Those of you who believe measurements aren't the whole story, do you have a hypothesis why that is?

[MarkS]

But you could also take the SoundLab speakers that I mentioned to Blumlein 88 above and pull them out into the room maybe 5-7 feet in front of the wall behind them, and toe them in a bit. They approximate a line source so there would be no significant early floor or ceiling reflections; they have a dipole null to the sides so they could be aimed to have no early sidewall reflections which reach the listening position; and the round-trip path length for the backwave energy is long enough that it won't contribute significantly to "small room signature" but WILL contribute to the desirable later-arriving reverberant energy.

I've heard SoundLabs set up that way (at a show many years ago) and they were spectacular!

Each speaker has two fairly directional arrays, one aimed at the listening area (toed-in aggressively to avoid the near-side walls) and the other aimed away from it, I can post photos if you'd like. The spectral balance of each array can be adjusted independently, and the level of the rear-firing array can also be adjusted independently.

This is very cool! These are your own DIY builds? Please do post photos! I suggest starting a new thread on them.

The number of speakers on the market with user-adjustable directivity is very small ...

 

[Wes]

More like Orlando, just before New Year's day.

I was thinking of the raid on Entebbe airport...

 

[dfuller]

Generally speaking, if measurements don't tell the whole story it's because we aren't using the right measurements. For example, Benchmark noting an audible difference from intersample overs that didn't necessarily show up in traditional THD+N measures. https://benchmarkmedia.com/blogs/application_notes/listening-vs-measurements

Speakers are the final frontier for measurements, though - largely electronics have been figured out.

 

[dc655321]

So here is the big question: Is a combination of flat direct sound plus non-flat reflected sound the ONLY valid way to achieve our preferred perceived frequency response?

Maybe not. Let me describe one theoretical alternative approach:

Suppose we prioritize minimizing the spectral discrepancy between the direct and reflected sound, such that the direct sound's frequency response becomes MORE like the (non-flat) preferred steady-state in-room frequency response, and the reflected sound becomes MORE like the direct sound (closer to flat)? In other words, have the two approximately "meet in the middle". Might there be a subjective benefit to having the spectral balance of the reflections be virtually identical to that of the first-arrival sound? Well, we don't know - it hasn't been researched (though I can think of arguments why it might be). Is there an obvious reason why this approach would not be viable? If so let me know, because such is not apparent to me.

Let me know your thoughts thus far, if you don't mind. (And no, I'm not necessarily talking about an omnidirectional system, though that is one possible approach.)

My thoughts are that maybe you're describing Geddes' approach (cf. diagram at top of pg 9)?
If so, I can't say whether that is strictly correct when assessing from a "true to source" perspective, as the purpose is dissimilar.

I tend to look at the problem of acoustic reproduction as being strictly fidelity to the source.
Some may see that as too simplistic. But, it is all that is strictly required IMO, and maybe the best we can do without exotic designs.

 

[Duke]

My thoughts are that maybe you're describing Geddes' approach (cf. diagram at top of pg 9).

GOOD CALL!!

I am indeed a disciple of Earl Geddes (at one time he and I were partners on the project which became the Summa), but he does not approve of some of the things I do, in particular my use of additional drivers to manipulate the reverberant sound.

I tend to look at the problem of acoustic reproduction as being strictly fidelity to the source.

That sounds good, but what exactly do you mean by "strict fidelity to the source"? If you mean "flat" response, then along which axes? My point being, given that drivers are generally variable-pattern transducers of electrical signals into pressure waves which propagate in a finite three-dimensional space, how exactly are you evaluating for "strict fidelity" to a two-dimensional (voltage modulations over time) signal?

I tend to look at acoustic reproduction as inclusive of room interaction as well as the psychacoustics of human hearing, the latter informing "where the goal posts are" for desirable acoustic characteristics.

Or to put it another way, imo a speaker should get two things right: The direct sound, and the reflections.

Or to put it a third way, imo the ONE thing a speaker should do is, create the perception of hearing live music.

 

[LTig]

[..] We have a body of research which shows that a combination of flat direct sound and non-flat reflected sound which sums to a steady-state in-room frequency response curve along the lines of those shown in post #13 results in a subjectively flat, or at least subjectively preferred, perceived frequency response. (The "trained listeners" curve is arguably the most reliable one.)

So here is the big question: Is a combination of flat direct sound plus non-flat reflected sound the ONLY valid way to achieve our preferred perceived frequency response?

