I'm trying to find more information about the "Position Data Path" parameter inside the Fields tab of the Smoke Object (Sparse). There doesn't appear to be any help documentation for that parameter:
https://www.sidefx.com/docs/houdini/nodes/dop/smokeobject_sparse.html [www.sidefx.com]
Thanks,
Luke
Help Docs for "Position Data Path"
567 3 2-
- Luke Letellier
- Member
- 241 posts
- Joined: 4月 2014
- Offline
-
- omarz
- スタッフ
- 184 posts
- Joined: 9月 2017
- Offline
That parameter can be used to set the path to a Position [www.sidefx.com] data on the object, which is like parenting the pyro object to another transform. Its use isn't really recommended... It was added as an aid for the Minimal OpenCL mode, to allow the Center of the simulation box to be animated. The Simple GPU Torch shelf tool (under Simple FX) shows this in action. The default Sparse mode isn't expected to work under the same tight constraints of the minimal solver, and can resize the simulation container to accommodate motion of the fluid.
You can dive inside the Pyro Solver SOP to see how things are hooked up if interested, but I wouldn't suggest you use it manually in a simulation.
The documentation will be updated with a description along these lines.
You can dive inside the Pyro Solver SOP to see how things are hooked up if interested, but I wouldn't suggest you use it manually in a simulation.
The documentation will be updated with a description along these lines.
-
- omarz
- スタッフ
- 184 posts
- Joined: 9月 2017
- Offline
To clarify the above, there are some advanced uses of the position data to reorient the container (similar to other fluid solvers in Houdini). The recommendation to steer clear is based on the fact that the sparse solver is pretty efficient at working in a default axis-aligned container. You can experiment with it, and orienting the container might actually be a good idea if you need to perform a dense simulation for some reason. For example, if wind is blowing your smoke in a diagonal direction, orienting the container with that direction might be a good idea (since the smoke then occupies the simulation box "lengthwise" instead of diagonally, which simulates a lot of empty space).
-
- Luke Letellier
- Member
- 241 posts
- Joined: 4月 2014
- Offline
-
- Quick Links