I call the program: 'Son of Maniac'. It is so called because it was inspired by a program written in 1956 by Kister, Stein, Ulam, Walden and Wells at the Los Alamos Scientific Laboratory in New Mexico. Their program played a miniature version of chess on a 6x6 board. There were no bishops, no castling and the pawns could only move one square on their first move. The program ran on a MANIAC computer with a speed of 11000 operations per second. This enabled the program to perform an exhaustive search to a depth of 4-ply in an average of 12 minutes.
The only recorded game (that I know of) played by the program was against a young lady member of the laboratory who was taught how to play chess a week earlier. The game went as follows:
White: MANIAC
Black: Human
1 d2-d3 b5-b4
2 Ne1-f3 d5-d4
3 b2-b3 e5-e4
4 Nf3-e1 a5-a4
5 b3xa4? Nb6xa4
6 Kd1-d2? Na4-c3
7 Nb1xc3 b4xc3+
8 Kd2-d1 f5-f4
9 a2-a3 Ra6-b6
10 a3-a4 Rb6-a6
11 a4-a5 ...
12 Qc1-a3 Qc6-b5
13 Qa3-a2+ Kd5-e5
14 Ra1-b1? Ra6xa5
15 Rb1xb5 Ra5xa2
16 Rb5-b1 Ra2-a5
17 f2-f3 Ra5-a4
18 f3xe4 c5-c4
19 Ne1-f3+ Ke5-d6
20 e4-e5+ Kd6-d5
21 e5xf6=Q Ne6-c5
22 Qf6xd4+ Kd5-c6
23 Nf3-e5#
So far I've just created an initial GUI which will obviously change as the program evolves(see image below).
By clicking on pieces and squares I can move the pieces on the board. It recognises pawn promotion and offers the choice of piece to promote to. Next I need to devise methods whereby the program can generate a list of its legal moves. From then on it will probably just get harder.
No comments:
Post a Comment