Wich Engine can Solve this Puzzle?

[Engin]

yes thats the point why engines can not find it easy i guess

[Engin]

many thanks Dan

good thats some of handfull engines can see some advantage

[yanquis1972]

Engine: Sting SF 8 (4096 MB)
by Tord Romstad, Marco Costalba, Joona Kii
18.00 0:01 -2.34-- 1.d8Q (5.823.334) 3826
18.00 0:01 -2.46-- 1.d8Q (6.513.937) 4099
18.00 0:01 -2.38 1.d8Q Nf7+ 2.Kd7 Nxd8 3.Kxd8 Ba5+ 4.Kd7 Bc7 5.Nxe3 Bxg3 6.Bc2+ Kg8 7.Nd5 Kf7 8.Kc6 h5 9.Kb5 h4 10.Nxc3 Nc7+ 11.Kxc5 h3 12.Ne4 Bf4 (6.876.216) 4236
19.00 0:01 -2.50-- 1.d8Q (8.193.641) 4690
19.00 0:01 -2.42 1.d8Q Nf7+ 2.Kd7 Nxd8 3.Kxd8 Ba5+ 4.Kd7 Bc7 5.Nxe3 Bxg3 6.Kc6 Kg6 7.Kb6 Nc7 8.Bc2+ Kf6 9.Kxc5 Bf2 10.Kc6 Bxe3 11.Kxc7 Ke5 12.Kc6 h5 13.Kb5 h4 14.Kc4 (8.982.480) 4951
20.00 0:02 -2.54-- 1.d8Q (11.909.476) 5714
20.00 0:02 -2.34 1.d8Q Nf7+ 2.Kd7 Nxd8 3.Kxd8 Ba5+ 4.Kd7 Bc7 5.Nxe3 Bxg3 6.Bc2+ Kg8 7.Nd5 Nb4 8.Bb3 Nxd5 9.Bxd5+ Kg7 10.Be4 Kf6 11.Kc6 Ke5 12.Bc2 Kd4 13.Kb5 Bf4 (12.657.645) 5895
21.00 0:02 -2.46-- 1.d8Q (16.626.556) 6655
21.00 0:02 -2.42 1.d8Q Nf7+ 2.Kd7 Nxd8 3.Kxd8 Ba5+ 4.Kd7 Bc7 5.Nxe3 Bxg3 6.Kc6 Kg6 7.Kb6 Nc7 8.Bc2+ Kf6 9.Kxc5 Bf2 10.Kc6 Bxe3 11.Kxc7 Ke5 12.Kc6 h5 13.Kb5 h4 14.Kc4 (19.226.559) 7123
22.19 0:03 -2.54-- 1.d8Q (22.848.294) 7573
22.19 0:03 -2.50 1.d8Q Nf7+ 2.Kd7 Nxd8 3.Kxd8 Ba5+ 4.Kd7 Bc7 5.Nxe3 Bxg3 6.Kc6 Kg6 7.Kb6 Nc7 8.Bc2+ Kf6 9.Nf5 Bf4 10.Nxh6 Ke6 11.Nf5 Kd5 12.Ne3+ Kd4 13.Nf5+ Kc4 14.Ne3+ (25.487.491) 7856
23.19 0:03 -2.62-- 1.d8Q (34.188.974) 8549
23.01 0:04 -2.38++ 1.d8Q (37.017.454) 8799
23.19 0:04 -2.58 1.d8Q Nf7+ 2.Kd7 Nxd8 3.Kxd8 Ba5+ 4.Kd7 Bc7 5.Nxe3 Bxg3 6.Kc6 Kg6 7.Kb6 Nc7 8.Bc2+ Kf6 9.Nf5 Bf4 10.Kxc5 h5 11.Kc4 Be5 12.Nh4 Ke6 13.Bf5+ Kd6 14.Bg6 (40.382.899) 8999
24.19 0:05 -2.70-- 1.d8Q (50.460.583) 9508
24.19 0:06 -2.58 1.d8Q Nf7+ 2.Kd7 Nxd8 3.Kxd8 Ba5+ 4.Kd7 Bc7 5.Nxe3 Bxg3 6.Kc6 Kg6 7.Kb6 Nc7 8.Bc2+ Kf6 9.Nf5 Bf4 10.Kxc5 h5 11.Kc4 Be5 12.Nh4 Ke6 13.Bf5+ Kd6 14.Bg6 (60.183.793) 9901
25.19 0:07 -2.66-- 1.d8Q (71.599.991) 10189
25.19 0:07 -2.74-- 1.d8Q (80.834.745) 10374
25.19 0:08 -2.66 1.d8Q Nf7+ 2.Kd7 Nxd8 3.Kxd8 Ba5+ 4.Kd7 Bc7 5.Nxe3 Bxg3 6.Kc6 Kg6 7.Kb6 Nc7 8.Bc2+ Kf6 9.Nf5 Bf4 10.Kxc5 h5 11.Kc4 Be5 12.Nh4 Ke6 13.Nf3 Bf6 14.Kd3 (89.518.587) 10500
26.19 0:19 -2.66 1.d8Q Nf7+ 2.Kd7 Nxd8 3.Nxe3 Ba5 4.Kd6 Bc7+ 5.Kd5 Bxg3 6.Bc2+ Kh8 7.Kc4 h5 8.Kxc3 h4 9.Bf5 Kg7 10.Nd5 Nb4 11.Kc4 Nxd5 12.Kxd5 Kf6 13.Bh3 Bf2 14.Bc8 (233.079.742) 11998
27.19 0:27 -2.66 1.d8Q Nf7+ 2.Kd7 Nxd8 3.Nxe3 Ba5 4.Nc4 Bc7 5.Na3 Nf7 6.Kc6 Bxg3 7.Kb6 Nc7 8.Nb1 Be5 9.Nxc3 Na8+ 10.Kb7 Bxc3 11.Kxa8 c4 12.Kb7 Ne5 13.Kb6 Bd4+ 14.Kb5 (332.420.871) 12293
28.19 0:47 -2.66 1.d8Q Nf7+ 2.Kd7 Nxd8 3.Nxe3 Ba5 4.Nd5 Kg7 5.Kc8 Ne6 6.Kb7 Nac7 7.Ne3 Nd8+ 8.Kc8 Nce6 9.Kd7 Kf6 10.Kd6 Bc7+ 11.Kd5 Bxg3 12.Kc4 Kg5 13.Nd5 c2 14.Bxc2 (607.929.647) 12668
29.19 1:10 -2.66 1.d8Q Nf7+ 2.Kd7 Nxd8 3.Nxe3 Ba5 4.Nc4 Bc7 5.Ne3 Nf7 6.Kc6 Bxg3 7.Kb6 Nc7 8.Bc2+ Kg8 9.Nf5 Be5 10.Kxc5 h5 11.Kc4 Kf8 12.Kd3 Nd5 13.Kc4 Ne7 14.Nh4 (895.472.953) 12782
30.03 1:16 -2.58++ 1.Nf6+ (978.342.869) 12812
30.01 1:20 -2.50++ 1.Nf6+ (1.034.377.125) 12864
30.01 1:22 -2.38++ 1.Nf6+ (1.063.265.989) 12852
30.01 1:23 -2.20++ 1.Nf6+ (1.076.610.003) 12851
30.01 1:25 -1.92++ 1.Nf6+ (1.099.932.939) 12853
30.01 1:26 -1.52++ 1.Nf6+ (1.111.242.186) 12858
30.01 1:27 -0.90++ 1.Nf6+ (1.126.708.527) 12868
30.01 1:29 0.00++ 1.Nf6+ (1.153.570.001) 12864
30.01 1:32 +1.37++ 1.Nf6+ (1.196.160.173) 12880
30.01 1:36 +3.42++ 1.Nf6+ (1.239.030.664) 12893
30.19 1:56 +4.52 1.Nf6+ Kg7 2.Nh5+ Kg6 3.Bc2+ Kxh5 4.d8Q Kg4 5.Qe8 c4+ 6.Kc6 Kf3 7.Bd1+ Kf2 8.Qh5 Kg1 9.Qe2 Bc5 10.Qxc4 c2 11.Bxc2 Nb4+ 12.Kxc5 Nxc2 13.Qxc2 Kf1 14.Qd1+ (1.513.657.492) 13003

[Arpad Rusz]

My version of this famous study:
http://ruszchessstudies.blogspot.hu/201 ... y-114.html

Finding 1.Nh5!! seems hard for the engines.

