Horns - Necessary to complete the Audiophile Journey?

[watchnerd]

The Journey is its own petard.

The journey is a small hat-shaped explosive device?

 

[richard12511]

@watchnerd

Have you decided yet? Why don't you get a JBL EON615 and send it to @amirm for testing. Guitar Center has a 180 day return policy and even though this is NOT a high-end speaker, it will give you a taste of horns.

View attachment 73343

Damn. a spin that good, with 127db output, 50Hz extension and all for $400? What's the catch?

 

[richard12511]

Yeah, many.

First of all, my guitar amp, which is not even a reproducer...it's an 'instrument' with tone and timbre unto itself.

For the actual PA speakers I have (Electro Voice):

• They trade frequency response for loudness -- they go down to 55 Hz, but have 127 dB max output with a 15" speaker
• They're designed to be used in multiples via a mixing board to balance out a whole room. Up close, they often don't sound very good. Bad driver integration at close range.
• Audiophile criteria like imaging and depth aren't really part of the design criteria, which can result in pre-recorded music sounding a little odd or flat, more 'canned'
• Super lame sounding at low volumes
• Big, burly, durable, and butt ugly

But with measurements that good, it would seem that most of the negatives you listed wouldn't apply here.

 

[watchnerd]

But with measurements that good, it would seem that most of the negatives you listed wouldn't apply here.

Go for it.

I already own a PA speaker.

And still too ugly for my living room and doesn't integrate with my current gear easily.

And except for the loudness, I'm not sure what the EON would be better at compared to my current monitors.

 

[Neale]

Yeah, many.

For the actual PA speakers I have (Electro Voice):

• They trade frequency response for loudness -- they go down to 55 Hz, but have 127 dB max output with a 15" speaker
• They're designed to be used in multiples via a mixing board to balance out a whole room. Up close, they often don't sound very good. Bad driver integration at close range.
• Audiophile criteria like imaging and depth aren't really part of the design criteria, which can result in pre-recorded music sounding a little odd or flat, more 'canned'
• Super lame sounding at low volumes
• Big, burly, durable, and butt ugly

I would like to see ASR test some of the top PA speakers .. they generally do what is listed above, don't sound lame, do cost >$5000 per box. Providing quality sound at near or mid-field is one thing; the ability to cover +2000 people with the same quality has advanced immensely in the last 10 years and is huge credit to smart people and advances in DSP.

 

[KSTR]

Damn. a spin that good, with 127db output, 50Hz extension and all for $400? What's the catch?

This a very entry level all-in-PA, with 2 mic/line inputs and some knobs for settings. They're made to the cheapest possible standards, just good and durable enough to get the message accross to the public, literally, with a good degree of intelligibility.

You'll find them with DJs and keyboarders hired to entertain a wedding or company party, and shops use this to praise their goods in presentations in front of their doors, all that kind of stuff.

The good radiation pattern is something to note though, achieved by the horn and a huge diffraction lens in front of the midwoofer, and JBL choose wisely to keep that innovative aspect of the design a much as possible in the actual product whereas they cheaped out on construction costs, given the extremely low retail price. With the huge quantities involved, the molded enclosuse makes sense, it helps to cut build and assembly cost. But it is not overly rigid nor well-damped. The drivers are cheap, the electronics are cheap. There is zero love involved in making those speakers and thats how they sound like in the end.

 

[watchnerd]

I would like to see ASR test some of the top PA speakers .. they generally do what is listed above, don't sound lame, do cost >$5000 per box. Providing quality sound at near or mid-field is one thing; the ability to cover +2000 people with the same quality has advanced immensely in the last 10 years and is huge credit to smart people and advances in DSP.

So you want to put a big line array in the Klippel?

I don't think it would fit.

 

[Duke]

Damn. a spin that good, with 127db output, 50Hz extension and all for $400? What's the catch?

But with measurements that good, it would seem that most of the negatives you listed wouldn't apply here.

Radiation pattern control was a thing in prosound long before it gained traction in high-end home audio. The audience needed to be covered, but with minimal sound going where it wasn't needed, the latter to preserve a decent direct-to-reverberant sound ratio insofar as reasonably feasible.

