Are speakers actually already a solved problem?

[Anton S]

... This leads me to wonder: Are speakers, much like amplifiers and DACs, now a solved problem—even for budget options? If so, what innovations remain to be explored? ...

Short answer, no. If speakers were a solved problem, the sonic differences among them would be minimal, and you'd be relegated to choices based primarily on cosmetics. That is still very much not the case.

 

[goat76]

By that token, neither are DACs or cables.

If an audibly significant difference shows up in measurements, that's good evidence.

What a bunch of 'fans' claim, isn't good evidence. It's anecdotes awaiting good evidence. Like measurements, or blind comparisons.

You're on Audio Science Review, how do you think science works?

Are you trying to build a strawman out of this?

In comparison to DACs and cables, the differences in the sound of two different loudspeakers are often large enough that most of us can be pretty sure that the sonic differences we hear are real, even in sighted listening. That's why I can comfortably put some trust in people who claim that one loudspeaker is perceived as brighter-sounding than another loudspeaker in direct comparison, especially when that same observation is often coming up in discussions about the sonic differences between these two brands.

In the case of my example with the sonic differences between Genelec and Neumann speakers, the differences people report are also confirmed when looking at the measurements that @kemmler3D provided. So people are both able to hear the difference between these two loudspeaker brands (in general, blind or not), and we can also see that the measurements prove that those sonic differences are truly there.


So back to the topic...
My point is that even if the two manufacturers are both aiming at the same goal of accurately reproducing sound, the sonic differences are large enough that most people will be able to hear them, and the reason for this is that these two manufacturers have different ideas of what accurate sound reproduction should sound like. One goes for wide dispersion and the other goes for a narrower dispersion which leads to them sounding different enough that most people will hear a pretty significant difference in a direct comparison.

Let's play with the thought that those sonic differences were also clearly heard if the direct sound was in isolation, then I'm sure many of us would likely be questioning which one of the loudspeakers has the somewhat "wacky" frequency response, and which one truly comes closest to accurate reproduction the sound. But such questioning doesn't seem to come up if the sonic differences are caused by differences in dispersion, then it just seems to be accepted even if it leads to fairly obvious differences in the overall sound.

That's one of the larger reasons why I don't see loudspeakers as a "solved problem". Two highly respected loudspeaker manufacturers with the same goal of accurately reproducing sound, and still, they have vastly different ideas of how the dispersion should be to reach that goal.

 

[goat76]

Spend some time here: http://soundmedia.jp/nuaudk/

And then reconsider.

Leaving aside the recording side of the chain: what is the ideal loudspeaker dispersion pattern?

Based on excellent resource above, we can hear that violins tend to project lots of upper-mid/treble energy upwards, while a cello projects those frequencies forwards.
Timpani are bright from above, too.

So, a conventional 6.5" 2-way speaker will probably represent a vocal or cello reasonably well, since the radiation pattern will share some characteristics.
For a solo violin, however, we might wish to have a small mid-high driver aimed upwards to throw more of that energy towards the ceiling.

We can conclude: the "ideal" dispersion pattern is heavily dependent on the program material.


Chris

It's not the task of the loudspeakers to simulate any single/particular instrument's dispersion pattern, that part is up to the placement and pickup pattern of the microphones used for capturing the music instruments in the first place. What we hear is the stereo illusion of the whole ensemble, and that "whole" is the task of the loudspeakers to reproduce in a convincing way that hopefully mirrors reality in the best possible way. I believe there should be such a dispersion pattern that reflects the reality more faithfully than others, and closer to how our hearing would hear the "whole" if we "were there".

 

[Duke]

It's not the task of the loudspeakers to simulate any single/particular instrument's dispersion pattern, that part is up to the placement and pickup pattern of the microphones used for capturing the music instruments in the first place. What we hear is the stereo illusion of the whole ensemble, and that "whole" is the task of the loudspeakers to reproduce in a convincing way that hopefully mirrors reality in the best possible way. I believe there should be such a dispersion pattern that reflects the reality more faithfully than others, and closer to how our hearing would hear the "whole" if we "were there".

Imo the question of what the loudspeaker dispersion should be is really a question of what the reflection field at the listening position should be. Imo there are two schools of thought:

1. The reflection field should come as close as reasonably possible to what one would experience from live instruments; and

2. The reflection field should enable the venue spatial cues already on the recording to be perceptually dominant over the "small-room signature" of the playback room.

