Morph3DMean - something odd is happening

I’m having an issue with the PtMorph3DMean component in Grasshopper (Rhino 6.29.20238.11501).

I get very strange geometry: the two base objects/modules are no longer recognisable anywhere in the final result. Instead I’m getting strange sheared and angular geometry which is being used to tile ALL the grid cells. My modules have the same (simple, square) footprint and height - and aren’t very different in design, so I’m surprised I’m getting something which bears no relation to either. The attractor input (W) seems to make no difference to what happens (changing magnitude does nothing) although the component obviously won’t work without it.
Unfortunately this component isn’t covered in the documentation - but I’ve worked through Frank Melendez’s example from ‘Drawing From the Model’. Again, I just get a geometric mess bearing no discernable relationship to the two simple base modules.
Attached is my gh file - which is a comparison between results from Morph 3D using one of the modules and an attractor, and the results from Morph 3D Mean using both modules and the attractor. Morph 3D with Morph 3D (31.6 KB) and a screen grab of the results

I do see this and it looks like a bug. I also see how the magnitude is changing the grid, but not updating the weights. I’ll investigate further.
However, when I simplified the surface or used a mesh module, it gave expected result.
Morph 3D with Morph 3D (39.4 KB)

Wow - thank you so much for the swift response. Happy to know that I haven’t made an obvious error (all too likely), and that there is something afoot. Also brilliant to know that there are alternative ways to sort it out. You’re a star!