Much faster fluid sims...

[JeffG2]

Narrow Band FLIP for liquid simulations

http://cgpress.org/archives/narrow-band-flip-for-liquid-simulations.html

Researchers from the Technische Universität München and IST Austria have presented a method for simulating liquids that only uses FLIP particles within a narrow band of the liquid’s surface, while efficiently representing the remaining inner volume on a regular grid. The new method drastically reduces the particle count and simulation times while yielding results that are nearly indistinguishable from regular FLIP simulations. According to the authors, the approach is easy to integrate into any existing FLIP implementation. More on Technische Universität München’s website.

[JeffG2]

Can this be incorporated into Phoenix?

[Paul Oblomov]

Try bifrost in latest maya 16.5. It is not THAT faster. Uses simlar tech.

[samuel_bubat]

I´m still waiting for this to be integrated in some product.
https://www.youtube.com/watch?v=rce92SZ1y60

[super gnu]

that looks very impressive.. i wonder what the drawbacks are. combine that adaptive resolution with the narrowband "surface only" concept, plus some lod based on viewing distance, and we could end up with realtime oceans with only 10 particles

[Mokiki]

I am still waiting for this: https://www.youtube.com/watch?v=7Who8EpbvCY
Speed improvements are all great and I gladly take them, but man am I ready for realistic surface tension models.

[super gnu]

i want both :P

[samuel_bubat]

I´ve seen so many promising technology research that unfortunately never made it into a usable software.
For example bdtree from 2004
http://graphics.cs.cmu.edu/projects/bdtree/
or imperfect shadow mapping from 2008
https://www.youtube.com/watch?v=Pdp3rfyFF14