The first would be a pathway to a convincing "they are here" presentation; and the second would be a pathway to a convincing "you are there" presentation. I'm not claiming one is right and the other wrong, nor am I claiming they are mutually exclusive, but I am claiming that they are two somewhat different paradigms.

 

[Ktacos]

Short answer, no. If speakers were a solved problem, the sonic differences among them would be minimal, and you'd be relegated to choices based primarily on cosmetics. That is still very much not the case.

Differences between neutral speakers with similar dispersion is already quite minimal. Mostly not the case because manufacturers suck at making speakers not because we don't know how to make them good.

 

[ChrisG]

It's not the task of the loudspeakers to simulate any single/particular instrument's dispersion pattern, that part is up to the placement and pickup pattern of the microphones used for capturing the music instruments in the first place. What we hear is the stereo illusion of the whole ensemble, and that "whole" is the task of the loudspeakers to reproduce in a convincing way that hopefully mirrors reality in the best possible way. I believe there should be such a dispersion pattern that reflects the reality more faithfully than others, and closer to how our hearing would hear the "whole" if we "were there".

I can see what you're getting at, but disagree.

First, some listeners prefer "the musicians are here" presentation. ie, for them, it has to be important that the reproduction system matches the polar pattern of the instrument(s).

Next, no matter what the recording engineer may or may not do, the fact is that violins send more mid-high energy upwards than forwards. ie, no matter if you want "they are here" or "you are there" presentation, some of the violin's mid-high energy should arrive at your ears from a higher-up position. That could be reflected off your own ceiling (for "they are here") or arriving from suitably-placed "height" channels with the venue's reflective properties imposed/recorded (for "you are there").



When it comes to mic positions, listening to some samples in the link should make it clear that choosing any single mic position is inherently compromised: instruments sound different at different angles. Whether you want it to sound like "they are here" or "you are there" is largely irrelevant: instruments throw different sound in different directions, and the reproduction of such must be incomplete with any conventional 2-channel playback.


It's nice to have some discussion which has gone beyond mere frequency response.

Chris

 

[dasdoing]

I think this is possible in principle with those ultrasonic beamforming devices? But as you note I am not sure how far along the fidelity is or can get.

they are called Parametric Speakers and yes, they use ultrasonic waves to create highly focused beams of sound.
The last time I looked at these they seemed to have totally unusable responses for HiFi.....but then they are not really produced for that, so there might be more possible than what these products achieve that are only produced to direct speech to a visitor of a museum or similar.

 

[goat76]

I can see what you're getting at, but disagree.

First, some listeners prefer "the musicians are here" presentation. ie, for them, it has to be important that the reproduction system matches the polar pattern of the instrument(s).

The “musicians are here” can never be the right approach from an high fidelity point of view, as that would require dedicated loudspeakers for each instrument, and those loudspeakers should in that case optimally have the same dispersion pattern as the instrument they are dedicated to.

There are no such recording, so that approach is doomed to fail.

Next, no matter what the recording engineer may or may not do, the fact is that violins send more mid-high energy upwards than forwards. ie, no matter if you want "they are here" or "you are there" presentation, some of the violin's mid-high energy should arrive at your ears from a higher-up position. That could be reflected off your own ceiling (for "they are here") or arriving from suitably-placed "height" channels with the venue's reflective properties imposed/recorded (for "you are there").

In my opinion, there are no way to have any realistic reproduction of those upward reflection from a stereo recording, no matter the loudspeakers dispersion pattern or approach.

When it comes to mic positions, listening to some samples in the link should make it clear that choosing any single mic position is inherently compromised: instruments sound different at different angles. Whether you want it to sound like "they are here" or "you are there" is largely irrelevant: instruments throw different sound in different directions, and the reproduction of such must be incomplete with any conventional 2-channel playback.

Using a single microphone for the recordings is almost never enough to capture the full spectrum of pretty much any instrument, that’s why multiple microphones are often used in many positions to capture the complete sound of an instrument and all the important dispersion energy it generates in a room to sound complete.