[peter]

Hi Arpad!

My version of this famous study:
http://ruszchessstudies.blogspot.hu/201 ... y-114.html

Finding 1.Nh5!! seems hard for the engines.

Very beautiful version of yours and this one doesn't have the detour of denying the Knight- fork capturing the Queen.

Sting yet gets it too:

7n/3P2k1/n2K3p/2p5/Bb3N2/2p1p1P1/8/8 w - - 0 1

Analysis by Sting SF 8 64bit:

1.Se6+
+- (5.61 --) Tiefe: 7 00:00:06 4kN
...
1.Se6+ Kg6 2.Lc2+ Kh5 3.Ld1+ Kg6 4.Lc2+
= (0.00) Tiefe: 7 00:00:07 9kN
...
1.Se6+ Kg6 2.Lc2+ Kh5 3.Ld1+ Kg6 4.Lc2+
= (0.00) Tiefe: 30 00:00:17 81721kN
1.Sh5+
= (0.08 ++) Tiefe: 31 00:00:20 125MN
...
1.Sh5+
+- (2.67 ++) Tiefe: 31 00:00:24 186MN
...
1.Sh5+
+- (20.69 ++) Tiefe: 31 00:01:04 668MN
...
1.Sh5+
+- (104.91 ++) Tiefe: 31 00:01:39 917MN
1.Sh5+ Kg6 2.Lc2+ Kxh5 3.d8D Sf7+ 4.Ke6 Sxd8+ 5.Kf5 e2 6.Le4 e1S 7.Ld5 c2 8.Lc4 c1S 9.Lb5 Sc7 10.La4 Sf3 11.Ld1 Se2 12.Lxe2
+- (104.91) Tiefe: 31 00:01:39 917MN

BTW 9...Nc6 isnstead of ...Nc7 in your solution line at the given link is one move longer to mate (14 after 9...Nc6 10.Bxc6 Nc7 instead of 13 after 9...Nc7).
Sorry, could not resist.

[peter]

Sting yet gets it too:

MateFinder:

7n/3P2k1/n2K3p/2p5/Bb3N2/2p1p1P1/8/8 w - - 0 1

Analysis by Stockfish MateFinder 64:

1.Ne6+ Kg6 2.Bc2+ Kh5 3.g4+ Kxg4 4.Bd1+ Kh3 5.Nf4+ Kg3 6.Ne2+ Kf2
+/- (1.34) Depth: 7/12 00:00:00 37kN
1.Ne6+ Kg6 2.Bc2+ Kh5 3.Bd1+ Kg6 4.Bc2+
= (0.00) Depth: 8/17 00:00:00 62kN
...
1.Ne6+ Kg6 2.Bc2+ Kh5 3.Bd1+ Kg6 4.Bc2+
= (0.00) Depth: 25/17 00:00:05 109MN
1.Nh5+
= (0.20 ++) Depth: 26/26 00:00:06 132MN
...
1.Nh5+
+- (2.02 ++) Depth: 26/26 00:00:06 133MN
...
1.Nh5+
+- (#52 ++) Depth: 27/50 00:00:10 255MN
1.Nh5+ Kg6 2.Bc2+ Kxh5 3.d8Q Nf7+ 4.Ke6 Nxd8+ 5.Kf5 e2 6.Be4 e1N 7.Bd5 c2 8.Bc4 c1N 9.Bb5 Nc6 10.Bxc6 Nc7 11.Ba4 Nb3 12.Bxb3 c4 13.Bd1+ Nf3 14.Bxf3#
+- (#14) Depth: 27/50 00:00:10 257MN
...
1.Nh5+ Kg6 2.Bc2+ Kxh5 3.d8Q Nf7+ 4.Ke6 Nxd8+ 5.Kf5 e2 6.Be4 e1N 7.Bd5 c2 8.Bc4 c1N 9.Bb5 Nc6 10.Bxc6 Nc7 11.Ba4 Nb3 12.Bxb3 c4 13.Bd1+ Nf3 14.Bxf3#
+- (#14) Depth: 48/50 00:00:17 422MN

[MikeB]

many thanks Dan

good thats some of handfull engines can see some advantage

also McBrain ( a Stockfish engine) if you turn the matefinder option on:

