Engine test: 41 mate in 4 puzzles from Classic Chess Puzzles

Discussion of chess software programming and technical issues.

Moderators: hgm, chrisw, Rebel

Chessnut1071
Posts: 313
Joined: Tue Aug 03, 2021 2:41 pm
Full name: Bill Beame

Engine test: 41 mate in 4 puzzles from Classic Chess Puzzles

Post by Chessnut1071 »

Tests everything except castling: ent passant, discovered check, double check, stalemate, and pawn promotion. Kenneth Howard even submitted one of his own classic gems. If anybody has access to some of the leading engines it would be nice to see the relative performance especially on time. enjoy!



FEN[121] = "n7/3p1p2/NpkNp1p1/1p2P3/3Q4/6B1/b7/4K3 w - - 0 1 "; // Sam Loyd (1856) 4-move mate [Qg1] - (5)
FEN[122] = "K6Q/1p6/pPq4P/P2p2P1/4pP1N/7k/n5R1/1n2BB2 w - - 0 1 "; // Sam Loyd (1867) 4-move mate [BxP] - (11)
FEN[123] = "8/8/8/3p4/1N1P2Kp/Bkp5/5Q2/4N3 w - - 0 1 "; // Sam Loyd (1867) 4-move mate [Qf8] - (13)
FEN[124] = "4K3/8/pp2B3/P7/4kpR1/2P1p3/4Q3/2r5 w - - 0 1 "; // Philip Klett 4-move mate [Rg2] - (19)
FEN[125] = "8/Q5p1/2Bpq3/KPn2pR1/3kpP1p/1P1p3b/3B1b2/7n w - - 0 1 "; // Philip Klett 4-move mate [Rg6] -(20)
FEN[126] = "2Q3Kn/p7/8/3NkPp1/p7/P1Np1pPr/1P1p4/1b2b2r w - - 0 1 "; // Philip Klett 4-move mate [Pb4] - (21)
FEN[127] = "4b1N1/2pr2pR/2rp1p2/1Q1B1kp1/2pP4/4P1K1/1n3PN1/R7 w - - 0 1 "; // Philip Klett 4-move mate [Rg1] - (22)
FEN[128] = "r1b5/1p1n3p/4p2K/p5Bp/3k4/Q4B2/2P5/5n2 w - - 0 1 "; // Philip Klett 4-move mate [Be7] - (23)
FEN[129] = "4N3/1p6/3B2nN/pPpk4/4R1Pp/P1p2P2/K1P1b3/3n4 w - - 0 1 "; // Philip Klett 4-move mate [Bh2] - (24)
FEN[130] = "3r1n2/2bp2p1/1p1kp1Pp/4N2p/Pp2B2Q/7P/1P3P1B/K7 w - - 0 1 "; // Philip Klett 4-move mate [Ba8, Bf3] - (25)
FEN[131] = "3K4/8/6R1/p3k1pp/2bp1R1p/n1b1P3/2p2N2/2B4B w - - 0 1 "; // Philip Klett 4-move mate [Bd2, RxNP] - (26)
FEN[132] = "5n2/r2p1p2/p1PBkn2/p3Ppp1/6b1/1N1pP3/1N1P3K/2R5 w - - 0 1 "; // Philip Klett 4-move mate [Na4] - (27)
FEN[133] = "8/1Qp5/2K5/1R5p/7P/5k2/8/8/ w - - 0 1 "; // William Shinkman (1890) 4-move mate [Ra5] - (32)
FEN[134] = "8/8/8/p7/R4B2/8/5k2/1B5K w - - 0 1 "; // Otto Wurzburg (1937) 4-move mate [Bf5] - (41)
FEN[135] = "2N5/8/N1B5/4k2p/8/7K/8/2Q5 w - - 0 1 "; // Miroslav Havel 4-move mate [Kh4] - (48)
FEN[136] = "8/1R6/8/3k3N/N5R1/8/8/4K3 w - - 0 1 "; // Miroslav Havel (1920) 4-move mate [Rg4-g7] - (50)
FEN[137] = "8/1R4p1/3kN3/5PR1/8/6K1/8/8/ w - - 0 1 "; // Miroslav Havel (1924) 4-move mate [Pf6] - (51)
FEN[138] = "8/8/2K3p1/4N2B/7k/5p2/Q7/8 w - - 0 1 "; // Miroslav Havel (1948) 4-move mate [Qb1] - (52)
