Page 1 of 4
Test Position
Posted: Thu Sep 10, 2009 3:29 pm
by treppner
A 90' 30 game between Deep Hiarcs 12 - Stockfish 1.4 on my machine. Stockfish plays a move that completely paralyzes Hiarcs. 46...Be8 is the normal move and black might have a better position. But 46...f5! is completely winning. Hiarcs was forced to resign a couple moves later.
[D] 5k2/5p2/2b1p1pN/1p2P1P1/5Q1p/1Pb4P/4KP2/1q1B4 b - - 0 46
Re: Test Position
Posted: Thu Sep 10, 2009 4:47 pm
by plattyaj
treppner wrote:A 90' 30 game between Deep Hiarcs 12 - Stockfish 1.4 on my machine. Stockfish plays a move that completely paralyzes Hiarcs. 46...Be8 is the normal move and black might have a better position. But 46...f5! is completely winning. Hiarcs was forced to resign a couple moves later.
[D] 5k2/5p2/2b1p1pN/1p2P1P1/5Q1p/1Pb4P/4KP2/1q1B4 b - - 0 46
No tricky evaluation needed to find this but Schola probably takes too long to find it for my liking:
Code: Select all
15 40 2496 24263128 Be8 Ng4 Qb2+ Kf1 Qa1 Ke2 Qa2+ Kf1 b4 Qf6 Qd2 Bf3 Bb5+ Kg2 Ke8
15 40 4425 42341258 Be8 Ng4 Qb2+ Kf1 Qa1 Ke2 Qa2+ Kf1 b4 Qf6 Qd2 Bf3 Bb5+ Kg2 Ke8
16 44 5348 51307759 Be8 Ng4 Qb2+ Kf1 b4 Qf6 Qd2 Bf3 Qc2 Kg2 Qxb3 Qh8+ Ke7 Nf6 Bd7 Qxh4
16 44 8634 82482936 Be8 Ng4 Qb2+ Kf1 b4 Qf6 Qd2 Bf3 Qc2 Kg2 Qxb3 Qh8+ Ke7 Nf6 Bd7 Qxh4
17 40 10884 103858330 Be8 Ng4 Qb2+ Kf1 b4 Qf6 Qa1 Qh8+ Ke7 Qf6+ Kd7 Qf3 Kd8 Qd3+ Bd4 Nf6 Bc6 Ke2
17 40 25778 243519232 Be8 Ng4 Qb2+ Kf1 b4 Qf6 Qa1 Qh8+ Ke7 Qf6+ Kd7 Qf3 Kd8 Qd3+ Bd4 Nf6 Bc6 Ke2
18 44 31335 297784575 Be8 Ng4 Qb2+ Kf1 Qa1 Ne3 Bd7 Ke2 Bxe5 Qxh4 Ke7 Bc2 Bc6 Be4 Qd4 f3 Kd6
18 82 77529 733012101 f5 exf6 e5 Qg4 Bb7 Kf1 Qd3+ Be2 Qd2 Kg1 Bd4 Kf1 Qc1+ Bd1 Bc8 Qf3 Qxg5
18 82 77529 733012101 f5 exf6 e5 Qg4 Bb7 Kf1 Qd3+ Be2 Qd2 Kg1 Bd4 Kf1 Qc1+ Bd1 Bc8 Qf3 Qxg5
19 182 95770 906659891 f5 ++
Andy.
Re: Test Position
Posted: Thu Sep 10, 2009 5:05 pm
by AdminX
Analysis by Bright-0.4a3 on a 2 GHz Core2Quad:
1...Be8 2.Ng4 b4 3.Ne3 Qa1 4.Kf3 Bd4 5.Be2 Bxe5 6.Qxb4+ Kg8 7.Kg2 Bc6+ 8.Bf3 Bxf3+ 9.Kxf3 Qh1+ 10.Ke2 Qxh3 11.Qe4 Bc3 12.Qg2 Qxg2 13.Nxg2
-+ (-1.53) Depth: 6/18 00:00:00 26kN
-+ (-1.66) Depth: 19/67 00:01:02 257mN
1...f5 2.exf6 e5 3.Qg4 Bb7 4.Nf5 Be4 5.Kf1 Bxf5 6.Qf3 Qd3+ 7.Qxd3 Bxd3+ 8.Be2 Bf5 9.f4 Bxh3+ 10.Kf2 exf4 11.Bxb5 Be6 12.Bc4 Kf7 13.Bxe6+ Kxe6
-+ (-4.09) Depth: 19/76 00:02:32 561mN
-+ (-4.09) Depth: 19/76 00:02:32 561mN
Re: Test Position
Posted: Thu Sep 10, 2009 7:29 pm
by jdart
I assume it will be found eventually but Arasan didn't get this after 5 minutes search on a quad.
--Jon
Re: Test Position
Posted: Thu Sep 10, 2009 9:25 pm
by jwes
Both Crafty and Toga do not find it after some hours, but if I make the first move, they both see it in a few seconds.
Re: Test Position
Posted: Thu Sep 10, 2009 10:27 pm
by bob
jwes wrote:Both Crafty and Toga do not find it after some hours, but if I make the first move, they both see it in a few seconds.
I've noticed this. My first thought was I am reducing something critical. I disabled reductions (pers 4 0 command) and ran again. No better. I then disabled the forward pruning (pers 5 0 command along with disabling reductions). Still no better. This might be a case of some aggressive check extension working, where I have found in real games that the give-check extension is good, the rest are either harmful or do not help.
I am studying it as it is an interesting question as to "why?"
Re: Test Position
Posted: Thu Sep 10, 2009 10:31 pm
by Karlo Bala
bob wrote:jwes wrote:Both Crafty and Toga do not find it after some hours, but if I make the first move, they both see it in a few seconds.
I've noticed this. My first thought was I am reducing something critical. I disabled reductions (pers 4 0 command) and ran again. No better. I then disabled the forward pruning (pers 5 0 command along with disabling reductions). Still no better. This might be a case of some aggressive check extension working, where I have found in real games that the give-check extension is good, the rest are either harmful or do not help.
I am studying it as it is an interesting question as to "why?"
Maybe Zugzwang?
Re: Test Position
Posted: Thu Sep 10, 2009 10:50 pm
by Spock
Zappa Mexico II x64 takes nearly an hour to find it on my quad
11/35 0:00 -0.97 1...Be8 2.Ng4 Qb2+ 3.Kf1 Qa1 4.Qf3 b4 5.Kg2 Qc1 (401.171) 1120
11/35 0:00 -0.97 1...Be8 2.Ng4 Qb2+ 3.Kf1 Qa1 4.Qf3 b4 5.Kg2 Qc1 (576.367) 1193
12/38 0:00 -1.17 1...Be8 2.Ng4 b4 3.Ne3 Qa1 4.Qxh4 Bb5+ 5.Kf3 Bc6+ 6.Ke2 Qa6+ (981.558) 1234
12/38 0:00 -1.17 1...Be8 2.Ng4 b4 3.Ne3 Qa1 4.Qxh4 Bb5+ 5.Kf3 Bc6+ 6.Ke2 Qa6+ (1.119.867) 1259
13/41 0:01 -1.09 1...Be8 2.Ng4 b4 3.Ne3 Qa1 4.Nc4 Bb5 5.Qe3 Bxc4+ 6.bxc4 (1.760.191) 1343
