Bullet Fluids

[Ocra]

Is there a bullet based SPH in the works at all ? If so, will forces be read to and from the rigidbodies ?

[pildanovak]

I thought it's allready there...?
http://www.youtube.com/user/rtrius?feature=watch

[majestik666]

as far as I know this is the latest implementation :

https://github.com/rtrius/Bullet-FLUIDS

based on fluids v.2
There is also fluids v.3 but don't think there's a bullet implementation for it as of now

[rtrius]

I'm currently looking into something else(rigid body MLCP solvers), so the
latest version is indeed at the github repository. It will most likely be at
least a few weeks before the next update.

The current version is more or less final as far as the interaction between
rigid bodies and SPH fluids is concerned. There are other issues that need
to be resolved, though, so it is still experimental. For example, the
emitter/absorber API needs to be replaced as the emit()/absorb() functions
are not called for each internal simulation step.

The SPH system will probably not reach stable/production until after
Bullet 3.x is released.


Although it is based on Fluids v.2, there have been many optimizations
and other improvements. For instance, single threaded CPU performance is
~3.5x faster and adding multithreading increases that to ~7x (excluding the
SPH stiffness increase). See issue #296 in the tracker for more details.
(http://code.google.com/p/bullet/issues/detail?id=296)