CD² (Constant Directivity and Critical Distance)

[Arindal]

the desired room properties of a 2100 hundred-person vinyard shaped concert hall for original music production on the one hand, and a large-ish home listening room with a disproportionally large near field for recording-repro on the other hand, are diametrically opposed.

Absolutely true, reverberation time and initial delay time are necessarily completely different, and so is real source localization vs. phantom localization.

Nevertheless my personal take on the matter is we can learn something from concert hall and studio acousticians.

early reflections are detrimental, because they blur the image. We can't separate them from the direct sound when under 6-10 milliseconds, resulting in them being perceived as part of the direct sound, aka time smearing.

I am not so sure about time smearing for such early reflection. They do affect localization, but only if being discrete enough and localizable at the same time. If reflections in this early window (sometimes referred to as the ´Haas effect´) are sufficiently diffused in time and angle, the are more helping perceived dynamics and loudness and not so much deteriorating localization.

Late reflections less so, because we can separate them, recognizing them as a separate event (echo), which adds space. Not always good, but better than smearing.

It very much depends on the time, direction and pattern of the reflectogram. If it is a meaningful reverb pattern, as our ears would expect it from a cathedral or reverberant concert hall, it is actually not as much ´smearing´, if only the direct sound and very early reflections are consistent and dominant. The word ´echo´ implies a discrete, delayed sound event which our brain can perceive as kind of a second wave of direct sound, and it is very bad for any occasion (read about the Munich Gasteig Philharmonie disaster if interested).

Dipoles and Line Arrays (and various other speaker concepts) address both. Less total room reflections on account of their radiaton patterns, but much less early reflections and only somewhat less late.

As mentioned, my experience with dipoles is mixed. They might be a solution to keep the directivity constant and the indirect soundfield tonally balanced (as the late S. Linkwitz pointed out), but particularly at higher frequencies, they tend to add a pretty discrete, pretty early second wavefront, which is reducing the critical listening distance from a practical perspective significantly.

The openings are to the sides, so a dipole null will be present there?

Null is in the direction of both openings having the same distance from the listening/measureming perspective. To the left and right on the sketch.

But they sure work well as tweeters crossed at 700Hz up to some 18kHz without issue. Some of the best tweeters he ever tested, said Chua (ampslab-spk.com). Few artefacts, great HD, he only did not publish of axis SPL, so i will mail him if he has any data on that.

I don't doubt this, such planar tweeters can withstand a lot of power, too. I just see the difficulties when it comes to bringing such to constant directivity. The active wavefront area is just very huge, and applying waveguide or anything is rather complicated. Maybe I have not seen the right solution to date.

 

[Y~BNA]

I am not so sure about time smearing for such early reflection. They do affect localization, but only if being discrete enough and localizable at the same time. If reflections in this early window (sometimes referred to as the ´Haas effect´) are sufficiently diffused in time and angle, the are more helping perceived dynamics and loudness and not so much deteriorating localization.

OK. I see some issues on which the information i gathered during 15~20yrs of casually reading on audio tells me slightly different from some of the points you're stating. Here goes:

For the Haas-effect (aka self-explanatory term precedence effect) to occur, sonic events must be further apart in time than the the 6-15ms below which we'll perceive it as a singular event (frequency and context dependent, but i haven't come across Haas below 10-30ms).
When speakers are, say, 1.5m in front of a 'rear wall', at any distance in front of the speaker the rear signal will have to travel 3m more than the direct signal, so that is already at the lower limit of the 6-15ms. Any reflections bouncing forward from floor or side walls will have an even shorter travel time. Whether and under which circmstances that is audible my be depending on several factors, but it IS smearing, and if audible, IS detrimental.

The late reflections can be detrimental too when in excess, but time smearing is not their profession. Everything beyond some 10ms will subconsciously be perceived as separate and thus psycho-acoustically utilized (as spacial information, etc.). Dipoles in my experience are very effective that way, in large(ish) rooms (which btw acoustically are still 'small').

All this is according to the information i've come across and made myself somewhat familiar with. Toole, Angus, Linkwitz, etc. seem to agree IIRC, so i'll consider it innocent until proven guilty. Wold that be OK? If YOU prove it guilty, i will be very impressed of course!

