3
$\begingroup$

I am attempting to replicate blenders wave modifier using geometry nodes but have run into a hurdle trying to implement damping(fall off), due to a limited understanding of maths.

enter image description here

I'm combining these two techniques to try and achieve this:

Control falloff geometry node setup

https://www.reddit.com/r/blender/comments/nm8yfc/geometry_node_setup_for_procedural_wave_animation/

This is my current attempt: enter image description here

The wave is working but the frame labeled "Not Sure" is the attempt at damping and it essentially does nothing.

EDIT:

After some experiments I've managed to get the damping to work as intended but it does not follow the empty. enter image description here enter image description here enter image description here

Any advise on where I've gone wrong in my graph, would be greatly appreciated as I've sunk quite a few hours into this.

Would it help if I shared the .blend?

Thank you!

Improve this question
$\endgroup$

1 Answer 1

Reset to default
5
$\begingroup$

The function, as graphed, is of t, which is [distance-to-0], (along the X axis). So we just have to substitute [distance-to-Empty] for t, giving ourselves the option of scaling and/or shifting it along the way:

enter image description here

You may want different parameters, but I think these cover all options,with the least interaction between them..

enter image description here

Improve this answer
$\endgroup$
8
  • 1
    $\begingroup$ Thanks so much for your time! These parameters should be perfect for what I'm trying to achieve. I will spend some time over the next few days trying to understand the maths behind it. $\endgroup$
    – special_frog
    Commented Sep 25, 2021 at 0:08
  • $\begingroup$ @special_frog Thanks! If you knock out all the nodes with parameters attached, the math is just the graphed expression. $\endgroup$
    – Robin Betts ♦
    Commented Sep 25, 2021 at 2:46
  • $\begingroup$ As a follow up to this question... Would this same setup work with proximity information instead of location? or even having two(or more) origins in order to have interference? $\endgroup$
    – special_frog
    Commented Sep 25, 2021 at 7:17
  • $\begingroup$ 1. This already is proximity information.. it uses the distance to the empty. Perhaps you could be a bit more specific.. but I feel if you want some more elaborate function-of -distance, that could just be an adjustment to the falloff section of the tree. 2. Sure you could have interference. Perhaps instead of setting the Z position of verts, you could add to the pZ. Then you could stack the modifiers. I'll have a go, if you like, $\endgroup$
    – Robin Betts ♦
    Commented Sep 25, 2021 at 7:44
  • 1
    $\begingroup$ That works great! I think distance from geometry would produce some pretty interesting results but is an additional level of complexity that is not necessary for my current project. I will be sure to credit you once I've made something with your modifier. Many thanks, this is super helpful! $\endgroup$
    – special_frog
    Commented Sep 25, 2021 at 8:40

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .