Are speakers actually already a solved problem?

[GXAlan]

Meaning, the real challenge of reproducing the sound of a grand piano is more about the limitations of stereo sound.

+1. It’s probably why Steinway Lyngdorf Model D works. Sadly it’s the same price as the piano!

So my main question is: Why can't the speaker manufacturers come to the same conclusion of how the ultimate dispersion characteristics should be?

Poll: What is your listening distance? (Fall/Winter 2024)

What’s your listening distance? If you have multiple systems, pick all that apply. (Used at least once a week) We have had previous polls in the past but let’s sample the membership again. Feel free to comment with typical SPL or details about your systems. Since many of us have headphones...

www.audiosciencereview.com

Listening distance is also a proxy for other stuff in the room and room size. The ideal dispersion is room and speaker position dependent. That’s why Meyer Sound offers the same speaker with different dispersion options.

 

[Alexander Gödde]

I think whether speakers are a solved problem is the wrong question. We need to solve speakers + room (or headphones + individual head).

 

[Haskil]

The other challenge is to take a grand piano, listen to it in a real room and compare it to the recording of a piano on any level of system. Even Amir will say that they generate a different sound. (It’s all due to the way the sound is produced and dispersion characteristics.)

Wouldn't it be more logical, I ask, to compare the sound of the grand piano to its recording made in the same room?

 

[Keith_W]

So my main question is: Why can't the speaker manufacturers come to the same conclusion of how the ultimate dispersion characteristics should be?

I just finished typing this post on my show report. I asked a speaker designer a similar question:

Me: "Can speakers designed for 2 channel be used for home theatre?"
Him: "It depends on how wide your seating area is. 2 channel speakers typically have more narrow directivity and a smaller sweet spot. Narrow directivity is great if you want to avoid side wall reflections, but not if you want to provide a good experience for more listeners who may be sitting quite far off axis".

So the answer is: different people have different rooms, and may need to cater for more than one listener. Also different taste as to how much room influence they want.

 

[goat76]

Wouldn't it be more logical, I ask, to compare the sound of the grand piano to its recording made in the same room?

The problem with a comparison like that is that a “double effect” of the reflections from the room would occur when listening to the playback of the recording made in the same room.

 

[kharan]

What does a grand piano sound in a 25 m2 room at about 3 m away from it? And how often is that the listening environment? I figure it's not that common.
(about 270 square feet 11 feet away)

Rather, I figure what we need to reproduce a grand piano in a 250 m2 room and 10 m away from it.

And what we actually reproduce is what the audio engineers thought sounded closest to the above on their reference system, or to be even more accurate, what we reproduce is what the engineers thought would translate into the above on a consumer system based on what they heard on their reference system.

That's before even considering sounds that have no physical instruments to refer to.

Then there are things like BACCH or the Smyth Realizer, which according to their owners make recordings sound more real, yet no one has ever mastered a recording with those DSPs in mind, they are unquestionably not faithful to the recording.

 

[Haskil]

What does a grand piano sound in a 25 m2 room at about 3 m away from it? And how often is that the listening environment? I figure it's not that common.
(about 270 square feet 11 feet away)

Rather, I figure what we need to reproduce a grand piano in a 250 m2 room and 10 m away from it.

And what we actually reproduce is what the audio engineers thought sounded closest to the above on their reference system, or to be even more accurate, what we reproduce is what the engineers thought would translate into the above on a consumer system based on what they heard on their reference system.

That's before even considering sounds that have no physical instruments to refer to.

Then there are things like BACCH or the Smyth Realizer, which according to their owners make recordings sound more real, yet no one has ever mastered a recording with those DSPs in mind, they are unquestionably not faithful to the recording.

An excellent pianist will not play the same way in a 25 m2 living room as in a large room... The pianissimo will not be the same... neither will the fortissimo. And a great pianist will give the illusion of getting the maximum sound level from the piano he plays in a small room even though he will not exceed double fortissimo...