I don't think we can expect fully realistic dispersion of the recorded instruments from simple 2-channel stereo reproduction, that’s maybe possible with Dolby Atmos if there are any recording engineer who go the full length of trying to capture a single instrument from many different angles.

Even for the recording of the electrical guitar cab for my recordings, I use two close microphones, one at a further away distance for more distant sound, and added to that a bit of artificial reverb to make it sound like how that single instrument sounds in a room. So it takes about four channels for just one instrument to even come somewhat close to the real sound.

It's nice to have some discussion which has gone beyond mere frequency response.

Chris

Yes, loudspeakers and sound reproduction is a very broad and highly interesting subject. It can be highly complicated stuff.

 

[ChrisG]

The “musicians are here” can never be the right approach from an high fidelity point of view

Disagree
It's just as valid as "you are there", but presents a totally different set of requirements.
It's a very difficult thing to achieve, but how many HiFi-people have said "It sounds like the musicians are here in the room!"?

It's clearly a requirement/target for some listeners.

I don't think we can expect fully realistic dispersion of the recorded instruments from simple 2-channel stereo reproduction

Which concludes my point: there is no "correct" dispersion pattern.

Even for the recording of the electrical guitar cab for my recordings, I use two close microphones, one at a further away distance for more distant sound, and added to that a bit of artificial reverb to make it sound like how that single instrument sounds in a room. So it takes about four channels for just one instrument to even come somewhat close to the real sound.

A reasonable approach, but only if you want the "you are there" sound. Even then, the room capture should be in Atmos


Chris

 

[goat76]

Disagree
It's just as valid as "you are there", but presents a totally different set of requirements.
It's a very difficult thing to achieve, but how many HiFi-people have said "It sounds like the musicians are here in the room!"?

It's clearly a requirement/target for some listeners.

It’s pretty clear to me how an “you are there” type of fidelity can be optimized by minimizing the amount of reflection from the listening environment (which is definitely not part of the recording), but how would you go about optimizing for a “they are here” approach?

The problem with the ”they are here” approach is that the more you let the listening environment be part of what is heard, the more fuzzy the separations between the recorded elements will be as the strong early reflections from the listening environment only “see” the two sound generating loudspeakers in comparison to all the separations that the stereo illusion contains.

 

[Buckchester]

Since the introduction of the CTA-2034 in 2015, scientific speaker design has made significant strides. The availability of more affordable measurement equipment—both hardware and software—for professionals and consumers alike has played a crucial role in this advancement.

Today, manufacturers like KEF, MoFi, Genelec, Neumann, and Kali offer speakers that are nearly perfect for their specific use cases and budgets. These speakers consistently demonstrate:

• Highly linear frequency response
• Excellent directivity
• Inaudible distortion, even at high SPL levels
• Deep bass extension relative to their size

Moreover, most of these companies publish their performance data, providing transparency and showcasing their commitment to honesty with customers.

This leads me to wonder: Are speakers, much like amplifiers and DACs, now a solved problem—even for budget options? If so, what innovations remain to be explored?

I’ve also noticed that the top-performing speakers today are predominantly coaxial. Do you think coaxial designs will eventually replace conventional tweeter-woofer arrays if they continue to prove superior?

I’m essentially thinking out loud here, but it’s an exciting time for music and movie enthusiasts.

Generally speaking, I would say yes it's solved.

 

[Oddball]

Actually, the speakers are the only remaining problem in the Hi Fi and HT world. Other problems were or are solved. Very good and great speakers are so expensive that it is prohibitive to most of the customers. That is a huge problem waiting to be solved.

 

[ChrisG]

how would you go about optimizing for a “they are here” approach?

By starting with an acoustically-pleasant space (whatever that means) and then playing back via loudspeakers which are intended to mimic the original instruments, positioned however the listener sees fit. Perhaps all of those loudspeakers should be attached to CNC arms, so they can be re-aimed/positioned arbitrarily.
True optimisation in that direction would resemble a science experiment, but I suspect the results could be very good.

In this mode, having the acoustic image localise to a loudspeaker isn't a problem. In fact, it's intended: one instrument, one playback speaker.
The imaging, then, should be perfect.

There are few serious attempts at this stuff. Here's one: https://www.moma.org/collection/works/87291
Though I'd suggest that the speakers used will just happen to be approximately the size/radiation pattern of a human voice, rather than chosen specifically for that purpose.