13/41 0:01 -1.09 1...Be8 2.Ng4 b4 3.Ne3 Qa1 4.Nc4 Bb5 5.Qe3 Bxc4+ 6.bxc4 (2.138.859) 1258
14/47 0:04 -1.02 1...Be8 2.Ng4 b4 3.Qc4 Qc1 4.Qd3 Qxg5 5.Qd6+ Kg7 6.Kf1 Bb5+ 7.Kg2 Qd2 8.Qxd2 Bxd2 (4.190.185) 1033
14/47 0:04 -1.02 1...Be8 2.Ng4 b4 3.Qc4 Qc1 4.Qd3 Qxg5 5.Qd6+ Kg7 6.Kf1 Bb5+ 7.Kg2 Qd2 8.Qxd2 Bxd2 (4.871.466) 1099
15/47 0:05 -1.02 1...Be8 2.Ng4 b4 3.Qc4 Qc1 4.Qd3 Qxg5 5.Qd6+ Kg7 6.Kf1 Bb5+ 7.Kg2 Qd2 8.Qxd2 Bxd2 (6.165.442) 1172
15/47 0:05 -1.02 1...Be8 2.Ng4 b4 3.Qc4 Qc1 4.Qd3 Qxg5 5.Qd6+ Kg7 6.Kf1 Bb5+ 7.Kg2 Qd2 8.Qxd2 Bxd2 (6.825.325) 1243
16/51 0:09 -1.08 1...Be8 2.Ng4 b4 3.Ne3 Qa1 4.Qxh4 Bb5+ 5.Kf3 Bc6+ 6.Ke2 Qa6+ 7.Nc4 Bxe5 8.Qg4 Qb5 9.Kf1 Bc7 10.Qh4 Kg8 (14.802.045) 1621
16/51 0:09 -1.08 1...Be8 2.Ng4 b4 3.Ne3 Qa1 4.Qxh4 Bb5+ 5.Kf3 Bc6+ 6.Ke2 Qa6+ 7.Nc4 Bxe5 8.Qg4 Qb5 9.Kf1 Bc7 10.Qh4 Kg8 (16.652.880) 1689
17/51 0:12 -1.08 1...Be8 2.Ng4 b4 3.Ne3 Qa1 4.Qxh4 Bb5+ 5.Kf3 Bc6+ 6.Ke2 Qa6+ 7.Nc4 Bxe5 8.Qg4 Qb5 9.Kf1 Bc7 10.Qh4 Kg8 (24.077.530) 1889
17/53 0:15 -1.08 1...Be8 2.Ng4 b4 3.Ne3 Qa1 4.Qxh4 Bb5+ 5.Kf3 Bc6+ 6.Ke2 Qa6+ 7.Nc4 Bxe5 8.Qg4 Qb5 9.Kf1 Bc7 10.Qh4 Kg8 (30.055.952) 1990
18/55 0:24 -1.08 1...Be8 2.Ng4 b4 3.Ne3 Qa1 4.Qxh4 Bb5+ 5.Nc4 Bxe5 6.Qe4 Bd6 7.Bc2 Bc5 8.Qf4 e5 9.Qf6 Bxc4+ 10.bxc4 (54.040.854) 2227
18/55 0:28 -1.08 1...Be8 2.Ng4 b4 3.Ne3 Qa1 4.Qxh4 Bb5+ 5.Nc4 Bxe5 6.Qe4 Bd6 7.Bc2 Bc5 8.Qf4 e5 9.Qf6 Bxc4+ 10.bxc4 (64.653.938) 2297
19/57 0:40 -1.08 1...Be8 2.Ng4 b4 3.Ne3 Qa1 4.Qxh4 Bb5+ 5.Nc4 Bxe5 6.Qe4 Kg8 7.Qb7 Bxc4+ 8.bxc4 Qa2+ 9.Ke3 Qxc4 10.Qa8+ Kh7 11.Qe4 (98.182.863) 2416
19/57 0:49 -1.08 1...Be8 2.Ng4 b4 3.Ne3 Qa1 4.Qxh4 Bb5+ 5.Nc4 Bxe5 6.Qe4 Kg8 7.Qb7 Bxc4+ 8.bxc4 Qa2+ 9.Ke3 Qxc4 10.Qa8+ Kh7 11.Qe4 (121.435.881) 2445
20/59 1:25 -1.06 1...Be8 2.Ng4 b4 3.Ne3 Qa1 4.Qxh4 Bb5+ 5.Nc4 Bxe5 6.Qe4 Kg8 7.Qb7 Bxc4+ 8.bxc4 Qa2+ 9.Ke3 Bc3 10.Qc8+ Kg7 11.Qd8 Qxc4 (223.558.178) 2604
20/59 1:55 -1.06 1...Be8 2.Ng4 b4 3.Ne3 Qa1 4.Qxh4 Bb5+ 5.Nc4 Bxe5 6.Qe4 Kg8 7.Qb7 Bxc4+ 8.bxc4 Qa2+ 9.Ke3 Bc3 10.Qc8+ Kg7 11.Qd8 Qxc4 (304.606.493) 2638
21/61 3:15 -1.16 1...Be8 2.Ng4 b4 3.Ne3 Bc6 4.Qxh4 Kg8 5.Nc4 Be4 6.Qg3 Bd4 7.h4 Kf8 8.Nd2 Qb2 9.Qf4 Bc3 10.Ke1 Bd5 (525.010.584) 2684
21/61 3:48 -1.16 1...Be8 2.Ng4 b4 3.Ne3 Bc6 4.Qxh4 Kg8 5.Nc4 Be4 6.Qg3 Bd4 7.h4 Kf8 8.Nd2 Qb2 9.Qf4 Bc3 10.Ke1 Bd5 (608.323.132) 2657
