Author: James T. Walker
Date: 07:22:17 03/06/01
Go up one level in this thread
On March 05, 2001 at 19:12:05, John Wentworth wrote: >On March 05, 2001 at 14:11:51, Sune Larsson wrote: > >> >> [D]3k4/1p4pP/4Rp2/3p3q/3N4/8/K2PP1PP/8 w - - 0 1 >> >> Kubbel, Shajmati 1925 >> >> In this position black can relax a bit. His main worry is the dangerous >> pawn on h7, but now he's ready to send it to the blissful hunting-grounds. >> Then he should be able to play for a win since white's pawns are somewhat >> loose?! No, not this time monsigneur! Look above - the name Kubbel spells >> brilliant co-operation of the white pieces. True - it's actually white >> to move and win. This one is about covering and taking squares. And shows >> what a rook and knight can do to a queen... >> >> >> Test: White to move and win. The time for your program to find the correct >> line, gives a clue about the performance in blitz, rapid game and >> tournament level. Solution further down. >> >> >> Sune >> >> >> >> >> >> >> >> >> >>Kubbel,L >>Shajmati, 1925 >> >>1.Re3 Qxh7 2.Ne6+ Kc8 3.Rc3+ Kb8 4.Nf8 Qf5 5.e4 dxe4 6.Rc5 Qg4 7.h3 Qe2 8.Nd7+ >>Ka7 9.Ra5+ 1-0 > >Hiarcs 7.32 on PIII 450 128mb > >2:35 On my Athlon 900 Hiarcs finds Re3 in 1:00 but NOT the winning line as posted (score -0.41). It also does not have a winning score at that point. The question was how long does it take to find the "winning line" which is different from the winning move. But my point is does that matter? It seems to me that if any program finds Re3 and thereafter plays the winning line when the opponent makes the "correct" response that should be good enough. By the way on my machine it takes Hiarcs 7.32 2:47 to find the "winning line" with a score of +2.11 which later climbs to +2.7. Jim
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.