Code: Select all

so name matefinder value true
p f 7n/3P2k1/n2K3p/2p5/Bb3N2/2p1p1P1/8/8 w - - 0 1 
so name hash value 4096
so name threads value 18
g i
info depth 1 seldepth 1 multipv 1 score cp 539 nodes 1010 nps 168333 tbhits 0 time 6 pv f4e6 g7h7 d7d8q
info depth 2 seldepth 4 multipv 1 score cp 600 nodes 3593 nps 359300 tbhits 0 time 10 pv f4e6 g7h7 a4c2 h8g6 d7d8q
info depth 3 seldepth 5 multipv 1 score cp 531 nodes 4711 nps 428272 tbhits 0 time 11 pv f4e6 g7h7 a4c2 h8g6 d7d8q c5c4 d6d5
info depth 4 seldepth 6 multipv 1 score cp 487 nodes 5137 nps 467000 tbhits 0 time 11 pv f4e6 g7h7 a4c2 h8g6 d7d8q e3e2
info depth 5 seldepth 8 multipv 1 score cp 470 nodes 7045 nps 541923 tbhits 0 time 13 pv f4e6 g7h7 a4c2 h8g6 d7d8q c5c4 d6d5 e3e2 d8d7 b4e7
info depth 6 seldepth 10 multipv 1 score cp 472 nodes 10586 nps 661625 tbhits 0 time 16 pv f4e6 g7h7 a4c2 h8g6 d7d8q c5c4 d6d5 e3e2 d8e8
info depth 7 seldepth 14 multipv 1 score cp 6 nodes 45781 nps 1476806 tbhits 0 time 31 pv f4e6 g7g6 a4c2 g6h5 c2d1 h5g6 d1c2
info depth 8 seldepth 14 multipv 1 score cp 6 nodes 52131 nps 1489457 tbhits 0 time 35 pv f4e6 g7g6 a4c2 g6h5 c2d1 h5g6 d1c2
info depth 9 seldepth 14 multipv 1 score cp 6 nodes 54549 nps 1515250 tbhits 0 time 36 pv f4e6 g7g6 a4c2 g6h5 c2d1 h5g6 d1c2
info depth 10 seldepth 14 multipv 1 score cp 6 nodes 69154 nps 1503347 tbhits 0 time 46 pv f4e6 g7g6 a4c2 g6h5 c2d1 h5g6 d1c2
info depth 11 seldepth 14 multipv 1 score cp 6 nodes 82900 nps 1429310 tbhits 0 time 58 pv f4e6 g7g6 a4c2 g6h5 c2d1 h5g6 d1c2
info depth 12 seldepth 14 multipv 1 score cp 6 nodes 106760 nps 1351392 tbhits 0 time 79 pv f4e6 g7g6 a4c2 g6h5 c2d1 h5g6 d1c2
info depth 13 seldepth 14 multipv 1 score cp 6 nodes 129024 nps 1330144 tbhits 0 time 97 pv f4e6 g7g6 a4c2 g6h5 c2d1 h5g6 d1c2
info depth 14 seldepth 14 multipv 1 score cp 6 nodes 513378 nps 1361745 tbhits 0 time 377 pv f4e6 g7g6 a4c2 g6h5 c2d1 h5g6 d1c2
info depth 15 seldepth 14 multipv 1 score cp 6 nodes 671516 nps 1407790 tbhits 0 time 477 pv f4e6 g7g6 a4c2 g6h5 c2d1 h5g6 d1c2
info depth 16 seldepth 14 multipv 1 score cp 6 nodes 1297237 nps 1684723 tbhits 0 time 770 pv f4e6 g7g6 a4c2 g6h5 c2d1 h5g6 d1c2
info depth 17 seldepth 14 multipv 1 score cp 6 nodes 1708414 nps 1861017 tbhits 0 time 918 pv f4e6 g7g6 a4c2 g6h5 c2d1 h5g6 d1c2
info depth 18 seldepth 14 multipv 1 score cp 6 nodes 1760287 nps 1886695 tbhits 0 time 933 pv f4e6 g7g6 a4c2 g6h5 c2d1 h5g6 d1c2
info depth 19 seldepth 14 multipv 1 score cp 6 nodes 6448567 nps 4534857 tbhits 0 time 1422 pv f4e6 g7g6 a4c2 g6h5 c2d1 h5g6 d1c2
info depth 20 seldepth 14 multipv 1 score cp 6 nodes 9660044 nps 6228268 tbhits 0 time 1551 pv f4e6 g7g6 a4c2 g6h5 c2d1 h5g6 d1c2
info depth 21 seldepth 14 multipv 1 score cp 6 nodes 10744612 nps 6753370 tbhits 0 time 1591 pv f4e6 g7g6 a4c2 g6h5 c2d1 h5g6 d1c2
info depth 22 seldepth 14 multipv 1 score cp 6 nodes 16370921 nps 9135558 tbhits 0 time 1792 pv f4e6 g7g6 a4c2 g6h5 c2d1 h5g6 d1c2
info depth 23 seldepth 14 multipv 1 score cp 6 nodes 49550702 nps 16588785 tbhits 0 time 2987 pv f4e6 g7g6 a4c2 g6h5 c2d1 h5g6 d1c2
info depth 24 seldepth 14 multipv 1 score cp 6 nodes 103282948 nps 20954138 tbhits 0 time 4929 pv f4e6 g7g6 a4c2 g6h5 c2d1 h5g6 d1c2
info depth 25 seldepth 14 multipv 1 score cp 6 nodes 182269219 nps 23373841 hashfull 325 tbhits 0 time 7798 pv f4e6 g7g6 a4c2 g6h5 c2d1 h5g6 d1c2
info depth 26 seldepth 28 multipv 1 score mate 14 nodes 407325197 nps 25964125 hashfull 623 tbhits 0 time 15688 pv f4h5
info depth 26 seldepth 28 multipv 1 score mate 14 nodes 407336134 nps 25964822 hashfull 623 tbhits 0 time 15688 pv f4h5 g7g6 a4c2 g6h5 d7d8q h8f7 d6e6 f7d8 e6f5 e3e2 c2e4 e2e1n e4d5 c3c2 d5c4 c2c1n c4b5 d8c6 b5c6 a6c7 c6a4 c1b3 a4b3 c5c4 b3d1 e1f3 d1f3

