Author: F. Huber
Date: 03:33:01 01/07/04
Go up one level in this thread
On January 07, 2004 at 00:41:35, K. Burcham wrote: > >Mr. Huber, notice difference in depth. > > [D] 3Q4/5q1k/4ppp1/2Kp1N1B/RR6/3P1r2/4nP1b/3b4 w - - > >Shredder 7.04 > > 7/27 0:01 +9.55 1.Bxf3 d4 2.Rb7 Qxb7 3.Bxb7 gxf5 4.Ra8 Be5 5.Qg8+ Kh6 >(547.202) 486.4 > 8/23 0:01 +9.80++ 1.Bxf3 d4 2.Rb7 Qxb7 3.Bxb7 exf5 4.Ra8 g5 5.Qxf6 Bb8 >(639.872) 481.8 > 8/23 0:01 +10.30++ 1.Bxf3 d4 2.Rb7 Qxb7 3.Bxb7 exf5 4.Ra8 Nf4 5.Qh8+ >(686.336) 482.6 > 8/27 0:01 +10.64 1.Bxf3 Nd4 2.Ra8 Qc7+ 3.Qxc7+ Bxc7 4.Nxd4 Bxf3 5.Rb7 Kh6 >6.Rxc7 (837.840) 487.3 > 9/25 0:02 +10.74 1.Bxf3 Nd4 2.Ra8 Qc7+ 3.Qxc7+ Bxc7 4.Nxd4 Bxf3 5.Rb7 Kh6 >6.Rxc7 Bg4 (1.147.922) 483.3 > 10/30 0:03 +10.99++ 1.Bxf3 Nd4 2.Ra8 Qc7+ 3.Qxc7+ Bxc7 4.Nxd4 Bxf3 5.Rb7 Kg7 >6.Rxc7+ Kh6 (1.753.798) 477.6 > 10/30 0:04 +11.49++ 1.Bxf3 Nd4 2.Ra8 Qc7+ 3.Qxc7+ Bxc7 4.Nxd4 Bxf3 5.Nxe6 f5 >6.Nxc7 (1.908.000) 477.0 > 10/30 0:04 +12.00 1.Bxf3 Nd4 2.Ra8 Qc7+ 3.Qxc7+ Bxc7 4.Nxd4 Bxf3 5.Nxe6 g5 >6.Nxc7 d4 7.Rab8 (2.315.300) 481.1 > 11/32 0:07 +12.25++ 1.Bxf3 Nd4 2.Ra8 Qc7+ 3.Qxc7+ Bxc7 4.Nxd4 Bxf3 5.Nxf3 f5 >6.Rb7 g5 7.Rxc7+ (3.595.104) 483.4 > 11/32 0:08 +12.75++ 1.Bxf3 Nd4 2.Ra8 Qc7+ 3.Qxc7+ Bxc7 4.Nxd4 Bxf3 5.Nxf3 g5 >6.Rb7 Kg6 7.Rxc7 (3.971.160) 481.3 > 11/36 0:11 +14.42 1.Bxf3 Nd4 2.Ra8 Qc7+ 3.Qxc7+ Bxc7 4.Nxd4 Bxf3 5.Nxf3 Kh6 >6.Rh4+ Kg7 7.Ra7 g5 8.Rxc7+ Kg6 (5.326.054) 481.4 > 12/36 0:18 +14.64 1.Bxf3 Nd4 2.Ra8 Qc7+ 3.Qxc7+ Bxc7 4.Nxd4 Bxf3 5.Nxf3 Kh6 >6.Rh4+ Kg7 7.Ra7 g5 8.Rxc7+ Kg6 9.Rhh7 Kf5 (8.759.772) 486.2 > 12/36 0:20 +14.65++ 1.Rb7 Bc7 2.Rxc7 Bxa4 3.Rxf7+ (9.783.777) 486.9 > 12/36 0:20 +15.00++ 1.Rb7 Bc7 2.Rxc7 exf5 3.Rxf7+ Kh6 4.Bxf3 Bxa4 5.Bxe2 Bc2 >(9.800.037) 486.9 > 12/36 0:20 +M7 1.Rb7 Qxb7 2.Bxg6+ Kxg6 3.Qg8+ Qg7 4.Qxg7+ Kxf5 5.Qg4+ Ke5 >6.Re4+ dxe4 7.Qxe4+ (10.022.800) 487.4 > 13/13 0:20 +M7 1.Rb7 Qxb7 2.Bxg6+ Kxg6 3.Qg8+ Qg7 4.Qxg7+ Kxf5 5.Qg4+ Ke5 >6.Re4+ dxe4 7.Qxe4+ (10.169.727) 486.8 >best move: Rb4-b7 time: 0:22.313 min n/s: 486.619 CPU 99.5% nodes: 10.850.151 > > >ChestUCI Ver.3.1 >Franz Huber > >CPU: 2210 AMD xp >FEN: 3Q4/5q1k/4ppp1/2Kp1N1B/RR6/3P1r2/4nP1b/3b4 w - - >Search for Mate in 20 ... (Hash=256MB) > 9.23 2:07 +M9 1.Rb7 272.0 >Search completed ... (Time=150.34s) >Mate in 9 found ! (1 Solution in 02:30) > 9/09 2:30 +M9 1.Rb7 Qxb7 2.Bxg6+ Kxg6 3.Qg8+ Kxf5 4.Qg4+ Ke5 5.Qh5+ Rf5 >6.f4+ Bxf4 7.Qxe2+ Bxe2 8.Re4+ dxe4 9.d4+ 266.2 >best move: Rb4-b7 time: 2:30.453 min n/s: 266.245 nodes: 40.028.393 Hi Mr. Burcham! That´s indeed very interesting - usually a normal chess engine is announcing a mate in _more_ moves than the shortest one, but I´ve never seen before an announcment of a mate in _less_ moves!? This ´mate in 7´ of Shredder 7.04 is DEFINITELY WRONG - at least as _we_ would interpret a ´mate in xx´: (maybe Shredder took its version number instead ;-)) This ´M7´ is one possible variation, but NOT if black defends himself OPTIMAL. You can try this with ChestUCI - simply enter the first moves (which are the same in Shredders and Chests main line) manually, and then let ChestUCI look for the best defense with ´DefenderMoves=true´, i.e.: after entering 1.Rb7 Qxb7 2.Bxg6+ Kxg6 3.Qg8+ start ChestUCI with ´DefenderMoves´: ChestUCI Ver.3.1 UCI: CPU: Celeron 400MHz FEN: 6Q1/1q6/4ppk1/2Kp1N2/R7/3P1r2/4nP1b/3b4 b - - Search for Mate in -10 ... (Hash=48MB) Search completed ... (Time=0.50s) Mate in -6 found ! (4 Solutions in 00:00) 1 00:00 -M1 Kh5 Qg4+ 1 00:00 -M1 Kh5 Rh4+ 4 00:00 -M4 Qg7 Qxg7+ Kxf5 Qg4+ Ke5 Re4+ dxe4 Qxe4+ 6 00:00 -M6 Kxf5 Qg4+ Ke5 Qh5+ Rf5 f4+ Bxf4 Qxe2+ Bxe2 Re4+ dxe4 d4+ As you can see, Shredders ´solution (?)´ is also shown here (´-M4´), but the optimal defense for black is the last line with ´-M6´, which is totally the (correct) ´mate in 9´ from the starting position! So this is one more example that would lead to the conclusion: Never trust any mate announcement of a ´normal´ chess engine - for this purpose there´s no other correct way than using a special mate solver, and in my opinion Chest is (one of) _the best_! :) Best regards, Franz.
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.