I also would suggest that your point about speakers not being able to be chosen based solely on measurements because most of us don't listen in anechoic or quasi-anechoic rooms is missing a key link its logical chain:
Thank you for your constructive and enlightening post!
I must apologise, I have done ‘the usual’, I put down part of my thoughts, taking the other part for granted.
I am happy to see measurements as done by the Stereophile, ASR, National Research Council of Canada, Princeton 3D3A Lab, etc. They are a given, I’d be sad if I woke up one day and there were no loudspeaker measurements.
Indeed, the response can be well predicted from the measurements for the average listening room (8 – 9 feet tall, between a square and a 1:2 rectangle, between 150 and 600 square feet, more often between 250 and 400, carpeted, 2x4 timber frame + drywall; in Europe that would be somewhat smaller size and more massive walls).
Where I have a problem is the room mode excitement by monopole speakers in mid- and upper bass region in a typical 8-8.5 feet tall, compact room. I had less of that in my previous habitats (had taller ceilings, then regular height with light ceiling or floor structures, till the current one - solid reinforced concrete slabs 8’3” apart. The vertical room mode is quite something, as are the reflections. That is why I mention dipole champions Gradient Labs Oy and S.Linkwitz, they adress the issue which is particularily problematic in my current living room.
Also, unless the situation has changed from J.Salmi‘s time (and nothing tells me it has), different loudspeakers which sound well and similar in anechoic conditions sound different in real-life flats and houses, some good and some less so.
This makes me wonder that if the speakers are not made to sound realistic in real-life rooms, yet wast majority of the loudspeakers are made according to the same scheme (dome tweeter, cone midrange and bass or just a midbass, bass in vented box … could it be just because this can be comparably easily calculated using Thiele-Small parameters?), what if we are ‘conditioned’ to accept that 2-2.5-3 way bassreflex box sound as “state of the art”, the correct way how the sound should bereproduced?
Most speakers are made that way for 30+ years, dipoles are mere oddballs. And our measuremnts pursue ‘ideals’ within the limits of 2-2.5-3 way bassreflex box paradigm. Can we even see/hear outside that bassreflex box?
For me the only way out is to go to a live performance. Not pop/ rock, as it is reinforced/interpreted through loudspeakers in the first place, just acoustic – voice, piano, strings, orchestra, opera (I do not listen to jazz too much).
Now, is there a good way to do that objectively, to objectively compare loudspeakers to a live performance? There must be, but I do not have it.
Remembering live performances to the best of my ability and trying different loudspeakers, subjectively in my room dipoles in bass seem more promising than monopoles.