All the TC drivers are amazing and top of the line.
No, they're not.
Some of them are excellent, but some of them are simply awful. It was in fact a horrible-sounding TC woofer (the Stryke Audio HE15) that led me down the path of concluding that low normalized inductance is a necessity for high-fidelity bass.
Not to mention the company's checkered history of insolvency (at times leaving venders such as John Janowitz holding the bag for defective parts, and taking money from venders such as Soundsplinter without supplying product - they've been through that cycle at least twice now), endemic problems with build quality (people having to re-machine bolt holes in LMS-Ultras lately, for instance), puzzling shipping decisions (e.g. leaving bags of washers improperly secured in VMP passive radiator boxes, thereby allowing them to move around and dent the cone), etc.
How do you handle the bottom frequencies having fewer producing sources?
Not sure what you mean. In the closed box system they all sum in phase below the modal region, so all contribute. The overall levels are different as well. I can't tell you which sub overloads first because I've never had any part of the bass system overload in the course of my listening.
In the system for which I showed measurements (all three subs had PR subs, which mean that they go dipole below their cutoff) I simply applied EQ to cut the bump of energy above the 40Hz range where the KEF HTB2's cut off.
TLS Guy and Gene for example think wide dispersion soft dome tweeters sound best. Linkwitz also uses soft domes for their vertical dispersion, they all go to live events as reference.
The Linkwitz approach, with a dome on each side driven out of phase from one another, does constrain directivity to some degree.
Ditto the dipole radiation of the BG Neo3 in John K's very interesting NaO Note design. (The NaO Note strikes me as the current SOTA in dipole speakers, along with AJ's fantastically cool big coax with monopole highs, dipole mids, and cardoid bass. I'd love to hear that speaker.)
Kevin Voecks uses metal domes rather than soft domes.
More salient, I think, is that the better Revel speakers employ shallow waveguides to match the mid and tweeter directivity in the crossover region.
Ds-21 prefers narrow directivity and compression drivers.
Not entirely true.
First, all of the mains speakers I currently own actually use domes. The current Tannoy Dual Concentric models don't really compression-load the tweeter, nor obviously does the KEF Uni-Q.
My views are basically identical to the ones you ascribed to AJ. (Actually, there is little daylight on most things audio between my views and AJ's views.)
Second, as I wrote earlier, my recent experience with the image thrown by wider-pattern (but still
consistent pattern over the midrange) has led me to question whether narrower is necessarily better. A consistent pattern in the midrange is, I think, absolutely required for a speaker to be considered high-fidelity.
He has viewed live events from a distance. He also hates rear wave radiation that Dennis and Linkwitz seem to prefer.
That's a bit strong. I do think that the benefit is not worth the placement constraints it imposes.
One speaker that impressed me greatly as a concept was one of AJinFLA's prototypes of the dipole, with his 12" coax and at the time just an 18" woofer, all dipole. It sounded fantastic. I would also never give up enough of my room to make such speakers work.
Is any single HF driver/setup possibly correct for all perspectives and instruments?
I think so, yes. Something that provides a consistent pattern and has declining sound power with increasing frequency will work well with all recordings, and optimally with good recordings.
The width of said pattern and the rate of decline in the sound power with frequency are areas where differences in room construction/finishing, listening distance, and listener preference come into play. Ditto vertical polars. If someone doesn't care what a speaker sounds like when they're standing, a more constricted vertical listening window is fine.
