Author: Steve Maughan
Date: 06:28:45 10/12/00
Go up one level in this thread
Bas, >I wouldnt suspect so. In my case the system of hashing doesn't seem to matter a >lot. I tried 8 probes, 4 probes and Crafty's system. No big differences there. >I tried to prove that Crafy's system did better at extreme depths, but even >there it doesn't matter a lot. > >As for the cutfirst rate, I do over 93% average on WAC. But for example in the >startposition I see 88% or so. I don't see how your hashing system would affect >this much. I've managed to solve the hash table problem - it wasn't really a problem, just the way I was counting the nodes. I've checked again and the cutfirst rate is actually 83% for the initial position, which I am not happy with! I can't understand why it is so low when I do the standard move order of Hash, Good Captures, Killer1, Killer2, History, the rest. Any ideas? I see you're entering the Dutch. Is this you x88 or your bitboard program? Are you still using C++ Builder? Regards, Steve Maughan
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.