Hello:
towforce wrote: ↑Thu Mar 28, 2024 5:27 pmYes: I've only played through this game quickly - but at a glance, white's moves look a bit rubbish.
[...]
A
proof game does not need to make sense, it only proves that a given position can be reached legally from the starting position.
I tried
Natch with success. There were not any solutions after 23 plies (12 moves by white and 11 moves by black) after 5.79 seconds of computation; but it found 3,022,086 solutions after 25 plies (13 moves by white and 12 moves by black)!
problem.txt wrote:r2qk2r/pp3ppp/4p3/5b2/1Pp1n3/1P1pPQ2/5PPP/RN1K1B1R
25 single moves
Please note the lack of castling flags in the input. The original position has full castling rights for the black side. Nonetheless, many solutions —if not all— do not waste time moving the black rooks or the black king, thus keeping the castling rights.
Code: Select all
.\Natch-3.3>Natch --hash-size 512 --language e --output solution.txt problem.txt
Solutions found: 3022086
Positions analysed: 143
Resolution time: 42.39 s.
The output is a TXT of 568 MB:
Code: Select all
Natch 3.3 Copyright (C) 1997,98,99,2001-2005,2016,2019-2020 Pascal Wassong
+---+---+---+---+---+---+---+---+
|*R | . | |*Q |*K | . | |*R |
+---+---+---+---+---+---+---+---+
|*P |*P | . | | . |*P |*P |*P |
+---+---+---+---+---+---+---+---+
| | . | | . |*P | . | | . |
+---+---+---+---+---+---+---+---+
| . | | . | | . |*B | . | |
+---+---+---+---+---+---+---+---+
| | P |*P | . |*S | . | | . |
+---+---+---+---+---+---+---+---+
| . | P | . |*P | P | Q | . | |
+---+---+---+---+---+---+---+---+
| | . | | . | | P | P | P |
+---+---+---+---+---+---+---+---+
| R | S | . | K | . | B | . | R |
+---+---+---+---+---+---+---+---+
r2qk2r/pp3ppp/4p3/5b2/1Pp1s3/1P1pPQ2/5PPP/RS1K1B1R
25 single moves. 12 + 14
The resolution begins at Fri Mar 29 10:42:38 2024
Solution 1 (position 1)
1.Sg1-f3 Sg8-f6 2.Sf3-g5 Pc7-c5 3.Sg5-e4 Pd7-d5
4.Pa2-a3 Sb8-d7 5.Pc2-c4 Pd5xc4 6.Pd2-d4 Pc5xd4
7.Bc1-g5 Pd4-d3 8.Bg5xf6 Sd7xf6 9.Pb2-b3 Sf6xe4
10.Pe2-e3 Bc8-f5 11.Qd1-f3 Pe7-e6 12.Ke1-d1 Bf8-b4
13.Pa3xb4
Solution 2 (position 1)
1.Sg1-f3 Sg8-f6 2.Sf3-g5 Pc7-c5 3.Sg5-e4 Pd7-d5
4.Pa2-a3 Sb8-d7 5.Pc2-c4 Pd5xc4 6.Pd2-d4 Pc5xd4
7.Bc1-g5 Pd4-d3 8.Bg5xf6 Sd7xf6 9.Pb2-b3 Bc8-f5
10.Pe2-e3 Sf6xe4 11.Qd1-f3 Pe7-e6 12.Ke1-d1 Bf8-b4
13.Pa3xb4
[...]
So I post a random proof game, which should be one of the many shortest proof games, for example the first one of the output file:
[pgn][Event "Shortest proof game"]
[Round "Solution 1 (position 1)"]
[Result "*"]
1. Nf3 Nf6 2. Ng5 c5 3. Ne4 d5 4. a3 Nbd7 5. c4 dxc4 6. d4 cxd4 7. Bg5 d3 8. Bxf6 Nxf6 9. b3 Nxe4 10. e3 Bf5 11. Qf3 e6 12. Kd1 Bb4 13. axb4 *[/pgn]
Homework: find the most reasonable, senseful game/s among the many solutions.
------------
Paloma wrote: ↑Wed Mar 27, 2024 9:18 pm
Ajedrecista,
your pgn has no header.
I think I found the reason of the lack of buttons: I started writing headers by hand, but since I am not used to, I wrote something like
[Event=""],
[Result="*"] and so on, with PGN viewer not working correctly, so I stripped the headers, resulting in the lack of buttons (
is this the expected response from the lack of PGN headers?). My first fault was to type equal signs in the headers just before the quotation marks, instead of spaces. The second one was to get rid off the headers.
It is amazing how films can anticipate things: there is one western called
Hang 'Em High (1968), whose title was translated in Spain into
Cometieron dos errores, that literally means 'They made two mistakes', that exactly applies to me in this case.
Regards from Spain.
Ajedrecista.
