life

(note the simulation was limited to 100 iteration for the sake of the .gif filesize)

Six years ago (September 2011), I needed a science fair topic. Having recently been introduced to and fascinated by John Conway's Game of Life, I formulated a hypothesis along the lines of "There exists an optimal initial population density to maximize generations until stabilization in John Conway's Game of Life." I decided to revisit this project now. You can read more about The Game of Life here, or use a simple online simulator here. Different branches off this idea and implementation decisions are given their own file, detailed below.

ncurses.py

This was my first implementation in the revival of this project. It runs a simulation of The Game of Life and displays each iteration on screen. The grid size is a constant which may be adjusted, as is the character which is printed for "living" cells. The time each iteration is shown before the next is specified in seconds by DELAY. You can also set a timeout in the form of MAX_ITERATIONS. This program uses the curses library to draw over the screen instead of printing many, many lines. Be sure your terminal is at least SIZE+2 lines high. When the simulation completes you may press Enter to run another, or any other key to quit.

small.py

This is a trimmed down version of the above. It cuts out all of the UI code (besides a single print statement) to focus on the clarity and readability of the core logic. make2DList is a helper method which accept arguments for the height and width of the 2D list (or array, if you prefer) to create, and a function to generate the value for each element. This last parameter is optional, by default it will simple result in an array full of False. iterate accepts a game state in the form of a 2D list and returns the following generation. simulate will, given a grid size and percent of cells to be initially alive, randomly generate and simulate a game of life. Note this fraction should be between 0 and 1, not 0 to 100. The function will return False if the population was eradicated, or True if it stabilized in a cycle of repeated iterations. If executed as __main__, it will run a hundred-thousand trials on each grid size 5x5 through 9x9, and each starting population density 0.1 through 0.9, and print the percentage of simulations that stabilized (did not die out).

ninty_percent.py

This program focuses on the question of "Do any simulations actually survive at 90% [initial population density]?" Looking at the previous programs I almost always saw a 0% survival at 90% of cells initially populated. (Note 100% is a trivial case as all cells will die of overcrowding in one generation.) This program will run through simulations at grid sizes 5x5 by 9x9 with a 90% starting percentage of cells alive. Once it completes running ten-thousand trials at each grid size, it uses matplotlib to display a bar chart with the percentage of simulations that survived on each grid size. It took about twenty seconds to run the fifty-thousand simulations on my computer.

science.py

I wrote this program really because I was thinking of ways to visualize data related to this project and just wanted to make a 3D bar chart, because it looks cool.

tkinter.py

I was having some trouble with matplotlib on Windows; so, knowing Tk is cross-platform, I wanted to use the tkinter module included in Python to visualize the simulations.

update.py

The aim of the program is to continually run more simulations and collect more data to move the bars to more accurate positions.