Computer Chess Club Archives


Search

Terms

Messages

Subject: Endgame Tablebases (was Re: Reti Endgame study)

Author: Stuart Cracraft

Date: 19:20:57 01/14/98

Go up one level in this thread


That endgame tablebase search you posted was certainly interesting.
Can you digress a little bit about what you did to integrate tablebases
in standard play. To the endgame tablebase layperson, it would seem
impossible: the files are so large as to not be comfortable in memory
and fseeking a disk file during tree search at end points or in the
search would be too slow. What's the trick?

Also, with maximum tablebases enabled, what do you think is the
increase in ratings point strength on ICS?

--Stuart

On January 14, 1998 at 21:13:28, Robert Hyatt wrote:

>On January 14, 1998 at 17:47:04, Bruce Moreland wrote:
>
>>
>>On January 14, 1998 at 17:31:10, Stuart Cracraft wrote:
>>
>>>Reti composed a famous endgame study with two outside
>>>passed pawns. I forget the position. It seems like,
>>>for limited search, it might be good for some types
>>>of tests.
>>>
>>>If you know it, please post it, along with your solution
>>>time both with and without hashtable, if possible.
>>
>>7K/8/k1P5/7p/8/8/8/8 w - - 0 1
>>
>>White to play and draw, solution is 1. Kg7, and this is easy.
>>
>>bruce
>
>
>two ways to solve it.  First with endgame databases:  :)
>              depth   time  score   variation (1)
>                1     0.09 -Mat14   c7 <EGTB>
>                1     0.09     ++   Kg8!!
>                1     0.09  -7.48   Kg8
>                1->   0.09  -7.48   Kg8
>                2     0.09     --   Kg8
>                2     0.09 -Mat13   Kg8 <EGTB>
>                2     0.09 -Mat14   c7 <EGTB>
>                2     0.09     ++   Kg7!!
>                2     0.09   0.00   Kg7 <EGTB>
>                2->   0.10   0.00   Kg7 <EGTB>
>
>and without:
>              depth   time  score   variation (1)
>                1     0.00  -7.28   c7 Kb7
>                1->   0.01  -7.28   c7 Kb7
>                2     0.02     ++   c7!!
>                2->   0.04  -6.99   c7 Kb7
>                3     0.05     --   c7
>                3     0.07  -9.18   c7 Kb7 c8=Q+ Kxc8 Kg8
>                3     0.08     ++   Kg8!!
>                3     0.09  -7.34   Kg8 h4 c7 Kb7
>                3->   0.10  -7.34   Kg8 h4 c7 Kb7
>                4     0.12     --   Kg8
>                4     0.13  -9.18   Kg8 Kb6 Kf7 Kxc6
>                4     0.15     ++   Kg7!!
>                4     0.16  -7.57   Kg7 Kb6 Kg6 h4
>                4->   0.17  -7.57   Kg7 Kb6 Kg6 h4
>                5     0.18  -7.34   Kg7 Kb6 Kg6 h4 c7
>                5->   0.20  -7.34   Kg7 Kb6 Kg6 h4 c7
>                6     0.22  -7.42   Kg7 Kb6 Kf6 h4 Ke5 h3 c7
>                6->   0.23  -7.42   Kg7 Kb6 Kf6 h4 Ke5 h3 c7
>                7     0.26  -7.45   Kg7 Kb6 Kf6 h4 Ke5 h3 Kd6 Ka6
>                7->   0.27  -7.45   Kg7 Kb6 Kf6 h4 Ke5 h3 Kd6 Ka6
>                8     0.36     ++   Kg7!!
>                8     0.41  -7.02   Kg7 Kb6 Kf6 h4 Ke5 h3 Kd6 Ka6 c7 Kb7
>                8->   0.43  -7.02   Kg7 Kb6 Kf6 h4 Ke5 h3 Kd6 Ka6 c7 Kb7
>                9     0.52     ++   Kg7!!
>                9->   0.57  -6.73   Kg7 Kb6 Kf6 h4 Ke5 h3 Kd6 Ka6 c7 Kb7
>               10     0.68     ++   Kg7!!
>               10     0.89  -0.03   Kg7 Kb6 Kf6 h4 Ke5 h3 Kd6 Ka7 Kd7 h2
>                                    c7 h1=Q c8=Q Qd5+ Ke7
>               10->   0.94  -0.03   Kg7 Kb6 Kf6 h4 Ke5 h3 Kd6 Ka7 Kd7 h2
>                                    c7 h1=Q c8=Q Qd5+ Ke7
>               11     1.44   0.00   Kg7 Kb6 Kf6 h4 Ke5 h3 Kd6 h2 c7 h1=Q
>                                    c8=Q Qd1+ Ke7 Qd2 Qd7 Qxd7+ Kxd7
>
>the above is on my notebook with a small default hash
>table...



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.