Hello! I've been working with the pipe generation tutorial Entagma has up on Houdini's website and I have an animated point cloud connecting points to make lines that connect the points on a piece of animated geometry. In the Solver node there's some vex code but it only generates the pipes and connects the points on the first frame rather than moving with the animated geometry.

When I plug in the point animation data using the first input node it messes up and stretches out the formation of the pipe lines. However if I move to a frame and refresh the solver it holds the shape and position.

For this reason is there any way to have the solver reset per frame and/or write out the solver per frame after the reset so I have a cohesive animated point cloud work through the solver?

Thanks for the help!

If I understand you correctly I think we have a related issue. The solver sop does not seem to take into account more than one single frame of the inputs. I am trying to use the solver create a growing number that I feed into the transforms at the object level, and then
the solver does not read the those increasing transforms every frame. I will try to write elsewhere about this because I haven't head of other s than you mentioning something related

You might not see this as I'm writing 6 years later hehe

Correct, the solver SOP takes the geo at first sim-frame and after that reads the processed/output geo back on each frame/time-step, like a foor loop over time.

However In the solver SOP, the nodes called input_1 -> input_4 are reading geo connected, so you can for example copy the position of your transforms onto the solved geo.