Author: Dann Corbit
Date: 19:32:24 11/29/99
Go up one level in this thread
On November 29, 1999 at 22:21:58, Stephen A. Boak wrote: [snip] >The only true test is to pit many top programs many times against many top >humans, under reasonably controlled conditions. I think this is an especially important comment. That is because not only do the programs play differently, but also the humans. In a match not too long ago, Judit Polgar faced a computer. Many people thought Polgar would clean-up, since she is so strong tactically. I think that is exactly why she did not do as well as expected. Her area of strength is exactly the same as the computers, and computers are better at tactics than most GM's even (I think) except in rare cases like Zugzwang which a GM will see but some programs will not. On the other hand, a great strategic thinker will probably pose problems for a computer. Similarly someone who is a master of closed positions. Also someone who is effective at deep sacrifices (such that the 10 ply or so the computer will see is not far enough to see the benefit). So a single GM against a single computer program is not as telling as a mixture.
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.