magnet-automaton is a simulation of a model of a magnet. Physical justification TBC.
The model we are using is a two-dimensional grid of 'cells'. Each cell has a positive or negative 'spin', which supposedly is relevant to magnets. At certain times, a cell is chosen from this grid to be updated according to our update rule. The update rule is as follows:
Let n be the number of 'orthogonal neighbours' (those directly north, east, south, west) to this cell. Typically n = 4, but could be 2 or 3 for corners and edges respectively.
Let d be the number of differing polarities among the cell's neighbours. So, for example, if a positive cell is surrounded entirely by negative cells, d = 4.
If d > n/2, flip the polarity of the cell.
As of now, we will be running a model where a cell is selected at random from all of the cells in the grid and updated according to the above rule. We are interested in both visualising the evolution of a randomly-initialised grid, and potentially statistically analysing its long-term behaviour. A couple of examples of the latter: will grids always reach a 'stable' state, where no further updates can be made? How many random updates will have to be done before such a stable state is achieved, as a function of grid size? These questions should be easy to investigate with an efficient implementation of this model.
Run the following in the root of the codebase in order to run the unit tests:
make tests
bin/testerThe project founders may contrbute (ideally addressing issues) by branching (typically from master) and making a pull request. All other contributors will need to fork the repository and make a pull request from the fork. This pull request MUST resolve an issue and be thoroughly reviewed before being merged in.
The indentation style for this project is non-negotiably Allman.