The good radiation pattern is something to note though, achieved by the horn and a huge diffraction lens in front of the midwoofer...

Often in prosound the desired pattern coverage is achieved partially through the use of diffraction devices. Diffraction horns can measure great in a spin-o-rama, but imo they tend to not sound very good. They tend to sound edgy and harsh, particularly as the volume level goes up. This is because diffraction is a linear distortion to which the ear has a non-linear response. People often mistakenly attribute harshness at high SPLs in a PA system to the compression drivers being driven into distortion when actually what's happening is the ear's increased sensitivity to diffraction at high SPLs.

Diffraction horns are an excellent example of an audible issue which is not revealed by a spin-o-rama. Nor is diffraction a harmonic distortion. It is a delayed (and presumably distorted, though I've never seen the spectrum of diffraction) very early reflection which need not be very strong in order to be audible and objectionable at high levels. Because it arrives at a later time than the undiffracted direct sound, it is not "masked" by the ear/brain system's masking mechanism, which is one of the reasons why it is unexpectedly and disproportionately audible and objectionable.

At least that's my understanding.

Eyeballing a photo of the horn in the Eon 615, it is not obviously a diffraction horn. BUT if there is a significant mis-match between the exit angle of the compression driver and the entry angle of the horn, that would be a significant source of diffraction. Again eyeballing the photo, it looks like the entry angle of the horn differs in the horizontal and vertical planes, which implies diffraction in at least one of them.

My personal preference is for horns which do not rely on deliberately diffractive features to achieve constant directivity. I am not aware of any such horns being tested by Harman, presumably because their focus was on evaluating competitors, and such horns are numerically insignificant in the home audio marketplace.

The first nine posts of this thread on another forum are imo an educational survey of factors affecting sound quality in horns. It's a few years old but arguably still relevant. In it you will find the term "higher order modes", which refers to reflections within the horn itself, of which diffraction is a common cause.

 

[GXAlan]

This a very entry level all-in-PA, with 2 mic/line inputs and some knobs for settings. They're made to the cheapest possible standards, just good and durable enough to get the message accross to the public, literally, with a good degree of intelligibility.

You'll find them with DJs and keyboarders hired to entertain a wedding or company party, and shops use this to praise their goods in presentations in front of their doors, all that kind of stuff.

The good radiation pattern is something to note though, achieved by the horn and a huge diffraction lens in front of the midwoofer, and JBL choose wisely to keep that innovative aspect of the design a much as possible in the actual product whereas they cheaped out on construction costs, given the extremely low retail price. With the huge quantities involved, the molded enclosuse makes sense, it helps to cut build and assembly cost. But it is not overly rigid nor well-damped. The drivers are cheap, the electronics are cheap. There is zero love involved in making those speakers and thats how they sound like in the end.

The spin data is impressive though. You would want to see what resonances or harmonic distortions are present as that would help translate between subjective measurements and audio experience. The woofer waveguide isn’t seen anywhere else.

The compression driver is still the same one in the Studio 530 I believe.
https://www.audiosciencereview.com/forum/index.php?threads/jbl-studio-530-speaker-review.12298/

The woofer is not seen elsewhere but does use the same cloth surround (as opposed to rubber or foam) seen in the 4319 (minus all of the magnet enhancements)
https://www.audiosciencereview.com/forum/index.php?threads/jbl-4319-studio-monitor-review.12986/

I am somewhat afraid that the EON615 will measure well and sound great at normal SPLs, because it will have meant that all my audiophile speaker purchases have been audiofoolery and I am paying for appearances only.

 

[jhaider]

I am not aware of any such [diffraction]horns being tested by Harman, presumably because their focus was on evaluating competitors, and such horns are numerically insignificant in the home audio marketplace.

There is one exception: as I understand it the Sprinkle waveguide (JBL M2/7/3/SCL/HDI) is a sophisticated diffraction horn.

 

[richard12511]

Go for it.

I already own a PA speaker.

