Thanks for the input. It is a very interesting idea. Since I don't have a transposition table but only a stupid tree, I can't apply the same mechanic. I can image that with a transposition table, nodes are very equally distri- buted, so pruning one of the 16 moves almost always is a good choice.
By the way, how big is your transposition table? Do you take some bits of the zobrist hash and use that as a direct hash? What information do you store in the hash table (is it equal to a normal tree node)? Regards, Isaac -- GMX FreeDSL mit DSL 6.000 Flatrate und Telefonanschluss nur 17,95 Euro/mtl.! http://dslspecial.gmx.de/freedsl-aktionspreis/?ac=OM.AD.PD003K11308T4569a _______________________________________________ computer-go mailing list [email protected] http://www.computer-go.org/mailman/listinfo/computer-go/
