The best move

[kgburcham]

[d] 2r1q1k1/r4p1p/b3pBp1/n3P1QP/p2p3R/P2p2PB/2P2P2/R5K1 w - -

Houdini

20/70 0:07 +6.08++ 1.Bg4 (121.204.777) 15718
20/70 0:09 +6.57++ 1.Bg4 (141.896.758) 15738
20/70 0:12 +7.86++ 1.Bg4 (192.056.475) 15863
20/70 0:17 +11.61++ 1.Bg4 (279.206.511) 16214
20/70 0:27 +22.69++ 1.Bg4 (484.778.498) 17446
20/70 0:33 +M14 1.Bg4 dxc2 2.Qh6 c1Q+ 3.Rxc1 Rxc1+ 4.Kh2 Rh1+ (627.055.414) 18721


Critter

19/51 0:07 +4.56++ 1.Bg4 dxc2 2.hxg6 fxg6 (98.436.767) 13015
19/51 0:09 +4.96++ 1.Bg4 dxc2 2.hxg6 fxg6 (118.636.195) 13094
19/51 0:11 +5.55++ 1.Bg4 dxc2 2.hxg6 fxg6 (148.774.402) 13128
19/52 0:14 +6.45++ 1.Bg4 dxc2 2.hxg6 fxg6 (188.451.660) 13123
19/52 0:18 +7.79++ 1.Bg4 dxc2 2.hxg6 fxg6 (242.791.800) 13188
19/54 0:24 +9.80++ 1.Bg4 dxc2 2.hxg6 fxg6 (326.742.732) 13310
19/54 0:34 +12.81++ 1.Bg4 dxc2 2.hxg6 fxg6 (470.840.577) 13513
19/58 0:44 +17.32++ 1.Bg4 dxc2 2.hxg6 fxg6 (615.138.093) 13815
19/58 0:48 +M14 1.Bg4 dxc2 2.Qh6 c1Q+ 3.Rxc1 Rxc1+ 4.Kh2 Rh1+ (693.246.715) 14414

Rybka

10.00 0:01 +M14 1.Bg4 dxc2 2.Qh6 c1Q+ 3.Rxc1 Rxc1+ 4.Kh2 Rh1+ (618.824) 574
11.01 0:01 +M14 1.Bg4 dxc2 2.Qh6 c1Q+ 3.Rxc1 Rxc1+ 4.Kh2 Rh1+ (719.866) 598
12.01 0:01 +M14 1.Bg4 dxc2 2.Qh6 c1Q+ 3.Rxc1 Rxc1+ 4.Kh2 Rh1+ (841.515) 619
13.01 0:01 +M14 1.Bg4 dxc2 2.Qh6 c1Q+ 3.Rxc1 Rxc1+ 4.Kh2 Rh1+ (1.021.254) 654
14.01 0:01 +M14 1.Bg4 dxc2 2.Qh6 c1Q+ 3.Rxc1 Rxc1+ 4.Kh2 Rh1+ (1.355.073) 700
15.01 0:02 +M12 1.Bg4 dxc2 2.Rc1 Be2 3.Qh6 Qf8 4.Qxh7+ Kxh7 (2.137.087) 773

[ZirconiumX]

FruitFly comes in under the tactical spectrum. (I borrowed the three ply fail high reduction, sorry Robert!)

Score : +2.01
Depth : 17/1
Time : 00:00:28
Nodes : 4220000
N/sec : 147810
1. Rxd4 cxd3 2. Rxa4 Bb5 3. Rg4 Qf8 4. Rb4 Nc6 5. Rxb5 Nd4 6. Bg2 h6 7. Qd2 Nxb5 8. Bxa8 Rxa8 9. Qxd3

Score : +1.80
Depth : 18/1
Time : 00:00:43
Nodes : 6340000
N/sec : 146702
1. Rxd4 cxd3 2. Rxa4 Bb5 3. Rg4 Qf8 4. Rb4 Nc6 5. Rxb5 Nd4 6. Bg2 Nxb5 7. Be7 h6 8. Qf6 Qe8

Score : +1.50
Depth : 19/1
Time : 00:03:04
Nodes : 30920000
N/sec : 167512
1. Rxd4

