Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: positions to search deep for hours at 500 cpu's

Author: Dann Corbit

Date: 14:54:17 09/19/03

Go up one level in this thread


On September 19, 2003 at 16:58:57, Dieter Buerssner wrote:

>On September 19, 2003 at 15:55:41, Dann Corbit wrote:
>
>>WAC 230.
>>As far as I know, it still isn't really decided if the key move really leads to
>>a win.  I saw some convincing draw argumentation.  I would really like to know
>>once and for all if it is a draw or not.
>
>I fear, without going deeper into the line(s), even 500 CPUs will not be able to
>give a definite answer, even after hours.

Perhaps Vincent's program will trace along the pv.
In one scheduling algorithm I am considering, one processor will follow the
current PV and play at twice the rate of the root processor.  The notion is to
fill the hash table with forward data and also to spot potentially deep
problems.

At any rate, I would like to see what his program might come up with.

The nolot positions would be another natural set, along with the more difficult
positions from LCT II.



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.