Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Rybka EGTB example

Author: G. R. Morton

Date: 14:04:54 01/27/06

Go up one level in this thread


On January 27, 2006 at 14:09:22, George Tsavdaris wrote:

>On January 27, 2006 at 13:49:15, G. R. Morton wrote:
>
>>On January 27, 2006 at 12:27:46, Jouni Uski wrote:
>>
>>>
>>>8/7p/8/p4p2/5K2/Bpk3P1/4P2P/8 w - - 0 1
>>>
>>>1.Kxf5 b2 2.Bxb2+ Kxb2 3.e4 a4 4.e5 a3 5.e6 a2 6.e7 a1Q 7.e8Q Qf1+
>>>  ±  (1.25)   Depth: 17   00:00:26  2132kN
>>>1.g4 fxg4 2.e4 Kc4 3.Kxg4 Kb5 4.Kg5 Ka4 5.Bc1 Kb5 6.e5 a4 7.e6 Kc6
>>>  +-  (4.09)   Depth: 17   00:00:59  6451kN
>>>1.g4 fxg4 2.e4 Kc4 3.Kxg4 Kb5 4.Kg5 Ka4 5.Bc1 Kb4 6.e5 a4 7.e6 a3
>>>  +-  (5.33)   Depth: 18   00:01:06  8045kN
>>>
>>>I have one CD = 700 MB tablebases.
>>>
>>>Jouni
>>
>>Very interesting, it's endgame positions like these that make even the best
>>engines look so dumb - a human sub-expert can easily see that g4 is the tempo
>>move. However, Shredder 8 has no problem here. Presumably Rybka versions will
>>eventually find quickly such endgame moves.
>>
>Do you want more quickly than 59 seconds.....? :-)

Actually yes I do, even I found it in less than 59 sec and any enigne slower
than me is a disgrace  :-)

However, I was using the Rybka 1.01 Beta 10d (3.4 Ghz P4 hash 640M) and I got
tired waiting while it thought long on moves like e3. My point, of course, is
not to put down the superb Rybka or any engine but just to express my amusement
that strong engines that generally an amateur like myself could not hope to
defeat can still look stupid in certain positions. This was true for a number of
top free engines I used (but not Pro Deo). Amazing.

GR



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.