| |
Falloff 0.0 to 9.8
Filling a Circle (2D)
Get a random xy point in a unit square.
Find distance to center.
If distance greater than 1.0 try again.
This will avoid processing
points that are in the "corners" of the unit square.
Calculate a target radius,
target_rad = pow( 1 - distance, falloff)
where falloff are values such as,
0.0 uniform radial filling of the unit circle,
1.0 linear radial gradient,
2.0 quadratic radial gradient.
Calculate a random number 0 to 1 (probability).
Only if the probability is less than target_rad is the
xy point accepted.
Listing 1 (python code)
|
import random, math
width = 0.004
def scatter2d(falloff, num, frame):
accepted = 0
fname = '/Users/malcolm/distribution2d.%d.rib' % frame
f = open(fname, 'w')
f.write('Points "P" [\n')
while accepted < num:
x = random.uniform(-1.0, 1.0)
y = random.uniform(-1.0, 1.0)
distance = math.sqrt(x * x + y * y)
if distance > 1:
continue
target_rad = math.pow( (1.0 - distance), falloff)
probability = random.random()
if(probability <= target_rad): # and math.fabs(y) <= target_rad):
f.write('%1.3f %1.3f 0 \n' % (x,y))
accepted += 1
f.write('] "constantwidth" [%1.3f]\n' % width)
f.close()
return accepted
incr = 0.2
for j in range(50):
random.seed(1)
falloff = j * incr
scatter2d(falloff, 40000, j)
print falloff
|
|