Author: Mike Hood
Date: 10:28:29 11/21/05
Take a look at the following position. Fritz 9 (using 5-piece tablebases) finds the mate in 34 after Rxc6 almost immediately and stops calculating. Fruit 2.1 needs 51 seconds to find a mate in 9. (Fruit actually chooses Bc5 after 23 seconds, but needs a further 28 seconds to realise this leads to mate). [D] 7K/1kp5/2p5/8/8/8/8/2R3BN w - - 0 1 Which move is better? That depends on what you're doing. If you're doing an analysis, Fruit's solution is better. If the computer is playing a game and is short of time, Fritz's solution is better, because Fritz will make all 34 moves in less than 1 second, mating his opponent before Fruit has managed to make even one move. Incidentally, this is a counter-example to the claims that using EGTB's in probing slows down an engine. Sometimes it does, but in this case it drastically speeds up the engine. Incidentally, if I were playing this game in a club match I'd have a completely different strategy. As an average quality club player I'd play for safety rather than efficiency. I'd play 1.Bh2 Kb6 2.Bxc7+ Kxc7 and then move my knight up. If Black offered a knight for pawn exchange I'd accept it, because I know that I can mate with king and rook against king in my sleep. Analysis by Fritz 9: 1.Ng3 Kb8 2.Bd4 Kc8 3.Rxc6 Kd7 4.Rc5 +- (8.20) Depth: 7/18 00:00:00 9kN 1.Ng3 Kc8 2.Bd4 Kd8 3.Ne4 Kd7 4.Nf6+ Kd6 5.Rd1 +- (8.26) Depth: 8/18 00:00:00 25kN 1.Rxc6! +- (8.42) Depth: 8/18 00:00:00 28kN, tb=3 1.Rxc6! +- (8.59) Depth: 8/18 00:00:00 28kN, tb=3 1.Rxc6! +- (#34) Depth: 8/18 00:00:00 31kN, tb=7 ==================================== Analysis by Fruit 2.1: 1.Ra1 +- (9.27) Depth: 1/1 00:00:00 1.Rb1+ Kc8 +- (9.40) Depth: 1/4 00:00:00 1.Be3 +- (9.46) Depth: 1/4 00:00:00 1.Bd4 +- (9.59) Depth: 1/4 00:00:00 1.Bd4 c5 2.Rxc5 +- (10.03) Depth: 2/5 00:00:00 1.Ng3 Ka6 2.Rxc6+ Kb7 +- (10.21) Depth: 2/5 00:00:00 1.Ng3 Ka6 2.Bd4 +- (10.25) Depth: 3/6 00:00:00 1.Ng3 Kc8 2.Rxc6 Kd7 +- (10.08) Depth: 4/8 00:00:00 1.Rb1+ Kc8 2.Bd4 Kd7 3.Ng3 +- (10.14) Depth: 4/9 00:00:00 1.Rb1+ Kc8 2.Ng3 Kd7 3.Bd4 Ke6 +- (10.07) Depth: 5/10 00:00:00 1.Ng3 Kc8 2.Bd4 Kd7 3.Kg7 +- (10.36) Depth: 5/10 00:00:00 1.Ng3 Kc8 2.Bd4 c5 3.Bxc5 Kd7 4.Rd1+ Ke6 +- (10.52) Depth: 6/12 00:00:00 16kN 1.Ng3 Kc8 2.Kg7 Kd7 3.Kf6 Kd8 4.Rxc6 +- (10.77) Depth: 7/14 00:00:01 52kN 1.Ng3 Kc8 2.Kg7 Kd7 3.Kf6 Kd8 4.Rd1+ Ke8 5.Bc5 +- (10.86) Depth: 8/18 00:00:01 145kN 1.Ng3 Kc8 2.Rxc6 Kd8 3.Rf6 Ke7 4.Kg7 Kd7 5.Bc5 +- (11.01) Depth: 9/19 00:00:01 406kN 1.Kg7 Kc8 2.Rxc6 Kb7 3.Rh6 c6 4.Ng3 Kc7 5.Bd4 +- (11.03) Depth: 9/19 00:00:01 525kN 1.Kg7 Kc8 2.Kf7 Kb7 3.Ke6 Kc8 4.Rxc6 Kb8 5.Bh2 Kb7 6.Rxc7+ Kb6 +- (11.27) Depth: 10/22 00:00:01 1038kN 1.Kg7 Kc8 2.Kf6 Kb7 3.Rb1+ Ka6 4.Ng3 Ka5 5.Ne4 Ka4 6.Bh2 Ka5 7.Bxc7+ Ka4 +- (11.50) Depth: 11/28 00:00:02 2540kN 1.Kg7 Kc8 2.Kf6 Kb7 3.Ke6 c5 4.Rb1+ Ka6 5.Bxc5 Ka5 6.Ng3 c6 7.Ne4 +- (11.70) Depth: 12/30 00:00:06 6200kN 1.Kg7 c5 2.Bxc5 Kc8 3.Bb6 Kd7 4.Bxc7 Ke6 5.Ng3 Kd5 6.Kf6 Kd4 7.Nf5+ Ke4 8.Re1+ Kd5 +- (12.01) Depth: 13/33 00:00:16 18122kN 1.Bc5 Kc8 2.Kg7 Kb7 3.Rb1+ Ka8 4.Bg1 c5 5.Bxc5 c6 6.Rb6 Ka7 7.Ng3 Ka8 8.Rxc6 +- (12.23) Depth: 13/34 00:00:19 22367kN 1.Bc5 Kc8 2.Kg7 Kb7 3.Rb1+ Ka8 4.Bg1 c5 5.Bxc5 c6 6.Rb6 Ka7 7.Ng3 Ka8 8.Rxc6 +- (12.23) Depth: 14/34 00:00:23 26760kN 1.Bc5 Kc8 2.Kg7 Kb7 3.Kf6 Ka6 4.Rb1 Ka5 5.Ng3 Ka4 6.Rb4+ Ka5 7.Ne4 Ka6 8.Nc3 Ka5 9.Ra4# +- (#9) Depth: 15/36 00:00:51 61780kN
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.