I have a huge collection of chess problems that are solved

[Dann Corbit]

Besides the Chest database, I have many tens of thousands collected from all over the place. I have used Chest in batch mode to verify the solutions. Most of them have attributions in the EPD records so that you can track the history of the position if you care to.

Since my FTP/Web site is still toast, does anyone want to host them?

[pedrox]

Note (Dann)

The files are compressed first with bzip2 and I put then in a zip file just because it is easier than one at a time.

Only through direct mate in 1 through direct mate in 9 have been 100% verified. I am working on DM10 now.

http://danasah.googlepages.com/matedb.zip

[jdart]

Are they all mate problems?

--Jon

[David Dahlem]

I download the collection, extracted the first file, dm001.epd and picked a random position.

8/3N4/n7/pk2N3/1n6/8/PPPPPPPP/2K5 w - - dm 1; id "Bult0170"; c0 "DAWSON, Thomas Rainer Inédit";

[d]8/3N4/n7/pk2N3/1n6/8/PPPPPPPP/2K5 w - - 0 1

Does "dm1" mean a mate in 1? Or 1 move for the attacked king?

Regards
Dave

[David Dahlem]

The second position picked at random. This is a mate in 10?

8/4p1p1/1p2P1P1/1P6/B6N/2P3PN/PP5p/K4Rrk w - - dm 10; id "MSE90505"; c0 "CUPPINI, Alessandro 5857 Sakkelet 6.-7.1993 3rd Mention";

[d]8/4p1p1/1p2P1P1/1P6/B6N/2P3PN/PP5p/K4Rrk w - - 0 1

Regards
Dave

[George Tsavdaris]

I download the collection, extracted the first file, dm001.epd and picked a random position.

8/3N4/n7/pk2N3/1n6/8/PPPPPPPP/2K5 w - - dm 1; id "Bult0170"; c0 "DAWSON, Thomas Rainer Inédit";

Does "dm 1" mean a mate in 1? Or 1 move for the attacked king?

Opcode "dm": direct mate fullmove count

The "dm" opcode is used to indicate the number of fullmoves until
checkmate is to be delivered by the active color for the indicated
position. It always takes a single operand which is a positive integer
giving the fullmove count. For example, a position known to be a "mate
in three" would have an operation of "dm 3;" to indicate this.

This opcode is intended for use with problem sets composed of positions
requiring direct mate answers as solutions.

[Dann Corbit]

I download the collection, extracted the first file, dm001.epd and picked a random position.

8/3N4/n7/pk2N3/1n6/8/PPPPPPPP/2K5 w - - dm 1; id "Bult0170"; c0 "DAWSON, Thomas Rainer Inédit";

[d]8/3N4/n7/pk2N3/1n6/8/PPPPPPPP/2K5 w - - 0 1

Does "dm1" mean a mate in 1? Or 1 move for the attacked king?

Regards
Dave

Here are the DM1 positions with solutions. For some reason, the solutinos did not get put into the output set for dm001.