It very much depends on the time, direction and pattern of the reflectogram. If it is a meaningful reverb pattern, as our ears would expect it from a cathedral or reverberant concert hall, it is actually not as much ´smearing´, if only the direct sound and very early reflections are consistent and dominant. The word ´echo´ implies a discrete, delayed sound event which our brain can perceive as kind of a second wave of direct sound, and it is very bad for any occasion (read about the Munich Gasteig Philharmonie disaster if interested).

So late reflections are not smearing, they provide spacial/directional (and more?) cues, especially when well imitating original-direct-sound-spectral-content. They can still be detrimental however, when outleveling direct SPL. The real echo in a Munich music hall is something else. Allthough late reflections are technically echos, in home situations (acoustcally small) they are only SUBconscously perceived as separate events (making them cues). AFAIK.

As mentioned, my experience with dipoles is mixed. They might be a solution to keep the directivity constant and the indirect soundfield tonally balanced (as the late S. Linkwitz pointed out), but particularly at higher frequencies, they tend to add a pretty discrete, pretty early second wavefront, which is reducing the critical listening distance from a practical perspective significantly.

Which is why i am considering using a 100mm wide 1200mm tall true line source above XO 700Hz (true below 8kHz on account of 45mm gap between units) in my next project. Their figure of eight will not be due to a dipole null but to cylndrical waveform. That usually falls apart above some 4~5kHz accordng to James Griffin, Merlijn van Veen, e.a. With the help of pressurized HF by a membrane behind small apertures vertically arranged, i hope to uphold it way into the second-highest octave, however. This is where you could definitely shoot me, but i am still planning to give that a try.

Null is in the direction of both openings having the same distance from the listening/measureming perspective. To the left and right on the sketch.

So functions as a true dipole like the MTM section? But summing and enjoying W-frame resembling efficiency improvement by the baffle? Did i understand you correctly?

I don't doubt this, such planar tweeters can withstand a lot of power, too. I just see the difficulties when it comes to bringing such to constant directivity. The active wavefront area is just very huge, and applying waveguide or anything is rather complicated. Maybe I have not seen the right solution to date.

They've been found testing 'outstandingly' by Michael Chua (ampslab-spk.com), "the best tweeters he ever tested". That sounds promising. I have twelve of 'em lying around at present. I came across some articles claiming the size of the membrane being much larger than the aperture surface some pressurizing would take place (as in horn tweeters), making the apertures into new original sound sources (as in horn tweeters).
I very much hope this is true, as you can imagine, because 6mm wide original sound sources would have a wildly wide dispersion pattern, which i would have to narrow back horizontally, using a waveguide if the true line source cylinder does not contain it!

Pfff...

 

[Y~BNA]

Your goals are a mix of already-achieved and impossible, in one case.

The latter sounds like an interesting challenge!

In a normal listening situation, direct sound will always be louder than early reflections, and late reflections will always outnumber, but never be louder than, early reflections.

That doesn't sound right to me. When using an open dipole in a mid-sized room, early reflections will be less and late reflections will also be somewhat less than with conventional speakers, but not as much less as said early reflections. Said late reflections will be dominated by those late enough to provide spacial cues, and their combined SPL will always be higher than the early reflections combined SPL. This is almost like "that's what i learned at school", only i learned it somewhere else. How do you figure the earlies will always be higher SPL?

The spectral content of late reflections depends on the design of the speaker foremost and the room second, although most home spaces tend to act the same way. I agree constant directivity or similar designs make the most sense, particularly if they maintain pattern control down to around 100Hz. Although, frankly, many speakers sound very good even if they deviate as long as the overall radiation pattern is relatively even and there are no resonances.

Not from experience, but from literature, i also gathered that if constant directivity is only approached to a reasonable degree, listeners already praise it for their clear and articulated sound.
Dipoles add the advantage, in your mentioned pattern control, that they radiate 4.8dB lower energy (so also reflected energy) compared to a monopole with same on-axis SPL. They also excite less room modes, so all around better controlled than boxed speakers.
In the HF it's another story, but i'm gonna try to address that by applying a dipole true line source from 700kHz and up, as described in previous post (right above this one).

You can manipulate the strength of early reflections with absorption panels and do something similar for late reflections by using secondary highpassed speakers firing away from you and into the walls and ceiling, preferably at an angle.

I plan on doing that, but only after exploiting all options available in the speaker design.

The ideal speaker doesn't care too much about the room. If i can maintain a null with its gradually diminishing SPL shade on both sides of the 90 degree plane, i will have removed a lot of reflections and especially a lot of early ones. plus the floor- and ceiling reflections will be part absent. Hopefully large part.

That part of the impossible you mentioned? Even getting it to work up to 7~8kHz i will be one happy camper!

Tx for letting us use your brain!

 

[Blockader]

The ideal speaker doesn't care too much about the room.

You can only make a speaker “room-independent” to any meaningful degree by making its directivity extremely narrow from roughly 80 hz and up. But that’s not some universally 'ideal' speaker, it’s a speaker built around one very specific goal.

And that goal comes with its own compromises. The real engineering move isn't trying to delete the room or chase an abstract ideal, it's figuring out what the design actually needs to do, then choosing a concept that hits those requirements while keeping the trade-offs under control.

One of the first requirements you have to nail down is directivity: how wide do you want the beamwidth to be? Then comes the second most important question: what max SPL are you aiming for?

 

[Arindal]

I can see a broad baffle lowering the baffle-step frequency, but it's not obvious to me how it would contribute to "dominant early reflections from the side." Can you elaborate, or give an example?

My point is that a particular direcitivty index does not allow any conclusion, which angles and in which windows you actually have a lot of SPL and in which you have attenuation. A dipole and a cardioid, for example, share the same idealized directivity pattern, but a dipole has the ideal null at +-90deg, the cardioid at 180deg. An omnidirectional source plus broad baffle have a polar pattern which is not equal to a cardioid of the same overall power d.i. - in idealized form sound would only be emanating to the frontal 2pi sphere, widely un-attenuated, and then sharply fall once you get to angles of 100deg+.

I am referring solely to practical experience with concepts employing a broad-radiating tweeter in its lowest bands, which is forced by a broad baffle from an almost omnidirectional directivity towards a half-sphere pattern, i.e. in the region of +-90deg. The energy which would usually be distributed to 90...180deg, is kind of reflected to the frontal hemisphere, but not equally and particularly the frontal listening window is weaker than expected from an ideal cardioid.

A good example is the Grimm LS1 which is an 8" plus 1" broad-baffle concept with pretty low crossover freq (in the region of 1.6K), so the tweeter despite from being equipped with a waveguide, shows a vast band of typical broad-baffle behavior. This leads to a pretty high SPL at +-60...100deg in this band (2-4K), leading to an unusually high level of calculated side-wall reflections:

Note the grey line. A high level of side-wall reflections is contributing to inferior localization stability, and in this case all early reflections are also particularly colorated due to inconsistent directivity.

This will most likely not be an issue under studio conditions with more of absorption or well-integrated diffusion. Concepts that employ horns, line sources, large fullrange drivers or dipoles to control mid-high directivity, are also not affected. They are by concept attenuating SPL in side and rear windows, so the broad baffle has not much to boost to the sides and is mainly used to shift the baffle step to lower frequencies or milden its effects.

How is the broad-baffle GGNTKT M3, which you mentioned earlier, in this respect? Seems to me its 140-degrees-wide horizontal pattern could be a bit too wide in some situations.

My experience with this one under different conditions is limited. Judging by the isobaric graphs provided by the manufacturer, I would not consider this to be a problem. While they claim to have a broad listening window of +-70deg horizontally (-6dB), the SPL is decreasing sharply and evenly across all relevant bands, the moment you leave this window. If you apply the same standards on the Grimm, we would be talking about roughly a +-100deg window.

If you look at the attenuation at 90deg (h) in the lowest band of the tweeter (around 2K), the Grimm shows an attenuation of in the region of -5dB which is still pretty loud, the GGNTKT in the region of -10dB. Reason being the combination of cardioid and horn at play, which are anyways keeping directivity index comparably high. So, in theory side wall reflections can be a bit high in case the speakers are in parallel orientation to the wall behind them and side walls are closeby. But my guess would be the speakers are not meant for this and some toe-in can easily be applied the moment side wall reflections deteriorate localization stability.

Would assume that GGNTKT is not using the broad baffle as some kind of ´180deg waveguide´ here, surely not in the treble bands, but it is a measure to ensure sufficient delay between front and rear woofer and maintain efficiency for the frequency bands they are cancelling each other out.

