Author: Ralf Elvsén
Date: 17:39:47 10/09/00
Go up one level in this thread
On October 09, 2000 at 20:26:52, Dan Newman wrote: >On October 09, 2000 at 17:14:41, Ralf Elvsén wrote: > >>On October 09, 2000 at 16:19:44, Dan Newman wrote: >> >>> >>>Mine (Shrike) is doing this in just under 3s (1.27M nodes) on a P3/600. >>>It's getting about a 35% hit rate on the hash table, so this may well >>>be the problem. Those hash table moves are really important in getting >>>good move ordering (and a low branching factor). >>> >>>I also keep stats on the fraction of cutoffs from the first move tried >>>and the fraction that are from hash table moves. In this case I'm >>>getting 92% of the cutoffs from the first move and 25% from the hash table. >>> >>>-Dan. >> >>Sorry if I'm dense: 25% of the times when the move stored in the hashtable >>was searched, it lead to a cutoff? I.e. you had a hashhit but >>the score and depth wasn't enough to exit the node? >> > >Sorry about the confusion. What I meant was that 25% of the cutoffs were >due to trying a hash table move. > >>Of the remaining cases (no hashhit) the first move tried >>caused a cutoff in 92% of the cases? > >Don't I wish :). It's just the number of first-move cutoffs divided by >the total number of cutoffs. Yes, I look even more stupid then I am. I wrote something else then what I meant. >This number includes the hash table moves >since I try them first--if they are available. It's meant as a sort of >measure of how good the move ordering is. > >-Dan. So after passing the hashtest, 92% of the cutoffs comes from the first move searched. Of these 25% came from hashtablemoves, i.e. no movegeneration needed. Hope I got it :) > >> >>I guess I'm totally wrong but I just want to demonstrate how >>confused I am :) >> >>Ralf
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.