FEN[139] = "1n6/2Bp4/p2N4/N1k5/2p1P3/R1P5/1P6/1K1b4 w - - 0 1 "; // J. Kohtz & C. Kockelkorn (1862) 4-move mate [Ra4]- (53)
FEN[140] = "2r5/2p3n1/5R2/4k1PP/3R2p1/2n1P3/1Q2P3/4b2K w - - 0 1 "; // Kohtz & Kockelkorn (1901) 4-move mate [Qb7]- (58)
FEN[141] = "8/8/1p6/1P1P1B2/k7/1NK5/2N5/8 w - - 0 1 "; // Kohtz & Kockelkorn (1901) 4-move mate [Ne3] - (62)
FEN[142] = "4nKn1/1Q6/2p2P2/1p6/6r1/2P1B3/1N1Np1b1/4k3 w - - 0 1 "; // J. Kohtz & C. Kockelkorn (1911) 4-move mate [Qf7] - (64)
FEN[143] = "8/np1N1N2/4p3/1p1b4/1b2kp2/p1R5/p3P3/n1Q2K2 w - - 0 1 "; // GodFrey Heathcote (1915) 4-move mate [Rg3] - (73)
FEN[144] = "bK1N2N1/5Qnr/2p4p/2pnk3/8/1p1BPP2/6rp/7q w - - 0 1 "; // GodFrey Heathcote (1914) 4-move mate [Bc4] - (74)
FEN[145] = "1QB3n1/BK4P1/4Pp1r/p2kb2p/7P/5Rp1/1PP2Pp1/8 w - - 0 1 "; // GodFrey Heathcote (1921) 4-move mate [Pe7] - (75)
FEN[146] = "n7/3rp1NK/b4pP1/1r6/4kb2/p4R2/qNP2R1P/1n5B w - - 0 1 "; // Sam Loyd (1857) 4-move mate [RxB+] - (93)
FEN[147] = "8/1pp5/2b5/4N3/2Q2B2/k1p4r/p5rp/K1R5 w - - 0 1 "; // Darso Densmore (1916) 4-move mate [Bh6] - (95)
FEN[148] = "2k5/K2pR3/RB1p1r2/1p4N1/1p4N1/1P6/6n1/3Br2q w - - 0 1 "; // Darso Densmore (1916) 4-move mate [Ra2] - (96)
FEN[149] = "4b3/3q4/8/4P3/4BP1p/1R1Pk2p/4N3/4K2R w - - 0 1 "; // Darso Densmore (1916) 4-move mate [Ba8] - (101)
FEN[150] = "1KB1n3/Q4p1N/1Npk3q/r1p5/2p2p1b/R1p4b/2P2P2/4R3 w - - 0 1 "; // Louis Jorkisch (1918) 4-move mate [Ra3-a1] - (103)
FEN[151] = "8/8/8/4p3/4p3/2k5/2P1K3/1R1R4 w - - 0 1 "; // J. Kohtz & C. Kockelkorn (1892) 4-move mate [Rb5] - (109)
FEN[152] = "4k3/5RR1/8/1n6/8/7p/7K/8 w - - 0 1 "; // Otto Wurzburg (1918) 4-move mate [Rh7] - (111)
FEN[153] = "8/8/8/3R4/8/5K2/4n2p/7k w - - 0 1 "; // Sam Loyd (1885) 4-move mate [Rd2] - (112)
FEN[154] = "8/8/3K4/6p1/3k3P/1R1P1R2/8/8 w - - 0 1 "; // GodFrey Heathcote (1911) 4-move mate [Rh3] - (115)
FEN[155] = "R7/8/1p4K1/1Pp4P/1P1p2pk/2pR4/P5P1/8 w - - 0 1 "; // Kenneth Howard (1925) 4-move mate [Ra3] - (116)
FEN[156] = "7R/8/8/4N3/1p5p/4K2p/8/4k3 w - - 0 1 "; // Wolfgang Pauly (1923) 4-move mate [Rc8] - (135)
FEN[157] = "6R1/8/8/3N4/p5p1/3K2p1/8/3k4 w - - 0 1 "; // Wolfgang Pauly (1923) 4-move mate [Re8] - (136)
FEN[158] = "3k4/5R2/8/8/2K5/B3N3/8/8/ w - - 0 1 "; // Wolfgang Pauly (1906) 4-move mate [Nf5] - (137)
FEN[159] = "3k4/7R/8/8/K2B1N2/8/8/8/ w - - 0 1 "; // Wolfgang Pauly (1906) 4-move mate [Nd3] - (138)
FEN[160] = "3k4/6R1/8/8/1N1B4/3K4/8/8/ w - - 0 1 "; // Wolfgang Pauly (1906) 4-move mate [Be3] - (139)
FEN[161] = "3k4/7R/8/8/1N1B4/2K5/8/8/ w - - 0 1 "; // Wolfgang Pauly (1907) 4-move mate [Nd3] - (140)
dangi12012
Posts: 1062
Joined: Tue Apr 28, 2020 10:03 pm
Full name: Daniel Infuehr