Maybe not. Let me describe one theoretical alternative approach:

Suppose we prioritize minimizing the spectral discrepancy between the direct and reflected sound, such that the direct sound's frequency response becomes MORE like the (non-flat) preferred steady-state in-room frequency response, and the reflected sound becomes MORE like the direct sound (closer to flat)? In other words, have the two approximately "meet in the middle". Might there be a subjective benefit to having the spectral balance of the reflections be virtually identical to that of the first-arrival sound? Well, we don't know - it hasn't been researched (though I can think of arguments why it might be). Is there an obvious reason why this approach would not be viable? If so let me know, because such is not apparent to me.

I think it is not viable. You forgot the timing aspect. As I understand Toole's book the FR of the direct sound (which reaches the ears first) defines the perceived sound quality of the speaker. The effect of the first reflections merely enhances the perceived width (Haas effect). Therefore I think a non flat on axis FR and a slightly falling smooth off axis FR won't be perceived like a flat on axis FR and a stronger falling smooth off axis FR. I would expect such a speaker to sound a little too warm.

 

[Trif]

I was never able to accept that many audiophiles believe there's something just mystical about human hearing that simply can't be captured by science. And frankly I don't really think they believe that. But at the same time I don't think I ever heard or read a hypothesis about it, no matter how far-fetched. OK, maybe there is the "typical measurements rely on steady state signals and average certain kinds of distortions" but that's pretty much dismantled. I really don't believe there's a black and white divide between engineering types and the ones that simply trust their hearing without any interest for scientific explanation, that's just an exaggeration of the Internet era, it does a perfect job of making all shades of gray appear black or white as we all know. I'm really hoping for an interesting discussion.

When I was a youngster, they showed me a straight line, told me it was perfect frequency response, and said their speaker would reproduce all music and speech faithfully. It didn't. Then another company showed me another line, this one with a few bumps. They said it was done in an anechoic chamber, and this big step forward.... still didn't sound right. I'll spare you the intervening 3/4 of a century but the fact remains that "science", as it's been portrayed (or is that "betrayed"?) in the media has a long, long history of claiming to understand sound recording and reproduction and then forcibly demonstrating quite the opposite.

I don't think you need to accept mystical pixies as real to understand that we measure what we can, not necessarily what we should, or would if we could. As our measuring techniques improve we're able to increase the accuracy of our manufacturing. We have "spatial imaging" now. We sure didn't have that in the '70s, given the junk they were stuffing in cabinets then.

Anyway, the point is, you've turned the question around. Asking the science minded types here for hypotheses about mystical.... wouldn't it be more to fun to go over to some golden earred site and ask them to hypothesize about why we measure things?

Never forget that we are listening softly! Usually, we are trying to reproduce a performance with 1/10th to 1/100th the acoustic power of the original.

Since he's here, HI DUKE!!!

Listen to Duke, he makes sense. Don't try to listen at realistic levels like I did.

HUH?!? WHAT?

Oh. The hypothesis. They're the sort of men that need to compensate in some way that can't be called out - after all, who can say what you really heard? It's safely tucked in your head. So they claim to hear things that only the anointed can hear and [you can insert the appropriate psych book chapters about cult mentality here]. These people are the market for impressive speakers that boom-tweet their way into your heart for an afternoon or so of carefully selected demo pieces. They only turn them on to show off for others.

 

[Duke]

I think it is not viable. You forgot the timing aspect. As I understand Toole's book the FR of the direct sound (which reaches the ears first) defines the perceived sound quality of the speaker. The effect of the first reflections merely enhances the perceived width (Haas effect). Therefore I think a non flat on axis FR and a slightly falling smooth off axis FR won't be perceived like a flat on axis FR and a stronger falling smooth off axis FR. I would expect such a speaker to sound a little too warm.

I'm very much aware of timing aspects; I just haven't written about them much in this thread. Reflections contribute to perceived timbre (singing in the shower being a classic example). If the perceived timbre was too warm, you'd make adjustments until it sounds right.

A few days ago I asked Floyd Toole this question:

"What are your thoughts about steady-state room curves for omnidirectional or polydirectional loudspeakers? Would there be any advantage to reducing the spectral discrepancy between the direct and reflected sound, something your old bioplar Mirage M-1's may have done?"

Here is his response:

"My personal experience is too limited to answer with any certainty, and nobody I know of has pursued this as a research question. The well-behaved multi-directional speakers I have heard sound just fine, and are a viable alternative to forward firing designs. They appear to add a bit more "space" to a stereo presentation, and for those mixes that have hard-panned L and R images they might present a 'softer' image. But real investigation is needed to give a certain answer." [emphasis Duke's]

Link for the skeptical.

One theoretical benefit of minimizing the spectral discrepancy between the direct and reverberant sound is this: The ear/brain system examines the spectral content of incoming sounds in order to correctly classify them as either new sounds or reflections (repetitions of a recent new sound). The greater the spectral discrepancy, the harder the ear/brain system has to work to make this classification. The brain has limited CPU power available, and the more it has to devote to subconscious tasks such classifying reflections, the less is available for higher cognitive functions such as focusing on and fully appreciating the music.

