I have a basic misunderstand with particleGPU node. My goal is to have a particle emission on top of a movie file (position and color connected to the source).
I try to input a movie file directly into particlesGPU node (1 : particle source 2 : particle color)
Color a represented but not in place.
particlesgpu is a 3d system, so i assume it needs 3d info. try putting an image in a point transform and put that into the source and sourcecolor inputs of the particlegpu.
i’m not sure how this would work with a movie. you can create your own pointcloud perhaps by feeding images into reorder and that into a point transform.
while testing you might want to make your particle system very basic ie a circle or square with no forces. try making it smaller as well.
ii would also try having a separate point cloud for position and color. if you want a grid of particles, i think you can make one with gradient ramps on x and y piped into the red and green , and just white into blue.all into point transform. this is off the top of my head, but give it a try.
Here is the workflow : Particles
Create a rectangle sprinkle point sopto then chopto connect to the particlesGPU input source Color
moviefilein into a fit to make it square (like UV) Color to Particles
use a limit to normalize the point position
connect the limit to the second input of the remap and the color fit into the first input.
Now everything is related to the pixel space. The chopto and the fit need the have the same resolution.
Then i connect the resolution to the spinkle with this expression in the Number of point : op(‘fit1’).par.resolutionw * op(‘fit1’).par.resolutionh.
That is almost working but… If go more than 512x512 the FPS drop down.
So i think it is not the good way of computing point.
