Here we sample the radius as random(). Turns out when you do that the points are flocking to the center! This basically happens, since the regions in the parametrization cover differently sized areas in the target region (circle). This is an important issue when you do random sampling.

Since I found out Liascript is now all in on GenAI, I thought about updating some code. Here is a visualization of sampling points in a circle. The colors encode the density of all points in each region. Here, we sample the radius as a sqrt(random()). Otherwise it looks like the next video!

More playing around with my fluid simulation. Added very simple animation loops. Collision with moving bodies could use some work, but is ok for smaller speeds and not too many particles. I'm having fun looking at it lol

Want some tea? Next up will be particle spawners, some force fields and some very basic automatic movement of colliders. That might be it though, I am now in the just looking at the simulation and being content phase

Added fluid dynamics! :) It does take a bit of parameter fiddling to keep it stable once you add >100k particles, but that is fine for now. Also helpful to not plug in constants with the wrong unit that are off by a factor of 1000, which I definitely did not do, nope

Another scene (programmer art!) with some non-transparent particles rolling down a tube. Please don't judge my menus... this is for fun! Also recordings made it hickup a bit. Next up, the particles will interact :)

Haven't done much recreational graphics programming lately! Saw some particle systems, which got me thinking of sticking together some things I haven't put together before! :) Step 1: Basic GPU particle system. Step 2: Collision with environment (not each other, something for that will come next)