My understanding is that, in extreme cases, spectral discrepancy between the direct and reflected sound may contribute to listening fatigue. On the other hand ime good loudspeakers which minimize the spectral discrepancy between the direct and reverberant sound are exceptionally relaxing to listen to long-term, assuming they are set up correctly.

 

[JJB70]

There is a deeper truth in that if you like an audio system and it lets you enjoy music then why does anything else matter? I have listened to music on some terrific systems,on soundbars gaming headsets, wireless speakers, lifestyle systems and much else and I find that it really doesn't affect my enjoyment of music that much. My hobby audio gear is a classic Sony ES setup, my everyday system is a Technics lifestyle system and most of the time I use my laptop or phone with headphones or IEMs, I am perfectly happy with all options.

 

[LTig]

I'm very much aware of timing aspects; I just haven't written about them much in this thread. Reflections contribute to perceived timbre (singing in the shower being a classic example). If the perceived timbre was too warm, you'd make adjustments until it sounds right.

A few days ago I asked Floyd Toole this question:

"What are your thoughts about steady-state room curves for omnidirectional or polydirectional loudspeakers? Would there be any advantage to reducing the spectral discrepancy between the direct and reflected sound, something your old bioplar Mirage M-1's may have done?"

Here is his response:

"My personal experience is too limited to answer with any certainty, and nobody I know of has pursued this as a research question. The well-behaved multi-directional speakers I have heard sound just fine, and are a viable alternative to forward firing designs. They appear to add a bit more "space" to a stereo presentation, and for those mixes that have hard-panned L and R images they might present a 'softer' image. But real investigation is needed to give a certain answer." [emphasis Duke's]

Link for the skeptical.

One theoretical benefit of minimizing the spectral discrepancy between the direct and reverberant sound is this: The ear/brain system examines the spectral content of incoming sounds in order to correctly classify them as either new sounds or reflections (repetitions of a recent new sound). The greater the spectral discrepancy, the harder the ear/brain system has to work to make this classification. The brain has limited CPU power available, and the more it has to devote to subconscious tasks such classifying reflections, the less is available for higher cognitive functions such as focusing on and fully appreciating the music.

My understanding is that, in extreme cases, spectral discrepancy between the direct and reflected sound may contribute to listening fatigue. On the other hand ime good loudspeakers which minimize the spectral discrepancy between the direct and reverberant sound are exceptionally relaxing to listen to long-term, assuming they are set up correctly.

I see your point. I don't dispute that minimal discrepancy may be good, I just dispute (based on Toole's book) that non flat on axis FR is a good target. So to minimize discrepancy one need to go the omni way. However I listened to omnis (by Falcon Lab, don't remember the model) on Highend 2019 in Munich and found them too bright for my taste, not relaxing at all.

 

[Duke]

I see your point. I don't dispute that minimal discrepancy may be good, I just dispute (based on Toole's book) that non flat on axis FR is a good target.

Fair enough! If I were in your shoes, I'd trust Toole's book LONG before I'd trust some bozo on the internet spouting what looks like speculation mixed with conjecture.

So to minimize discrepancy one need to go the omni way. However I listened to omnis (by Falcon Lab, don't remember the model) on Highend 2019 in Munich and found them too bright for my taste, not relaxing at all.

Omni is one way of minimizing the discrepancy, but there are others. See post #98.

 

[Jim Matthews]

I was thinking of the raid on Entebbe airport...

Have you *been* to Orlando?

Other than Israeli commandos, it's much the same.

 

[Jim Matthews]

There is a deeper truth in that if you like an audio system and it lets you enjoy music then why does anything else matter? My hobby audio gear is a classic Sony ES setup, my everyday system is a Technics lifestyle system and most of the time I use my laptop or phone with headphones or IEMs, I am perfectly happy with all options.

You mean you're not about fixing it until it's broken?
How did a shameless heretic like you get in here?

 

[escksu]

Nope.

Whatever rocks your boat. I only state my points. You do not have to agree with them.

 

[JSmith]

How about "How does a person know if the speaker the reviewer reviewed has any resemblance to the one the person is buying"?

Once again you don't... you trust in the quality of the brand. It's the same for many things... like baby milk powder, people trust in the brand and rely on the breakdown of the ingredients and nutrition on the package. From there people need to assume it is what it says it is and feed it to their child.

Recently a company was caught swapping the op-amp in a product among other issues... that brand lost some of it's shine due to same.

Unless you go to the factory and watch your specific device being made or you know how to thoroughly inspect the internals... what else does one do otherwise? They rely on measurement reviews, user experience and reputation (customer service and build quality).



