Author: Russell Reagan
Date: 21:17:55 04/07/02
Go up one level in this thread
On April 07, 2002 at 23:59:15, Uri Blass wrote: >On April 07, 2002 at 23:47:59, Russell Reagan wrote: > >>Given a game, are there any methods of analyzing the game for the purposes of >>determining whether one of the players is a computer? I've heard people say that >>ICC uses some method to detect cheating, but I can't really figure out how they >>would do that, unless they just have a computer running many top programs and >>try to find players who are consistently in line with the moves a particular >>program plays. >> >>Any thoughts? >> >>Russell > >if the player use exactly the same time for every move you can suspect that it >is a computer. > >If the player thinks a long time about an obvious capture you can suspect that >it is a computer. > >If the player plays no tactical mistakes in complicated position that it is >natural to expect humans to blunder then you can also suspect that it is a >computer. > >Uri Good ideas. Do you think that if ICC does attempt to detect cheaters that they use a "selective search" and look for good candidates that are cheating? For example, they surely do not have the computing power capable of analyzing every game played on ICC. So do you think that they would only look at those players who either have very good records or have complaints filed against them? Also it might be a good idea to have some random selection, such as having a computer analyze random games and occassionally it will hit upon a player who warrants further investigating. Sounds interesting to me, kinda like the chess police :) Russell
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.