Author: Roland Pfister
Date: 03:54:04 05/13/98
Go up one level in this thread
On May 13, 1998 at 06:28:11, Howard Exner wrote: >On May 12, 1998 at 15:56:29, blass uri wrote: > > >>39...Nh5 does not win >>40.Qf2 Qe5+ >>41.g3 Nxg3 >>42.Rd5 Nf1+ >>43.Kg1 Qe6 >>44.d7 Qg6+ >>45.Kh1 Ng3+ >>46.Kh2 and black has to do draw by Nf1+ (Ithink the computer >>did not see that Rh1+ 47.Kg2 Ne4+ 48.Kxh1 Nxf2+ Kh2 >>win for white) >>Fritz5 also does not see in reasonable time Nh5 does not win > >This draw discovery I think makes this position even a better >one for testing. Perhaps the best move is now Re3 instead of >Nh5. It would be interesting to see how many programs play Nh5 >and when they again decide to reject the move. Is this a classic example >of the horizon effect? I had Patzer run all night (11 hours) on a UltraSparc 167MHz to see when it will find Nh5. From depth 6 to 14 (when time was over) it showed Re3 with a draw evaluation (-0.09). After reading Uri's post I entered Nh5 and it showed also a draw. It follows exactly his moves. So why do other programs think that Nh5 wins?
This page took 0.01 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.