Author: Jonathan Lee
Date: 09:41:32 04/05/01
Here are my reasons we should have testpositions: (not in any priority order)
1. In the position are all moves legal moves?
2. Is the position using the least amount of time?
3. Is the position using the least amount of nodes?
4. In the position did it arrive at the best answer or
is the horizon effect still there or
was it inaccurate in the first place?
5. If there is no consensus between 20 (more or less) different chess programs,
then why?
Did the testpositions (temporarily) disappear because it was a fad, a
statistical necessity, or just coincidence?
Jonathan (122 message)
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.