Beyond Linearity: Why Speaker Dispersion Matters Far More Than People Expect

[Lion♡]

For those participating in the thread, at approximately how many milliseconds do your initial first reaction times fall, and how are they distributed in terms of intensity?
It would be good if you could also share that information.

 

[Old_School_Brad]

But likewise I expect reflected sound to be less bright than direct sound, because that's how sound works in nature. Early reflections that call attention to themselves with a noticeable peak well above the bass region sound unnatural to me.

My Revel F12's have the intentional wide, smooth directivity that was Revel's design objective from the beginning. I can sit in the sweet spot and point to the instruments with my eyes closed, but I can also do that if I move over a couple of feet. That wide directivity did nothing to undermine staging clarity, it seems to me, but it does seem to enlarge the sweet spot usefully in my experience.

Rick "learned about staging after buying those Revels" Denney

There might be some elements of psychoacoustics or expectation bias at play here.

Wide dispersion inherently leads to more reflections, and in most typical rooms, that tends to blur imaging precision compared to speakers with controlled, narrow directivity.

 

[Heinrich]

What I read here recently is based an the expectation, that stereo should offer an illusion. There is nothing more wrong in the business as that. A recording is to be perceived as an abstraction like a painting (always) is. In case yo do not accept this, you will eventually, granted. For the peace of mind, for the sake of logic.

 

[custom]

For those participating in the thread, at approximately how many milliseconds do your initial first reaction times fall, and how are they distributed in terms of intensity?
It would be good if you could also share that information.

It would be interesting to see a REW impulse graph of two loudspeakers with different dispersion in the same room.

 

[anphex]

Thank you, please L and R separately for each.


Also averaging enabled:

View attachment 439989

I used exponential averaging with a factor of 0.88 and 24th-order smoothing—aren’t these results sufficient?

 

[Sokel]

I used exponential averaging with a factor of 0.88 and 24th-order smoothing—aren’t these results sufficient?

You don't need smoothing, what you need is to change the setting above that, from Spectrum to RTA 1/48 and do L and R independently.
Check @thewas 's post for the right settings.

 

[Duke]

One of the key features of a "good" spinorama is that the spectral response off axis is naturally related to the response on axis. There is an expectation of a downward slope that increases as the measurement moves away from the axis, but that is realistic. High frequencies tend to be absorbed in reflective materials more readily than low frequencies, especially in the presence of acoustic treatments.

Something which has not been extensively explored is, what happens when there is very little spectral discrepancy between the first-arrival sound and the reflection field. I presume this has not been deliberately studied because so few loudspeakers exhibit this sort of behavior, due to pattern-narrowing as we go up in frequency being so ubiquitous. Still, here and there we find speakers whose dispersion effectively minimizes the spectral discrepancy between the first-arrival sound and the reflection field, such as the JBL M2, Dutch & Dutch 8c, and @sigbergaudio's lineup.

FWIW, in my opinion there are both sound quality and spatial quality benefits arising from dispersion characteristics which minimize the spectral discrepancy between the direct sound and the reflection field.

 

[Lion♡]

It would be interesting to see a REW impulse graph of two loudspeakers with different dispersion in the same room.

You're right. You can see differences even in the impulse response, and I remember seeing a significant difference due to the radiation characteristics of the Revel compared to the Arendal in a previous user's measurements.
Of course, it would be even better to observe the impulse response in a binaural measurement.

 

[boxerfan88]

FWIW, in my opinion there are both sound quality and spatial quality benefits arising from dispersion characteristics which minimize the spectral discrepancy between the direct sound and the reflection field.

This is quite an interesting area … if the direct sound spectra & reflected sound spectra is the same/similar … what would happen?

Would the room impact reduce/disappear?
Would the soundstage become bigger?
Hmm…

 

[Duke]

This is quite an interesting area … if the direct sound spectra & reflected sound spectra is the same/similar … what would happen?

Ime the timbre of voices and instruments seems more natural and the presentation seems more relaxing over long listening sessions. Also, the sound quality tends to be more consistent throughout the listening area, as the perceived spectral balance doesn't as change much with variations in listening axis and/or distance.

Ime the spatial quality has a better chance of being "you are there", wherein the sense of space on the recording is dominant; as opposed to "they are here", wherein the playback room's spatial signature is dominant. Of course other factors (not the least of which is the recording itself) are involved.

Would the room impact reduce/disappear?

Room interaction still matters enormously, and imo the earliest room reflections are especially strong conveyors of the playback room's "small room signature", so having similar direct & reflected spectra is not a spatial quality "silver bullet". But if the (spectrally-correct) reflections retain their overtone structures as they decay - which assumes the room isn't overdamped - the reverberation tails on the recording arrive intact from many directions and therefore have a good chance of effectively conveying the sense of space on the recording.

