Alexandre B A Villares
Preparing a bit for teaching my #CellularAutomata class a few times this semester.#WIP #ConwaysGameOfLife with #numpy
Find the sketch-a-day archives and tip jar at: https://abav.lugaralgum.com/sketch-a-day
Code for this sketch at: https://github.com/villares/sketch-a-day/tree/main/2026/sketch_2026_02_28 #Processing #Python #py5 #CreativeCoding
Feb 28 at 8:13pmWeb
Show threadLorraine LeeFeb 28
@villares Huh, reminiscent of GameOfMinesweeper.
Show threadLorraine LeeFeb 28

@villares Google Scholar never disappoints

scholar.google.com/scholar?q=cā€¦

Google Scholar

Show threadAlexandre B A VillaresFeb 28
@lori ha! great!
Show threadAlexandre B A Villares šŸFeb 28

@rzeta0 my friend, what do you think about the code using np.roll() (in the linked sketch)?

I'm afraid it would be a bit hard to explain on a short class. So I'll use the simpler inefficient function...

The next big step in efficiency would be using #scipy I think:

def count_live_neighbors(status):
"""Counts the number of neighboring live cells"""
kernel = np.array([
[1, 1, 1],
[1, 0, 1],
[1, 1, 1]
])
return scipy.signal.convolve2d(status, kernel, mode='same', boundary="wrap")
Show threadTariqFeb 28

@villares

Hi - I few times I have done something like this I have just avoided the "rolling around the edge" issue by just having a larger array and only plotting an internal smaller window.

but your method of adding up the "rolled" method clever and in my opinion simpler than the "kernel" methods that I did once try.

my guess is that some students will prefer the rolling method, and some might prefer the "kernel" convolve2d method.

Also - inefficient doesn't matter too much I think when teaching the main ideas .. .optimisation can come later after the students understand the idea.

Thank you for sharing this - I hadn't considered this use of the roll() method ! Very cool!

Show threadTariqFeb 28

@villares

looking at your code a second time, I think I would do this

1. explain the simplest (slowest) method so the key idea of neighbour count is understood

2. only then explain some optimisations eg roll() or convolve2d

this is just my opinion

Show threadAlexandre B A Villares šŸFeb 28

@rzeta0 yeah, I'll do that.

Even then, the wrapping with the % operator is always a bit of a time sink to explain I'm afraid... I might do the "board shrinking" thing subtracting rows and columns from the loop at the borders.

Show threadAlexandre B A Villares šŸFeb 28
@rzeta0 I'm thinking about teaching the one where I use a function that loops 8 neighbors for each cell (commented out on the sketch) and then maybe showing/linking to the others (roll and scipy)...
Show threadTariqFeb 28

@villares

that makes sense to me

you can offer the roll/scipy as something optional for students to look into if they wish