Author: Martin Giepmans
Date: 11:45:15 01/11/02
Go up one level in this thread
On January 11, 2002 at 12:52:29, David Rasmussen wrote: >On January 11, 2002 at 12:10:32, Martin Giepmans wrote: > >> >>I don't know if I understand your question. >>How can you reach depth 8 when a mate is found at depth 1 or 2? >>Doesn't your engine stop in such a case? > >Yes, normally, but not in a test suite. > >>Even if you want to look for a shorter mate, searching *deeper* is of course >>useless ... > >I am not using for a shorter mate, I am just letting it search, to see how it >performs. Forget that this is a mate position. It is just a test position to >test extensions and search rules. In my case search explodes at higher depths >(>7). > >>Anyway, if the search explodes after a mate is found, I guess there must also be >>something wrong with the hashtable. >> > >Why? Even if the mate is in the hashtable, other moves are still followed deeper >with each iteration. And with extensions, some of these lines could be extended >very long, if I am not careful. And this is what happens. I don't see why that >implies a hash table problem. The hash table wont fix bad extensions. > >/David It depends on how you implement things. Once a mate in 3 is found (pv of 5 ply) any longer branch cannot possibly be a mate in 3 or less, so the engine can always stop when it finds a legal move at ply 6 (or even at ply 4 if it is only looking for shorter mates). If you do this (I guess you don't?) no explosion can occur in this position, unless the engine somehow forgets the matescore (which could indicate that the hashtable or something else that "remembers" is broken). Martin
This page took 0.01 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.