Author: Robert Hyatt
Date: 20:06:31 10/22/00
Go up one level in this thread
On October 22, 2000 at 22:29:06, Mark Longridge wrote: >Does the result learning in crafty understand the difference between >losing because of a lockup (which will sometimes occur with unattended games) >and losing because of being outplayed? Surely if crafty is flagged on move >12, this is does not have the same significance as getting checkmated on move >90? It should never "lock up" (it _never_ does for me). If it is locking up, you ought to find the cause. But if it does "lock up" then there is no learning anyway, since something hung or crashed. > >I'd say you could dispense with the result learning in such a situation, >but is this actually the case? > >I've seen this happen with other programs as well... > >Usually what happens is right after the book moves are finished, there is a very >slight possibility of a lockup, and with no operator present, naturally this >results in a loss by time.
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.