 

[kharan]

An excellent pianist will not play the same way in a 25 m2 living room as in a large room... The pianissimo will not be the same... neither will the fortissimo. And a great pianist will give the illusion of getting the maximum sound level from the piano he plays in a small room even though he will not exceed double fortissimo...

Agreed, my point trying to reproduce the full sound output of a grand piano wasn't necessary for good reproduction. It's an instrument meant to fill a concert hall. We only need to reproduce the volume of that piano (and how the pianist is playing it) in a typical listening position, and that's a lot easier for a speaker system.

And that's without taking into account the sound engineers who will make mastering decisions so that their recording sounds good on airpods, on sound bars and a classic Hi-Fi system alike. And that different engineers will end up making different choices.

 

[Purité Audio]

Wouldn't it be more logical, I ask, to compare the sound of the grand piano to its recording made in the same room?

As has been touched on, propagation from an actual instrument is very different from a pair of loudspeakers, microphone position makes a huge difference, one can only really hope for the finest reproduction of the recording .
Keith

 

[Haskil]

Agreed, my point trying to reproduce the full sound output of a grand piano wasn't necessary for good reproduction. It's an instrument meant to fill a concert hall. We only need to reproduce the volume of that piano (and how the pianist is playing it) in a typical listening position, and that's a lot easier for a speaker system.

And that's without taking into account the sound engineers who will make mastering decisions so that their recording sounds good on airpods, on sound bars and a classic Hi-Fi system alike. And that different engineers will end up making different choices.

A grand piano is intended to be played in a room that extends from the grand salon to the Colon Theater in Buenos Aires... Like a little violin or a cello.
The crap that bad record publishers do is another story.

 

[Pearljam5000]

Headphones are the closest to being a solved problem as you take the room out of the equation
I don't see that happening with speakers anytime soon

 

[Scytales]

The problem with a comparison like that is that a “double effect” of the reflections from the room would occur when listening to the playback of the recording made in the same room.

Absolutely.

The aforementioned Cabasse manufacturer acknowledged that issue. Thus, the live vs recording comparisons where made in acoustically suitable rooms, but the audience where briefed nevertheless about that issue.

I translate from French the text of a Cabasse leaflet of the time about this kind of comparison : "What causes can lead to audible differences [between live and the reproduction of the recorded sound]? [...] Reverberation plays an important role: recorded by the microphones, it will appear twice when listening to the Hi-fi system but only once when listening to the orchestra live."

 

[rank]

That's why we use soundboard.

 

[BullBuchanan]

The other challenge is to take a grand piano, listen to it in a real room and compare it to the recording of a piano on any level of system. Even Amir will say that they generate a different sound. (It’s all due to the way the sound is produced and dispersion characteristics.)

Do we know that is a playback problem and not a recording problem? There are a million speakers that have come out over the decades, but most professional recordings use the same handful of microphones that have existed for decades.

Is it possible that top of the line speakers in 2024 can reproduce a Nuemann u87 recording of a piano perfectly, but that u87 (which came out in 1967) just may not be capturing it as it was in the room?

That's even a pretty extreme example. Most pro audio gear is leagues below what hifi enthusiasts use. Plenty of engineers are mixing and mastering on headphones like ATH MX 50s, DT-1990s, Yamaha HS-8s etc. If that gear can't reproduce the recording gear 1:1, then how were they expected to account for that in the recording? At what point are we hearing the limitations of recording capabilities vs limitations of sound reproduction?

 

[dshreter]

The best speakers are also still extremely expensive. Not until the best speakers are like an iPhone, where an average consumer can purchase the best performance with the best design and best ergonomic considerations would I consider this solved.

There are other aspects like cardioid and adaptive designs. A textbook spinorama makes for the “best” general purpose designs but it isn’t necessarily optimally performing in any given real room.

 

[kemmler3D]

The best speakers are also still extremely expensive. Not until the best speakers are like an iPhone, where an average consumer can purchase the best performance with the best design and best ergonomic considerations would I consider this solved.

