Author: Robert Hyatt
Date: 08:51:46 03/02/06
Go up one level in this thread
On March 02, 2006 at 01:41:01, Gerd Isenberg wrote: >On March 01, 2006 at 19:52:09, Robert Hyatt wrote: > >>On March 01, 2006 at 18:25:24, Gerd Isenberg wrote: >> >>>What about an additional boolean (0..1) history table dimension, >>>true if Parent is first child of grandparent? >> >>Have not thought about it, to be honest. But isn't that recursive? i.e. for >>ply=3 that would be true for first move, false otherwise, but then at ply=4 >>wouldn't the same thing happen recursively??? > >during the pv. > >> >>Or maybe after 1:15 of lecturing on virtual memory page replacement, my head's a >>bit foggy? :) > >I thought on distinguishing late cutoffs at "expected cut-nodes" versus >"unexpected" cutoffs at all-nodes... aha, now the meaning is clear. And it is at least something to think about and test...
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.