1. The base code comes with a simple life form, the Mycoplasma. Your first task is to make it more interesting. Modify its rule set so that [it follows Game of Life rules].
2. Using the Mycoplasma as an example, develop at least two new forms of life. Their rule set should be different from the Mycoplasma.
Two new types of cells have been created: Isseria and Helicobacter. Mycoplasma's default colour is orange, Isseria's default colour is magenta and Helicobacter's default colour is cyan.
Helicobacter follows these rules:
- If the cell has exactly one live neighbour, it will die
- If the cell has exactly three live neighbours, it will live on to the next generation
- If the cell has more than three live neighbours, it will die due to overcrowding
- Lastly, if the cell is dead, it will come alive if it has exactly three live neighbours
Isseria follows these rules:
- If the cell has exactly two live neighbours, it has a chance of living on to the next generation
- If the cell has exactly four live neighbours, it has a chance of living on to the next generation
- Otherwise, it will die
When cells are created, they have a field which points to an enum referring to their cell species. This is used to keep track of the state of the cell without replacing the object. Enums are used for type safety as set values are defined, so that there is no room to type an invalid species (like would be the case if a String field is used to keep track of the species). It also reduces coupling as another solution would be checking the colour of the cell instead of a species field, however, if the colour is changed, this would need to be changed in every if-statement. Changes are now able to be made to the attributes without affecting other parts of the code.
– At least 1 life form should incorporate colour in its rule set, i.e., its colour may change while alive – At least 1 life form should exhibit different behaviours as time progresses
The colour of the Helicobacter cell is dependent on the current generation. A method darkenHeliColour(generation) is used to decrease the RGB value of the Helicobacter's colour, based on the current generation. This gives a slow gradient of Helicobacter cells becoming darker and darker as the simulation goes on. This darkened colour is used to signify a change in behaviour - the Helicobacter's increased resistance to infected cells. As the colour darkens, the probability that a Helicobacter cell gets infected when surrounded by infected cells decreases, simulating increasing immunity to infection as time progresses.
1. Simulation GUI Controls
To the GUI, five components have been added:
- The ability to draw cells to the field
When the GUI is first opened, the field is completely empty. At the bottom of the panel is a JComboBox where you can choose which type of cell to draw onto the field. After selecting a cell, you are able to click and drag the cell, manually populating the field. To draw cells to the screen, the FieldView (inside SimulatorView) implements a MouseMotionListener. This is used to get the position of the mouse. Once a user clicks, the mouse position is converted to a Location object which refers to the position on the field to fill in. The cell at this location's colour is then changed to the species specified.
The components below cannot be interacted with unless the Simulator is ran with a specified number of generations (you must call runLongSimulation() or simulate(generation)).
- The ability to change the playback speed
A slider at the bottom of the GUI has been added to change how fast the simulation is running. This is implemented by using the delay() method. The slider gives a value between 0 and 100. This is converted to a percentage and multiplied by 300 ms. For example, if the delay slider is set all the way to the left, there will be a 300 ms delay, but if the delay slider is set all the way to the right, there will be no delay. All slider positions in between are a percentage of 300 ms.
- Toggle simulation button
This button toggles whether the simulation is running or not. The text on the button changes depending on the state of the simulation. If the simulation has not been started yet, the text is set to "Start". If the simulation is running, the text is set to "Pause". If the simulation has been paused, the text is set to "Resume". The changing of the state of the simulation is implemented using a paused field. If the paused field is false, the simulation runs as usual. However, if paused is true, the simOneGeneration() method which progresses the simulation is not called in the simulate() method.
- Reset simulation button
This button resets both the simulation and the simulation controls pane. To reset the simulation, the generation is set to 0, and all cells on the field are set to dead (clearing the field). The simulations control pane is reset by changing the text of the toggle button to "Start" and resetting the value of the playback speed slider to 50%.
- Populate field button
This button acts as a special type of reset button. The board is reset, but the entire board is filled instead of made empty. If the user does not wish to draw the initial conditions of the field, the populate field button fills the entire board randomly with all types of cells. This is best used to see the interactions between the cells as the entire board is filled.
2. Non-deterministic cells
The rules that the Isseria cells follow are triggered based on a probability. Given the rule "If the cell has exactly two live neighbours, it has a chance of living on to the next generation", there is a 60% chance that this rule will execute. Furthermore, given the rule "If the cell has exactly four live neighbours, it has a chance of living on to the next generation", there is a 50% chance that this rule will execute. Non-determinism is implemented by using the Random.getNextDouble() method where an if-statement ensures that the next line of code is changed only if below the desired decimal value (e.g. Random.getNextDouble() < 0.6 for the former rule).
Non-determinism is also present in the symbiosis and disease challenge tasks. They are implemented in the same way as described above and are used to provide a probability of changing to a new cell, for the disease to spread, etc. Whenever non-determinism is implemented and implicitly mentioned in the rest of the report, (non-determinism) will be present next to the explanation to make it more explicit.
3. Symbiosis
There is a parasitic relationship between Helicobacter and the other two types of cells, Isseria and Mycoplasma, where Helicobacter benefits, and all others are harmed.
If an Isseria or Mycoplasma cell has between one and three neighbours that are Helicobacter, there is a probability (non-determinism) that the cell will change its colour to the Helicobacter colour as well as the Species enum to HELICOBACTER, and thus become Helicobacter cell (no object replacement). This state change to becoming a Helicobacter cell is solely determined by the number of Helicobacter neighbours, which causes it to change its colour to the Helicobacter colour and its Species enum to HELICOBACTER..
There is also a mutualistic relationship between Mycoplasma and Isseria cells.
If a dead Mycoplasma cell has at least one living Mycoplasma neighbour, and at least one living Isseria neighbour, the dead cell will come alive in the next generation as an Isseria cell. This simulates the "breeding" of neighbouring Mycoplasma and Isseria cells. This is implemented by getting neighbours of the current cell. If there is at least one Mycoplasma neighbour and at least one Isseria neighbour, the cell being processed changes its colour and enum, so thereby its state (no object replacement). This benefits Mycoplasma as its dead cells can come alive, and it benefits Isseria as the dead cell comes alive as an Isseria cell.
However, there is a chance (non-determinism) that when the two cells "breed", the new cell comes alive, but as an infected. This simulates a sexually transmitted disease between the cells. This is harmful to all life forms as the infected cell can infect any cell.
4. Disease
As mentioned before, when a Mycoplasma and Isseria cell are the neighbours of a dead Mycoplasma cell, there is a chance that the dead Mycoplasma changes to an infected state. Once the cell has an infected state, in addition to turning red and its enum updating to INFECTED, its behaviour is also changed by no longer being able to breed, and for other cells to be infected if it has a certain number of infected neighbours. This is achieved by the cell getting all of its living neighbours and filtering by which are infected. If it has at least one infected neighbour, there is a chance (non-determinism) that cell's state also becomes infected. This simulates a highly contagious infection, where only one neighbour needs to be infected. Dead cells can also have a chance (non-determinism) to come alive as an infected cell if three of its neighbours are infected.
Infected Mycoplasma cells are made to be even more contagious as an extra rule is added for infected Mycoplasma cells. Mycoplasma cells live on to the next generation if they have exactly two or three neighbours, as described in the rules mentioned before. However, if they are infected, they have a chance (non-determinism) to live on to the next generation no matter how many neighbours they have. This probability of living on is dependent on the current generation (more infectious at smaller generations when the infection is new, and less infectious as time progresses).