The topic of distortion perception is a fascinating one. In my opinion, it makes sense to focus on those distortions that are subjectively objectionable, but not worry about those that are of no audible consequence.
At this point I'm completely unconvinced by Roger Paul's claims. I do not believe that frequency can be modulated by modulating intensity in an electronic circuit. And if such modulation is occuring on a low-level scale, I do not believe that it is of any audible significance. Based on some of Roger Paul's examples, I think it would be obscured by a well-established characteristic of human hearing called "masking". Briefly, masking refers to the ear's tendency to completely ignore a low-level signal that is close in frequency to a simultaneous high-level signal.
If Roger Paul is dealing with changes in gain of 1/100th of a dB or less (as he claims), then any hypothetical doppler-type frequency-bending (which I do not believe takes place) would be completely ignored by the ear.
Duke
dealer/manufacturer