Author: Robert Hyatt
Date: 11:21:07 05/13/04
Go up one level in this thread
On May 13, 2004 at 13:39:10, Jeff GAZET wrote: >>>Thanks, i found : This bug occured only when i stopped the search with "move >>>now". In this case, in alphabeta() and quiesce(), i set stop_search=true and >>>returned immediatly. As those functions must return a value, i returned 0. >>>This put bad values in the hash table. >>>So i clear the table if the search as been stopped in alphabeta() or quiesce(). > >>Poor solution. Just don't "store" anything once time runs out. Anything you >>have _already_ stored is just fine. But when you stop searching, return a >>special value or check the "time-out" global flag and just don't store into the >>hash table once that has happened. >>Otherwise you will be clearing the hash table after most every search and that >>is not a good approach. > >Yes, that's what i did in fact, as the table was cleared every time. >So, i don't insert informations into the table if the time is out. >How do other engines ? those with no threads ? >Thanks. that is what I do, threads or not. An "interrupted" search is no good. Anything backed up to the root prior to the interruption is perfectly fine, but after the interruption no new information can be trusted.
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.