more info on McBrain:
https://github.com/MichaelB7/Stockfish/ ... s/tag/v2.3

The matefinder code in McBrain is from Joe Ellis' fork of Stockfish labeled MateFinder - but in McBrain it is a UCI option check off and on through the GUI or at the command line as opposed to being hardcoded in the executable. In McBrain, the code costs about 40 ELO in very short micro time controls ( 10 sec /game) - that is probably a higher cost in ELO than if you just used MateFinder by Ellis. At longer time controls it will be less of cost - but I'm not sure about how much. In earlier tests I ran on Matefinder, I have a vague recollection of it costing about 15 ELO, that was quite a while ago - but my sense is that it would still cost 10 to 15 ELO today. It's a trade off, but certainly Stockfish based engines like Matefinder and Sugar serve a very, very useful purpose.

Just trying too be helpful - hopefully no one flies off the handle because I mentioned McBrain in this post. Stockfish is the strongest chess engine in the world and all of us that appreciate a free engine open source like stockfish are very much indebted to all the volunteers and programmers who have made Stockfish freely available to everyone with the source so people like me and others can add some features ( play by ELO, Cerebellum book by Tom Zipproth, Matefinder by Elis etc or even combined a number of features into one engine as I have done with McBrain - and share that work with the world for a free. It's been misconstrued that my intent with MCBrain was to make it stronger than Stockfish - well that's impossible since the strength of McBrain is 100% derived from Stockfish and all the work that the Stockfish team does. Naturally I may tweak somethings here and there , and occasionally I may have a lucky run - bit let's not kid ourselves - McBrain is a stockfish engine and that is the sole reason McBrain can play chess at very high level. I just hope people enjoy McBrain as much as as I do - and the best part - it's 100% free. And guess what - I just updated McBrain with latest Stockfish commits and it just got stronger - but again not because of me - but due to large number of programmers and volunteers that have help make Stockfish what it is today and what it will be in the future. Enjoy!

Edit : The run above with McBrain was done with an unreleased version of McBrain - but the only difference are the most recent commits pushed to master , plus the ~1% speed pickup (excellent patch!) that Vondele proposed that has not yet been push - but I am 100% certain it will not be long before it is pushed.