Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Gambit 2.0 plays like Botvinnik!

Author: Robert Hyatt

Date: 13:10:17 04/02/01

Go up one level in this thread


On April 02, 2001 at 15:39:45, Uri Blass wrote:

>On April 02, 2001 at 15:01:24, Robert Hyatt wrote:
>
>>On April 02, 2001 at 14:36:41, Richard Heldmann wrote:
>>
>>>Deep Shredder finds it in 00:00:00 seconds. <yawn>
>>
>>
>>So does Crafty.. but it quickly changes its mind to something else.  It
>>quickly sees 0.00 but then finds something better.  It later sees that
>>Ba3 is not just a draw but a win.
>>
>>I think several others see a 0.00 score instantly, but beyond that it depends
>>on how deep you search to see the _real_ win.  IE my 13 ply score was almost
>>+4.  GT had a score of +1, which means it hadn't seen the full consequences
>>of Ba3 completely, until another iteration or so.
>
>It found the full consequence of Ba3 at the same iteration(iteration 14)
>see http://www.icdchess.com/forums/1/message.shtml?161242
>
>The first score was probably the score after failing high without solving the
>fail high problem and when it solved the fail high it could see a big score of
>more than +4.
>
>Uri


OK.  My program is a bit clearer in the way it outputs scores.  It doesn't give
a "score" for fail highs since that looks confusing.  It says "++" which means
the score is significantly higher than before, but it doesn't yet know how high
until the fail high is resolved...



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.