Chris

 

[EERecordist]

Since the introduction of the CTA-2034 in 2015, scientific speaker design has made significant strides. The availability of more affordable measurement equipment—both hardware and software—for professionals and consumers alike has played a crucial role in this advancement.

Today, manufacturers like KEF, MoFi, Genelec, Neumann, and Kali offer speakers that are nearly perfect for their specific use cases and budgets. These speakers consistently demonstrate:

• Highly linear frequency response
• Excellent directivity
• Inaudible distortion, even at high SPL levels
• Deep bass extension relative to their size

Moreover, most of these companies publish their performance data, providing transparency and showcasing their commitment to honesty with customers.

This leads me to wonder: Are speakers, much like amplifiers and DACs, now a solved problem—even for budget options? If so, what innovations remain to be explored?

I’ve also noticed that the top-performing speakers today are predominantly coaxial. Do you think coaxial designs will eventually replace conventional tweeter-woofer arrays if they continue to prove superior?

I’m essentially thinking out loud here, but it’s an exciting time for music and movie enthusiasts.

Speakers have really improved with new materials.

(Analog) transducers are harder than electronics. Microphones, phono cartridges, and speakers continue to evolve.

There is only so much demand. Although there are large mergers, the brands are still distinct, and there is still a large speaker driver manufacturing ecosystem - they seem to be stable and not merging. It is the same with fine microphones.

 

[Anton S]

Speakers have really improved with new materials.

(Analog) transducers are harder than electronics. Microphones, phono cartridges, ans speakers continue to evolve.

There is only so much demand. Although there are large mergers, the brands are still distinct, and there is still a large speaker manufacturing ecosystem - they seem to be stable and not merging. It is the same with fine microphones.

Well said.

 

[Oddball]

Well I see that most posters are just ignoring the reality which I can't. So bumping this up - speakers are the real problem as they are so expensive. Nothing has been solved as the price point where it has been solved is still way too high.

Since I acquired my first batch of reference speakers some 15 years ago nothing has really changed. Spend another 10K for LCR and 12K for subs to get the system up to reference levels. So not accounting for depreciation and in today's money, I have north of 50K in speakers. Sure - all problems are solved but then isn't the price sticker the real problem?

 

[mhardy6647]

Q: "Are speakers actually already a solved problem?"
A: Yes.

 

[Oddball]

 

[krabapple]

Short answer, no. If speakers were a solved problem, the sonic differences among them would be minimal, and you'd be relegated to choices based primarily on cosmetics. That is still very much not the case.

We 'know' from research, what performance characteristics listeners associate with 'good speaker sound'. It forms the basis for recommending speakers from Klippel data, as Amir does here, for example. To that extent, speakers are a solved problem. (It doesn't mean that the listerner experience cannot be made even better -- e.g., by adding channels)

But not all speaker manufacturers aim for those characteristics, and not all speakers can physically achieve them (e.g., small speakers really cannot output the bass that larger ones can, and bass performance accounts for ~30% of listener preference for a speaker).

So the correct comparison is between speakers with the same performance aims. Whether the audible differences among them would be minimal is actually a proposition that has not been rigorously tested, to my knowledge. (another way of coming at this, would be to make different speakers ostensibly sound 'the same' using DSP, and see if that is achievable in practice)

No one would ever expect all speakers -- which have to accomodate a large variety of use cases -- to sound 'minimally different'. But we know how to make speakers that are likely to 'sound good' under fair comparison conditions.

 

[Blumlein 88]

Well I see that most posters are just ignoring the reality which I can't. So bumping this up - speakers are the real problem as they are so expensive. Nothing has been solved as the price point where it has been solved is still way too high.

Since I acquired my first batch of reference speakers some 15 years ago nothing has really changed. Spend another 10K for LCR and 12K for subs to get the system up to reference levels. So not accounting for depreciation and in today's money, I have north of 50K in speakers. Sure - all problems are solved but then isn't the price sticker the real problem?

I don't think this is quite right. We have speakers that perform much better than in the past at a given price point. Not all do and some of the most expensive disappoint. That is due to not all manufacturers being up to date on speaker design. I also think the very top will always be moderately expensive because of size needed for large rooms high sound levels and low distortion. Even there if chosen carefully you get better for less.