If this was actually an issue, it would have shown up in the research when they were looking at the differences in listener preference between mono and stereo listening.
It would, if anyone would do such research with focussing on tonal balance preference with modern loudspeakers today. But that was never the case. Floyd Toole explained that the underlying research leading to his mono theory were complete in 1985 (presumably with more or less imperfect loudspeakers) and the most recent verification in 2008 was also comparing loudspeakers with significant flaws, focussing on the discrimination effect under which conditions the flaws were most likely to be detected (if I understood Dr. Toole correctly).
This is why every music producer and broadcaster thoroughly checks to confirm the mono mix sounds good - they know that true stereo sweet spot listening is extremely rare.
Cannot confirm this from decades of experience in pro audio, broadcast tech research and related fields. Mono compatibility checks are surely executed (for FM radio and vinyl mastering this is mandatory), but they are mostly meant to make sure a mono summation downmix would not lead to cancellation effects, phasing or lack of bass/lower midrange. It would come as a total surprise of mono monitoring would be taken as a reference to tonal manipulations, particularly above 1K, as this would almost certainly lead to degrading stereo tonality if any issues were found. You cannot have both perfect.
While it is true that music is vastly consumed in mono or under quasi-monaural conditions (such as portable bluetooth speakers), I don't think this represents critical listening or is to be taken as a reference. Most of people would not bother to hear slightly equalized vocals in mono, maybe would not even judge accurate tonality or notice the difference.
I forgot to include the link to the post for the Salon 2 and Array 1400 measurements:
The Array is definitely off the tolerance band that would define more or less constant directivity, expected to show exactly the outcome I have described (and experienced in exemplary manner with an Array 800 model back then).
Revel is difficult to judge, and I have never listened to it in a room that would allow any judgement or a definitive answer to the question if the outcome of uneven directivity could be equalized to satisfaction or not.
Would it be possible for you to demonstrate the results for other participants in the tests, along with the speakers in question and their measurements?
Such test should be doable and I would love to contribute details the test design. I would suggest to compare two speakers which are very similar in on-axis response, size, recommended listening distance and geometry while showing significant differences in off-axis behavior.
Presumably the relative timing of these spectrally colored reflections would matter with respect to the so-called precedence effect and fusion interval,
Certainly accurate. We should take into account that the precedence effect is already at play with the first sound events forming the direct sound, so our brain will compare the tonally of events to follow with the tonal signature of that first wavefront, if that makes sense. Depending on delay and meaningful tonal differences (meaningful in the way that either is matches the outcome of differently reflective surfaces or the HRTF with sound coming in at different angles), our brain would try to create a full picture of where the reverb came and how the room looks like. This is best judged with transient sound events in a certain time window after the initial event has decayed (so it would not continuously mask the reverb events making ´the reverb disappear´, as member
@gnarly has put in when trying the test tracks I had suggested).
since truly diffuse sound and reverberant field would seem to be very challenging, if not virtually impossible, to achieve in typical listening rooms.
In studio control rooms, this turned out to be achieveable. Under home listening conditions it is a bit more difficult and might require further attention to speaker directivity, listening distance, diffusive elements on walls near the loudspeakers. The idea is not to attenuate the reverb in the listening room completely (as it was done in studios some decades ago), but to bring it to a level and congruence with the direct sound that it would not deteriorate localization stability nor perception of the reverb in the recording.
Could you please provide some examples?
MEG and PMC were actively promoting this concept, but it is not hard to see that other studio monitor manufacturers are seemingly following a similar idea, for example Adam Audio, Kii Audio, PSI, Eve and others.
What do you consider to be examples of "true constant directivity speaker"?
The most obvious examples are those actively promoting the concept or being founded on that idea: Kii Audio, GGNTKT, Dutch&Dutch, Linkwitz, the smaller MEGs and all those being into curved line sources (more or less following Keele´s ideas). Interestingly despite from Linkwitz there are quite some manufacturers of dipoles coming to a similar result solely in terms of directivity (I do not mean to recommend a specific brand or modeL): Martin Logan, Magnepan, Quad ESL, Spatial Audio, Ecouton.
Interestingly, some hi-fi or high end audio manufacturers seemingly follow this idea, at least in the most important frequency bands, without making much noise about it. As mentioned, for me the TAD Labs models with bigger coaxial driver were the initial eye-opener what this concept can really achieve and why I found it so superior compared to continuously increasing directivity models. There are actually lots of different speakers offering constant directivity in a limited band, e.g. 0.8-7K, leaving it to the listener to decide if the bands below or above can be equalized accordingly without overly compromising the direct sound tonality. Examples by Genelec, TAD Labs, Elac, Magico turned out to fall into this category.
my father was a sound engineer at SR (Swedish public service radio ) and later at SVT ( Public service TV ) and then had his own company .he later became an video editor.
