Can any chess engine or mate finder solve the following:

[Ignacio]

Very well, Marek, bravo Sting!
Just as for the output- pv, 31.Be7 still might corst two moves, Bh6 not, seem to be other detours in rest of pv too, so as for shortest length to mate it could be shorter, but that doesn't matter here so much neither, I guess.

Out of edit- time again.
On the other hand, e.g. 2...h5 might give away some moves for Black again compared to 2...Ka8.

Best pv could be something about 57 moves to mate, I guess.

Try Chest Database.

[peter]

Try Chest Database.

Thanks, found it there as well as Google did by FEN-search as #56.
Yet still don't know the author

[Ignacio]

Try Chest Database.

Thanks, found it there as well as Google did by FEN-search as #56.
Yet still don't know the author

Blathy
http://www.yacpdb.org/#search/ODg4ODg4O ... LzEvMA==/1

[peter]

Blathy
http://www.yacpdb.org/#search/ODg4ODg4O ... LzEvMA==/1

Thanks a lot again, Ignacio!

Having found it by the same very fine database- search at yapdb.org, I return to give the link, I see you already gave it

[Dann Corbit]

Where is it from, Dann, must be a study?

Each Epd with annotations that tell where it came from:
[d]1Q4R1/8/P1b2p2/2p2k2/2P5/5p2/1r5P/4R1K1 b - - acd 44; acs 292; bm f2+; c0 "level: easy-1"; c3 "f2+"; c4 "(Krogius_EofCM) #1485 white_ahead"; ce -5592; dm 9; id "ECM.547"; pm f2+; pv f2+ Kf1 Bg2+ Rxg2 fxe1=Q+ Kxe1 Rxb8 a7 Rd8 Ke2 Ke4 Rg4+ Ke5 Rg7 f5 h4 Re8 Kf3 Kf6 Rb7 Ra8 Kf4 Kg6 Rc7 Rf8 h5+ Kh6 Rxc5 Re8 Ra5 Ra8 Ra4 Kg7 Ra6 Kh7 Kxf5 Rf8+ Ke4 Re8+ Kd5 Rd8+ Kc6 Rc8+ Kb7 Rf8 a8=Q Rf7+ Kc6 Kg7 Ra7 Rb7 Rxb7+ Kf6;
[d]8/kn4b1/P2B4/8/1Q6/6pP/1q4pP/5BK1 w - - acd 60; acs 285; bm Qxb2; c3 "Bc5+"; ce 224; dm 45; id "E_E_T 015 - D&L&L vs D&L&S"; pm Qxb2; pv Qxb2 Bxb2 axb7 gxf1=R+ Kxf1 g2+ Kxg2 Kxb7 Kf3 Kc6 Bf4 Kd7 Kg4 Ke6 Kg5 Kf7 Kh6 Bf6 Bg3 Bg7+ Kh7 Bc3 h4 Kf6 Bf2 Kf5 Kh6 Kf6 Be3 Be5 h3 Kf5 h5 Kf6 Kh7 Kf7 h6 Bb2 Bb6 Be5 Bc5 Bb2 Ba7 Be5 Bg1 Bb2 Be3 Be5 Bf2 Bb2 Bg1 Be5 Bb6 Bb2 Bc7 Bd4 Bg3 Bf6 Bd6 Bd4 Bf4 Bf6 Bg3 Bb2 Bf2 Be5 Ba7 Bf6 Bb8 Bd4 Bd6 Ke6 Bf4 Kf7 Bg5;
[d]8/pp4p1/4k1P1/3p4/3P4/2P5/4K2P/8 w - - acd 52; acs 599; bm c4; c0 "level: med-10"; c3 "c4"; c4 "(JenoBan_TofE) #193 white_wins"; ce -32607; dm 24; id "EGP 0214"; pm c4; pv c4 Kf6 c5 Kxg6 Ke3 Kf5 Kd3 a5 Ke3 a4 Kd3 a3 Kc2 Ke4 Kc3 a2 Kb2 Kxd4 Ka1 Kxc5 Kxa2 Kd4 Kb2 Ke4 Ka2 Kf4 Ka3 Kg4 Ka2 Kh3 Ka1 Kxh2;
[d]k1b5/1p1p1p1p/1PpPpP2/2B5/8/2p1p2P/2PbP3/3K4 w - - acd 127; acs 352; bm Bd4; ce 0; dm 56; id "yacpdb id 277997"; pm Bd4; pv Bd4 e5 Bxe5 Kb8 Bd4 h6 Be5 Ka8 Bd4 h5 Be5 Kb8 Bh2 Ka8 Be5;
[d]r4BK1/6P1/1p1p3P/3P4/2k5/p2N1q2/P7/8 w - - acd 54; acs 253; bm Ne5+; ce -32611; dm 24; id "yacpdb id 37311"; pm Ne5+; pv Ne5+ dxe5 h7 Ra7 h8=Q Qxd5+ Kh7 e4 Qg8 Qxg8+ Kxg8 e3 Bd6 e2 Bg3 Rxg7+ Kxg7 b5 Be1 b4 Bh4 b3 Be7 b2 Bxa3 e1=Q Bxb2;

