Designing Speakers For Music As You Would Experience It Live - Interview with Roy Delgado

[antcollinet]

I do think this is an interesting phenomenon though, because we're all familiar with hearing music coming from the window and having a sense of whether it's live or not. I'd love to know more about what causes this and whether / to what extent it translates into in-room listening.

Never gonna get that study done until you can get a CEA/ANSI standard window approved.

 

[MattHooper]

Finally - for the customer - how do they find out that this particular manufacturers house sound is what they want or need, especially since you can't hear what you are buying until you get it into your room? Other than by expensive trial and error?

I fail to understand how this can be a winning strategy for anyone.

I just don’t think that’s as difficult as you seem to imply. I haven’t found it difficult to grasp a manufacturer’s house sound when auditioning the same brand of speaker in different locations. The B&W speakers seem to have an obvious house sound and similar traits every time I’ve heard them in different stores.

I identified some consistent traits in various brands from store auditions, which held up when I got those brands in my own room. I don’t think there’s anything unusual about this.

 

[Justdafactsmaam]

Just as in the McGurk effect you KNOW that the sound isn't changing, but you have a lifetime of understanding speech that tells your subconscious that *that* mouth shape sounds like Ba - and *this* one sounds like Fa -

A lifetime? More like millions of years of evolution. We are born wired this way. Can’t be learned or more importantly unlearned

 

[ta240]

What are these "other approaches" and "other goals"? Can they be quantified by measurements, or only by listening?

There are people that don't care the least about things being quantified or measured when they buy audio. They listen to it, they like it, they buy it, then they move on with their life. Ironically, many of them are far more satisfied when they listen to their music.

It is weird that within this forum bubble, it is assumed by most that the entire world desires or at the very least should desire, the same thing they do.

Finally - for the customer - how do they find out that this particular manufacturers house sound is what they want or need, especially since you can't hear what you are buying until you get it into your room? Other than by expensive trial and error?

I fail to understand how this can be a winning strategy for anyone.

Actually, many of these speakers like Klipsch are available locally with free returns. If you don't like it, take it back. My two purchases bought from recommendations here were not returnable; one went in the trash and the other was resold at a noticeable loss. My speakers I bought at BB because I loved the way they sounded could have been easily returned with zero shipping cost, but I still loved them at home.

An objectivist tends to believe that correct knowledge and interpretations of measurements, and the use of blind listening tests, without any visual prejudice, can give a rational and exhaustive description and evaluation of the audio performance of a system.
This is a rational and scientific approach to 360-degree audio evaluation, which is in contrast to a more classic, audiophile narrative, which tells of something that the human ear can perceive and that no measuring instrument could capture, for example example differences that are said to be perceived between cables that show identical measurements in every parameter, or miraculous audiophile DACs that despite normal measurements would have magical properties of purity and quality that well-measuring but economical DACs would not have (punctually without blind tests with matched levels to demonstrate that you feel a real difference).

I had a friend that loved the Socrates quote "An unexamined life is not worth living" as she would frequently talk about, think about and worry about things that she had done or not done. I often thought to myself "An over examined life is one not lived". And as such, an obsessed over sound system is probably enjoyed less than many vastly less perfect ones.

So, I counter your argument (which was more words than I cared to take in) with

"Unga Bunga, music good, me listen now"

 

[Duke]

My main point is that describing your subjective perception of the sound you heard while wandering along a corridor not even in the room with the speaker is not a convincing way to convey to anyone else very much about the speaker.

I don't expect you to believe or accept any observation I make as having any meaning for you.

What hearing the speaker outside the room often conveys to me (not to you) is, an approximation of what its power response sounds like.

Of course it is theoretically possible for a speaker with normal directivity (radiation pattern generally narrowing as we go up in frequency) AND a tipped-up on-axis response to actually have a more balanced frequency response from outside the room. In this scenario, the radiation pattern + tipped-up on-axis response could presumably "average out" to something more balanced than what we hear in the room with the on-axis response dominating.

I do think this is an interesting phenomenon though, because we're all familiar with hearing music coming from the window and having a sense of whether it's live or not. I'd love to know more about what causes this and whether / to what extent it translates into in-room listening.

IF a speaker sounds good from the listening position, AND if that speaker also sounds similar (and similarly good) from outside the room with no line-of-sight to the speakers, the implication is that there is relatively little spectral discrepancy between the direct sound and the reflection field. Note that this may not be possible to evaluate from outside the room if the room is heavily damped.

Similarity between the timbres of the direct sound and the reflections is desirable from the standpoint of both sound quality and spatial quality. Floyd Toole on the subject: "... the subjective effect of [the in-room] reflections became more desirable if they exhibited timbre similar to the direct sound... When the direct and delayed sounds are similar the localizations are more precise, and timbre is less degraded." (emphasis mine)

I think there may also be a reduction on long-term listening fatigue when the direct sound and the reflections are spectrally similar. Let me explain:

The ear/brain system is constantly analyzing incoming sounds in order to classify them as either "new" sounds or "reflections". If it's a reflection arriving within 40 milliseconds or so of the first arrival, then the "precedence effect" suppresses the reflection's directional cues, but the ear/brain system still picks up the loudness and timbral cues. I speculate that, beyond some threshold, the greater the spectral discrepancy between "new sounds" (the direct sound) and reflections, the harder the ear/brain system has to work to correctly identify the reflections as such, and if the discrepancy is large enough then, over time, this "increased CPU usage" can result in listening fatigue. I asked Floyd Toole about this on another forum and he said in effect that this hasn't been studied yet, but he didn't shoot it down. Here is his reply (emphasis mine):

"The precedence effect really boils down to the direct sound dominating localization. Perception of secondary images associated with delayed sounds (reflections in our context) defines the breakdown of the precedence effect. When the precedence effect is functioning, we are aware only of the direct sound as far as localization is concerned, but later sounds can affect loudness and timbre. This is discussed in some detail in my book, starting in Section 7.6.4. It is very signal dependent.

"One of the factors known to cause delayed sounds to be more apparent is a difference in spectrum. So the perceptual inhibition (NOT masking) of delayed sounds as just described is related to the spectral similarity of the direct and reflected sounds. Delayed sounds that are spectrally similar to the direct sound are most successfully inhibited, as far as localization is concerned. Nobody I am aware of has chased down the quantitative differences necessary to be useful in the context of sound reproduction. All that I can say is that loudspeakers exhibiting "similar" direct and off-axis performance receive elevated sound quality ratings, while those misbehaving off axis do not. Those are also the speakers the have the fewest audible resonances, and that tend to "disappear" behind the screen in double-blind tests.

"Direct sound dominates in terms of sound quality as was shown in my early papers in 1985-86, and later, and it is also dominant in terms of sound localization if it is not corrupted by spectrally dissimilar reflected sounds. Does it require more "CPU power" when off axis misbehavior exists? I can only speculate that if sound quality is degraded - which is evident in every program played - I can imagine that it would eventually become an annoyance, even if it is a subconscious one. This is where the science that I am aware of fades."

So spectral similarity between the direct sound and the reflections contributes to good sound quality and good imaging, and it might contribute to long-term fatigue-free listening.

And what a speaker sounds like from outside the room, with no line-of-sight to the speakers, can be one way of evaluating the spectral balance of the reflection field largely independent of the direct sound.

In my opinion.

 

[rynberg]

Understand or just you don’t agree?

If a speaker manufacturer deliberately engineers a non-linear house sound for its products and does it well by controlling directivity and keeping distortion low, then I see nothing wrong. This can still be good engineering.

In my opinion it’s a good thing if customers have choice in how their music is reproduced.

What is not good is a manufacturer spouting pseudo scientific nonsense to promote its badly engineered product even if appeals to some people.

I'm not the poster in this reply, but I agree entirely with you. I just think there are many examples of Klipsch speakers, particularly their "vintage" line, that exhibit poor engineering. Perhaps many models are regularly "on sale" and are a better value at that lower price, but I find most of their models to be overpriced at their normal prices.

 

[rynberg]

Another ASR who doesn’t understand there’s more than one approach and more than one goal for speaker designers.

I'm not allowed to disagree with the goal of making all content sound like it's being played through cheap PA speakers? I mean, I would suggest people just spend $1k on a set of powered 12" EV PA speakers instead of $4,000 for a Klipsch vintage line model.

 

[rynberg]

You obviously know your stuff. What would be your very top choice of passive speaker, bar none, in the $2000.00 to $2500.00 price range? I'd like to consider having a listen based on your experience and recommendation.

There is no "top choice" of speaker, it depends on your use case, personal preferences of soundstage precision vs. width, whether you'll use a sub, if you have EQ available, etc.

 

[Matt_Holland]

I fail to understand how this can be a winning strategy for anyone.

I wasn’t suggesting that it will be a winning strategy, just that good engineering can be used to deliver the performance aims.

We should acknowledge that another large hifi speaker manufacturer, B&W, has a non-linear house sound (boosted HF) and they sell bucket loads of speakers. So, like with Klipsch, it can be a winning formula.

 

[kemmler3D]

I can speak to this from experience, to an extent.

Understand. Toole et al demonstrated a (large?) majority preference for a particular in room response. So if a manufacturer is going to deviate from that, then from day one they are going to be limiting their market share.

House sound is far from the only limiting factor on a given brand's market share. Marketing budgets, access to distribution, and the degree of competition in one's niche actually have a much bigger impact, among other things.

