I performed the flow calibration test, printing the one-wall cube and measuring the thickness of each wall. Turns out they're all quite different!
Any thoughts on what may cause this? Is this something others have noticed. If there's something to be said about this, it might make an interesting addition to the document. It also requires marking the axes before removing from the build plate, so that the orientation is not lost.
Thanks for this great site.