Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Who can solve this stupidly easy position... (solved)

Author: Uri Blass

Date: 08:37:03 07/19/02

Go up one level in this thread


On July 19, 2002 at 11:24:57, Alejandro Dubrovsky wrote:

>On July 19, 2002 at 11:07:15, Omid David wrote:
>
>>On July 19, 2002 at 09:56:34, Alejandro Dubrovsky wrote:
>>
>>>On July 18, 2002 at 19:01:25, Omid David wrote:
>>>
>>>>Which program can declare a draw in the following position:
>>>>
>>>>[D]8/4k3/8/p1p1p1p1/P1P1P1P1/8/2B1K3/8 w
>>>
>>>right evaluation in any case.  just check if the hash depth is greater than 100,
>>>then it must be a draw
>>
>>Good job!
>>However in a little more complicated positions the programs might not be able to
>>reach the depth of 100 plies!
>
>sure. but there isn't anything much more complicated where a static evaluator
>would be reasonably expected to declare a draw either.  eg
>[D]4k3/p4rr1/Pp1p4/1PpPp3/2P1Pp1p/5PpP/1RBR2P1/4K3 w
>
>smallpotato here doesn't even get to 20 in any nice time, but there are programs
>that would get to 100 rather quickly i believe, and be able to call it.  i don't
>think there would be many positions with more pieces than this that a static
>evaluator would be able to declare as a draw

I do not believe that there are programs that can
get 100 plies quickly here.

You need to consider options like sacrificing the rook and program search
capture the rook first so they have no chance to get 100 plies.

Uri



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.