PDA

View Full Version : Maya N-particles for LW?



Chrusion
11-17-2014, 10:56 AM
Are there ANY particle systems for LW that's similar to Maya N-particles? I don't have access to Maya, nor the budget to purchase a license, just to do simulations where you fill a container/drum with 1 million particles and mix them with 8 dozen paddles.

ernpchan
11-17-2014, 11:01 AM
Short answer is 'No'.

You might be able to do something similar with tiny Bullet spheres. I was able to do a Bullet sim where a bucket loader was scooping up balls and dumping them.

Trying to do it with 1 million objects is really pushing it though.

MSherak
11-17-2014, 12:45 PM
Just download the latest version of Blender. It's free and has Particle Fluids. Nice thing is you can load them back into Lightwave.


http://www.blender.org/
http://www.clintons3d.com/plugins/lightwave/
http://www.splotchdog.com/index.php?option=com_content&view=article&id=12:bfs&catid=11:lightwaveplugins&Itemid=121



-M

Chrusion
11-21-2014, 07:51 PM
ernpchan,

How did you create the initial pile of balls? Arrayed a sphere in an object layer, then use parts mode in bullet?

So far I'm having a bit of luck using PFX. I made a dense, curved mesh that floats just above the curved surface of a lower poly collision container. I emit a ramped burst of particles using frame mode over 20 frames, with the peak rate of 8,000. The result is around 88K particles that fall an inch or two onto the high friction collider surface with 0.1 fix strength. I then start a set of 96 paddles on the outside of a drum turning at 8 rpm that push the particles into "rows" for the next set of offset paddles, that push the row of particle over into the path of the next row of offset paddles, yada, yada. The particles are self-interacting using Bounce at 50% and 50% viscosity.

Each frame takes 1 -2 minutes to calc. The result is not totally physical accurate, but it does have hints of realism. In the end, it's FAR better than a static mesh with an animated displacement map trying to simulate a volume of gravel in a drum, although I'm still using the static mesh underneath the mass of particles (instanced with a sharp, angular stone obj.) to provide coverage for gaps in the particle density, to make it appear there's thickness and volume.

ernpchan
11-21-2014, 09:18 PM
Yeah, I made a bunch of balls via an array, did a bullet sim based on parts. And then did a save transformed on the mesh to make my pile. Then used that geo in my bucket sim.

Chrusion
11-25-2014, 09:36 AM
The end result...
http://youtu.be/Sx2rF0U0xf0

ernpchan
11-25-2014, 09:49 AM
Wow, very cool.

lardbros
11-25-2014, 10:14 AM
That's really very good!! Nicely done :D