Author: Mogens Larsen
Date: 14:10:18 02/01/01
Go up one level in this thread
On February 01, 2001 at 16:54:24, Dann Corbit wrote: >I disagree. Hmm... seems like Uri is losing his patent on this one. >The testing should continue with 17.07 until an accurate ELO is >clearly established. If you always move to the strongest available version, you >will never learn how strong it is. Since crafty is updated frequently, you will >never complete the thousands of games necessary. The completed set of games >should also prove useful to analyze possible weaknesses. That is an obvious reason, but a newer version should be included nonetheless. Especially to latest version of the strongest free program is doing against commercial competition. In my opinion that takes precedence over establishing a semi-accurate ELO rating. That calibration role should be taken by programs that haven't been upgraded for various reasons for a long time, ie. Shredder 2, Hiarcs 7.32 and so on. The possible weakness argument might fly, but that depends on the changes made since then. Hyatt himself would know if he wants to look at 17.07 or (soon) 18.01 games. Mogens.
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.