 

[Arindal]

Regarding Haas-Effect:

The threshold for precedence is indeed somewhere in the approx. 30ms region, but please note these experiments have been conducted with real source localization and more or less spectrally identical, broad-band signals. That is not the case with stereophony, where we are dealing with phantom sources (at times midrange and tweeter not at the same position), combfiltering effects caused by very early reflections, and oftentimes signals with spectral differences and additional early reflections (side walls, ceiling, floor, mixing console). Not much of imagination is needed to picture a signal which will this or that way lead to a ´smeared´ localization, as for example in case of some pronounced sibilants, the early reflection from the side wall ´overrides´ the tweeter + ceiling + desk localization.

The late reflections can be detrimental too when in excess, but time smearing is not their profession. Everything beyond some 10ms will subconsciously be perceived as separate and thus psycho-acoustically utilized (as spacial information, etc.). Dipoles in my experience are very effective that way, in large(ish) rooms (which btw acoustically are still 'small').

Just from practical experience, I found diffusion in time and angle always helps keeping reflections subtle even under conditions beyond the critical listening distance. A true dipole adds one source of early, discrete and localizable wavefront, and particularly in the treble bands the existing concept like Linkwitz 521 are rather bipoles than dipoles, as wavelengths become to short and tweeters are to far apart to ensure perfect cancellation unter 90deg.

So late reflections are not smearing, they provide spacial/directional (and more?) cues, especially when well imitating original-direct-sound-spectral-content. They can still be detrimental however, when outleveling direct SPL.

That is more or less matching my experience, yes. Under home listening conditions ´outleveling direct SPL´ might be a common thing, so it is in my understanding important to keep the added reverb pattern subtle and ´hiding behind the reverb on the recording´.

Which is why i am considering using a 100mm wide 1200mm tall true line source above XO 700Hz (true below 8kHz on account of 45mm gap between units) in my next project.

Sounds like a pretty large diaphragm area to me, making it very difficult to keep directivity constant. 45mm gaps between tweeters to me sound way too far apart for 8K, you might be facing some issues with lobing or ´phasey´-sounding treble. Only experiments can show. I would suggest to use only one tweeter in the highest two octaves, i.e. 4K upwards, and blend in the others towards lower frequencies in order to maintain the desired vertical pattern.

Have heard a presentation by Erik Wiederholtz as of Perlisten who was doing something very similar with several dome tweeters, forming a miniaturized line array and waveguide at the same time. If I recall it correctly, he mentioned something like a gap of 20mm between the domes and a threshold frequency of 5K, above which he insisted on having only the central tweeter active.

Nevertheless, your concepts sounds very promising, magnetostatic planar tweeter have proven to work excellent in a line source for sound recinforcement, and I would really like to know if you make any progress. All the best!

So functions as a true dipole like the MTM section?

Yes.

 

[amirm]

Isn't Constant Directivity the holy grail of speaker building? Read Toole,

No. There is no requirement for directivity to be constant. From Dr. Toole's book:

"A good loudspeaker for this purpose would therefore be one that has two
qualities: wide dispersion, thereby promoting higher levels of reflected sound,
and a relatively constant directivity index so that the direct-sound and reflected sound
curves have similar shape."

We want off-axis response to be similar to on-axis but it can narrow down as frequencies climb.

You actually made a grander statement: "i am hearing musicians, not a loudspeaker-cone in a coffin"

There is no such thing in any research. This is marketing material promoted by people selling dipole, etc. types of speakers. No research backs your claim with these speakers which I find to always sound spatial whether that is in the source or not. And at any rate, a recording has nothing to do with a live session with musicians.

Keele, Vanderkooy, Beranek, Angus, Linkwitz, etc. etc. I could go on for quite a while really. Anyone who has published on loudspeakers in rooms basically.

Please don't make claims like this without direct quotes. I know of no researcher claiming what you said above in quotes.

 

[amirm]

Doesn't managing the room help, by having less reflected energy in the first place?

No. Ultimate dead room is an anechoic chamber or outdoors. Neither sounds as good as a real concert hall which is full of reflections.

 

[Arindal]

wide dispersion, thereby promoting higher levels of reflected sound,

As a wide dispersion is also contributing to a higher level of early reflections, particularly side-wall reflections, how is this going together with keeping localization stable and precise? Or is a precise localization not a goal of excellent sound quality? I agree to your claim that recordings are not meant to sound like a live session, but recordings do have their own qualities of localization, and under studio conditions localization provided by the monitors is excellent in most of cases, so why should home listening deviate from that? You yourself say that ´sounding spatial whether that is in the recording or not´ you consider being a flaw.