22/63 5:53 -1.24 1...Be8 2.Ng4 b4 3.Ne3 Bc6 4.Qxh4 Kg8 5.Nc4 Qa2+ 6.Kf1 Qa1 7.Qg4 Bd5 8.Ke2 Qa2+ 9.Kf1 Qa1 (939.106.096) 2653
22/63 7:32 -1.24 1...Be8 2.Ng4 b4 3.Ne3 Bc6 4.Qxh4 Kg8 5.Nc4 Qa2+ 6.Kf1 Qa1 7.Qg4 Bd5 8.Ke2 Qa2+ 9.Kf1 Qa1 (1.195.258.261) 2642 TB:39
23/65 12:41 -1.20 1...Be8 2.Ng4 b4 3.Ne3 Bc6 4.Qxh4 Kg8 5.Nc4 Qa2+ 6.Kf1 Qa1 7.Ke2 Bxe5 8.Qg4 Bd5 9.h4 Bd4 10.Kf1 e5 11.h5 Be6 (2.056.920.042) 2699 TB:39
23/65 57:25 -5.79 1...f5 2.exf6 e5 3.Qg4 Bb7 4.Nf7 Qa2+ 5.Kf1 Qd2 6.Qe2 Kxf7 7.Qxd2 Bxd2 8.Be2 Bc6 9.Bd3 Bxg5 10.Be2 Bxf6 11.f3 Bd5 12.Kf2 b4 13.Bd1 g5 14.Bc2 g4 (9.121.796.065) 2647 TB:94
23/65 58:02 -5.79 1...f5 2.exf6 e5 3.Qg4 Bb7 4.Nf7 Qa2+ 5.Kf1 Qd2 6.Qe2 Kxf7 7.Qxd2 Bxd2 8.Be2 Bc6 9.Bd3 Bxg5 10.Be2 Bxf6 11.f3 Bd5 12.Kf2 b4 13.Bd1 g5 14.Bc2 g4 (9.221.033.750) 2647 TB:119
Re: Test Position
Posted: Thu Sep 10, 2009 10:51 pm
by bob
Karlo Bala wrote:bob wrote:jwes wrote:Both Crafty and Toga do not find it after some hours, but if I make the first move, they both see it in a few seconds.
I've noticed this. My first thought was I am reducing something critical. I disabled reductions (pers 4 0 command) and ran again. No better. I then disabled the forward pruning (pers 5 0 command along with disabling reductions). Still no better. This might be a case of some aggressive check extension working, where I have found in real games that the give-check extension is good, the rest are either harmful or do not help.
I am studying it as it is an interesting question as to "why?"
Maybe Zugzwang?
Didn't try disabling null-move, as it didn't look like a zugzwang position with queens and other pieces still on the board. I'll give it a try to see...
Re: Test Position
Posted: Thu Sep 10, 2009 10:57 pm
by Tord Romstad
bob wrote: This might be a case of some aggressive check extension working, where I have found in real games that the give-check extension is good, the rest are either harmful or do not help.
I don't think so. Crafty extends checks far more aggressively than Stockfish, unless you have changed something recently. Stockfish extends checks by only half a ply, except at PV nodes.