Score : +1.61
Depth : 20/1
Time : 00:03:49
Nodes : 37810000
N/sec : 165073
1. Rxd4 cxd3 2. Rg4 Qf8 3. Rxa4 Bb5 4. Rb4

And when I *show it* h5, I get:

Score : >= +2.18
Depth : 17/1
Time : 00:00:23
Nodes : 3710000
N/sec : 159455
1. h5

Score : +2.48
Depth : 18/1
Time : 00:00:45
Nodes : 6810000
N/sec : 148419
1. h5

Score : +2.18
Depth : 19/1
Time : 00:01:29
Nodes : 12940000
N/sec : 144097
1. h5

Score : +2.48
Depth : 20/1
Time : 00:04:04
Nodes : 38640000
N/sec : 158274
1. h5

Matthew:out

[lech]

Interestingly after 28...Ra7 Fischer played the brilliant 29. Bg2! but he should have played Bg4!!! instead which is even stronger.

Why ? Sting prefers 29.Bg2! too.
[d] 2r1q1k1/r4p1p/b3pBp1/n3P1QP/p2p3R/P2p2PB/2P2P2/R5K1 w - - 2 29

Code: Select all

Sting SF PV (2 threads)
info depth 20 seldepth 34 multipv 1 score cp 282 nodes 6384128 nps 361891 time 1
7641 pv g5h6 e8f8 h5g6 f8h6 h4h6 f7g6 h3e6 g8f8 e6c8 d3c2 c8a6 a7a6 h6h7 d4d3 f6
g5 a5b3 a1f1 d3d2 g5d2 b3d2 h7c7 d2f1 g1f1 a6b6 c7c2 f8f7 f2f4 b6b3 c2c7 f7e6 c7
c6 e6f5
info depth 21
info currmove g5h6 currmovenumber 1
info nodes 8250624 nps 373432 time 22094
info currmove c2d3 currmovenumber 2
info nodes 12594800 nps 366383 time 34376
info currmove h5g6 currmovenumber 3
info nodes 12883485 nps 366779 time 35126
info currmove h4d4 currmovenumber 4
info currmove g1h2 currmovenumber 5
info nodes 13286325 nps 367147 time 36188
info currmove a1d1 currmovenumber 6
info currmove h3f5 currmovenumber 7
info currmove a1c1 currmovenumber 8
info currmove h4g4 currmovenumber 9
info currmove h5h6 currmovenumber 10
info currmove c2c3 currmovenumber 11
info currmove h4f4 currmovenumber 12
info currmove h3g2 currmovenumber 13
info depth 21 seldepth 0 multipv 1 score cp 290 lowerbound nodes 20399466 nps 36
3050 time 56189 pv h3g2 e8f8 g2e4 d3d2 h5g6 f7g6 a1d1 f8f7 e4g6 f7g6 g5g6 h7g6 h
4h8 g8f7 h8h7 f7g8 h7a7
info nodes 20399466 nps 362953 time 56204
info currmove h3g2 currmovenumber 1
info depth 21 seldepth 0 multipv 1 score cp 298 lowerbound nodes 22732887 nps 36
3627 time 62517 pv h3g2 e8f8 g2e4 d3d2 h5g6 f7g6 a1d1 f8f7 e4g6 f7g6 g5g6 h7g6 h
4h8 g8f7 h8h7 f7g8 h7a7
info nodes 22732887 nps 363447 time 62548
info currmove h3g2 currmovenumber 1
info depth 21 seldepth 0 multipv 1 score cp 311 lowerbound nodes 26527991 nps 36
2612 time 73158 pv h3g2 e8f8 g2e4 c8c5 c2d3 a5b3 h5g6 f7g6 e4g6 b3a1 g6e4
info nodes 26527991 nps 362612 time 73158
info currmove h3g2 currmovenumber 1
info depth 21 seldepth 0 multipv 1 score cp 329 lowerbound nodes 30128187 nps 36
4562 time 82642 pv h3g2 a6b5 g5h6 e8f8 h6h7 g8h7 h5g6 h7g8 h4h8
info nodes 30128187 nps 364492 time 82658
info currmove h3g2 currmovenumber 1
info depth 21 seldepth 0 multipv 1 score cp 356 lowerbound nodes 33534994 nps 36
6622 time 91470 pv h3g2 a6b5 g5h6 e8f8 h6h7 g8h7 h5g6 h7g8 h4h8
info nodes 33534994 nps 366622 time 91470
info currmove h3g2 currmovenumber 1
info depth 21 seldepth 0 multipv 1 score cp 397 lowerbound nodes 37806442 nps 36
5714 time 103377 pv h3g2 a6b5 g5h6 e8f8 h6h7 g8h7 h5g6 h7g8 h4h8
info nodes 37806442 nps 365657 time 103393
info currmove h3g2 currmovenumber 1
info depth 21 seldepth 0 multipv 1 score cp 458 lowerbound nodes 45278007 nps 35
6556 time 126987 pv h3g2 a6b5 g5h6 e8f8 h6h7 g8h7 h5g6 h7g8 h4h8
info nodes 45278007 nps 356556 time 126987

