Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Shredder 8 Bug (probably already known)

Author: Jorge Pichard

Date: 14:42:50 02/23/04

Go up one level in this thread


On February 23, 2004 at 15:02:47, Uri Blass wrote:

>On February 23, 2004 at 13:33:12, John Merlino wrote:
>
>>Michael Byrne posted some analysis by Shredder 8 yesterday that ended with these
>>two PVs:

That is strange, I just run two infinite Analysis on my Athlon 1.2 GHz for 09:40
and it gave me two different results:

[d]2r2rk1/p3bb1p/2n1Q1p1/q2pP3/3P1P2/p1NB1NR1/1P4P1/1K1R4 w - - 0 24

New game
2r2rk1/p3bb1p/2n1Q1p1/q2pP3/3P1P2/p1NB1NR1/1P4P1/1K1R4 w - - 0 1

Analysis by Shredder 8:

1. +- (5.58): 1.Bxg6 Bxe6 2.Bxh7+ Kh8 3.Rh1 Bh4 4.Bd3 Rxf4 5.Nxh4 Rf1+ 6.Rxf1
Nxe5 7.dxe5 a2+ 8.Nxa2 Bf5
2. -+ (-6.94): 1.Rxg6+ hxg6 2.Qg4 Qb6 3.Rd2 Nxd4 4.Nxd4 Rxc3 5.Nf5 Rb8 6.Nh6+
Kf8 7.Qxg6 Bxg6 8.Nf5
+-  (5.61)   Depth: 15/47   00:09:40  140720kN
(Pichard, MyTown 23.02.2004)



New game
2r2rk1/p3bb1p/2n1Q1p1/q2pP3/3P1P2/p1NB1NR1/1P4P1/1K1R4 w - - 0 1

Analysis by Shredder 8:

1. +- (6.23): 1.Bxg6 Bxe6 2.Bxh7+ Kh8 3.Rh1 a2+ 4.Ka1 Rxf4 5.Bd3+ Bh4 6.Nxh4
Rf1+ 7.Rxf1 Bf5 8.Rxf5 Nxe5 9.dxe5
2. -+ (-6.93): 1.Rxg6+ hxg6 2.Qg4 Qb6 3.Rd2 Nxd4 4.Nxd4 Rxc3 5.Nf5 Rc1+ 6.Kxc1
a2 7.Nxe7+ Kg7 8.Kd1 a1Q+ 9.Ke2
+-  (5.61)   Depth: 15/47   00:09:40  140720kN
(Pichard, MyTown 23.02.2004)



>>24.Bxg6 Bxe6 25.Bxh7+ Kh8 26.Rh1 Bh4 27.Bd3 Rxf4 28.Nxh4 Nxd4 29.Nf3+ Bh3
>>30.Rhxh3+ Rh4 31.Rxh4#
>>  +-  (5.11)   Depth: 15/45   00:03:50  85550kN
>>24.Bxg6 Bxe6 25.Bxh7+ Kh8 26.Rh1 Bh4 27.Bd3 Rxf4 28.Nxh4 Rxd4 29.Nf5+ Rh4
>>30.Rxh4#
>>  +-  (5.61)   Depth: 15/46   00:04:40  100640kN
>>
>>Obviously, those '#' symbols are incongruous with the eval. Has this been
>>mentioned before?
>>
>>The post is at www.talkchess.com/forums/1/message.html?350488.
>>
>>jm
>
>
>I saw this king of behaviour with some programs.
>Shredder is not the first program when it happened to it and I saw it with
>Junior and Crafty in the past.
>
>I guess that it is no bug and there are 2 possibilities:
>
>1)Shredder does not detect mate in the evaluation
>2)The pv is some not reliable moves that shredder collected from the hash tables
>instead of having pv array.
>
>Uri



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.