Author: David Rasmussen
Date: 06:30:49 09/18/02
Go up one level in this thread
On September 18, 2002 at 07:38:32, James Swafford wrote: >On September 18, 2002 at 05:19:49, David Rasmussen wrote: > >>On September 17, 2002 at 21:31:17, James Swafford wrote: >> >>> >>>I'm skeptical about my hash efficiency - it seems to me I'm not >>>getting enough hits from my hash tables. >>> >>>I wonder if some of you would post some numbers for me to compare >>>by running the following 3 positions: >>> >>>r4k2/q4npp/P2P4/2p1R3/2N5/6PP/Q5K1/8 w - - bm Re7; id "ECM.1028"; >> >> Time used: 00:00:10.00 >> nps: 512962 nodes: 5134756 Q-nodes: 390756 (7%) >> move order: 94% >> hash probes: 1394483 hash hits: 400348 (28%) hash usage: 13% >> pawn hash probes: 2687859 pawn hash hits: 2683772 (99%) >> EGTB probes: 0 EGTB hits: 0 (0%) > >Is hash usage a sum of cutoffs, or does it include nodes in which >you get a move to try as well? > Hash usage is just the saturation of the hashtable. Only 13% of the entries were written to at all, in this run. That indicates that the hashtable doesn't need to be this big, for such a short run. Hash probes counts how many times I probe the hash table at all, and hits count how many times there is a signature match with the draft being ok. This means that there is an exact score, an upper bound or a lower bound. An exact score is always used. A bound may not always be useful. So even if there's a hit, it might not produce a cutoff. On the other hand, in all cases where there is a signature match, a good move might be available from the hashtable, for move ordering purposes. How often there is a move or not, or how often ordering this first results in a cutoff, I don't know. /David
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.