High Poly Destruction simulation`

#1

Hi all,
I have a little R&D Project that I’ve been working on that has reached a point where I can not think of how to progress.

The Project starts with a shoebox, which fills with fluid (realflow) the fluid then rises out and morphs into the shape of a football boot. With a little trickery it then blendshapes into a Zbrush model of itself all dried up and cracked. From there I would like the dried up cracked model to disintegrate revealing the actual shiny boot underneath.

Though The disintegration bit is proving to much to handle as it is a large poly heavy model that came from a realflow mesh.

I have tried loading it into houdini to fracture it, turning to ncloth and tearing it plus a few other cheeky bits.

Though the computer can’t really handle a simulation like this in any workable timescale.

Does anyone know if there’s a anything I could do to maybe reduce the polycount inside the animation before attempting the simulation?
I know with ncloth I could simulate something on a lower poly version then have the high poly follow but I assume this wouldn’t work with tearing?

I’m looking for results similar to this
https://vimeo.com/61050621
Though it doesn’t need to be so detailed.

Many Thanks

Jason.

#2

There is nothing I can see in that shape that would require it to need that many polys. Just retopo that mesh into something much simpler and then do your dynamics to that new properly prepared geometry. If you need some kind of crazy detail generate the displacement maps from z brush and apply them to your cleaned up mesh.

#3

Hi, It needs to be the exact same topology as the real flow mesh in order to seamlessly blendshape into the dried up effect. I could displace from it, but it’s still a high poly realflow mesh.

#4

You could try a poly reduce node in Houdini, and apply it to the geo for the entire animation of the real flow mesh and animate it’s strength so the end result is a compromise of shape and res to blend into the z brush shape, and see what that gives you. Otherwise I think you might need to go with a comp route to help the blending bit.

#5

This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.