Has anyone got any real details?
To what extent was it "cheating" is what I'd like to know.
It would be reasonable to save the finite supply of urea liquid for the times when the engine emits most NOx, under heavy acceleration (i.e driver demand, pedal position) and possibly also at lower revs. Similarly one might wait till the engine is warm otherwise the urea wont turn to gas within the exhaust and would be wasted. There is plausible denial so far, I would say.
If it measures actual rpm profiles and acts only on those that are a few percent within the standard test profiles, then that is definitely "test detection".. Similarly again, if there is a flag set for "test" conditions, and this remains set for some time, possibly the whole ignition cycle after a "cadence detection" of some sort, then sure, they're busted.
Where was it, actually, between these two levels?
Reg readers need to know.