And still too ugly for my living room and doesn't integrate with my current gear easily.

And except for the loudness, I'm not sure what the EON would be better at compared to my current monitors.

If you already own a PA speaker that you don't like the sound of, then it necessarily must measure badly. This speaker measures excellently, so it will sound excellent(ie nothing like the PA speakers you've heard). That's the point of these spinoramara style measurements, is it not?

 

[watchnerd]

If you already own a PA speaker that you don't like the sound of, then it necessarily must measure badly. This speaker measures excellently, so it will sound excellent(ie nothing like the PA speakers you've heard). That's the point of these spinoramara style measurements, is it not?

Huh?

When I was referring to my "current monitors", I was not speaking of my Electro Voice PA speakers, but my living room monitors (Dynaudio Contour 20).

a) I'm not shopping for a PA speaker for my living room, or elsewhere

b) I use the PA I have for outdoors. Is Spinorama optimized for outdoors?

Also, I've heard the EON line in person, although not that model.

They sound like budget PA speakers, albeit better than average for the price.

If you're convinced that Spinorama tells you everything you need to know about speaker quality, perhaps you can buy the EONs.

 

[richard12511]

Huh?

When I was referring to my "current monitors", I was not speaking of my Electro Voice PA speakers, but my living room monitors (Dynaudio Contour 20).

I wasn't talking about your current monitors. I was talking about your outdoor PA speakers you use. You said they don't sound very good, that means they also must not measure very good. Given that these do measure very good, they won't sound anything like your PA speakers, and likely much, much better. Looking at the Countrour 20 measurements, the JBLs should sound slightly better at equal volume, but also be capable of playing much louder cleanly. This is going by the science done by Toole and colleagues.

If you're convinced that Spinorama tells you everything you need to know about speaker quality, perhaps you can buy the EONs.

Spinorama doesn't tell you everything, but the main thing it doesn't tell you - max output - these speakers also have in spades. It's not just me, Floyd Toole himself has said(what the science shows thread on AVS) that spinorama data combined with max output data is sufficient to characterize the sound of loudspeaker, so you're also disagreeing with Toole and the best audio science we have. The second most important thing besides output that the spinorama misses is separate horizontal and vertical directivity lines, but with this speakers waveguide, that shouldn't be a problem.

I probably will buy them, actually. Definitely worth a try.

 

[watchnerd]

I wasn't talking about your current monitors. I was talking about your outdoor PA speakers you use. You said they don't sound very good, that means they also must not measure very good. Given that these do measure very good, they won't sound like your PA speakers.

I don't really care about my improving my outdoor PA speakers because:

a) I use them maybe 4-6 times a year

b) At a gig, they have to work with the mixing board we have. The EVs I own have a Dante card that works with our mixing board.

c) Venue acoustics are the big wild card, and the speakers aren't used in isolation, but with other PA speakers

d) At a performance, most people care about your ability as a band / as musicians

e) At a tailgate party, nobody cares about sound quality, they're just getting drunk and eating BBQ

Spinorama doesn't tell you everything, but the main thing it doesn't tell you - max output - these speakers also have in spades.

Compared to home speakers, pretty much any PA speaker has better max output.

Is our goal to get the loudest speakers possible for home use?

 

[KSTR]

I am somewhat afraid that the EON615 will measure well and sound great at normal SPLs, because it will have meant that all my audiophile speaker purchases have been audiofoolery and I am paying for appearances only.

It will certainly not sound bad and JBL will sure have tried to make "best possible compromises" within that budget, it's just that one should *not* expect to have a very refined speaker here but you might indeed get a thrill from the dynamics and effortlessness these speakers will still have in spades.

 

[Duke]

My personal preference is for horns which do not rely on deliberately diffractive features to achieve constant directivity. I am not aware of any such horns being tested by Harman...

There is one exception: as I understand it the Sprinkle waveguide (JBL M2/7/3/SCL/HDI) is a sophisticated diffraction horn.

What I meant to say was that was, that I am not aware of any non-diffraction horns being measured by Harman. My apologies for the ambiguous wording.

