Copy link to clipboard
Copied
Hi, i am trying to achieve a seemingly simple task using AE 2022 and Trapcode Particular v18. I have a higher goal in mind but am also just curious as to why this is not working as i expect it to… trying to learn more about the inner workings of particular and its logic…
I have a simple composition, HDTV 1080 25, one red BG Solid, one White Solid with a circle mask applied and one solid with Particular on it (starting with standard settings), emitting black Particles.
Emitter Type is Layer, Layer Emitter is set to my White Solid, using Masks.
Velocity, Velocity random, Velocity distribution, and Velocity from Emitter are all Zero. Emitter Size Z is also Zero.
Desired Outcome: After a few seconds in the timeline i want the whole White Circle covered with black Particles, without any white parts showing. I don’t want to set Size any higher than 5, ultimately even lower.
Problem: at 100 Particles/s things work as expected. As i crank the P/s up however, White spots are left on the Circle. Even on highest P/s setting. I would expect the circle to be entirely black eventually.
What baffles me even more is that i rendered both 100P/s and 50000P/s with completely different Random Seeds and compared them. Again at 100P/s as expected, the Particles are emitted in completely different positions. But at 50000P/s the two renders are completely the same after about 3sec.
It is almost as if some kind of static fractal field obscuration layer is applied, where Particles are never emitted, no matter what… what gives?
Then i went ahead and changed the Layer Type to Text/Mask and selected my Circle as the Layer:
Granted, the White Spots are smaller and fewer now but even now the Circle is not completely Black. First i thought that this was due to Particles being emitted randomly, but again, i changed the random seed and compared the two results: First, Particles are emitted in completely different random Places, but at Time 15s the two images are exactly the same, with the same spots left white. I simply don’t understand it and it’s driving me nuts!
So just for the fun of it I made Tests with the Box emitter and the result is the same: regardless of random Seed, there are some spots where Particular just does not want to emit… at random Seed 100000 the picture at 15s is the same as with random seed 5 (also why are particles on the edges of the box less dense??)
Random Seed 100000:
Random Seed 5:
I know that i could achieve my desired goal with Trapcode Form but i want to know why Particular is behaving as it does here… Can Someone please explain to me what is happening? Thanks
This is to be expected. In a computer, random is never fully random and in the given case there is no special provision for perfect coverage. Particular just doesn't know anything about the emitter's source pixels beyond defining its boundaries. This is really nothing new and would even happen in sophisticated simulation environments in 3D programs where you would typically solve it by having inter-particle collision, sticky collisions on the target objects and an overall high viscosity in the s
...Copy link to clipboard
Copied
This is to be expected. In a computer, random is never fully random and in the given case there is no special provision for perfect coverage. Particular just doesn't know anything about the emitter's source pixels beyond defining its boundaries. This is really nothing new and would even happen in sophisticated simulation environments in 3D programs where you would typically solve it by having inter-particle collision, sticky collisions on the target objects and an overall high viscosity in the simulation plus possibly extra options and controllable micro forces for avoiding jitter or "nudging" particles. You can try to set up something similar in Particular using the fluid dynamics physics model most likely, but of course this will bring the simulation speed down and hamper interactivity a lot.
Mylenium
Copy link to clipboard
Copied
Thank you for your detailed answer, Mylenium. I think i can live with that now i know its to be expected but still the fact that with layer emitters only about 2/3 of the layer is covered is, i think, quite ridiculous... I kind of achieved my goal for now with a box emitter set to dynamic form and a layer mask, but because invisible particles are emitted in the transparent areas, this is pretty heavy on the system and also not ideal for aux systems... hope they change how layer emitters work, it has to be doable..
Find more inspiration, events, and resources on the new Adobe Community
Explore Now