Unified Particle Physics for Real-Time Applications

[manny]

Just wanted to share this interesting paper with you guys:
- paper: http://blog.mmacklin.com/flex/
- related blog post: http://blog.mmacklin.com/2014/05/15/new-siggraph-paper/

I think having a unified system for collision detection is pretty awesome as this particle based model allows you to simulate pretty much any body, whether convex or concave.
However according to the authors there are certain limitations, like simulating large flat surfaces.
But still, having just one way to simulate fluids, gas, cloth and rigid bodies is awesome.

[Julio Jerez]

I agree, if we are going to simulatate more than rigid bodies, the rigid body model asumtion is too restristive and brake down iin many situation.
the lagragians model of rigi body beradoun when simulation very large, and veryu small models.
and is incapable to model fluisd, Gasses, or plasticity.

on the other hand teh rigid body nmodel is eassy, and faster.

The one problems I the pariocles simulation is the they are no teiing teh comple true, the never deal with lareg matrices losing cohenerance
but in general I agree teh futore is in Particles model tah can simulate everything,
and with high performaces computing, it seem to be teh natural extention of physics simulation.

I will check ou the paper, thank you.