We want off-axis response to be similar to on-axis but it can narrow down as frequencies climb.

Sounds like a ´contradictio in adiecto´. If off-axis response is deviating from direct sound response more than just a bit, treble bands will inevitably be attenuated by several decibels compared to bass and lower midrange, which in my understanding does not meet the definition ´similar response´ anymore. From my experience, this constitutes a colorated reverb field in the listening room, hence ´dull´ tonality of the reverb, as well as the impression of envelopment changing.

The logical conclusion would be that neutral reverb tonality in your understanding is not necessary for sound quality, or am I getting anything wrong here?

 

[amirm]

As a wide dispersion is also contributing to a higher level of early reflections, particularly side-wall reflections, how is this going together with keeping localization stable and precise?

Have you ever listened to live music? Is it localized and precise?

 

[Arindal]

Have you ever listened to live music? Is it localized and precise?

Depending on the concert hall and seat in question, yes it is. Not only from the conductor´s perspective which I had the chance to listen from several times.

 

[Shorty]

Depending on the concert hall and seat in question, yes it is. Not only from the conductor´s perspective which I had the chance to listen from several times.

Depending on the concert hall and seat in question, for the majority of seats it isn’t. In my experience.

 

[Arindal]

Depending on the concert hall and seat in question, for the majority of seats it isn’t. In my experience.

I recommend to attend a performance in one of the venues offering excellent localization and transparency over a considerable number of seats. Have already recommended two in this thread (one philharmonic concert hall and an opera theatre).

Even if you are talking about a hall which regrettably does not offer excellent localization for many seats, I see little reason why the reproduction of a recording under home conditions should be worse than the standard the mixing engineer has set in his or her control room. Which in the vast majority of cases I am aware of, is aiming at the best seats in the hall, sometimes the conductor, and in some cases even better than those two (an aesthetic goal of classical recordings which was widely discussed in the 1970s and 1980s).

 

[audiofooled]

In rooms ceiling bounce can be quite problematic and IMO effects of it are usually underestimated. Vertical directivity may be just as important as horizontal, caveat being that, preferable horizontal dispersion width and frequency range within which it should be constant or not necessarily so is still going to be a subject of debate, whereas vertical control in CBT's may be possibly recognized as preferable, with a matter of some research.

An example of virtually no ceiling bounce to speak of:

JBL CBT 70J-1 Review (Constant Beam Transducer)

This is a review and detailed measurements of the special JBL CBT 70J-1 (upper unit) constant beam transducer design. It was kindly purchased new and shipped to me by a member. It costs US $1047 on Amazon including free delivery. You can get it in black or white colors. Note: our company...

www.audiosciencereview.com

@amirm 's comments at the end of the review say a lot about in room listening experience one may be getting if this is the case.

 

[Arindal]

Vertical directivity may be just as important as horizontal

Agreed to the statement that vertical directivity might be just as important as horizontal, but for a different reason. Flaws in vertical directivity bear a higher risk of causing imbalanced reverb tonality, as ceiling and floor reflections are usually reaching the listener earlier, and are more likely to be a product of angles (as seen from the loudspeaker axis) closer to the listening window. In contrary, horizontal directivity is more important when it comes to localization, as our ears are much more sensitive to horizontal localization compared to vertical one.

preferable horizontal dispersion width and frequency range within which it should be constant or not necessarily so is still going to be a subject of debate, whereas vertical control in CBT's may be possibly recognized as preferable, with a matter of some research.

As in most situations, a more or less broad horizontal directivity is desirable, as there might be several listeners, or the listener might be moving his/her head within the listening window, in cases a generally higher d.i. is necessary, narrower vertical dispersion is the way to go. With conventional loudspeakers concepts, it is simply more difficult to achieve this, with the exception of line source or line arrays.

An example of virtually no ceiling bounce to speak of:

Pretty narrow vertical dispersion indeed, but the vertical listening window is narrowing down even more above 7K. And it comes at the hefty price of horizontal lobing including a significant step in directivity, midrange resonances and treble being subject to almost chaotically lobing due to many dome tweeters. I have not listened to the speaker in question, but have heard several similar concepts delivering a pretty unsatisfying upper treble.

