, how could this happen, I wonder ?"

Therefor it is very unclear what "strength" is, looking stupid in some positions
does not mean at all that someone/thing is weak. It might look from our point of
view weak when engines do not understand some for us trivial stuff, but there
are enough things that make humans look weak compared to computers. As long
computers play the game of chess in such a different way then humans do, this
will last forever.

Greets, Thomas

P.S.: How do we measure human strength ? Is there a different method then games
and game performance ? Which is widely used ?



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.