I'm under the impression that anechoic chambers are considered to be not particularly enjoyable for music playback. Somebody please correct me if I'm mistaken.
From own experience, conducting quite a number of experiments inside an anechoic chamber, I would say your impression is accurate when we are talking about listening to music recordings in stereo without meaningful reverb. Meaningful in the sense that it is telling our brain something about room properties, tonal directivity of the instruments, depth-of-field and alike. The highlighted proximity is really annoying, like a row of mono horns yelling at you.
The picture changes, though, with acoustic recordings containing a meaningful reverb pattern of the concert venue. Would say you stop feeling like being in an anechoic chamber, except for the missing envelopment and ambience. The frontal stereo image can be very good, even showing decent depth-of-field.
In general I would say close-to-anechoic conditions are not a reasonable goal when listening to stereo. I am aware that some 30 or 40 years ago some people were having the opposite opinion, coming up with overdamped, very dry listening rooms.
optimizing for sufficient delay was better, both in sound quality and in spatial quality.
Can fully confirm this from own experiments. We should note, that adding discrete reflections coming in from a particular angle, is never a good idea, is you want to reproduce anything that's resembling the reverb on the recording. Delay is much better, although it might add a feeling of distance and reduced proximity to phantom sources, if reflections are too strong. That's typical for dipoles, if the wall behind the speakers is adding too much of a discrete reflection pattern. To counter this, one should reduce the listening distance, which is oftentimes not possible or kind of counterintuitive when listening to narrow radiation speakers. That was impression particularly with fullrange dipoles like Linkwitz 521, employing a rear-firing tweeter.
The delays were at least ten milliseconds relative to the direct sound, and the "non-optimum" arrival angle was from the front of the room, and moreso from above the speakers than from the same plane as the speakers. There were one or two wall-bounces in the reflection path so the arrival directions were kinda smeared.
Sounds like a good concept, but maybe discrete reflections still dominate the reverb pattern even if they bounce two- or threefold. Very much depends on the absorption pattern of wall and ceiling. What I found to be a good idea is combining rear-firing and lots of diffusion. It might even reduce the necessity for additional delay.
My personal method of double-checking this, is extensively listening to à capella tracks with several voices, alternating overly dry ones (like Barbershop style) with overly reverberant mixes (sacred choir music or alike). You get a pretty quick feeling for one of the mix styles being off the charts, either the close-mic´d, dry ones too close and annoying, or (which is often the case) the classical ones too distance, giving this ´I-am-sitting-in-the-last-row-of-the-cathedral´-feeling. If that makes sense, it is just my method.
Excluding rock music, of course. One of my test tracks is the opening track of ELP's Trilogy, with an enormously loud bass drum foot-roll by Carl Palmer. I use that to test the integrity of the foam surrounds on my Advent woofers.