What astonishes me is the overall directivity index is just around a conventional 7 dB. Would have expected it to be much higher as a result of narrow vertical dispersion, but obviously the horizontal dispersion is unusually wide over vast frequency bands.

 

[Suono]

Poiché un'ampia dispersione contribuisce anche a un livello più elevato di riflessioni iniziali, in particolare quelle sulle pareti laterali, come si concilia questo con il mantenimento di una localizzazione stabile e precisa? Oppure una localizzazione precisa non è forse un obiettivo per un'eccellente qualità del suono? Concordo con la tua affermazione secondo cui le registrazioni non dovrebbero suonare come una sessione dal vivo, ma le registrazioni hanno le loro qualità di localizzazione, e in condizioni di studio la localizzazione fornita dal monitor è eccellente nella maggior parte dei casi, quindi perché l'ascolto domestico dovrebbe discostarsi da questo? Tu stesso affermi che "suonare in modo spaziale, che sia presente o meno nella registrazione" lo consideri un

case by case when dealing with the first early reflections...the solution can never be the reduction of the emission amplitude

 

[Duke]

My point is that a particular direcitivty index does not allow any conclusion, which angles and in which windows you actually have a lot of SPL and in which you have attenuation. A dipole and a cardioid, for example, share the same idealized directivity pattern, but a dipole has the ideal null at +-90deg, the cardioid at 180deg. An omnidirectional source plus broad baffle have a polar pattern which is not equal to a cardioid of the same overall power d.i. - in idealized form sound would only be emanating to the frontal 2pi sphere, widely un-attenuated, and then sharply fall once you get to angles of 100deg+.

Agreed - the actual pattern shape matters, not just the calculated DI number. From which directions (and when) are significant first reflections arriving? It depends on the pattern shape and in-room set-up.

I am referring solely to practical experience with concepts employing a broad-radiating tweeter in its lowest bands, which is forced by a broad baffle from an almost omnidirectional directivity towards a half-sphere pattern, i.e. in the region of +-90deg. The energy which would usually be distributed to 90...180deg, is kind of reflected to the frontal hemisphere, but not equally and particularly the frontal listening window is weaker than expected from an ideal cardioid.

Yes, that makes sense - THANK YOU!! The wide baffle constrains the pattern width to 180 degrees at the lower end of the tweeter's range, BUT there is still an EXCESS of energy at large off-axis angles in that frequency region.

Just from practical experience, I found diffusion in time and angle always helps keeping reflections subtle even under conditions beyond the critical listening distance. A true dipole adds one source of early, discrete and localizable wavefront, and particularly in the treble bands the existing concept like Linkwitz 521 are rather bipoles than dipoles, as wavelengths become to short and tweeters are to far apart to ensure perfect cancellation unter 90deg.

Agreed.

Sounds like a pretty large diaphragm area to me, making it very difficult to keep directivity constant. 45mm gaps between tweeters to me sound way too far apart for 8K, you might be facing some issues with lobing or ´phasey´-sounding treble. Only experiments can show. I would suggest to use only one tweeter in the highest two octaves, i.e. 4K upwards, and blend in the others towards lower frequencies in order to maintain the desired vertical pattern.

Have heard a presentation by Erik Wiederholtz as of Perlisten who was doing something very similar with several dome tweeters, forming a miniaturized line array and waveguide at the same time. If I recall it correctly, he mentioned something like a gap of 20mm between the domes and a threshold frequency of 5K, above which he insisted on having only the central tweeter active.

Imo problems arise when combining line-source-approximating behavior with point-source-approximating behavior over different frequency regions in the same loudspeaker. As you know, SPL falls off by 6 dB per doubling of distance from a point source, but only by 3 dB per doubling of distance from a line source. So not only will the frequency response of the direct sound change with distance, but the reflection field's spectral balance will be dominated by the frequency region where line source behavior is most closely approximated.

Have you ever listened to live music? Is it localized and precise?

If I understand David Griesinger correctly, listening distance is a major factor in whether or not sharp localization of sound images occurs in a live music setting; once you are even a few feet beyond a certain distance, which he calls the "Limit of Localization Distance (LLD)", sharp localization collapses.

So imo the answer to your second question would be, like for so many other things in audio, "it depends..."