Imo the arrival time of the strong onset of reflections also plays a significant role in whether the presentation trends towards the "you are there", or the "they are here", end of the sense-of-space spectrum. For instance a large room with long first-reflection path lengths presents a weaker playback-room signature and therefore the sense of space on the recording is more likely to be perceptually dominant.

Would the soundstage become bigger?

Probably not in width, but potentially in depth and/or in sense of envelopment/immersion, ime.

Hmm…

I realize it's a different paradigm from the one established and supported by decades of solid research. But I don't think it makes any disproven leaps of faith; only unproven ones.

 

[Duke]

For those participating in the thread, at approximately how many milliseconds do your initial first reaction times fall, and how are they distributed in terms of intensity?
It would be good if you could also share that information.

Do you mean "first reflection times"?

 

[Lion♡]

Do you mean "first reflection times"?

yes. I thought it would be even better if we discussed everything while considering factors like ITDG from early reflections, as well as the different impulse response patterns from various speakers in different spaces.

 

[Heinrich]

Imo the arrival time of the strong onset of reflections also plays a significant role in whether the presentation trends towards the "you are there", or the "they are here", end of the sense-of-space spectrum. For instance a large room with long first-reflection path lengths presents a weaker playback-room signature and therefore the sense of space on the recording is more likely to be perceptually dominant.

If you look closer, the first reflections would also reveal the speakers‘s location, disturbing stereo-based imagination.

Anyway, how would humanity prove all the speculations raised here? Do we think of a research program in the upcoming future?

 

[rdenney]

It seems to me that stereo illusions would get smeared if all the spectral cues were preserved in the reflections. Performance spaces are specially treated to prevent that, especially because the delay would usually be larger. The downward slope maybe gives a different enough signature that our brains interpret it as “reflection” and handle it correctly. But when smooth it avoids certain large sections of the spectrum calling attention to themselves in the reflection.

Speakers that disperse the full spectrum probably work well in some spaces.

Rick “arm-waving” Denney

 

[sigbergaudio]

It seems to me that stereo illusions would get smeared if all the spectral cues were preserved in the reflections. Performance spaces are specially treated to prevent that, especially because the delay would usually be larger. The downward slope maybe gives a different enough signature that our brains interpret it as “reflection” and handle it correctly. But when smooth it avoids certain large sections of the spectrum calling attention to themselves in the reflection.

Speakers that disperse the full spectrum probably work well in some spaces.

Rick “arm-waving” Denney

If I understand what you are suggesting correctly, this feels like the opposite of the current assumptions. It's when the reflections are similar enough to the direct sound that we are able to understand it as part of the original signal, rather than noise / something that distorts the original signal. This is not perceived as smearing.

But I can share a possibly related an interesting feedback I often get from our speakers:

People hearing them for the first time instinctively / at first gets an impression that imaging is a bit imprecise, but then after listening a little while the perception changes. There's a sudden realization that everything sounds, for a lack of better way to put it, larger, but still precise. So you can say still point and say "the singer is there" or "the violinist is there" but that virtual space that they occupy is larger than normal. Not in a "it's somewhere over there" way, it's still precisely placed, but just ..larger. And so is the sweet spot. To a larger degree you can move around from the narrow sweetspot and maintain the illusion.

It seems to be a combination of a coaxial driver, even off-axis dispersion and the cardioid pattern that comes together to form this (imo very appealing) experience - as this is very similar in both the designs we have with these features.

 

[rdenney]

I would think that if first reflections are such a key part of the image, the symmetry of the room (physically and in the frequency domain) would become critical. That would not work in many situations.

But I recall that Toole was interested in spectral similarity but not spectral equality. I’ll have to go back and read it again.

Rick “back to sound board duties” Denney

 

[Duke]

If you look closer, the first reflections would also reveal the speakers‘s location, disturbing stereo-based imagination.

Strong early ipsilateral (same-side-wall) reflections tend to broaden the apparent source width, arguably thereby making the speaker's location less noticeable. But imo these same reflections do tend to degrade the image precision.

Anyway, how would humanity prove all the speculations raised here? Do we think of a research program in the upcoming future?

I highly doubt that funding will ever be available for loudspeaker research like it was in the past.

 

[sigbergaudio]

I would think that if first reflections are such a keg part of the image, the symmetry of the room (physically and in the frequency domain) would become critical. That would not work in many situations.

But I recall that Toole was interested in spectral similarity but not spectral equality. I’ll have to go back and read it again.

Rick “back to sound board duties” Denney

I am no expert in acoustics, so I base most on this of what I've found through experimentation. I haven't found symmetric placement to be critical to either imaging or soundstage.

The song below is an interesting example. It has extreme panning with some phase trickery as well. On a speaker system with a really wide soundstage, the panning on this track sounds more extreme than they do with headphones, with the synth sounds in the beginning of the track panning beyond 90 degrees to each side. I've heard this work and appear similar both to the right and left even with asymmetrical distance to the left and right wall.

 

[Duke]

yes. I thought it would be even better if we discussed everything while considering factors like ITDG from early reflections, as well as the different impulse response patterns from various speakers in different spaces.

I have deliberately tried to push the arrival times of strong early reflections back in time as far as I reasonably can. My room is would be about 17 by 13 feet if it was a clean rectangle, but it is not. My target is to have a "reflection-free interval" following the direct sound of about 10 milliseconds; 15 milliseconds would be better, and 20 milliseconds better still.

Based on physical distances, my floor bounce arrives at 2.4 milliseconds after the direct sound, and my ceiling bounce at 4.4 ms. My understanding is that the floor and ceiling bounces are perceptually rather benign, and both reflection points are outside my speaker's -6 dB coverage pattern limits (though the floor reflection is not far outside of it).

My speakers are fairly directional and are toed-in aggressively such that the first ipsilateral reflections are quite weak, being well outside the speaker's coverage pattern. They arrive at about 2.8 milliseconds.

So the first strong lateral reflections are actually the contralateral ones, off the opposite side walls. These arrive at about 9.8 milliseconds.

The reflection off the wall behind my head arrives at about 10.6 milliseconds.

Each speaker has an additional up-and-rear firing driver whose path length includes a ceiling bounce, and its output arrives at about 11.5 milliseconds.

So while there are reflections which arrive well before my 10 millisecond target, they are either benign or weak or both. The strong onset of reflections is delayed until ballpark 10 milliseconds.

 

[Duke]

It seems to me that stereo illusions would get smeared if all the spectral cues were preserved in the reflections.

The ear/brain system primarily derives it directional cues from the first .68 milliseconds of a sound, but derives its "sense of space" cues from all subsequent reflections as well. Strong early reflections can influence the directional cues, as when strong early same-side-wall reflections widen the soundstage. But imo these strong early reflections DO smear the image precision, so it's a tradeoff.

I'm not sure whether spectrally equivalent early reflections have a greater or lesser effect on the apparent source width. Too much dissimilarity would be disable the ear's ability to correctly identify reflections as such. It is not clear to me that too much similarity would be detrimental because I think spectral similarity (or equality) would maximize the ear's ability to correctly identify reflections as such. (The ear/brain system looks at the spectra of incoming sounds in order to classify them as reflections or as new sounds, and if the latter, the ear computes their arrival direction largely from the arrival time differential between the two ears.)

Performance spaces are specially treated to prevent that, especially because the delay would usually be larger. The downward slope maybe gives a different enough signature that our brains interpret it as “reflection” and handle it correctly. But when smooth it avoids certain large sections of the spectrum calling attention to themselves in the reflection.

I understand that the conventional wisdom holds the downward-sloping spectrum of typical off-axis energy to be a feature rather than a bug because it approximates the downward-sloping spectrum of the reflection field in a large concert hall.

Imo there is a factor which this approach doesn't take into account, and that factor is... wait for it... wait for it... just a bit longer... time.

Imo the in-room reflections arrive far too early for a significant downward-spectral-tilt to be natural-sounding. My understanding is that the perceived spectral balance is a weighted average of the spectra of the direct sound and the reflections, and my experience (in a home audio setting) has been that a significant discrepancy between the two can result in listening fatigue. But this is just an opinion.

I would think that if first reflections are such a key part of the image, the symmetry of the room (physically and in the frequency domain) would become critical.

With conventional wide-dispersion speakers, and without suitable room treatment, yes. Symmetry in the early reflections matters.

There are ways to mitigate the influence of the first reflections, both from a spatial and from a sound-quality standpoint, in situations where the room is not physically symmetrical. These include what might be called compensating room treatment, and avoiding illuminating the typical first-reflection zones in the first place (the latter being the approach I use).

I recall that Toole was interested in spectral similarity but not spectral equality.

That is my recollection as well. I'm not aware of any studies which advocated aspiring to spectral equality. But I think that's because spectral equality is quite rare and relatively expensive and/or impractical to achieve, and not because it's an inherently bad idea.

If you are open to anecdotes I'll put on my asbestos suit and spout a few.

Rick “arm-waving” Denney

Rick “back to sound board duties” Denney

Duke "experiencing acute middle-name envy" LeJeune