Here is YACPDB:
http://www.yacpdb.org/
You can paste a position in the slot next to the "Add" button and then press "Search" and it will try to look up the position.

[Dann Corbit]

YACPDB says the last one is a draw.

[Dann Corbit]

They are all supposed to be checkmates.

Every engine I tried said that these positions are lost:
1Q4R1/8/P1b2p2/2p2k2/2P5/5p2/1r5P/4R1K1 b - -
8/pp4p1/4k1P1/3p4/3P4/2P5/4K2P/8 w - -
r4BK1/6P1/1p1p3P/3P4/2k5/p2N1q2/P7/8 w - -

These two look like draws, according to the engines that I tried:
8/kn4b1/P2B4/8/1Q6/6pP/1q4pP/5BK1 w - -
k1b5/1p1p1p1p/1PpPpP2/2B5/8/2p1p2P/2PbP3/3K4 w - -

Try lowering the zugzwang verification depth and recompiling SF. I bet it works.

That's a really good idea.
I made a UCI Option for it.

[Dann Corbit]

They are all supposed to be checkmates.

Every engine I tried said that these positions are lost:
1Q4R1/8/P1b2p2/2p2k2/2P5/5p2/1r5P/4R1K1 b - -
8/pp4p1/4k1P1/3p4/3P4/2P5/4K2P/8 w - -
r4BK1/6P1/1p1p3P/3P4/2k5/p2N1q2/P7/8 w - -

These two look like draws, according to the engines that I tried:
8/kn4b1/P2B4/8/1Q6/6pP/1q4pP/5BK1 w - -
k1b5/1p1p1p1p/1PpPpP2/2B5/8/2p1p2P/2PbP3/3K4 w - -

Try lowering the zugzwang verification depth and recompiling SF. I bet it works.

That's a really good idea.
I made a UCI Option for it.

Here is a modified version, if anyone else wants to play with it.
https://www.dropbox.com/s/lfp54yivt3tft ... sh.7z?dl=0

[peter]

Try lowering the zugzwang verification depth and recompiling SF. I bet it works.

That's a really good idea.
I made a UCI Option for it.

Here is a modified version, if anyone else wants to play with it.
https://www.dropbox.com/s/lfp54yivt3tft ... sh.7z?dl=0

Thanks, Dann, but as for the Blathy- study it doesn't succeed, neither does a SF not using any nullmove at all I tried. That one isn't a question of nullmove only

[lech]

Very well, Marek, bravo Sting!

Seems to be what Arpad Rusz calls cyclic zugzwang, if I remember correctly.
...

Peter, after a small modification of Blathy study (without h pawns), I think Sting 5 is not able to solve it without "a lazzy King" code. Am I sure?

[d]k1b5/1p1p1p2/1PpPpP2/2B5/8/2p1p3/2PbP3/3K4 w - - 0 1

Now white King should go to take the Pawn f7.