@tmuikku has written about the application of this principle to home audio. Paraphrasing, he finds that there is a listening distance at which image precision transitions from being precise to being noticeably less precise, and this distance can be found by ear.

Here's a presentation on the subject by David Griesinger; he's not talking about home audio but imo the information is still relevant:

Of course whether or not sharp image localization is a priority in a home audio setting, and where it sits in one's hierarchy of priorities, is a matter of personal preference.

 

[Suono]

Of course whether or not sharp image localization is a priority in a home audio setting, and where it sits in one's hierarchy of priorities, is a matter of personal preference.

In high fidelity, I think it's a priority. The more a system focuses the stereo image and solidifies it, the more faithful it is to the recording and its tonal balance. Focus cannot be considered an extra, nor a peculiarity subject to personal taste.
The more solid it is, the more guaranteed the reproduction of information is

 

[Duke]

They've been found testing 'outstandingly' by Michael Chua (ampslab-spk.com), "the best tweeters he ever tested". That sounds promising. I have twelve of 'em lying around at present. I came across some articles claiming the size of the membrane being much larger than the aperture surface some pressurizing would take place (as in horn tweeters), making the apertures into new original sound sources (as in horn tweeters).

I very much hope this is true, as you can imagine, because 6mm wide original sound sources would have a wildly wide dispersion pattern, which i would have to narrow back horizontally, using a waveguide if the true line source cylinder does not contain it!

Have you looked at the factory data sheet? It shows the off-axis response being in the same ballpark as a 1" dome tweeter, albeit closer to constant-directivity in the top octave than would be typical for a 1" dome. Anyway, the pattern narrows north of 4 kHz or so by much more than one might expect from a single 6mm-wide aperture.

On the other hand imo the horizontal radiation pattern looks a bit wider and more uniform than one might expect from the planar diaphragm without the apertures, so I think they are making a worthwhile improvement in the radiation pattern.

... my project is not based on a listening position, but rather on illuminating as good as possible a rather large open plan room where people live, eat, cook, sit in the couch and listen to music during much of that. People who love music so much they don't have enough time sitting in a sweet spot in their listening room..

Cheers!

... i hope the direct sound outlevels the reflected sound (by a lot, preferably), the late reflections WAY outnumber and/or outlevel the early ones, and said reflections tend to uphold the spectral content of the direct sound. Until i find a better definition, these three will be among the things i strive for.

I think your goals are good and your overall game plan is good, and even if you don't achieve all of your goals perfectly I think you'll still end up with a really sweet system. Don't get overly caught up in what the Directivity Index number is; imo what you want is a long enough time delay between the arrival of the direct sound and the strong onset of reflections so that they don't perceptually "fuse". And imo 10 milliseconds is long enough (though more would be better).

I do have a couple of thoughts, and if they're not useful just ignore them:

First, I think you will definitely need waveguides on your stack of planar drivers to avoid having far too much early reflection energy off the same-side walls. I suggest making your best estimate of how much toe-in you can get away with, and then choosing your waveguided radiation pattern width such that you are not strongly "illuminating" that same-side-wall. Imo it would be desirable for your first strong horizontal-plane reflections to be the long-path, across-the-room reflections off the OPPOSITE side wall. So I'm thinking your target radiation pattern width might end up being in the range of 60 to 90 degrees wide, depending on the specifics of your room and speaker placement options..

Second, what are your plans for the backwave? If you are able to position the speakers at least 5 feet out from the front wall, you can probably let the planar array's backside disperse as wide as it pleases. That being said, imo there is an argument for waveguiding the backwave as well, such that when you "aim" the front wave you are simultaneously "aiming" the backwave.

I wish I had a solution for the vertical spacing issue.

You might make some triangular cardboard cut-outs representing different radiation pattern widths and set them down on the floor where the speakers will go and play around with different "toe-in" angles, to help you visualize what it will take to avoid strong early same-side-wall reflections. Imo THOSE are the enemy that would thwart your plans.

 

[amirm]

In high fidelity, I think it's a priority.

It controlled testing, listeners tend to like the diffused imaging they get with wider dispersion, instead of the more focused ones. This is why so much of "audiophile music" is "wet" with added reverb and such. It is a cute effect when that kind of imaging happens but it comes at the expense of a wider sound stage. Indeed, this is why people like dipoles as that takes this concept to an extreme.