Hey guys, i'd like to apply an ocean evaluate (Spectrum) over an already cached flip sim i have of boat wakes. But i'd like to have that ocean spectrum deformatino only applied outside of a falloff surround the boat wakes or boat colliding object so the wakes sim still matches the shape around the colliding boat accurately. I can't see any option on the Ocean evaluate node to use a mask. Can anybody point the way please?

Cheers,

A.

There is a stable point count/@ptnum between the input and output geometry of the ocean evaluate node. So you can also just put a wrangle down after the ocean evaluate and use a lerp() function to mix between the original point positions and the displaced point positions. You can do this based on a float attribute mask that you paint on the geometry, or any other masking technique for that matter You can also mask the spectrum itself and then feed that masked spectrum into the ocean evaluate node. Hope this helps, cheers!

Robbert
There is a stable point count/@ptnum between the input and output geometry of the ocean evaluate node. So you can also just put a wrangle down after the ocean evaluate and use a lerp() function to mix between the original point positions and the displaced point positions. You can do this based on a float attribute mask that you paint on the geometry, or any other masking technique for that matter You can also mask the spectrum itself and then feed that masked spectrum into the ocean evaluate node. Hope this helps, cheers!

The tank for the sim is moving, so the geo so i'm not painting manually would work even using the UVs (which might stick, not sure). But i do have the vorticity and veloicty resulting from the sim saved in the compressed cache. I guess i could use this as a base for that masking method you're talking about... i'd probably blur that a lot to fade it around but i guess that could work. I'm just thinking that mixing / blending those very dense geos is going to be quite slow(??). Let's give it a try.

Cheers,

A.

Adriano

Robbert
There is a stable point count/@ptnum between the input and output geometry of the ocean evaluate node. So you can also just put a wrangle down after the ocean evaluate and use a lerp() function to mix between the original point positions and the displaced point positions. You can do this based on a float attribute mask that you paint on the geometry, or any other masking technique for that matter You can also mask the spectrum itself and then feed that masked spectrum into the ocean evaluate node. Hope this helps, cheers!

The tank for the sim is moving, so the geo so i'm not painting manually would work even using the UVs (which might stick, not sure). But i do have the vorticity and veloicty resulting from the sim saved in the compressed cache. I guess i could use this as a base for that masking method you're talking about... i'd probably blur that a lot to fade it around but i guess that could work. I'm just thinking that mixing / blending those very dense geos is going to be quite slow(??). Let's give it a try.

Cheers,

A.

instead of using the points you could create a volume mask and use that to mask out any points that are inside of the volume. That way you can move it with your sim. If I understand correctly, you want to mask out an area near the boat. In that case, you could take the boat, make it into a vdb, dilate the vdb a bit, maybe smooth it out, make a nice falloff with a volume VOP. You might be able to use it with the ocean evaluate directly, or use a volume sample function to read the mask from the volume to the points.

The particle fluid surface node has options to mask by vorticity and velocity. This is generally used for similar masking, the ocean spectrum is then computed at rendertime. There is an ocean surface material in the material palette that you can inspect to figure that stage out.