popGrains for flowers bent by the wind. Problem..

   1933   2   1
User Avatar
Member
118 posts
Joined: Feb. 2016
Offline
I'd like to apply some wind force to a flower.
I've tried to use the popGrains node without much success




Have a look at the reference image, my goals are simple:
1) A wind force should bend and move the flower.
2)The movement of the whole flower should be influenced also by the weight/size of the ovary (green sphere).
3) the branches/petals attached to the ovary should move and bend accordingly

So far I sort of managed to move and bend the main stem inside a popnet with popGrains and attach the ovary-sphere to the end point of the stem. I cheated a bit to simulate the weight of the sphere, by dealing down the weight of the end point of the stem.

To simulate the branches/petals attached to the ovary, I tried to reproduce this setup found in Grain Solver for wires [www.tokeru.com] (file by Matt Estelahere [www.tokeru.com] ).
I got crazy results when I tried to use the @targetP attribute to drive the stiffness of my wires. I guess there is some conflict going on. Also probably it's not the way to go…

Anybody want to have a look at the attached project file?
Thanks!
Edited by Andr1 - Feb. 22, 2018 15:52:12

Attachments:
FlowerGrain_problem.hiplc (582.9 KB)

User Avatar
Member
8526 posts
Joined: July 2007
Online
I'd suggest Wire Solver for this as Grain solver doesn't solve for angular constraints
Tomas Slancik
FX Supervisor
Method Studios, NY
User Avatar
Member
118 posts
Joined: Feb. 2016
Offline
Hey Tomas thanks for the tip,

I tried the wire solver way and got better results indeed.
I used this trick [ihoudini.blogspot.de] to simulate the reciprocal connection between the wire and the attached sphere.
It reads internal force (@finternal) of the attach point on the wire, and apply it to the sphere with a “point force” node.

Problem:
How to do the same with multiple wires/spheres?

My idea was to use a Sop Solver node and read the @finternal and @P for every attach point and apply it to their relative sphere.
Somehow I should find a way to simulate a point force for ever attach point on the spheres. (But I don't really know how to do it).

Anyway, I'm already stuck before even trying: when I connect the Sop Solver (along with rbd solver) to a multiple, spheres go upwards.

I attached the hip file, in case you or anybody else is kind enough to have a look..
Thanks!
Edited by Andr1 - Feb. 24, 2018 04:10:20

Attachments:
Q_multipleStemsAndSpheres.hiplc (403.1 KB)

  • Quick Links