Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: One test position

Author: Will Singleton

Date: 17:25:32 06/26/03

Go up one level in this thread


On June 26, 2003 at 13:22:53, Uri Blass wrote:

>On June 26, 2003 at 12:28:45, Will Singleton wrote:
>
>>That's a tough one, for a lot of engines.  As an experiment, I added a little
>>code to return a draw score if, during any search, there appears to be a
>>"pseudo-repetition" (a series of pointless checks).
>>
>>The standard Amateur doesn't find e5 in 10 minutes.  The experimental one finds
>>it almost immediately.  This is probably an anomaly, but it does work in some
>>situations.  I'll have to look into this some more.
>>
>>6 0 62 163450 Qg3 Kc6 Qc3 Kd6 Qg3
>>7 -28 153 406041 Qg3 Kc6 Qc3 Kb6 Qxd4 c5 Qxd5 Qxd5 exd5 Bg4
>>7 0 231 613431 e5 Kxe5 Qxg7 Ke6 Qf7 Ke5 Qf4 Ke6 Qf7
>>8 0 370 993897 e5 Kxe5 Qxg7 Ke6 Re1
>>9 0 807 2220633 e5 Kxe5 Qxg7 Ke6 Re1 Ne2 Rxe2 Kd6 Qg3 Kc6 Qc3 Kb6 Re6 Bxe6 Qe3
>>10 0 2247 6141050 e5 Kxe5 Qxg7 Ke6 Re1 Kd6 Qg3 Kc6 Qc3 Kb6 Re6 Nxe6 Qe3
>>11 0 7629 15942395 e5 Kxe5 Qxg7 Kd6 Qxd4 Qg8 Rf6 Be6 Qb4 Kd7 Qb5 Ke7 Rf7 Bxf7
>>Qb4
>
>The target of white is to win and not to draw.
>How much time does amateur needs to see a positive score?
>
>Uri

The experimental rep code is probably unsound, so the score isn't right.  I
thought the idea might be of interest to others.

Will




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.