Author: Dann Corbit
Date: 12:06:07 08/20/02
Go up one level in this thread
On August 20, 2002 at 15:03:55, Dieter Buerssner wrote: >On August 20, 2002 at 14:46:51, Omid David wrote: > >>What you described results in a very poor program. What I suggested was a >>sophisticated search algorithm to build the weakest possible program, so that no >>one can ever lose to it ;-) > >I think, you should use your method together with other piece values - perhaps >similar to the ones suggested by Albert. Without thinking too deep about this, I >would guess, that a contest for the "best bad" program would with reasonably bad >playing programs allways yield in draws. What I mean is, sophistication of >algorithms would not help at all, and it seems to me almost impossible, to force >the opponent to mate me. In such a game, few more rules would be needed. For >example, loss on time and resigning should be a win for me (and therfore tried >to be avoided). Other than time control, just change the sign of the evaluation for a normal chess program and I think you will have it. Can't say that I would care to play against it. Sounds like a bore.
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.