Salva is a bit overkill for realtime fluid..?
see https://slembcke.github.io/WaterWaves e.g.
What are 
doing for these effects?
#Water surface simulation?
zealous jasper
sand nimbus
I experimented a bit with a compute shader FFT water simulation a while back: #showcase message
zealous jasper
That's awesome.
It's not attached to Rapier, is it?
pliant adder
Have you seen bevy_water?
zealous jasper
Have you seen bevy_water?
no, looking, thank you.
That's excellent. Any other implementations or is that where the community mostly looks for water?
sand nimbus
That's awesome.
It's not attached to Rapier, is it?
No it's not attached to Rapier. It's not really a fluid simulation at all. It just generates an ocean wave spectrum (based on wind, water height etc.) and then turns it into displacement using FFT.
placid vector
If you're looking for something extremely simple and would consider rolling your own, I'm happy to share some thoughts and code. I working with this low poly water and buoyancy prototype https://claudijo.itch.io/pirate-sea-jam
zealous jasper
^ that Pirate sea jam is impressive, thank you for sharing.
I wonder if "water" is not just too broad. In my use case, I don't need an ocean. I'm interested in being able to have streams that merge - so like a tributary down a slope that pours into a creek. When the player steps in the creek it should ripple from the disturbance. But should also be flowing.
So, laminar flow is more important than ocean spectrum in my use case, I think.
I realize this is a gigantic topic, so perhaps it's better to ask more conceptual / design questions at this point. How would y'all approach the above use case?
zealous jasper
see also: #general message
placid vector
Interesting problem area. Would love to hear what you end up doing and how.
zealous jasper
Should we have a #water channel?
I would be interested to dedicate some energy to sharing articles, approaches, design requirements, etc.
Currently considering the following:
- laminar flow 2d, pipe, uniform grid
- laminar flow 2d, pipe, non-uniform grid (pipe obstructed by static rigid Shape)
- (2) with kinematic providing impulse
For implementation:
- no particle simulation, as in Salva IIUC
- water mesh's cells have: static positions (mesh is "rigid"), non-compressible fluid; only track pressure (~flux) along face normals.
- n cells aprox. 2**10
Brainstorming:
- combine bulk transfer for "under water" with sum-of-sines for "surface water".
- show surface effects of kinetic player controller influence: combine Bessel function impulse with sum-of-sines for the surface model.
- show surface effects of bulk transfer by translating surface vertices along z axis
I want to combine earthquakes (bulk) with local (Bessel) + periodic (sine) surface waves.
Shoot for the 
moon I guess.
zealous jasper
Looks like the (2007) wave particles -> (2017) wave packets -> (2020) wave curves line of research (https://dl.acm.org/doi/10.1145/3386569.3392466) is a much more clear description of the babble ^ above.
zealous jasper
This paper uses the terminology "bulk" and "surface" as well: https://dl.acm.org/doi/10.1145/3592098