Author: Dieter Buerssner
Date: 06:56:20 12/21/03
Go up one level in this thread
On December 21, 2003 at 09:05:42, Omid David Tabibi wrote: >On December 21, 2003 at 08:53:39, Jaime Benito de Valle Ruiz wrote: > >>I've never read any thread about this before, so forgive me if this is a >>recurring topic. >>Many people have noticed that the Principal Variation shown by Shredder is >>sometimes ridiculous, to the point of giving out its queen and a rook for >>nothing, or allowing a silly mate. > >My understanding is that unlike most other engines when failing high or low >Shredder does display a full PV extracted from hash tables (whereas other >engines just show the first move). These moves have naturally bounded scores, >and so are not reliable. That is also one of the main problems in extracting PV >using MTD(f) search. But other than that, Shredder's PVs are accurate. I guess, Shredder is always displaying a PV reconstructed from hash (not only in fail high/fail low situations). Between search and display time, hash contents can change, and the displayed PV can look stupid. My engine also displays variations when failing high. This is done by updating a typical PV-array also in the case of fail low (where most engines probably do not update the PV-array). Typically the displayed PV makes sense, and is understandable as a "refutation line". When the search ends in a HT-cutoff, things are different, and I see stupid continuations after that cutoff, when reconstructing more from the variation via HTs at display time. Regards, Dieter
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.