There are other aspects like cardioid and adaptive designs. A textbook spinorama makes for the “best” general purpose designs but it isn't necessarily optimally performing in any given real room.

Agree.

DACs are a "solved problem" because even a lower-mid-tier DAC has transparent sound. It's a trusim around here that "any half-decent" amp or DAC is transparent.

This is not how it is with speakers, not even close.

Are there multiple examples of speakers that seem to show similar, near-ideal performance on certain parameters? Yes.

Is "any half-decent speaker" in 2024 going to deliver inaudible distortion, high dynamic range (>100dB) and ideal FR and dispersion? Hell no. You can go out to any electronics store, pick the middle-priced option, and end up with bad FR, distortion AND dispersion.

I am also wondering if "inaudible distortion" is really a common state of affairs at all. We're used to transducers having considerable THD and IMD. But anecdotally people say the Purifi drivers (which employ patented means of reducing IMD) sound different.

I would agree with others who have said the physics / science is verging on settled, but the market is far from a "spinning our wheels to make inaudible distortion more inaudible" state like the electronics market is.

I also agree that we're whistling past the dynamic range graveyard here. There are not many cheap speakers that can do >100dB bass, let alone 110dB. While we can debate the wisdom or necessity of such high SPL, there are many who will strenuously argue it's necessary for realism, and it's really hard to argue the difference wouldn't be audible...

Speakers aren't "solved" IMO until you can get the total package at a consumer-friendly price from multiple brands... certainly well under $1K per pair.

 

[anphex]

The best speakers are also still extremely expensive. Not until the best speakers are like an iPhone, where an average consumer can purchase the best performance with the best design and best ergonomic considerations would I consider this solved.

There are other aspects like cardioid and adaptive designs. A textbook spinorama makes for the “best” general purpose designs but it isn’t necessarily optimally performing in any given real room.

Extremely? You would consider KEF R-Series expensive in relation the performance they got? Also remember passive speakers are basically FOR LIVE. So the comparision with an iPhone that gets thrown out every other year doesn't stand.

I also don't really know how the price should become even lower. Consider material, design, manufacturing, logistics, shipping, long-term-revenue for the company. Considering this, Kef R-Series are stupid cheap, ridiculously so.

Also the only working cardiod speakers I know are ACTIVE, which is basically cheating and not really speaker design imho.

 

[Paffi]

Also the only working cardiod speakers I know are ACTIVE, which is basically cheating and not really speaker design imho.

I think that should not be generalized - as far as I know, the cardiod implementation of ME Geithain is different to the one in more "modern" designs (D&D, GGNTK etc.). Despite beeing somehow active (amplifier built in), no DSP "cheating" is implemented.

 

[JustJones]

Not until the best speakers are like an iPhone, where an average consumer can purchase the best performance with the best design and best ergonomic considerations would I consider this solved.

If you're using iPhone as a generic holder I agree but as someone who would never own an iPhone , yes I've used them to me they suck so I would say smart phones not specific to one brand.

 

[sigbergaudio]

Extremely? You would consider KEF R-Series expensive in relation the performance they got? Also remember passive speakers are basically FOR LIVE. So the comparision with an iPhone that gets thrown out every other year doesn't stand.

I also don't really know how the price should become even lower. Consider material, design, manufacturing, logistics, shipping, long-term-revenue for the company. Considering this, Kef R-Series are stupid cheap, ridiculously so.

Also the only working cardiod speakers I know are ACTIVE, which is basically cheating and not really speaker design imho.

For clarity, neith our design (Sigberg Audio Manta) nor I suspect the Dutch&Dutch 8C or Geithain use the DSP directly to create the cardioid effect, it is passive / mechanical in nature. So while the speakers are active, they didn't have to be to create the cardioid effect.

EDIT: I also think it is an interesting perspective to consider active speakers to be "cheating". Care to elaborate?