Sverchok loop particles paths

Support Particles and Physics Simulations
, ,
1

Hi forum, I’m writing for help on how to loop a particle system’s particle paths. I have a simple particle system confined in a space that has geometry for collision. Basically a particle system inside a cube. None of the particles die in this setup. The problem I am trying to solve is of how to loop the particles paths. Each particle’s starting position must match it’s final position. I have a working node setup with Sverchok where the particle’s positions are saved to a multi cache node then this data is used to make a cyclic polyline. I then create new geometry that follows each polyline’s curve frame. The curve frame parameter is being driven by the frame info node. However, for some reason that I still don’t understand, the computation of each curve’s changing frame is very slow, lagging and choppy. Instead of the frame animating smoothly along the curve it jumps from one position to the next. I also get an intermittent error on the curve frame node saying “curve has zero curvature at some points; impossible to calculate correct frenet frames for such points”. This error doesn’t happen when the timeline is playing. It only happens when the animation is paused and I jump around to different frames in the timeline. I have a stackexchange question with more info on the current setup: https://blender.stackexchange.com/questions/281242/how-to-loop-a-confined-particle-system-setup-up-with-collisions-and-zero-deaths
Here is my blend file:
density.blend (2.3 MB)

Thanks for the help.

2

I’ve done this with Geometry Nodes instead of using particles, and have created a ‘particle’ type setup for geometry nodes, and this allows for seamless loops of “particles”.

Let me look into your specific scenario a little more, but I think this could work for you. Looking into the stackexchange link, then I’ll be back with more info for ya :+1:

Edit:
I see some fundamental issues with the concept of what you are trying to do. I think it’ll always look wonky for a systems points to return points to their original positions Unless particles are orbiting or have a simple ‘somewhere+back’ type movement. I could do it without the collision with my setup no problem, but I don’t have an answer for with collision