compare seven programs, one position

[kgburcham]

[D] r1bq1rk1/pp2bppp/1n2p3/3pP3/8/2RBBN2/PP2QPPP/2R3K1 w - -

12 threads
except Crafty, 8 threads
HT off
Hiarcs limited to 2048 hash

Houdini 1.5 x64 (4096 MB)

23/67 1:10 +0.72++ 1.Bxh7+ (2.280.399.034) 32143
23/67 1:12 +1.14++ 1.Bxh7+ (2.316.324.265) 32109
23/67 1:16 +2.46++ 1.Bxh7+ (2.449.957.248) 32159
23/73 1:37 +5.54 1.Bxh7+ Kxh7 (3.169.368.107) 32454
24/73 1:40 +5.51 1.Bxh7+ Kxh7 (3.247.038.189) 32458

Stockfish 2.0 JA 64bit (4096 MB)

31/17 2:15 +0.84++ 1.Bxh7+ Kxh7 (2.192.294.782) 16127
31/17 2:16 +1.17++ 1.Bxh7+ Kxh7 (2.204.003.177) 16128
31/17 2:18 +1.81++ 1.Bxh7+ Kxh7 (2.242.174.028) 16141
31/13 2:34 +3.11++ 1.Bxh7+ Kh8 (2.536.508.072) 16397
31/37 2:44 +4.96 1.Bxh7+ Kxh7 (2.710.082.848) 16429
32/44 2:58 +5.21 1.Bxh7+ Kxh7 (2.936.976.745) 16462

HIARCS 13.2 MP (2048 MB)

21/52 2:56 +0.57 1.Bxh7+ (1.006.578.225) 5701
21/52 3:01 +1.06++ 1.Bxh7+ (1.039.859.420) 5715
21/52 3:14 +2.06++ 1.Bxh7+ (1.119.818.651) 5761
21/52 3:46 +4.02 1.Bxh7+ Kxh7 (1.349.367.864) 5950
22/52 4:09 +4.01 1.Bxh7+ Kxh7 (1.504.030.806) 6040
23/52 4:28 +4.01 1.Bxh7+ Kxh7 (1.626.689.194) 6064
24/52 5:02 +4.13 1.Bxh7+ Kxh7 (1.853.975.231) 6122

Deep Rybka 4 SSE42 x64 (4096 MB)

21.09 3:08 +0.67++ 1.Bxh7+ Kxh7 (265.890.623) 1408
21.01 3:09 +0.82++ 1.Bxh7+ Kxh7 (266.394.277) 1406
21.01 3:11 +1.22++ 1.Bxh7+ Kxh7 (268.612.958) 1404
21.01 3:23 +2.02++ 1.Bxh7+ Kxh7 (287.791.709) 1417
21.01 3:52 +3.62++ 1.Bxh7+ Kxh7 (331.054.120) 1426
21.01 5:27 +3.62++ 1.Bxh7+ Kxh7 (471.338.188) 1438


Zappa Mexico II (4096 MB)
default settings

20/59 8:16 +0.53 1.Nd4 g6 (6.694.694.189) 13494

Zappa Mexico II (4096 MB)
null move disabled

16/51 8:54 +0.60 1.Nd4 Bd7 (6.803.667.362) 12733


Deep Shredder 12 x64 (4096 MB)

21/58 7:30 +0.13++ 1.h4 (4.307.930.609) 9567

Crafty 23.01:

1.Qc2 g6
= (0.25) Depth: 21/42 00:09:09 6865mN

[gladius]

GarboChess JS has no clue here. It has no king safety, and probably searches about 100x slower than these other engines, so probably expected .

Ply:11 Score:50 Nodes:7068640 NPS:92122 Nd4 f5 exf6 Bxf6 Qh5 g6 Qg4 e5 Ne6 Qd6 Bf5

However, if you make the first two moves along the PV, it comes up with a reasonable continuation, until it gets greedy.

Ply:14 Score:-1834 Nodes:7286470 NPS:97054 Ng5+ Bxg5 Qh5+ Bh6 Bxh6 g6 Qh3 Kg8 Bxf8 Qxf8 Qh4 Bd7 Rc7 Rc8 Rxc8

Going up to the move where it gets greedy, it finds Rg3, but only at ply 14.

[d]r1bq1rk1/pp3p2/1n2p1pB/3pP3/8/2R4Q/PP3PPP/2R3K1 w - -[/d]

Ply:14 Score:-679 Nodes:14532676 NPS:93876 Rg3 Nd7 Bg5 f6 exf6 Rxf6 Rf3 e5 Rfc3 Nc5 Qh4 Ne4 Rxc8 Rxc8

It finally finds a continuation with a positive score if we move way down the PV:

[d]r1bq1rk1/pp1n4/4ppp1/3pP1B1/8/6RQ/PP3PPP/2R3K1 w - -[/d]

Ply:12 Score:1365 Nodes:4663398 NPS:109667 exf6 Rxf6 Qh5 Kf8 Qh7 Ke8 Rf3 Qe7 Qh8+ Rf8 Qh6 Nf6 Qxg6+

[Jouni]

The second best problem solver (after guess who)

Analysis by Critter 0.90 (2CPU):

1.Bxh7+
± (0.82) Depth: 20/59 00:01:45 268mN
1.Bxh7+
± (0.93) Depth: 20/59 00:01:46 271mN
1.Bxh7+
± (1.11) Depth: 20/59 00:01:48 275mN
1.Bxh7+
± (1.37) Depth: 20/59 00:01:51 281mN
1.Bxh7+
+- (1.76) Depth: 20/59 00:01:58 300mN
1.Bxh7+
+- (2.35) Depth: 20/59 00:02:11 335mN
1.Bxh7+
+- (3.23) Depth: 20/59 00:02:33 386mN
1.Bxh7+ Kxh7 2.Ng5+ Bxg5 3.Qh5+ Kg8 4.Bxg5 f6 5.exf6 Rxf6 6.Bxf6 Qxf6 7.Rf3 g6 8.Qg4 e5 9.Rxf6 Bxg4 10.Rxg6+ Kf7 11.Rxg4 Kf6 12.Rc7 d4 13.h4 d3 14.Kf1 Nd5 15.Rxb7 Rc8 16.Ke1 Nf4
+- (3.51) Depth: 20/77 00:03:09 469mN
1.Bxh7+ Kxh7 2.Ng5+ Bxg5 3.Qh5+ Kg8 4.Bxg5 f6 5.exf6 Rxf6 6.Bxf6 Qxf6 7.Rf3 g6 8.Qg4 e5 9.Rxf6 Bxg4 10.Rxg6+ Kf7 11.Rxg4 Kf6 12.Rc7 d4 13.h4 d3 14.Kf1 Nd5 15.Rxb7 Rc8 16.Ke1 Nf4
+- (3.63) Depth: 21/81 00:03:32 528mN, tb=20

tb=20 so Critter's has analyzed position until ending!?!?

Jouni