[d]2b5/1p1pp1pp/1pP5/p7/8/P2PPPP1/RBNkBRpP/QK1nrqrn w - - acn 2; acs 0; bm Bc1#; c0 "VANE BOR Inédit"; ce 32766; dm 1; id "Bult0608"; pv Bc1#;
[d]5K2/2P1P1P1/2Ppkp2/3ppp2/2P1P1P1/8/8/8 w - - acn 10; acs 0; bm c8=B# c8=Q# cxd5# e8=Q# e8=R# exd5# exf5# g8=B# g8=Q# gxf5#; c0 "T. Petroviæc original"; ce 32766; dm 1; id "ZOA1E896"; pv g8=Q#;
[d]5K2/8/2qk4/2nPp3/3r4/6B1/B7/3R4 w - e6 acn 1; acs 0; bm dxe6ep#; ce 32766; dm 1; id "ChestDB.0000"; pv dxe6ep#;
[d]7B/1p1p3p/1R4pP/1P1Pp3/bP6/QP2P3/B1PP1N2/k3K2R w - - acn 2; acs 0; bm Bxe5# Ke2#; c0 "K.Fabel+N.Petrovic; 2001.04.26"; ce 32766; dm 1; id "e1a11507"; pv Bxe5#;
[d]7K/6PB/7k/8/5N2/7N/1p6/b7 w - - acn 1; acs 0; bm g8=N#; c0 "DESPRES, M."; ce 32766; dm 1; id "Bult0051"; pv g8=N#;
[d]7n/BBP2P1P/8/P1PpK3/P5RR/5k2/Pn2NPN1/3Q2b1 w - d6 acn 10; acs 0; bm cxd6ep#; ce 32766; dm 1; id "ChestDB.0001"; pv cxd6ep#;
[d]8/2N3p1/5b2/k1B2P2/pP4R1/8/K1nn4/8 b - b3 acn 1; acs 0; bm axb3ep#; ce 32766; dm 1; id "ChestDB.0002"; pv axb3ep#;
[d]8/2Ppp1pp/2p3P1/8/pp6/P3PPP1/RB1kBRp1/QK1nrqrn w - - acn 2; acs 0; bm Bc1#; c0 "VANE BOR Inédit"; ce 32766; dm 1; id "Bult0607"; pv Bc1#;
[d]8/8/8/4Q3/8/N1K5/pr1P4/k1B5 w - - acn 1; acs 0; bm Bxb2#; c0 "Jurij Antonov; 2000.12.15"; ce 32766; dm 1; id "c3a13122"; pv Bxb2#;
[d]8/p1ppppp1/1p5p/4P3/3B4/4P1P1/PRPP1P1P/k3K2R w - - acn 1; acs 0; bm Ke2#; c0 "DAWSON, Thomas Rainer The Problem 1914"; ce 32766; dm 1; id "breu0690"; pv Ke2#;
[d]b2bknRn/qP1pPrP1/1p1KRpp1/BP1Pp3/1p1p4/4NP2/p7/8 w - - acn 10; acs 0; bm exf8=Q# exf8=R#; c0 "DAWSON, Thomas Rainer Revue Roumaine 05/1933 In Memoriam W. Pauly"; ce 32766; dm 1; id "Bult0609"; pv exf8=Q#;
[d]b6n/2b1p2K/2B1Q1p1/4p1kp/4P3/B3rpP1/3NP3/8 w - - acn 2; acs 0; bm Bxe7#; c0 "LEU, Gheorghe Telescacco 92"; ce 32766; dm 1; id "ts980083"; pv Bxe7#;
[d]rb6/k1p4R/P1P5/PpK5/8/8/8/5B2 w - b6 acn 1; acs 0; bm axb6ep#; ce 32766; dm 1; id "ChestDB.0003"; pv axb6ep#;

[Dann Corbit]

The second position picked at random. This is a mate in 10?

8/4p1p1/1p2P1P1/1P6/B6N/2P3PN/PP5p/K4Rrk w - - dm 10; id "MSE90505"; c0 "CUPPINI, Alessandro 5857 Sakkelet 6.-7.1993 3rd Mention";

If it is supposed to be a mate in ten then there is a bug, because Nf2#.

[d]8/4p1p1/1p2P1P1/1P6/B6N/2P3PN/PP5p/K4Rrk w - - 0 1

Regards
Dave

Here is a mate in 9 that took almost 8 hours to solve:
[d]2r1k3/6pr/p1nBP3/1p3p1p/2q5/2P5/P1R4P/K2Q2R1 w - - acn 3013537185; acs 27873; bm Rxg7; ce 32750; dm 9; id "ChestDB.3753"; pv Rxg7 Qxe6 Rxh7 Ne7 Bxe7 Qg6 Bc5 Qxh7 Qd6 Rxc5 Re2+ Re5 Rxe5+ Kf7 Re7+ Kf8 Qd8#;

[David Dahlem]

I download the collection, extracted the first file, dm001.epd and picked a random position.

8/3N4/n7/pk2N3/1n6/8/PPPPPPPP/2K5 w - - dm 1; id "Bult0170"; c0 "DAWSON, Thomas Rainer Inédit";

Does "dm 1" mean a mate in 1? Or 1 move for the attacked king?

Opcode "dm": direct mate fullmove count

The "dm" opcode is used to indicate the number of fullmoves until
checkmate is to be delivered by the active color for the indicated
position. It always takes a single operand which is a positive integer
giving the fullmove count. For example, a position known to be a "mate
in three" would have an operation of "dm 3;" to indicate this.

This opcode is intended for use with problem sets composed of positions
requiring direct mate answers as solutions.

My point, all the engines i've tried on this "dm1" position call it a "Mate in 2".

Regards
Dave

[Dann Corbit]

It is a mate in 2. Here is the correct result for this position. There are definitely lots of problems in the set that are not correctly specified. Some of them are probably transcription errors and some of them are probably errors in the original problems.

[d]8/3N4/n7/pk2N3/1n6/8/PPPPPPPP/2K5 w - - acn 6; acs 0; bm c4+; c0 "DAWSON, Thomas Rainer InTdit"; ce 32764; dm 2; id "Bult0170"; pv c4+ Ka4 Nb6#;