Slicer Sample Window Test

I've seen a few designs (including my own) exploiting control of printing at individual layer precision. Examples include progressive bridging, stacking prints and print-in-place moving parts.

I'm not sure how consistent various slicers are when it comes to sampling a 3D model in a series of 2D slices. So I decided to make a few test models to check. Sample models are for 0.2mm layer height.

The first question which comes to mind is how the model is sampled for each layer. Given a layer has a thickness, how do we interpret the volume of the layer as a 2D slice? Naively we could take the intersection of a plane through our model, but at what height? Initial experiments in Cura indicate approximately the midpoint but with some bias and potential rounding. If the sample window is too narrow (0.02mm for 0.2mm layer height) the results get a bit "noisy" but at 0.05mm the transition is crisp (by adjusting Z).

Other models from:
https://www.thingiverse.com/thing:3731
https://www.th