Re: Engine test: 41 mate in 4 puzzles from Classic Chess Puzzles

Post by dangi12012 »

I see your array starts at 121. Why dont you release all positions at once? - my guess is that the comment will tell us what type it is and that there are more gems in the other positions as well :)
Worlds-fastest-Bitboard-Chess-Movegenerator
Daniel Inführ - Software Developer
Chessnut1071
Posts: 313
Joined: Tue Aug 03, 2021 2:41 pm
Full name: Bill Beame

Re: Engine test: 41 mate in 4 puzzles from Classic Chess Puzzles

Post by Chessnut1071 »

dangi12012 wrote: Wed Nov 03, 2021 11:52 pm I see your array starts at 121. Why dont you release all positions at once? - my guess is that the comment will tell us what type it is and that there are more gems in the other positions as well :)
Look in tab 2. Those were listed last week.
User avatar
Roland Chastain
Posts: 659
Joined: Sat Jun 08, 2013 10:07 am
Location: France
Full name: Roland Chastain

Re: Engine test: 41 mate in 4 puzzles from Classic Chess Puzzles

Post by Roland Chastain »

Mater find a mate in 2 for position 127 (b5b2), a mate in 1 for position 130 (e5d3), and no mate for position 155.

For the mate in 1, this is a bug in the program, but the mate in 2 seems correct to me. Am I wrong?

For the position 155, I don't know.
Qui trop embrasse mal étreint.
Andrew
Posts: 231
Joined: Thu Mar 09, 2006 12:51 am
Location: Australia

Re: Engine test: 41 mate in 4 puzzles from Classic Chess Puzzles

Post by Andrew »

Roland Chastain wrote: Thu Nov 04, 2021 9:20 am Mater find a mate in 2 for position 127 (b5b2), a mate in 1 for position 130 (e5d3), and no mate for position 155.

For the mate in 1, this is a bug in the program, but the mate in 2 seems correct to me. Am I wrong?

For the position 155, I don't know.
Hi Roland, for position 127, after Qxb2 then Kg6 and it's at least 4 more moves to mate (have tried multiple engines).

Position 155 also looks fine with multiple engines.

For position 130, there is a single response e5.

Andrew
User avatar
Roland Chastain
Posts: 659
Joined: Sat Jun 08, 2013 10:07 am
Location: France
Full name: Roland Chastain

Re: Engine test: 41 mate in 4 puzzles from Classic Chess Puzzles

Post by Roland Chastain »

Andrew wrote: Thu Nov 04, 2021 11:18 am Hi Roland, for position 127, after Qxb2 then Kg6 and it's at least 4 more moves to mate (have tried multiple engines).
Hi Andrew. Indeed. Thank you for your answer.
Qui trop embrasse mal étreint.
Joerg Oster
Posts: 952
Joined: Fri Mar 10, 2006 4:29 pm
Location: Germany
Full name: Jörg Oster

