Author: James Swafford
Date: 07:22:38 12/01/04
Go up one level in this thread
On December 01, 2004 at 05:43:19, Renze Steenhuisen wrote: > > >Hi All, > >I implemented a basic repetition detection algorithm, but unfortunately it seems >to cause more nodes to be expanded! Is this a weird fenomenon, or is it OK? > >Cheers! > > Renze It might be ok. Suppose you're searching move A then move B. Move A is a draw by rep. Move B will return a score of +3 pawns. Also, let's say the search for move A will return a score of +5 pawns if you disable repetition detection. With draw by rep disabled, you'll get your +5 pawns for move A, then quickly search move B (lots of cutoffs). With draw by rep enabled, move B will become the new "best move", so you'll likely search far more nodes. -- James
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.