The problem with the ’mechanical engineering’ view is when you define a "goal" or a "standard" then it can easily be seen if that goal is met and an unassailable judgement made. No consideration is given to the absolute magnitude or the underlying importance of the goal to begin with. As Lew pointed out above....if vanishingly low distortion is set as the goal then the system with the lowest distortion must always be best.
Regarding the distortion comment (since I can’t seem to just let that lay there...), it matters what kind of distortion it is. This is part of why the measurements are important- some kinds of distortions are unpleasent and other kinds are innocuous to the ear. I’ve talked probably too much about this elsewhere so won’t go into it more than that.
Regarding your comment about ’no consideration is given...’ as far as I can make out this simply isn’t true. What is true is we have a lot of designers that lack education that would be useful in their field. Lacking that they rely on stories to get around the elephant in the room. Human nature being what it is, often we can be convinced to believe those stories too.
what happens when the modulation is so extreme that it is actually considered mistracking? Is it at all possible that not being 100% rigidly coupled could cause the recovery from mistracking to be more benign? Now consider that mistracking is not an all or nothing type of proposition and is constantly happening and I seriously have to question the importance of the rigid coupling anyone demands. Everyone is allowed their choice of compromise and the choices by Viv Labs and the supporting anecdotal information is really interesting to me. Productive discussion about the factors involved is helpful to all and it seems at least plausible to me that the lack of rock solid coupling and anti-skate could make say 5° of tracking error on the Viv Labs more sonically benign than a lesser amount of TAE on a ’traditional arm’. One other thing that strikes me as odd in this all is that if TAE is truly the sole arbiter then why do the shorter underhung arms seem to be preferred?
Having run an LP mastering studio I can tell you that the limitation of modulation in the groove and the most of the distortion of the LP are all in the playback side rather than record. This places the performance (or lack thereof) almost entirely in the hands of the end user.
Mastering engineers know this, and so they really don’t put so much modulation in the groove as to cause a reasonable pickup to mistrack (we used an old SL1200 with a Grado Gold for our ’reasonable pickup’; if it could track the cuts we were making we knew we were in good shape). That’s a pretty old lesson, going way back to the Living Stereo era, when RCA cut the original Pines of Rome that tended to knock the rather primitive pickups of the time right out of the groove. Put simply, what you propose in your first question isn’t a thing.
If the coupling isn’t there as I have talked about, one of the results is more mistracking. IOW it works opposite of what you propose. This is simply because the arm is putting more energy into the cantilever.
Most audiophiles I know really want to get as close to the musical event as they can. The idea that the rig is designed to not do that is anathema. At any rate if the system has the rigid coupling (and deadness) as I wrote about, one thing that is instantly audible is how much better the bass is, which, if compared to CDs or RtR tape of the same recording, will be shown to be every bit as good, much to the chagrin of both camps’ advocates. But its more than that, with the rigidity also comes a more transparent midrange and smoother highs (particularly at volume), since its less susceptible to airborne vibration. This is one of those things that is not just easy to measure; its also easy to hear.
Regarding your last question, how do you know that underhung arms are actually preferred? Do you know of a poll regarding such??