Author: scott farrell
Date: 23:22:21 08/21/02
Go up one level in this thread
If you look harder you'll see the answer is in the PV. about the 3rd move in the PV changes dramitcally. Most chess programs will do 3x more time for next ply, unless there is some dramatic change in the PV, it sort of has to REDO some of the plies that previously had the incorrect moves in the PV. This is a general behaviour of Iterative Deepening and is VERY common. I found 'internal iterative deepening' helped that problem in my program. I wonder if gandalf does not use IID. Scott sfarrell@icconsulting.com.au sfarrellc (ICC)
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.