Author: Mike S.
Date: 10:16:39 12/23/01
Go up one level in this thread
On December 23, 2001 at 01:59:24, Jim Monaghan wrote: >This position is Capablanca-Kostic, 1919 >[D]8/4bkp1/5p1p/r1pR1P1K/1pP4P/1P2B3/5P2/8 w - - 0 1 >Black's in a bad way. To win White just needs to penetrate to g6 with his king. >1. Rd7 Ke8 2. Rd3 Kf7 3. Rd5! >(zugzwang, same initial position but with black to move) > >If 3... Bf8 4. Rd7+ Be7 5. Rc7! >(...) On fast machines does any program see the main idea and stick with >it. Or is this a good test position? I think it's interesting because test positions where this motif of passing the right to move in the same position occurs, are rare. I don't have a very fast machine either (P3/700): Shredder 5.32 seems to find a simpler way to pass the right to move to Black (in "nearly" the same position so to speak): 1.f3, and the variation follows the Kostic line until Kh5-g6, but later Shredder prefers 1.Rd7 Ke8 2.Rc7 again in it's pv. For some time in between, it followed the idea with Rd5-d7-d3-d5, but didn't stick to it. Analysis by Shredder 5.32: 1.Bxc5 Bxc5 -+ (-1.59) Depth: 1/2 00:00:00 -+ (-3.11) Depth: 1/2 00:00:00 1.Rd7 Ke8 2.Rd1 Ra3 3.Rd3 µ (-1.02) Depth: 1/2 00:00:00 = (0.01) Depth: 5/10 00:00:00 1.Kg4 Ra3 2.Rd3 Ra1 3.f3 Rh1 4.h5 Re1 5.Kf4 = (0.02) Depth: 5/10 00:00:00 ² (0.46) Depth: 9/18 00:00:00 89kN 1.f4 Ke8 2.Kg6 Ra3 3.Rd3 Kf8 4.h5 Ra2 5.Rd2 Ra3 6.Rb2 Bd6 ² (0.47) Depth: 9/18 00:00:00 120kN ² (0.44) Depth: 12/24 00:00:02 436kN 1.Kg4 Ra3 2.Rd3 Ra1 3.f3 Rb1 4.h5 Bf8 ² (0.45) Depth: 12/24 00:00:03 591kN ² (0.44) Depth: 14/26 00:00:10 2378kN 1.f4 Ke8 2.Kg6 Ra3 3.Rd3 Kf8 4.h5 Ra2 5.Bd2 Ra1 6.Rg3 Bd6 ² (0.45) Depth: 14/26 00:00:12 2568kN, tb=1 ² (0.36) Depth: 16/28 00:00:19 4268kN, tb=6 1.Kg4 Ra3 2.Rd3 Ra1 3.f3 Ra2 4.h5 Rg2+ 5.Kh3 Re2 ² (0.37) Depth: 16/28 00:00:25 5437kN, tb=12 ² (0.38) Depth: 16/28 00:00:29 6374kN, tb=12 1.Rd7 Ke8 2.Rc7 Kd8 3.Rb7 Ra3 4.Kg6 Ke8 5.Rb5 Rxb3 6.Bxc5 Bxc5 7.Rxc5 Rc3 8.Kxg7 Ke7 9.h5 b3 10.Rb5 ² (0.39) Depth: 16/28 00:00:32 7232kN, tb=12 ² (0.40) Depth: 18/30 00:01:13 16717kN, tb=12 1.f3 Ra3 2.Bxc5 Bxc5 3.Rxc5 Rxb3 4.Rc7+ Kf8 5.Kg6 Rxf3 6.Rf7+ Ke8 7.Kxg7 Kd8 8.Rxf6 Ra3 ² (0.41) Depth: 18/30 00:01:32 20831kN, tb=56 ² (0.48) Depth: 19/31 00:02:24 32721kN, tb=82 1.Rd7 Ke8 2.Rc7 Kd8 3.Rb7 Ra3 4.Kg6 Rxb3 5.Kf7 Bd6 6.Ke6 Kc8 7.Rxg7 Rd3 8.Bxh6 Be5 9.Be3 Rd6+ 10.Kf7 ² (0.49) Depth: 19/31 00:02:41 36121kN, tb=82 ² (0.67) Depth: 22/34 00:06:54 90907kN, tb=213 ----- Fritz 7a: Analysis by Fritz 7: 1.Kg4 Ke8 2.h5 Ra3 3.Rd3 Ra8 4.f3 Ra1 5.Kf4 Rh1 6.Kg4 ² (0.38) Depth: 10/22 00:00:00 148kN ² (0.29) Depth: 12/26 00:00:01 399kN 1.f3 Ra3 2.Bxc5 Bxc5 3.Rxc5 Rxb3 4.Rc7+ Kf8 5.Rb7 Rxf3 6.Kg4 Rc3 7.Rxb4 Ke7 ² (0.32) Depth: 12/26 00:00:01 622kN ² (0.38) Depth: 14/28 00:00:05 2135kN 1.Rd7 Ke8 2.Rc7 Kd8 3.Rb7 Ra3 4.Kg6 Rxb3 5.Kf7 Bd6 6.Ke6 Bc7 7.Bxc5 Rh3 8.Bxb4 ² (0.41) Depth: 14/28 00:00:06 2514kN ± (0.85) Depth: 21/38 00:04:01 103110kN ----- It's not quite clear to me at the moment, why both programs expect after 1.Rd7 Ke8 2.Rc7, the move 2...Kd8 (which allows White's king to go to g6), instead of the immediate 2...Ra3. Fritz 7a thinks Ra3 is 2nd best at this point. >Thanks and Merry Christmas to all here, Same to you all all others! Regards, M.Scheidl
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.