Author: leonid
Date: 11:32:04 11/20/01
Go up one level in this thread
>> In mine, I was not able to use >>old "best move" for each ply when mine start search in 5 move, after completed >>search in 4. > >You were "not able"? Why is that? Reason is partially in the fact that I have no separation between mate solver and second part that work like chess playing part. When first part fail to find immediate mate, completely different logic start its work of finding this time not victorious but only acceptable move. Two parts in chess program use many same variables, mainly for "best moves inside of each ply". To have less mess with all this, all variables are initiated once again, after the mate solver work is done. Also, when I tried to use only mate solver and nothing more, and my "best moves for each ply" were not cleared between two searches, I found total time is slightly worst, and not better, that when search is done just at single depth. As I see now, after what you have said, it was more my way of looking and not seeing something there that impossibility to make sequential search (iterative deepenig) profitable with each deeper search. Good news! Cheers, Leonid
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.