#numpy feels a bit like another, harder, language to learn beyond #python :(
A first attempt at a "sand simulator" but it seems too hard...
#py5 #Processing #creativeCoding
import numpy as np
import py5
def setup():
global grid
py5.size(500, 500)
grid = np.zeros((py5.width, py5.height), dtype=bool)
def draw():
py5.set_np_pixels(grid * 255, 'L')
update_grid()
def update_grid():
for i in reversed(range(grid.shape[0]-1)):
row, next_row = grid[i], grid[i+1]
sand = row == 1
next_row_empty = next_row == 0
falling = sand & next_row_empty
row[falling] = 0
next_row[falling] = 1
def mouse_dragged():
x = py5.mouse_x % py5.width
y = py5.mouse_y % py5.height
if py5.is_key_pressed:
grid[y-2:y+3,
x-2:x+3] = 1
else:
#grid[y, x] = 1
grid[y][x] = 1
py5.run_sketch()
sketch_2022_05_20 (see next toot for the most recent version)
Naive sand simulator… #Processing #Python #numpy #py5
import numpy as np
import py5
def setup():
global grid
py5.size(500, 500)
grid = np.zeros((py5.width, py5.height), dtype=bool)
def draw():
py5.set_np_pixels(grid * 128, 'L')
update_grid()
def update_grid():
for i in reversed(range(grid.shape[0]-1)):
row, next_row = grid[i], grid[i+1]
sand = row == 1
next_row_empty = next_row == 0
falling = sand & next_row_empty
row[falling] = 0
next_row[falling] = 1
sand_right = row[1:] == 1
next_left_empty = next_row[:-1] == 0
falling_left = sand_right & next_left_empty
row[1:][falling_left] = 0
next_row[:-1][falling_left] = 1
sand_left = row[:-1] == 1
next_right_empty = next_row[1:] == 0
falling_right = sand_left & next_right_empty
row[:-1][falling_right] = 0
next_row[1:][falling_right] = 1
def key_pressed():
if py5.key == ' ':
grid[:] = np.zeros((py5.width, py5.height), dtype=bool)
def mouse_dragged():
x = py5.mouse_x % py5.width
y = py5.mouse_y % py5.height
if py5.is_key_pressed:
grid[y-4:y+5,x-4:x+5] = 1
else:
grid[y,x] = 1
py5.run_sketch()
sketch_2022_05_20 update!
I couldn't resist fiddling a bit more, and it came out more realistic!
Code at https://abav.lugaralgum.com/sketch-a-day