LÖVE

Poisson disc sampling, an algorithm for generating tightly-packed points which are all some minimum distance from one another.

Based on: this video - the first few minutes are also a good description.

A visualization of the selection process with some color added based on the 'age' the point was picked. and the actual points table displayed as... points

This is pretty useful stuff!

Thanks! It started out as a simple starfield generator, but the more 'knobs' I add and 'dials' I turn, I'm seeing a lot of possibilities here. Since I really need to get around to learning shaders, I think the first project will be something that takes in the points and outputs 'nebula' type shading based on point density/age.

I like this explanation:

https://youtu.be/flQgnCUxHlw

My implementation (move mouse to create new circles)
press space to switch the view
press S to reset the field
press G to hide he grid
https://github.com/darkfrei/love2d-lua- ... c-sampling

https://www.youtube.com/watch?v=PlQuPHga82Q

Situation, if the Poisson Disc Sampling allows just one connection to the old points.
https://youtu.be/cqcDK5IPvhM



https://github.com/darkfrei/love2d-lua- ... c-sampling

@darkfrei
Your implementation doesn't look quite as random and evenly distributed as the original. I see more patterns and lines among the samples.

ReFreezed wrote: ↑Wed Oct 19, 2022 1:34 pm @darkfrei
Your implementation doesn't look quite as random and evenly distributed as the original. I see more patterns and lines among the samples.

You are right, it's not the same: in the original PDS the generated point will be chosen randomly, but here the generated point will be chosen near the mouse position is only.
But you can simulate the original simulation if you move mouse from starting point to the outside.

Points: Connections: https://youtu.be/SdPaPliAi7s