I agree that the Sprinkle horn is a very sophisticated, low-coloration diffraction horn.

An example of what I would call a "non-diffraction horn" would be Earl Geddes' Oblate Spheroid, which would more precisely be called a "waveguide", as its design prioritizes radiation pattern control over acoustic amplification. As does the Sprinkle.

Strictly speaking the Oblate Spheroid is not completely free of diffraction, but my understanding is that its contour is an optimimum shape for minimizing disturbance of the waveform (the Prolate Spheroid being the another optimum shape).

My armchair-quarterback analysis of the Sprinkle waveguide is that the complex shape spreads out diffraction effects and manages internal reflections, presumably to push them down below the threshold of audibility. I have not heard the M2 but have heard Peter Noerbaek's M2!5 and imo it is superb.

 

[watchnerd]

It will certainly not sound bad and JBL will sure have tried to make "best possible compromises" within that budget, it's just that one should *not* expect to have a very refined speaker here but you might indeed get a thrill from the dynamics and effortlessness these speakers will still have in spades.

+1

Having heard the EONs, you're not getting the equivalent of a JBL 708P active studio monitor for an 80% discount.

JBL isn't stupid....

 

[Duke]

Floyd Toole himself has said (what the science shows thread on AVS) that spinorama data combined with max output data is sufficient to characterize the sound of loudspeaker, so you're also disagreeing with Toole and the best audio science we have.

The spinorama may be the single most powerful predictive tool in existence today, but imo the science of psychoacoustics is still an open canon. I don't think either Toole or Olive would claim that the spinorama tells the whole story.

From one of Sean Olive's landmark papers, A Multiple Regression Model for Predicting Loudspeaker Preference Using Objective Measurements: Part II – Development of the Model:

"LIMITATIONS OF MODEL

"The conclusions of this study may only be safely generalized to the conditions in which the tests were performed. Some of the possible limitations are listed below.

"1. Up to this point, the model has been tested in one listening room.

"2. The model doesn't include variables that account for nonlinear distortion (and to a lesser extent, perceived spatial attributes).

"3. The model is limited to the specific types of loudspeakers in our sample of 70."

 

[watchnerd]

The spinorama may be the single most powerful predictive tool in existence today, but imo the science of psychoacoustics is still an open canon. I don't think either Toole or Olive claim that it tells the whole story.

From Sean Olive's landmark papers, A Multiple Regression Model for Predicting Loudspeaker Preference Using Objective Measurements: Part II – Development of the Model:

"LIMITATIONS OF MODEL

"The conclusions of this study may only be safely generalized to the conditions in which the tests were performed. Some of the possible limitations are listed below.

"1. Up to this point, the model has been tested in one listening room.

"2. The model doesn't include variables that account for nonlinear distortion (and to a lesser extent, perceived spatial attributes).

"3. The model is limited to the specific types of loudspeakers in our sample of 70."

+1

I haven't heard a speaker designer say that it's the sum total of what needs to be done.

After all, small dB differences (low Q) over multiple octaves can affect the perceived voicing of a speaker, even if it looks basically flat on a graph. This is easy to simulate with EQ, and subjectively obvious if one plays with wide spectrum tilt EQ.

Even @amirm listens to speakers after he measures them....

 

[Duke]

After all, small dB differences (low Q) over multiple octaves can affect the perceived voicing of a speaker, even if it looks basically flat on a graph. This is easy to simulate with EQ, and subjectively obvious if one plays with wide spectrum tilt EQ.

There are areas where I'll often deliberately introduce a dip in the frequency response. This is because the ear interprets distortion as a tonality. For instance, the woofer's surround resonance may be imperceptible in the frequency response curve, but ime that doesn't mean its effects are inaudible and not worth minimizing via a bit of EQ.

Even @amirm listens to speakers after he measures them....

I trust Amir's ears more than I trust the preference score prediction which is calculated from the measurements. When I have read reviews wherein there was an apparent discrepancy between the predicted preference rating and Amir's ears-on evaluation, I trust Amir.