JSmith

 

[Freeway]

Spin-sonics, nano-acoustics waves and phononic components. Predicted by Lord Rayleigh 1885.
https://phys.org/news/2021-07-spin-...&utm_medium=email&utm_campaign=daily-nwletter

 

[escksu]

I'm a little confused, and curious... If imaging information is not in the source content (signal from the speakers), which should be measurable (however involved), then does the brain manufacture (synthesize) information that is not there? Feelings I can grasp as not being measurable, though often enough you can point to an external cause if present (some event in life), but figuring out an image from a pair of speakers it seems like you have only the signal from them plus whatever your brain provides. So your brain must add something (unm E easurable) to the signal, is that correct?

I confess I have always considering the sound field as calculable and measurable, especially since things like MATLAB and COMSOL plus more advanced measurement software became available, but again based upon very different image processing theory. And of course limited in complexity -- an arbitrary room is hard to calculate, though is usually measurable.

"Last night I saw, upon the stair, a little man who wasn't there..." - Don

Perhaps you should google oh how our brain can be fooled. You might have come across some illusions which fools our brain. Some examples of illusions. More than just eyes, ears and touch too.

https://www.vox.com/science-and-hea...llusion-science-humility-reality-polarization

https://www.hear.com/useful-knowledge/auditory-illusions/

https://www.scientificamerican.com/article/touching-illusions-2008-05/

Try this yourself:

https://en.wikipedia.org/wiki/Spinning_dancer

This is a famous illusion. In reality, the dancer is not spinning. This is purely 2D animation (there is no depth). but your brain created the illusion that she is spinning. Some sees is clockwise, some counterclockwise. Some can make her change direction (clue: try focusing on the "reflection" at the bottom).

This is to show that our brain do at times add/remove information on its own or process information differently from what we expected.

Btw, you can try plugging 1 ear and see if there is a difference between hearing with 1 ear and with both ears. Also try covering 1 eye (clue: your perception of depth changes).

 

[DonH56]

Perhaps you should google oh how our brain can be fooled. You might have come across some illusions which fools our brain. Some examples of illusions. More than just eyes, ears and touch too.

https://www.vox.com/science-and-hea...llusion-science-humility-reality-polarization

https://www.hear.com/useful-knowledge/auditory-illusions/

https://www.scientificamerican.com/article/touching-illusions-2008-05/

Try this yourself:

https://en.wikipedia.org/wiki/Spinning_dancer

This is a famous illusion. In reality, the dancer is not spinning. This is purely 2D animation (there is no depth). but your brain created the illusion that she is spinning. Some sees is clockwise, some counterclockwise. Some can make her change direction (clue: try focusing on the "reflection" at the bottom).

This is to show that our brain do at times add/remove information on its own or process information differently from what we expected.

Btw, you can try plugging 1 ear and see if there is a difference between hearing with 1 ear and with both ears. Also try covering 1 eye (clue: your perception of depth changes).

I get all that; I am no expert, but not completely ignorant. The question is, is imaging simply our brain fooling us, or is that a practical, measurable element to it?

I think we're going in circles, I'll bow out. Thank you for the information!

 

[dc655321]

That sounds good, but what exactly do you mean by "strict fidelity to the source"? If you mean "flat" response, then along which axes?

Strict fidelity can only mean one thing, output equals input, along the direct axis.
If a speaker's frequency response is compromised for near-field use, that's a no-go
And your notion of mixing direct and off-axis to a sum-of-least-offense would not do it for me (though it's likely not a massive compromise).

I tend to look at acoustic reproduction as inclusive of room interaction as well as the psychacoustics of human hearing, the latter informing "where the goal posts are" for desirable acoustic characteristics.

Or to put it another way, imo a speaker should get two things right: The direct sound, and the reflections.

Or to put it a third way, imo the ONE thing a speaker should do is, create the perception of hearing live music.

Like I've tried to say (poorly), I don't place much weight on human perception of transducer output.
If the input is reproduced faithfully along some axis, and not butchered on the off-axes, that's about all I could ask for.

Expecting a speaker to transform all inputs into some mystical live music experience seems foolish to me.
Kinda cool, but the variability of rooms and (especially!!!) recordings will squash that fantasy.

 

[Blumlein 88]

I get all that; I am no expert, but not completely ignorant. The question is, is imaging simply our brain fooling us, or is that a practical, measurable element to it?

I think we're going in circles, I'll bow out. Thank you for the information!

Q-sound proves stereo speakers can fool our brain. It can image objects behind or well beside you over two channels only.

Now that doesn't mean all imaging is an illusion. Our normal hearing in the real world is capable of letting us pinpoint where something is with pretty good precision. The main factors in this ability are used by normal stereo to help us hear what is an audio illusion. We hear sound coming from a position where there is no sound source.