Re: Engine test: 41 mate in 4 puzzles from Classic Chess Puzzles

Post by Joerg Oster »

Roland Chastain wrote: Thu Nov 04, 2021 11:47 am
Andrew wrote: Thu Nov 04, 2021 11:18 am Hi Roland, for position 127, after Qxb2 then Kg6 and it's at least 4 more moves to mate (have tried multiple engines).
Hi Andrew. Indeed. Thank you for your answer.
Matefish (work in progress ...) on Position 127:

Code: Select all

position fen 4b1N1/2pr2pR/2rp1p2/1Q1B1kp1/2pP4/4P1K1/1n3PN1/R7 w - - 0 1
go mate 4
info string No mate in 1 found
info string No mate in 2 found
info string No mate in 3 found
info string Success! Mate in 4 found!
info depth 7 seldepth 7 multipv 1 score mate 4 nodes 1778690 nps 1572670 hashfull 0 tbhits 0 time 1131 pv a1g1 g5g4 g2h4 f5g5 f2f4 g4f3 g3h3
bestmove a1g1 ponder g5g4
Position 130:

Code: Select all

position fen 3r1n2/2bp2p1/1p1kp1Pp/4N2p/Pp2B2Q/7P/1P3P1B/K7 w - - 0 1
go mate 4
info string No mate in 1 found
info string No mate in 2 found
info string No mate in 3 found
info string Success! Mate in 4 found!
info depth 7 seldepth 7 multipv 1 score mate 4 nodes 111667 nps 1488893 tbhits 0 time 75 pv e4a8 d8a8 h4b4 d6d5 b4d2 d5e4 d2d3
bestmove e4a8 ponder d8a8
Position 155:

Code: Select all

position fen R7/8/1p4K1/1Pp4P/1P1p2pk/2pR4/P5P1/8 w - - 0 1
go mate 4
info string No mate in 1 found
info string No mate in 2 found
info string No mate in 3 found
info string Success! Mate in 4 found!
info depth 7 seldepth 7 multipv 1 score mate 4 nodes 136302 nps 1434757 tbhits 0 time 95 pv a8a3 c5b4 a3b3 c3c2 d3d4 c2c1n b3h3
bestmove a8a3 ponder c5b4
Didn't know about Mater. :)
Jörg Oster
User avatar
Roland Chastain
Posts: 659
Joined: Sat Jun 08, 2013 10:07 am
Location: France
Full name: Roland Chastain

Re: Engine test: 41 mate in 4 puzzles from Classic Chess Puzzles

Post by Roland Chastain »

Joerg Oster wrote: Thu Nov 04, 2021 12:34 pm Matefish (work in progress ...) on Position 127:
Thank you for the informations.
Joerg Oster wrote: Thu Nov 04, 2021 12:34 pm Didn't know about Mater. :)
It seems to have a bug. :)
Qui trop embrasse mal étreint.
User avatar
Ajedrecista
Posts: 2025
Joined: Wed Jul 13, 2011 9:04 pm
Location: Madrid, Spain.

Re: Engine test: 41 mate in 4 puzzles from Classic Chess Puzzles.

Post by Ajedrecista »

Hello Bill:

Code: Select all

FEN[130] = "3r1n2/2bp2p1/1p1kp1Pp/4N2p/Pp2B2Q/7P/1P3P1B/K7 w - - 0 1 "; // Philip Klett 4-move mate [Ba8, Bf3] - (25)
Chest 5.2 gives five solutions instead of two:

