next »

[nova++]

So, I'll try and boil this down to be as simple as possible. I am trying to place random points upon a uniform set of triangles in 3D space. Specifically on the triangles; per-vertex isn't good enough for what I need.

So far, my attempts have resulted in a considerable bias toward the center of the triangle, and I am wondering what other methods there are that might work better.

This is the best I've gotten so far. The triangle's points are A, B, and C, the random values (0 to 1) are R1, R2, and R3.

P = (R1 * A + R2 * B + R3 * C) / (R1 + R2 + R3)

Which looks like this:



So... yeah. I've been sniffing around but I haven't found anything thus far that does what I need

...It's also entirely possible I've screwed up some of my randomization, so I need to look into that angle soon as well. That would be just my luck. <- tests complete, it's definitely the scattering algorithm, not the randomization

[BorisTheBrave]

Here's my code for doing it:

https://github.com/BorisTheBrave/blue-noise-particles/blob/master/blue_noise_particles.py#L190

Edit: This way looks neat, too: https://math.stackexchange.com/questions/18686/uniform-random-point-in-triangle

[nova++]

Oh man, thanks. Barycentric coordinates, that was exactly the seed I needed to find this: https://adamswaab.wordpress.com/2009/12/11/random-point-in-a-triangle-barycentric-coordinates/

Which looks like:



Now it's perfect!