Does Neighbor Count support limiting the number of particles found per frame during i

[cgdvs]

Hi,Tyson.
  I frequently run into a major pain point when using the Neighbor Count attribute test. With a fixed number of particles, I want to adjust how fast the search animation propagates, in other words, the speed of particle spread. When particles are stationary, adjusting the search radius cannot control this propagation speed. If the radius is too small, no particles will be detected at all.
Therefore, I really hope there can be a filter option to limit the number of particles found per search. Even with a large search radius, this filter could restrict the output to only two selected particles per frame into the new event. This would let me precisely control the propagation speed of the Neighbor Count effect. What do you think?

[tyFlow]

Can't you just use the timing rollout to do that? Ex: set the nth value to something like 2,3,4,etc...so that propagation doesn't happen each frame, thus slowing it down?

[cgdvs]

Can't you just use the timing rollout to do that? Ex: set the nth value to something like 2,3,4,etc...so that propagation doesn't happen each frame, thus slowing it down?

I want to control the propagation speed of the spread effect. This is different from making the operation run every few frames. Running it at frame intervals causes the animation to stutter, and even with variation values, the result still looks unnatural. What I need is smooth, freely adjustable speed to make the spread faster or slower. Limiting the number of particles processed each frame is only my initial idea, and I would be glad to hear if you have a better solution.
I previously achieved this using the Retimer operator: I increased the total particle count, reduced the search radius, and then used Retimer to control the propagation speed. The drawback is that all other particle effects have to be created in a separate tyFlow setup.
What I ultimately want is for the particle search in Neighbor Count to finish within a specified time range. It could be completed within 10 frames, or over 100 frames. This duration needs to be adjusted to be faster or slower based on camera and director requirements. I encounter this requirement very often, so I bring it up to you again.

[tyFlow]

I see - you could instead add a filter like "Random 0-1" and set the value to something like .1 - then enable "seed by time". That will effectively cause only a random selection of 10% of particles to be processed each frame. By adjusting the value up/down between 0-1, you can control how large a substep of particles will be processed, thereby speeding or slowing the propagation.

[cgdvs]

I see - you could instead add a filter like "Random 0-1" and set the value to something like .1 - then enable "seed by time". That will effectively cause only a random selection of 10% of particles to be processed each frame. By adjusting the value up/down between 0-1, you can control how large a substep of particles will be processed, thereby speeding or slowing the propagation.

That sounds like a theoretically good approach! When I first considered it, I thought a percentage filter might not work well. I didn’t think about using seed by time back then. With seed by time added, the result is somewhat similar. However, since it relies on random seeding, there will always be some particles that slip through. Still, this works as a helpful workaround. Thanks for the great idea.