Code: Select all

FEN: 3r1n2/2bp2p1/1p1kp1Pp/4N2p/Pp2B2Q/7P/1P3P1B/K7 w - - 0 1

ChestUCI:
FEN: 3r1n2/2bp2p1/1p1kp1Pp/4N2p/Pp2B2Q/7P/1P3P1B/K7 w - -   (10+11)
Stellungs-Analyse:  C0/R0/K2/P6/X10   W:8/35
Suche nach Matt in 10 ...  (Hash=64MB)
   4	00:00	           0	0	+M4	Be4d3
   4	00:00	           0	0	+M4	Be4f3
   4	00:00	           0	0	+M4	Be4g2
   4	00:00	           0	0	+M4	Be4h1
   4	00:00	           0	0	+M4	Be4a8
Suche abgeschlossen ...  (Zeit=0.05s)
Matt in 4 gefunden !  (5 Lösungen in 00:00)
  4/4	00:00	      35.200	718.367	+M4	Be4a8 Bc7b8 f2f3 b4b3 Ne5f7+ Kd6c5 Bh2g1+
  4/4	00:00	      35.200	718.367	+M4	Be4h1 Bc7b8 b2b3 b6b5 Qh4xb4+ Kd6c7 Qb4c5+
  4/4	00:00	      35.200	718.367	+M4	Be4g2 Bc7b8 b2b3 b6b5 Qh4xb4+ Kd6c7 Qb4c5+
  4/4	00:00	      35.200	718.367	+M4	Be4f3 Bc7b8 b2b3 b6b5 Qh4xb4+ Kd6c7 Qb4c5+
  4/4	00:00	      35.200	718.367	+M4	Be4d3 Bc7b8 Ne5f7+ Kd6c6 Nf7xd8+ Kc6c5 Qh4c4+
I have not tried other problems, but there could be more solutions.

Regards from Spain.

Ajedrecista.
Chessnut1071
Posts: 313
Joined: Tue Aug 03, 2021 2:41 pm
Full name: Bill Beame

Re: Engine test: 41 mate in 4 puzzles from Classic Chess Puzzles.

Post by Chessnut1071 »

Ajedrecista wrote: Thu Nov 04, 2021 7:29 pm Hello Bill:

Code: Select all

FEN[130] = "3r1n2/2bp2p1/1p1kp1Pp/4N2p/Pp2B2Q/7P/1P3P1B/K7 w - - 0 1 "; // Philip Klett 4-move mate [Ba8, Bf3] - (25)
Chest 5.2 gives five solutions instead of two:

Code: Select all

FEN: 3r1n2/2bp2p1/1p1kp1Pp/4N2p/Pp2B2Q/7P/1P3P1B/K7 w - - 0 1

ChestUCI:
FEN: 3r1n2/2bp2p1/1p1kp1Pp/4N2p/Pp2B2Q/7P/1P3P1B/K7 w - -   (10+11)
Stellungs-Analyse:  C0/R0/K2/P6/X10   W:8/35
Suche nach Matt in 10 ...  (Hash=64MB)
   4	00:00	           0	0	+M4	Be4d3
   4	00:00	           0	0	+M4	Be4f3
   4	00:00	           0	0	+M4	Be4g2
   4	00:00	           0	0	+M4	Be4h1
   4	00:00	           0	0	+M4	Be4a8
Suche abgeschlossen ...  (Zeit=0.05s)
Matt in 4 gefunden !  (5 Lösungen in 00:00)
  4/4	00:00	      35.200	718.367	+M4	Be4a8 Bc7b8 f2f3 b4b3 Ne5f7+ Kd6c5 Bh2g1+
  4/4	00:00	      35.200	718.367	+M4	Be4h1 Bc7b8 b2b3 b6b5 Qh4xb4+ Kd6c7 Qb4c5+
  4/4	00:00	      35.200	718.367	+M4	Be4g2 Bc7b8 b2b3 b6b5 Qh4xb4+ Kd6c7 Qb4c5+
  4/4	00:00	      35.200	718.367	+M4	Be4f3 Bc7b8 b2b3 b6b5 Qh4xb4+ Kd6c7 Qb4c5+
  4/4	00:00	      35.200	718.367	+M4	Be4d3 Bc7b8 Ne5f7+ Kd6c6 Nf7xd8+ Kc6c5 Qh4c4+
I have not tried other problems, but there could be more solutions.

Regards from Spain.

Ajedrecista.
if that's FEN 136, Miroslav Havel's 4-move mate every square on that diagonal is a solution. There may be one other 4-move mate with multiple solutions. Back then they didn't have computer assistance.