Carrara Water Particles Animation Experiment and Questions

Joepingleton

View the animation on YOUTUBE.COM by clicking here.
This is a quick experiment with meta-balls particles in Carrara 8.5 Pro. I also tried a Caustic render on the simulation.

The main problem is that the particles are met-balls from a particle emitter and they are set to collide with all objects in the scene. But they seem to "leak" thru the glass. I tried adding a mesh inside the glass and that helped a little. But particles are still slipping through the objects. Any suggestions on how I might fix this?

Thanks for looking.

tbwoq

I have not found a way to stop particles leaking with certain shapes except maybe using invisible planes for the metaballs to settle on. Try a plane shape(s) at the very bottom of the glass where the particles begin to leak.

thoromyr

Just doing animation of simple meshes, no metaballs, I get lots of pass through. I've tweaked the physics settings this way and that and modified meshes and it just doesn't seem to be reliable.

I haven't had time to test it, but one theory I have is that the collision detection works better when objects have non-simple movement. What I mean is this: put some objects in mid air and a plane beneath them. When the sim runs the objects will drop straight down at the mesh and (at least with the scene I was working on) they would all pass through the top face, but they would be perturbed and start to spin. The objects would then (mostly) collect and settle on the *bottom* face *inside* the object.

My (untested) idea is to put an intervening plane in to cause the perturbation before the objects hit the top face so that, hopefully, they would not penetrate it. Mostly its untested because in this specific case it didn't *really* matter they were inside the object, I just wanted a collection of naturally settled objects on a plane so deleting the object and rendering final frame was all that mattered.

I don't know how this would go with your metaball problem, but I thought it worth mentioning.

Dartanbeck

I imagine you've already tried upping the density of the glass?

Joepingleton

Thanks for the suggestions.
In my experiment I set the density of everything to 100.

I tried adding a intervening plane and it stops them from entering the glass, but i was hoping to have them fill the glass.

Dartanbeck

Ahhh... finally watched the video and I see what you mean.
That shouldn't be happening.
100? I was thinking maybe just a bit higher than 1.
I wonder if it would work to set a temporary plane to the bottom of the inside of the glass, just to get the z axis value (height) and enter that in the particle emitter for ground level? So you'd just delete the plane. But I wonder why they're escaping through the mesh like that.

evilproducer

joeping said:

Thanks for the suggestions.
In my experiment I set the density of everything to 100.

I tried adding a intervening plane and it stops them from entering the glass, but i was hoping to have them fill the glass.

They can't fill the glass as it is not a true fluid simulation.

For animating something like that it would require a combination of a glass with some kind of modeled and/or morphing liquid and an emitter for the pour and maybe another for a splash.

Joepingleton

Thanks for the feedback Evil,
I feared that was the case, but I thought that I was just missing something.

evilproducer

Sorry to give you the bad news. I know that has been a feature request in the past.

I believe the issue is that the particles and objects don't really have a volume, so there is no way to fill using meta-balls.

McGuiver

All is not lost.....Float has a free plugin called pycloid which will do fluid simulations. It will fill your glass with fluid.

You can find it here: http://www.f1oat.org/pycloid/index.html

DUDU

McGuiver said:

All is not lost.....Float has a free plugin called pycloid which will do fluid simulations. It will fill your glass with fluid.

You can find it here: http://www.f1oat.org/pycloid/index.html

We are precisely under discussion about Pycloid on Carrarators.
There are big problems of crashes, error messages and disappearance of the particles.
What a pity that FD (Frederic Rible), its creator, is not there any more to continue this marvellous plugin.
When we look at his demos, it's extraordinary what we can do with that !

Dartanbeck

Frederic isn't there? What do you mean?

Hey McGuiver, still have that water demo you did with PyCloid before?

DUDU

I think that Fred is in the depths of Lightwave…:down:

Dartanbeck

Bummer, I say! :down:

McGuiver

Dartanbeck said:

Frederic isn't there? What do you mean?

Hey McGuiver, still have that water demo you did with PyCloid before?

Here's one: http://vimeo.com/52197357

I never had any problems with the pycloid plug-in.....I am still using C 8.1.....maybe the issues are with 8.5 ?

Joepingleton

PyCloid looks like it maybe the solution. I need to learn how to use it. Right now my experiments with it "leaks" more than this experiment.

Any good tutorials on how to use it?

DUDU

I work on C8 also, have you a mini tuto to explain how you make.
I am on w7/64…
I start from a scene of FD (Waterfall), when I change a parameter, there is an error message or Carrara crashes.
If I replace the plate by a ground, there is no more particules emission…
Is there a script somewhere which we can reload?

tbwoq

Hi joeping.

The tests I ran were geared toward stopping the leaking only, and as others pointed out, Carrara's particles don't have any volume, just hot points. There is a feature in the particle emitter advanced tab that is supposed to push the particles apart to create a type of volume(not liquid), but it has been broken since at least C5.

I agree with what EP suggests and would morph/animate some liquid in the glass while pouring the particles. It should be convincing enough and a few others on the forum have used this method.