Author: Robert Hyatt
Date: 21:12:57 02/05/99
Go up one level in this thread
On February 05, 1999 at 21:39:50, William Bryant wrote: >My program play abnormal moves when I try to exit the search when the alloted >search time is over. > >I check for time at the start of search (but not in search_root() and not in the >QSearch), >When this iteration of search exits with a score of zero, I break from searching >any further moves or from any further updating of the PV. The search should >still return the best move previously found prior to >time being flagged, shouldn't it? > >Any help with placement of time code in the search would be helpful. > >William >wbryant@ix.netcom.com the main thing is to be _certain_ that you can tell when a time-exit is taken so that you do _nothing_ to the PV at all. That is usually the cause of errors... because if you just return 0, that will look like a legit score and wreck things...
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.