Variable viscosity

[jstrob]

If we could get variable viscosity we sould be able to simulate stuff like lava like this:

[Ivaylo Katev]

yes, but the current viscosity method is not suitable for mapping.

[jstrob]

And like this:

[jstrob]

yes, but the current viscosity method is not suitable for mapping.

Could we use something like distance from a null or something like that instead of using mapping?

[Ivaylo Katev]

i didn't mean exactly mapping, it just requires all the liquid to have the same viscosity.
anyway, we can't continue to use grid based solver and to meet the competition, the particle based solvers can reach better quality in the liquid area, and we have to introduce particle some kind of particle based solver.

[jstrob]

That would be awesome! Keep up the good work!

[flat]

May I expect the particle based solvers in the future ?
They are excited news.

[Ivaylo Katev]

actually we started to work on it immediately after the release, see one image made with it.

[möbius]

oh boy oh boy, awesome! but why is there still a PHX-Grid visible in the viewport on your image?

[Da_elf]

couldnt age be a factor in viscosity?

[Ivaylo Katev]

oh boy oh boy, awesome! but why is there still a PHX-Grid visible in the viewport on your image?

we don't want to break the backward compatibility, and actually there is no need to do this. the old solver remains, and the new one is also able to export its content as grid. this gives us many advantages. first, the grid content can be compressed much better than the particles, second you can render it without having frost, using the tools that we already have.

couldnt age be a factor in viscosity?

if the viscosity is mappable, any value can be a factor. tn our grid based solver this is currently impossible.

[TSR_trix]

FLIP solver... now thats what im talking about!
You guys rock.

[Paul Oblomov]

Maybe something like this ? SPH but looks promising.
http://www.52-technology.com