HW bashing is just plain silly. OK, got that out of the way.
I heard the 3D arm in a VERY fine system that I know very well. It is mounted on a Classic 4, and that rig replaced a Forsell Reference; same cartridge (Koetsu). The owner does not want to deal with the complexity of the Forsell. Obviously, it is almost impossible to seperate the sonic change wrought by the arm from that of the table; and that of the particular synergy of each pairing. Having said, here are my impressions:
Tonally, low resonant signature is most certainly a characteristic of the sound that I heard. There is a striking reduction in the audibility of those narrow (and not so narrow) frequency bands that seem highlighted in relation to the rest of the frequency spectrum, and that results in sound that seems extremely well organized tonally. Soundstaging is VERY stable and precise; and perhaps as a result of the tonal evenness and purity (maybe), the soundstage is smaller front-to-back as well as left-right with smaller individual images; but, all extremely well organized. Overall, the sound is leaner with less weight.
But, and for me, this is a big "but". While the smaller soundstage does not bother me, the tonal aspects of the new sound do to a degree. I find there to be a sameness of timbre and tonal character to every lp played, with a subtle tonal politeness that I respected but didn't excite me. I don't understand this, since reduction in resonance should allow the differences in recordings to be more obvious, not less so. Live music is not always polite, it can and should sound downright nasty sometimes. Dynamic politeness is intrinsically linked to tonal quality, and particularly in the case of a turntable where speed stability is so important. Speed stability with the Classic 3 is first rate, but in spite of this the sound, for me, did not have quite the dynamic get-up-and-go and sheer explosiveness that I remember with the Forsell. Is it the table, the arm? I don't know. Very fine sound, but once again, system context is the key. Personally, I would not assume that reducing "resonant signature" in one component will necessarily yield sound that is closer to "real" without seriously considering system context.