Author: Peter Berger
Date: 11:42:06 03/06/02
Go up one level in this thread
On March 04, 2002 at 17:43:29, Albert Silver wrote: >The following position occurred in Spassky-Beliavsky, Reykjavik, 1988 > >2b1qrk1/5p1p/pBn3p1/1p2p3/4P2N/bBP1Q3/P4PPP/3R2K1 w - - > >Spassky found a tremendous move 25.Nxg6!! and won the game. I don't think any >engines will have much luck with it but feel free to try it. I've analyzed it in >detail and can verify that it is the strongest move and correct, so as a further >challenge, see if you can find the *strongest* continuation after the possible >defense: 25...hxg6 26.Qh6 Be6!? > > Albert Gerrit Reubold's Bringer 1.9 needs 25 minutes on my AMD 1.33 GHz which still looks like a reasonable solution time to me. 0:22:35.8 (15/44) 376443060 0.62 1.Nf3 Bg4 2.Bd5 Nb8 3.c4 Nd7 4.Qxa3 Nxb6 0:24:59.4 (15/44) 417095484 0.63 1.Nxg6 (Mat=100,50=0) 0:29:27.1 (16/47) 495277913 0.74 1.Nxg6 hxg6 2.Qh6 Be6 3.Rd3 f5 4.Rh3 Kf7 5.Rg3 Ne7 6.exf5 Bxb3 7.fxg6+ Ke6 0:52:26.5 (17/50) 895107271 0.79 1.Nxg6 hxg6 2.Qh6 Be6 3.Rd3 f5 4.Rh3 Kf7 5.exf5 Bxb3 6.Qxg6+ Ke7 7.Rh7+ Rf7 8.axb3 Rxh7 9.f6+ Kd7 10.Qxh7+ (Mat=-35,50=0) 1:48:05.9 (18/52) 1878169802 0.78 1.Nxg6 hxg6 2.Qh6 Be6 3.Rd3 f5 4.Rh3 Kf7 5.exf5 Bxb3 6.fxg6+ Ke7 7.Qg5+ Rf6 8.Rh7+ Ke6 9.Qg4+ Kd5 10.Rd7+ Bd6 (Mat=-375,50=9) 5:49:48.5 (19/59) 6424556986 0.67 1.Nxg6 hxg6 2.Qh6 Be6 3.Rd3 f5 4.Rh3 Kf7 5.exf5 Bxb3 6.Qxg6+ Ke7 7.Rh7+ Rf7 8.axb3 Rxh7 (Mat=-555,50=0) The evaluation is not that conclusive but it was interesting to compair with Gambit Tiger 2.0 aggressive which showed a similar only faster behaviour and found it in 6-7 minutes at ply 14. Similar was that even which much higher depth the evaluation didn't rise further than right after it was found. Both seem to get it by simply searching deep (for example Nxg6 was below 10th place in the move order at ply 15 for Bringer and 3rd in Tiger's for ply 14). Although there were many results posted most were with a shorter thinking time. It would be interesting how long it takes other engines if they are given enough time to work it out a little longer. Regards, pete
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.