nCloth animated inputMeshAttract + rigidity = problem

#1

Hi

I got a particle instancer turned into geos and comnibed and turned into ncloth

I animate inputMeshAttract so a dynamic collisions will happen with the ground. Nice

But as soon as I turn up rigidity (and polyShell on) the sim start to explode uncontrollably. I need the objects to appear rigid.

Anyone else encounter this problem? If so what did you do to fix it?

Thanks

F

#2

Perhaps it would help to increase substeps and lower inputAttract.

#3

thanks for your reply, Duncan,

I’ll do some test…

#4

Hi Duncan,

The suggestion did not seem to work. In my various tests I also found that as I do not need to animate the inputMeshAttract at all. as long as the input mesh has animation and rigidity turned up on the nCloth, it becomes unstable and explode. Is there any way to work around it?

I use nucleus spaceScale of 0.02.

objectSpace on nCloth,
rigidity > 0

everything else is pretty much default

Thanks

F

#5

What about self collision? Is it possible that that could affect something? Collision conflicts can cause cloth objects to explode. But I don’t know if that applies here.

#6

No Self Collision on the nCloth. Still no solution … :frowning:

#7

I tried something a little similar to the setup you described and it worked OK for me in Maya2012 and Maya2014 (didn’t test in 2013). I did notice that making InputAttractDamp greater than 1.0 caused instability in 2012. We changed interpolation of inputAttract in 2014 so it should in general work better.

I’ve attached the scene… two objects that are keyframed then poly combined to a single mesh and made into nCloth with rigidity and inputAttract.

#8

I played with the scene a bit more… it was not hard to get some crazy rotations and instability. At this point I think it may be the way rigidity is computed… there can be problems specifically when the input mesh shape is deformed, which you are relying on for your animation. If I can think of any solutions I’ll post them, but you might want to look at a different approach.

#9

Thanks a lot, Duncan, I really appreciate your input. I find it really hard to get a descent setup without breaking somewhere along the line. Thanks again

F

#10

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.