Hello, I've been studying this webinar on flip fluids and seem to have run into a bit of a problem.

https://youtu.be/Q62_p_rlZKM?t=1h16m31s [youtu.be]

Whenever I try using the dual rest solver vop to create uvs, nothing seems to happen. No matter what parameters I adjust, the uvs just end up at 0. I even downloaded the example files and the method still doesn't seem to do anything. Has the workflow for this type of thing changed since this webinar was created or am I just missing something obvious? Thank you.

Been trying to figure out the same. The TEXTURE VOP seems to be different in 16.5, so I'm guessing the workflow has changed. Would be interesting to find a solution for this. Thanks!

Has anyone got more info on this, trying to use the dual rest solver in a vop and having no luck, whats the correct setup to do this manually?

flamingBurrit0
Hello, I've been studying this webinar on flip fluids and seem to have run into a bit of a problem.

https://youtu.be/Q62_p_rlZKM?t=1h16m31s [youtu.be]

Whenever I try using the dual rest solver vop to create uvs, nothing seems to happen. No matter what parameters I adjust, the uvs just end up at 0. I even downloaded the example files and the method still doesn't seem to do anything. Has the workflow for this type of thing changed since this webinar was created or am I just missing something obvious? Thank you.

That uv from dual rest solver is a bunch of nonsense that would never work. The ‘working’ example shows the uv being plugged into a disconnected texture vop which does nothing. So the working file must have been relying on the uvquickshade's default orthographic texture projection along the y axis and not using rest at all. What he was trying to do was map the rest x and z to uv coordinates, which might be acceptable for some situations such as creating static meshes for games, but because dual-rest is enabled, the uv's would reset back to the reference value periodically and would result in visual popping for animations. If you want that, skip the dual rest solver and create a bind for ‘rest’, use a swizzle of “0 2 1” to map x and z to u and v, and connect the swizzled output to a bind export for ‘uv’.

This is assuming rest is even coming through the simulation to the surface correctly in the first place. I would start there to debug if it still doesn't work.