Author: Dann Corbit
Date: 12:57:20 01/23/01
Go up one level in this thread
Over 10 games, with two evenly matched programs even 10-0 or 0-10 would not be astonishing. Tiger is probably somewhat stronger than crafty. Therefore, a lopsided result is even less surprising. A 9.9-0.5 result for crafty would be more surprising, yet not astonishing. A single, small set of measurements is very inconclusive. That's why the SSDF runs hundreds of games before they emit a single peep. Very wise of them. Of course, they are trying to demonstrate strength. If you are just running a contest to find a winner, then the number of games is not important. The winner is easily determined. But if you want to find out how strong something is, it will take a very large number of games.
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.