Can the inference engine detect moving point of component instance while editing


Anyone know a good way to get a snap from two planes of mirror image component while editing? It seems that inference does not “see” a moving piece of the same component while it is moving.


My go to for that is to draw an edge between the two outside the context, then move to the midpoint.


It would be more intuitive (and thus SketchUp-like) if it could, but it can’t and I think coding such a behavior could be a little tricky. Basically you need to get the trajectory of the move operation, transform it to the other coordinate system, see if there is an intersection and find a solution for what move distance results in both points overlapping (not necessarily half the distance between as the instances can be differently scaled).

I also use the temporary edge and midpoint approach.


If you have enabled length snapping, set precision higher and length to half.
Zoom in for better control:


Thanks all. Yes it would be a bright shiny happy world if the inference engine worked that way. In the real world I’ll keep using external guides. Cheers :beers: