Author: Uri Blass
Date: 12:31:33 10/24/03
Go up one level in this thread
On October 24, 2003 at 15:10:25, Afzal Siddique wrote: >On October 24, 2003 at 10:21:01, Mihaly Szalai wrote: > >>[D]8/6p1/3N4/8/4B3/4K1kp/5p1p/8 w - - 0 1 >> >>This position is nice and not too difficult. (White wins.) >>E.g. Ruffian solves it immediately. >>You should give it a try with your engine. >> >>Thanks >>Mihaly >> >>PS. Junior is baffled once again. > > >Hi, > >Suprisingly Junior 5.0 finds it in an instant > >New game >8/6p1/3N4/8/4B3/4K1kp/5p1p/8 w - - 0 1 > >Analysis by Junior 5.0: > >1. +- (4.74): 1.Nf5+ Kg4 2.Kxf2 Kf4 3.Ng3 Kg5 4.Bh1 Kg4 5.Nf1 Kf4 6.Nxh2 g6 >7.Bf3 >2. = (0.00): 1.Ke2 h1Q 2.Bxh1 Kh2 3.Ne4 Kxh1 4.Kf1 > >(, MyTown 23.10.2003) No Junior5 does not find it The solution is based on the idea 1.Ke2 h1Q 2.Bxh1 Kh2 3.Kf1 Kxh1 Junior5 does not consider underpromotions so it find wrong positive score for another move. 1.Nf5+??? Kg4 2.Kxf2 h1N is a draw but Junior5 cannot find it. Junior cannot find the solution because it has more chess knowledge than Ruffian. Junior knows that King and knight with no more material cannot win against king and pawns. Ruffian does not know it so it searches and find the win after 1.Ke2 h1Q 2.Bxh1 Kh2 3.Kf1 Kxh1. It is a waste of time to give Junior studies because it is designed to fail in studies by not considering illogical lines(Junior5 even does not consider underpromotions). 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.