Author: Howard Exner
Date: 09:15:57 09/01/98
Go up one level in this thread
On September 01, 1998 at 08:42:30, Amir Ban wrote: >On September 01, 1998 at 07:17:44, Howard Exner wrote: > > >>5. Rebel missed a chance to immediately transpose to a won ending with >>the moves 1. Qxe6+ Qxe6 2. Rxe6 Kxe6. Black's passed d pawn can be >>easily stopped while white's pawns are unstoppable. Instead Rebel played >>the liguidating sequence 1. Qb7+ Kd4 2. Qb6+ Kd3 3. Qxe3+ Rxe3 4. Rxe3 Kxe3. >>Note the better location now of black's king. Black should be able to draw this >>now by advancing his d pawn. >>4R3/5Q2/3pr3/p2k1p2/5P2/1P2q1PP/P6K/8 w - - id "Rebel 9.0 - Hiarcs 6.0"; am >>Qb7+; >> > >This is wrong. Qb7+ is the right move, and it wins a rook: Indeed. Qb7+ wins but in a game I would play Qxe6 in seconds as the endgame can be won in your sleep. This is what first caught my eye in this position, namely that the human route to victory is so elementary while the computer program misses it. Qb7+ followed by the easy win of a rook also wins easily. I should have had the test position from move #3 with avoid move Qxe3. > >1.Qb7+ Kd4 2.Qb6+ Kd3 3.Qb5+ Kc2 4.Qc4+ or 3...Kd2 Qd5+. > >How did Rebel manage to miss this ? Rebel8 for the longest time evaluates Qxe3? as +10.00. It has a real difficult time as it must think that white Queens while black cannot. Thanks for spotting this.
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.