Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Fritz wrestling with a winning move?

Author: Ernst Walet

Date: 12:40:58 12/31/01

Go up one level in this thread


On December 30, 2001 at 17:12:13, pavel wrote:

>On December 30, 2001 at 17:07:05, pavel wrote:
>
>>On December 30, 2001 at 16:55:41, Jeroen van Dorp wrote:
>>
>>>[D]r1b2r2/3pNpkp/3pn1p1/2pN3P/2PnP3/q3QP2/4BKP1/1R5R w - - 0 1
>>>
>>>This is a problem: white to move and win.
>>>When looking at it, (as a human, of course:)) you immediately start thinking
>>>about h6+ or Qh6+ followed by hxg6+ looking at the two rooks comfortably with an
>>>open h-file, and the knights "checking" the mobility of the black king.
>>>
>>>All engines I tested started however with Qd2, and almost all switched to Qh6+
>>>very soon on my PIII@840: Gandalf 5, 47 seconds, CT 14.0 34 seconds, Shredder 6
>>>32 seconds , Hiarcs 7.32 in 54 seconds, Nimzo 8 saw the mate in 20 seconds,
>>>Junior 7 in 39 seconds, and Fritz 7(both MMX (7a) and no-MMX)... didn't find it
>>>but after some 6 minutes, and found the mate in 9 minutes 41 seconds.
>>>
>>>What could be the reason?
>>>Are there people here willing to check it with Fritz on faster machines? BTW not
>>>only looking for the *move* Qh6+ but for the move *and mate* with Qh6+
>>>
>>>
>>>[Event "?"]
>>>[Site "?"]
>>>[Date "2001.12.30"]
>>>[Round "?"]
>>>[White "?"]
>>>[Black "?"]
>>>[Result "*"]
>>>[Opening ""]
>>>[Setup "1"]
>>>[FEN "r1b2r2/3pNpkp/3pn1p1/2pN3P/2PnP3/q3QP2/4BKP1/1R5R w - - 0 1"]
>>>
>>>1.Qh6+ Kxh6 ( 1...Kh8 2.hxg6 Qxf3+ 3.gxf3 Nxf3 4.Qxh7# ) 2.hxg6+ Kg5 ( 2...Kg7
>>>3.Rxh7#) 3.Rh5+ Kxh5 $8  4.f4+ Qf3+ ( 4...Kh4 5.Rh1+ Qh3 6.Rxh3# ) 5.Bxf3+ Nxf3
>>>6.Rh1+ Nh4 7.Nf6+ Kh6 8.Rxh4+ Kg7 9.Ne8+ Rxe8 10.Rxh7+ Kf8 11.Rxf7# *
>>
>>
>>How much nps do you get with Fritz7a on this position?
>
>
>It took me 11.15mins to find it in my PIII 1Ghz 512mb ram.
>
>wonder why. I got around 530+nps.
>
>pavs

Hastable size?

Ernst.

New game
r1b2r2/3pNpkp/3pn1p1/2pN3P/2PnP3/q3QP2/4BKP1/1R5R w - - 0 1

Analysis by Fritz 7:

1.Qd2 g5 2.Nb6 Nf4 3.h6+ Kh8
  ±  (1.32)   Depth: 6/24   00:00:00  246kN
1.Qd2 g5 2.Nb6 Nf4 3.h6+ Kh8 4.Nxa8 Qxa8
  ±  (1.32)   Depth: 7/30   00:00:00  373kN
1.Qd2 g5 2.Nb6 Rb8 3.Nbxc8 Rfxc8 4.h6+ Kh8 5.Nxc8 Rxc8
  ±  (1.38)   Depth: 8/26   00:00:01  936kN
1.Qd2 g5 2.Nb6 Rb8 3.Nbxc8 Rfxc8 4.h6+ Kh8 5.Nxc8 Rxc8 6.Rhd1
  +-  (1.54)   Depth: 9/31   00:00:04  2491kN
1.Qd2 g5 2.Nf5+ Kg8 3.Nxd6 Nf4 4.Nxf4 gxf4 5.Rh4 Qa2 6.Rb2 Qa1 7.Rg4+ Kh8
  +-  (1.41)   Depth: 10/33   00:00:09  6079kN
1.Qd2 g5 2.Nf5+ Kh8 3.Nxd6 Qa2 4.Rb2 Qa3 5.e5 Nf4 6.Nxf4 gxf4 7.Rhb1
  +-  (1.50)   Depth: 11/31   00:00:31  19777kN
1.Bd3!
  +-  (1.54)   Depth: 11/36   00:00:54  34452kN
1.Bd3 g5 2.e5 g4 3.Rh4 g3+ 4.Kxg3 Kh8 5.exd6 Ba6 6.Kh2 Rfe8
  +-  (1.69)   Depth: 12/39   00:02:22  89234kN
1.Qh6+!
  +-  (1.72)   Depth: 12/39   00:03:02  115257kN
1.Qh6+!
  +-  (1.88)   Depth: 12/39   00:03:02  115416kN
1.Qh6+!
  +-  (2.19)   Depth: 12/39   00:03:02  115616kN
1.Qh6+!
  +-  (2.82)   Depth: 12/39   00:03:03  115922kN
1.Qh6+ Kxh6
  +-  (#11)   Depth: 12/40   00:06:44  250243kN
1.Qh6+ Kxh6
  +-  (#11)   Depth: 12/40   00:06:44  250378kN

(Celeron 1.2GHz, 832MB hash, 3,4 and 5 piece EGTB's 31.12.2001)




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.