Author: Sune Fischer
Date: 13:07:50 04/07/03
Go up one level in this thread
On April 07, 2003 at 15:14:49, Russell Reagan wrote: >On April 07, 2003 at 14:25:25, Sune Fischer wrote: > >>No I think Yace is faster now. > >I can't find any positions where Yace is faster. It's close, but not faster. >Could be different on different processors I suppose. I don't know if Dieter has made that version public (yet), but we had a little contest tuning for perft and Yace ended up being 10-15 times faster. The trick is to hash the nodes, but since the "rules" doesn't say anything about how you should calculate perft, a hash is perfectly legal. Movei was already using its own bag of tricks to calculate perft fast, so hashing was just the next step. It might be possible for Movei to reclaim the title if Uri implemented a hash also, but until then I think Yace is the current speed holder. Yace calculated perft 9 in about 3 hours, perft 9 was sort of the goal in that little contest :) -S.
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.