Author: James T. Walker
Date: 14:10:35 11/21/02
Go up one level in this thread
On November 21, 2002 at 10:20:45, Roy Eassa wrote: >On November 20, 2002 at 18:30:54, James T. Walker wrote: > >>On November 20, 2002 at 18:14:06, Roy Eassa wrote: >> >>>On November 20, 2002 at 17:00:19, James T. Walker wrote: >>> >>>>On November 19, 2002 at 19:18:55, Roy Eassa wrote: >>>> >>>>>White to move and win. >>>>> >>>>>1q1k4/2Rr4/8/2Q3K1/8/8/8/8 w - - 0 1 >>>>> >>>>>Shredder 6.02 and Fritz 6 can't find it; most of the other top engines see it >>>>>instantly. It's zugzwang, of course; I knew Fritz was susceptible but I'm >>>>>surprised by Shredder. >>>> >>>>Fritz 6 & Fritz 7 find Mate in 33 instantly on my computer. Maybe you are >>>>missing some tablebases? >>>>Jim >>> >>> >>>I don't use 'em! >> >>Then don't blame the program for making endgame errors. :-)) >>It's really hard to ignore perfect knowledge when it's available. >>Jim > > >The winning line should be trivial for any modern program to *calculate*. It probably is if you turn off null move pruning. However you then make many other moves which were trivial not trivial any more. Everything in life/world is a compromise. Jim
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.