Author: William Penn
Date: 21:17:51 06/05/05
Go up one level in this thread
On June 05, 2005 at 15:18:35, Kurt Utzinger wrote: >On June 05, 2005 at 15:05:15, Eduard Nemeth wrote: > >>On June 05, 2005 at 14:45:37, Kurt Utzinger wrote: >> >>>On June 05, 2005 at 12:53:45, Eduard Nemeth wrote: >>> >>>>[D]6n1/3p4/8/1P1Pk1n1/4P3/B5K1/8/8 w - - >>>> >>>>b6! wins easy, but S9 cannot find it!!! >>> >>> >>> Interesting example to demonstrate that there >>> are situations where too much pruning may be >>> worse. On the other hand: all other engines >>> which find the move b5-b6 within seconds have >>> almost no chance to win a match vs Shredder 9. >>> The overall playing strength of a program is >>> more important than a single position which >>> can't be solved -:) >>> Kurt >> >>In my opinion S9 should not only ENG-ENG games win, he should be good for home - >>and endgame analysis too! ;) > > Hi Eduard > Just right, but what are you claiming about > > New game, 180'/40+60'/20+30' > 6n1/3p4/8/1P1Pk1n1/4P3/B5K1/8/8 w - - 0 1 > > Analysis by Shredder 9: > > 1. +- (4.45): 1.b6 Nxe4+ 2.Kh4 Nd6 3.Bxd6+ > 2. ² (0.54): 1.Bb2+ Kd6 2.Kf4 Nxe4 3.Kxe4 Kc5 4.Bg7 d6 5.Bf8 > > (, MyTown 04.06.2005) > > As you can see, Shredder 9 has no difficulty in playing > the correct move if you use increase the number of > analysing lines. > Kurt I can confirm your observation with Shredder 7.04, Shredder 8, and Shredder 9 UCI. I also discovered this peculiarity, but I don't understand why. I conclude that it is always necessary to check an analysis using multiple lines to make sure this strange behavior isn't involved. I would like to know more about this Shredder phenomenon. Is this the only known example? Is it only related to endgames? ... WP
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.