Author: Sven Reichard
Date: 10:13:31 10/10/99
Hi leonid, sorry, wasn't around for a couple of days, so I missed the thread. My idea was a bit different from Dann's, and I don't know if that would help you. i) Fix a (sufficiently large) set of test positions with hopefully unique solutions (single best moves); here Dann's work comes into play in order to ensure the quality of this set; ii) find out (and publish) how several engines do on those (if they find the solution, how long they take). This would enable the interested to get a more detailed comparison between various engines, and help programmers evaluate their own code. This was in reply to somebody's question whether or not to take testing positions to evaluate programs. However, there was no positive feedback so far, so I dropped the matter. Sven.
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.