Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Position from Hydra's game

Author: Dieter Buerssner

Date: 13:06:30 05/27/05

Go up one level in this thread


On May 27, 2005 at 15:47:47, John Merlino wrote:

>Pretty amazing if Hydra really did announce Mate in 16 during the game, no?

Few years ago, Chrilly Donninger explained to me personally, that it is not
possible, to get correct mate scores with hash tables (!!??). Later, he
explained, that it does not matter, to get correct mate scores (which I would
agree to, when it is really guaranteed, that the engine will deliver mate in all
circumstances. Still I would not like a mate in x score, when it really turns
out to be only mate in x+n with positive n. I of course, prefer to have may
engine only show mate in n, when it is a strict lower or upper bound.) In this
context, can anybody show the mate in 16, or do we have to doubt it? When Hydra
is still using "vague" mate scores, I would not be too impressed. Also, no
mention, if Hydra used TBs. With TBs finding mate in 19 seems not too difficult;
I assume many engines will find it in reasonable time. In the concrete position,
probably almost any engine will win - so it does not really matter, what score
it shows; be it +7 or +20 or mate in x.

Regards,
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.