[Jack Lad]

Why ? Sting prefers 29.Bg2! too.
[d] 2r1q1k1/r4p1p/b3pBp1/n3P1QP/p2p3R/P2p2PB/2P2P2/R5K1 w - - 2 29

Bg2 is mate in 44 whereas Bg4 is mate in 12. I hope that explains why.

[Jack Lad]

Rybka
15.01 0:02 +M12 1.Bg4 dxc2 2.Rc1 Be2 3.Qh6 Qf8 4.Qxh7+ Kxh7 (2.137.087) 773

How the hell can Rybka find that in just 2 secs?

What score does it give for Bg2 instead?

[Terry McCracken]

Rybka
15.01 0:02 +M12 1.Bg4 dxc2 2.Rc1 Be2 3.Qh6 Qf8 4.Qxh7+ Kxh7 (2.137.087) 773

How the hell can Rybka find that in just 2 secs?

What score does it give for Bg2 instead?

He has a very powerfull 12 core system!

[kgburcham]

Rybka
15.01 0:02 +M12 1.Bg4 dxc2 2.Rc1 Be2 3.Qh6 Qf8 4.Qxh7+ Kxh7 (2.137.087) 773

How the hell can Rybka find that in just 2 secs?
What score does it give for Bg2 instead?

[d] 2r1q1k1/r4p1p/b3pBp1/n3P1QP/p2p3R/P2p2PB/2P2P2/R5K1 w - - 0 1

Ok, Terry says big hardware.
I used my small pc and set cores=1.
Now look at the result after a reboot with clear hash.
Get someone else to run the same test and see the result.

Rybka

1.Bg4 Nb3 2.cxb3 Rac7 3.Qh6 Qf8 4.hxg6 Qxh6 5.Rxh6 fxg6 6.Bxe6+ Kf8 7.Bxc8 Bxc8
+- (5.58) Depth: 10 00:00:02 398kN
1.Bg4 dxc2 2.Qh6 c1Q+ 3.Rxc1 Rxc1+ 4.Kh2 Rh1+ 5.Kxh1[] Bb7+
+- (#14) Depth: 11 00:00:04 1072kN
1.Bg4 dxc2 2.Rc1 Bb7 3.Qh6 Qf8 4.Qxh7+ Kxh7 5.hxg6+ Kxg6 6.Bh5+ Kh7 7.Be2+ Qh6
+- (#11) Depth: 12 00:00:05 1277kN
1.Bg4 dxc2 2.Rc1 Be2 3.Qh6 Qf8 4.Qxh7+[] Kxh7 5.hxg6+[] Kxg6 6.Bxe2[] Rc3 7.Bh5+[] Kh7
+- (#12) Depth: 13 00:00:06 1491kN
1.Bg4 dxc2 2.Rc1 Be2 3.Qh6 Qf8 4.Qxh7+[] Kxh7 5.hxg6+[] Kxg6 6.Bxe2[] Rc3
+- (#12) Depth: 14 00:00:07 1800kN
1.Bg4 dxc2 2.Rc1 Be2 3.Qh6 Qf8 4.Qxh7+[] Kxh7 5.hxg6+[] Kxg6 6.Bxe2[] Rc3
+- (#12) Depth: 15 00:00:09 2427kN