Skip to content

Latest commit

 

History

History
50 lines (36 loc) · 3.65 KB

README.md

File metadata and controls

50 lines (36 loc) · 3.65 KB

Rust libraries and programs for solving impartial games and calculating their nimbers, developed by Piotr Beling. Currently, only the normal play convention is supported, but support for misère games is planned.

Included libraries:

Included programs:

  • igsolve (crate, doc) - a console-based application for calculating nimbers with igs;
  • ogsolve (crate, doc) - a console-based application for calculating nimbers with ogs.

Installation

Programs can be compiled and installed from sources. To do this, a Rust compiler is needed. The easiest way to obtain the compiler along with other necessary tools (like cargo) is to use rustup.

Once Rust is installed, to compile and install a program with native optimizations, just execute:

RUSTFLAGS="-C target-cpu=native" cargo install <program_name>

for example

RUSTFLAGS="-C target-cpu=native" cargo install igsolve

The results obtained so far

Here I provide the nimbers calculated using my software.

Cram (under the normal play convention)

4 5 6 7 8 9 10 11
4 0 2 0 3 0 1 0 1
5 - 0 2 1 1 1 2 0
6 - - 0 5 0 1 0 ?
7 - - - 1 3 1 ? ?

The table shows the nimbers of initial cram positions for different board sizes computed so far. Note that the symmetry strategy implies that even-by-even boards are losing and therefore of nimber 0 (the second player can win by responding with moves symmetrical to the center of the board).

To the best of my knowledge, the bold values (for the largest boards: $9 \times 7$, $8 \times 7$, $9 \times 6$, $10 \times 5$, $11 \times 4$) were calculated by me and published here for the first time.

The nimbers of the smaller boards were earlier computed by Glop, which is the solver developed by Lemoine and Viennot. Nimbers of most boards no larger than $5 \times 7$ were first given by Martin Schneider in his master's thesis entitled Das spiel juvavum in 2009. Uiterwijk reported that the $11 \times 5$ board is losing (and thus of nimber 0) in his paper Solving Cram Using Combinatorial Game Theory (Advances in Computer Games: 16th International Conference, ACG 2019, Macao, China, August 11–13, 2019). I have verified the correctness of all the nimbers in the table with my solver.

Publications and citations

When using my software for research purposes, please cite the following paper which details the key algorithms used: