Which anechoic response curve predicts the preferred loudspeaker? (fun exercise)

[Sancus]

i am in the "only sweet spot sounds good" team, so on axis for me is above all

You still hear the off-axis response in the form of reflections, whether you are in the sweet spot or not. So that's not how it works.

 

[ernestcarl]

You still hear the off-axis response in the form of reflections, whether you are in the sweet spot or not. So that's not how it works.

Yep. Off-axis response matters

I listen to music and watch the news on TV off-axis all of the time. If the off-axis sound of my speakers in my rooms were all horrible/bad, I would rather turn everything off relegating myself to listen to the HVAC or fan and fridge noise most of the time while at home.

 

[dasdoing]

You still hear the off-axis response in the form of reflections, whether you are in the sweet spot or not. So that's not how it works.

not i my case since I am in the "treated first reflections walls" team also

 

[dasdoing]

Yep. Off-axis response matters

I listen to music and watch the news on TV off-axis all of the time. If the off-axis sound of my speakers in my rooms were all horrible/bad, I would rather turn everything off relegating myself to listen to the HVAC or fan and fridge noise most of the time while at home.

I hate reflections. I want to kill them. And I did lol

 

[ernestcarl]

But that is a generalization... you underestimate our ear-brain system's ability to adapt to different environments and acoustic spaces. I do understand that some people with hearing damage or age-related hearing impairment aren't able to adapt anymore as easily, finding most reflections as being more confusing and annoying than helpful/contributing to the overall auditory experience.

 

[Sancus]

not i my case since I am in the "treated first reflections walls" team also

As far as I understand, this doesn't actually kill reflections unless you made your room into an anechoic chamber. It just reduces them. I would guess an extremely poor off-axis response is still audible with treatments, just less so. Depending on the treatments, of course, and the frequency range they affect.

 

[preload]

Nobody else wants to make predictions based on the listening window and horizontal off-axis response curves?

 

[Duke]

Thanks for posting the additional information, preload. B looks the best to me. Beams north of 10 kHz but imo that's live-with-able. I'd be tempted to try a rear-firing tweeter in that top octave, just for fun, to see if adding a wee bit more energy to the reverberant field up there was beneficial.

I don't like the excess off-axis energy of A in the 3-8 kHz region. I'd suggest nearfield listening and/or acoustic treatment to minimize the relative contribution of the off-axis energy, so that the (impressively smooth) on-axis response is dominant. Otherwise I think A could sound a bit forward, aggressive, and/or fatiguing over time. Imo you want to listen to A on-axis; it just gets worse and worse off-axis.

The off-axis energy of C looks to me like it "zigs" where the on-axis "zags", so the one may fill in the other somewhat. (I could easily be mistaken about this because I'm not used to wrapping my head around those normalized off-axis curves; wish John Atkinson wouldn't normalize them.) I would NOT want to listen to this speaker on-axis or nearfield, but rather by ear would try to find the off-axis angle where it sounds best. I think this speaker might sound better from far away than up close, as it looks to me like the off-axis response is better than the on-axis response. The farther away you are the more the off-axis response dominates the perceived tonal balance. Imo those on-axis peaks are both in bad places, hence my instinct to have the off-axis sound dominate.

All of these comments ignore the bottom octave. C has a lot of upper-bass energy which may complicate subwoofer integration.

So with the expectation that there is an off-axis angle from which C sounds pretty good, and assuming we're not listening nearfield, my pick order is B-C-A. If we're listening nearfield and/or in a highly damped room I waffle between putting A or B first, probably A by a small margin, but C would definitely be last. Personally I am not a fan of highly damped rooms, and don't want to be limited to nearfield listening.

All of that being said, unfortunately I am not very confident of my ability to interpret normalized off-axis curves nor predict preference between different problematic curves. And there could be broad, gentle (very low-Q) trends which are virtually invisible to my untrained eye but not to the ear.

I'm guessing A is a fairly "conventional" cone-n-dome speaker (a small two-way?), and somebody worked very hard to optimize its on-axis response. B may be unconventional, as imo its off-axis response south of 10k is unusually smooth... if it's conventional, somebody did an excellent job on the system design and crossover. And I'm thinking (hoping!) there's an off-axis angle and/or a different listening height where C performs better than the measured on-axis curve predicts, OR that the apparently wonky off-axis response really does complement the wonky on-axis response and vice-versa.

Edit: Everywhere I've said "on-axis", I SHOULD have said "30 degree listening window."

 

[richard12511]

I think without subs some may enjoy C the most because of that bass slam(though B digs deeper).

I think B looks the best. It has by far the smoothest directivity up to 10k or so, after which it seems to struggle, but 10k+ content is rare and not that important(imo).

I'll say B > A >= C (harder to pick between A and C)

 

[dasdoing]

As far as I understand, this doesn't actually kill reflections unless you made your room into an anechoic chamber. It just reduces them. I would guess an extremely poor off-axis response is still audible with treatments, just less so. Depending on the treatments, of course, and the frequency range they affect.

they were installed using ETC. also I use studio monitors which are designed to have narrow dispersion. that's actualy the main diference between normal speakers and monitors. monitors are designed with treated walls in mind

 

[Sancus]

they were installed using ETC. also I use studio monitors which are designed to have narrow dispersion. that's actualy the main diference between normal speakers and monitors. monitors are designed with treated walls in mind

I dunno what we're talking about anymore! You bought speakers based on their constant, narrow directivity. So you do care about off axis response, because that is literally an example of off axis response mattering.

 

[dasdoing]

But that is a generalization... you underestimate our ear-brain system's ability to adapt to different environments and acoustic spaces. I do understand that some people with hearing damage or age-related hearing impairment aren't able to adapt anymore as easily, finding most reflections as being more confusing and annoying than helpful/contributing to the overall auditory experience.

it's about the stereo image (center clarity); the ability to identify clear panning positions. also the center get's realy muddy with the reflections since what plays in the center get's dubble the amount of reflections compared to what plays hard panned

 

[dasdoing]

I dunno what we're talking about anymore! You bought speakers based on their constant, narrow directivity. So you do care about off axis response, because that is literally an example of off axis response mattering.

lol. this started when I said the on axis is what most matters to me. then I explained why.

 

[SECA_alan]

B wins over A for me, having recently spent time looking at such charts and choosing new Loudspeakers.

B has better directional control, a nice wide dispersion which is what I wanted in my family room. I also prefer the look of the bass roll off, assuming the hump is the typical Stereophile artifact.

I'm not saying I could blindly identify the speakers in any given order though!

 

[Plcamp]

“Even if you can hear it, there is so little energy in music above 10khz that its relevance is fairly minor, IMO. It doesn't contribute anything to spatial qualities. The peaking in "A" from 3 to 7khz, a very audible region, is a bigger sin than anything happening above 10khz.”

Looking at it again, might be that A has a larger diameter woofer than b, thus the dispersion not as good a match near crossover?

 

[preload]

B wins over A for me, having recently spent time looking at such charts and choosing new Loudspeakers.

B has better directional control, a nice wide dispersion which is what I wanted in my family room. I also prefer the look of the bass roll off, assuming the hump is the typical Stereophile artifact.

I'm not saying I could blindly identify the speakers in any given order though!

How bout C?

 

[Duke]

Looking at it again, might be that A has a larger diameter woofer than B, thus the dispersion not as good a match near crossover?

I think A's midwoofer is not particularly large but its cone may be rather stiff.

I think B has a bit larger diameter woofer than A (based on 300 Hz dispersion), and I think it's crossed over to a midrange driver or to a wide-pattern horn. I suspect the latter because I don't see signs of a subsequent crossover to a tweeter in the off-axis response until 10 kHz ballpark, and that might be a bullet tweeter up there. Maybe it has a whole bunch of different-sized drivers so that the transition from one to another is smooth off-axis, but I don't see how it could be that smooth.

 

[MediumRare]

No. Resonances? Time domain? Vertical dispersion? Huge distortion in the mid range? Can’t tell any of those things that could make these speakers problematic.

if I told you a meal was properly salted, the right temperature, and had no active bacterial content, does that tell you anything about the taste or if you would like it?

 

[SECA_alan]

How bout C?

As has been noted, there's limited information available here. I suspect it is a very good loudspeaker, but the listening window in the midrange isn't so well behaved, and I don't like the look of the bass peak. I don't want to infer too much, if it is a stand mount compared to a floor standing model, then it is doing very well.

 

[Duke]

Resonances?

Right there you have put your finger on one of the problems with Atkinson's normalized curves. In an un-normalized curve, a resonance shows up as a ridge of off-axis energy. However if it also shows up as an on-axis peak, and if the off-axis curves are normalized relative to the on-axis, then the existence of the resonance is concealed by the resulting smoothness of the off-axis curves.

Edit: Everywhere I've said "on-axis", I SHOULD have said "30 degree listening window."