|On this page|
This node works with the Lattice From Volume node and deformation nodes.
The Lattice From Volume node creates a lattice geometry from a volume or set of volumes.
This node automatically applies the equivalent deformations to the volume(s).
The Lattice From Volume marks the points of the lattice geometry with the source voxel they come from (the
iz integer point attributes, which together specify the 3D location of the original sample on the voxel grid), and a
rest attribute which records the point’s original position in space. This node compares the deformed lattice point positions to their “rest” positions and uses those transformations to deform the input volumes.
This node always outputs VDB volumes.
In a SOP network, generate or import the volumes you want to deform.
Press ⇥ Tab, type
volume deformand press Enter. Click to place the nodes.
The Volume Deform tool creates both a Lattice From Volume and Volume Deform node at the same time, with the Lattice From Volume node already wired into the Volume Deform node’s second input.
Wire the source volumes into the Lattice From Volume node’s first input. Connect another wire from the source volumes into the Volume Deform node’s first input.
The default Output type on the Lattice From Volume node is Points, which is fastest. However, some deformer nodes cannot deal with unconnected points, and expect connected/polygonal geometry (for example, the Soft Transform SOP). To be able to use those types of nodes, set Output type on Lattice From Volume to Polyline.
This node will apply the same deformations to the volumes in the first input.
Tips and notes
You can downsample the lattice for faster but lower-quality deformation:
Connect a VDB node to the Lattice From Volume node’s second input. In the VDB node’s parameters, set the Voxel size to a value larger than the actual volume’s voxel size.
(When its second input is connected, the Lattice From Volume node will use the voxel size of the volume in that input to control the lattice resolution, rather than matching the resolution of the input volume(s).)
The first input should contain the volumes you want to deform. You should wire the same volumes into the Lattice From Volume node’s first input.
Deformation cage point geometry. This geometry should come from the output of a Lattice From Volume node, before being deformed by other SOPs (for example, the Bend SOP), then wired into this input. This node will apply the same deformations to the volumes in the first input.
Which volumes are going to be deformed by the lattice. This should match the equivalent parameter on the Lattice From Volume node.
When this is on (the default), the node uses cubic interpolation to give smooth, high-quality deformation, especially for low-res volumes. If you turn this off, the node uses linear interpolation. You might want to turn this off because linear is much faster, and the quality difference may not be noticeable if you have a high-resolution volume/lattice.
Perform Partial Update
When this is off, the node replaces the incoming volumes with the deformed result. Areas of the original volumes that are not captured by the lattice will be lost. If you turn this on, the node merges parts of the incoming volumes that are not captured by the lattice with the deformed output. You can use this capture and deform only part of the volumes.
Scale Density Where Compressed
If the deformed volume is a fog volume, realistically make it less dense where deformation stretches it out, and more dense where deformation compresses it. If you turn this off, the deformation does not affect the fog density.
Max Density Scale
When Scale Density Where Compressed is on, you can turn this on and set a limit of how much the density can change.
When this is on, collapse constant tiles in the VDB output, and look for areas within this amount of difference from the background value and deactivate them.