A Series Of FLIP Parameter Tests

[Atom]

Hi All,

I am trying to visualize the relationship between some of the FLIP parameters on the FlipObject and the FlipSolver.

These tests are conducted on the default network create by the shelf tools. Only the parameters listed in each image have been changed.

Overall I think I like a higher Particle Radius Scale, locking in on 4.0 up from the 1.2 default. It seems like the fluid clings to the surface of the collision object in a more natural way with a particle radius scale of 4.0.

If you want to thin out the fluid subtract 1 from the Particles Per Voxel expression in the FlipSolver Reseeding tab.

ceil(pow(ch("../flipfluidobject1/gridscale"), 3))-1

If you want to thicken your fluid add one to the above expression. If you add more than 1 you can introduce a nice churning effect.

ap_3_sphere_flip_test.hipnc

Edited January 28, 2017 by Atom

[Skybar]

Not sure what your goal is, but if you are aiming for natural you are gaining lots of volume out of thin air in most of these?

[Atom]

I am trying to go for a character splashing/stomping around at the top of a waterfall. I was looking for a way to increase volume a bit without lowering particle separation. Currently I supply a shape to the FluidObject as my initial fluid source then it all runs over the falls and thins out over time. Originally I used the shape as an emitter connected as a VolumeSource to the FlipSolver but then the sim got too full of particle that it maxed out my memory and slowed to an unusual crawl.

Is there some node I can drop down that will simply manage the entire particle count from a single value?

What if I want to maintain a million particles throughout a long animation?

 

Edited January 28, 2017 by Atom