Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Avoid move position - testing repetitions

Author: Dieter Buerssner

Date: 10:19:30 01/10/03

Go up one level in this thread


On January 10, 2003 at 12:54:02, Frank Phillips wrote:

>Thanks for the suggestions.  In fact I tried all those things.  No hash, no
>extentions and no (hash + no extensions).
>
>After 1 hour 15 mins (of normal engine configuration) I have Re3+ and +0.65.

Do you mean +0.65 for black or for white? If for some reason (which I could not
understand) your engines evalutes black better, you won't see it. Also, do you
give 0.0 for the first repetition (at least deeper in the tree)? If not, it will
proablaby be too deep. As Uri said, switching off extensions, will hurt. You
could switch off any extensions beside check. With 1 ply extension for check,
the line should be about 11 ply deep. I did not think carefully about it, but
null move may delay things. The "silent" move Kd2 will be needed, after which
null move may fail high with not enough depth. Also you might want to try to
start a little bit later in the line I have given. It should almost be possible,
to investigate the problem by a search tree dump (but a lot of work ...).

>Here is a similar one I cannot get either.  It have come from CCC, but I cannot
>remember.
> 7k/4QP2/7p/6rP/8/1K6/6r1/8 b - -

I get 0.0 after 30 seconds at ply 14 (P4, 2.53 GHz). This one is harder for Yace
than the above one.

Good luck,
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.