Then - how do they decide what their house sound is going to be? How do they know which particular sound is going to suit a decent proportion of those who are left? Unless they repeat Toole's work, it seems the only way they have is to come up with something that sounds good to them?

This one is easy - look at what sells, and base your house sound on products from that brand. Apply some judgment and taste if you want.

Finally - for the customer - how do they find out that this particular manufacturers house sound is what they want or need, especially since you can't hear what you are buying until you get it into your room? Other than by expensive trial and error?

I fail to understand how this can be a winning strategy for anyone.

It's a winning strategy because among speakers with technically good performance according to Toole's work, the sound of each speaker is broadly similar and so you're forced to compete more on price, raw performance, durability / longevity, etc. These things all cost money above and beyond simply reaching good frequency response and directivity. If you are forced to compete directly with Genelec, KEF, Neumann, etc. you will have to work extremely hard to eke out a small market share.

However, if you can compete on house sound and matters of taste in house sound, you may find a niche that loves it, where everyone else hates it.

It's better to have 80% of 5% of the market than 1% of 100%. The principle of making something opinionated (even if that opinion is "wrong") where people either love it or hate it, rather than something that is generally acceptable to most people, is a sound business strategy.

To @Matt_Holland 's point, you see this in practice all over the place. Bose, B&W and Beats all made huge sums with highly colored sound, although I'd argue that's more from marketing than actual performance. However, you also see die-hard Klipsch, ZMF, Zu, Triangle, and other fans. They're employing a strategy of cornering a niche through actual (although deviant from the science) performance.

 

[MattHooper]

I'm not allowed to disagree with the goal of making all content sound like it's being played through cheap PA speakers? I mean, I would suggest people just spend $1k on a set of powered 12" EV PA speakers instead of $4,000 for a Klipsch vintage line model.

Of course you can disagree.

But the fact you have a certain goal does not make it a "mistake" when someone else has a different goal.

You prefer (it seems) as neutral a speaker as possible. Other people may not, and that includes the variety of goals and approaches among speaker designers.
Personally, as someone who enjoys comparing the qualities of live sound to reproduced as part of my criteria, I'm very glad there are some speaker designers who use a similar criteria. I've enjoyed designs by such companies, and I'm glad there are different strokes being produced for different folks.

 

[garyrc]

it is quite conceivable for a subconscious bias can make something sound other than the sound waves hitting your ears.

Thanks for a very good point. I agree. However, one word in your post, and those of others, pushed a button of mine.

Because,

We're losing language.

There were several posts somewhere on this forum concerning "subconscious" vs., "unconscious," but I can't find most of them at the moment.

Here is what the APA dictionary says:

I. Unconscious
n cognitive psychology, a mental process that occurs without a person being explicitly aware of it and largely outside of conscious control. Also called subliminal process.
n psychoanalytic theory, a psychical process that takes place in the unconscious; for example, repression.

II. Subconscious
Updated on 04/19/2018
n. a lay term that is widely used to denote the unconscious or preconscious mind as described by Sigmund Freud or the general idea of subliminal consciousness. It is also popularly associated with autosuggestion and hypnosis. Because of its imprecision, the term is now generally avoided by psychologists.

I was a Psych major, then went to graduate school in Psych, later worked 21 years in a Psych dept., and never heard the word "subconscious" spoken by faculty. It was always "unconscious."

Some other time, if they come up on this forum, I'll tackle the misuse of words such as "factoid," "schizophrenia" and "acoustic suspension.."

 

[MarkS]

You prefer (it seems) as neutral a speaker as possible. Other people may not, and that includes the variety of goals and approaches among speaker designers.
Personally, as someone who enjoys comparing the qualities of live sound to reproduced as part of my criteria, I'm very glad there are some speaker designers who use a similar criteria.

OK, I'm completely mystified by this comment. As someone who attends a LOT of live classical performances, my goal is to have my system sound as much as possible like the live sound that I hear at these performances. And I don't see how anything other than a neutral speaker would be in service of that goal.

How does a designer choose to deviate from neutrality in such a way as to make the sound "more live"? Please explain that design process to me.

 

[antcollinet]

Psych, later worked 21 years in a Psych dept., and never heard the word "subconscious" spoken by faculty. It was always "unconscious."

Language evolves - it is not a static thing. And as your dictionary shows, the word "subconcious" now has basically the same meaning as unconscious.

Further - you may have a regaional variation. English (UK) dictionaries don't make the distinction you outline above. In fact "unconscious" in the Cambridge Dictionary refers much more to the state of being ("he was unconscious when they found him"), than the workings of the brain.

 

[Purité Audio]

OK, I'm completely mystified by this comment. As someone who attends a LOT of live classical performances, my goal is to have my system sound as much as possible like the live sound that I hear at these performances. And I don't see how anything other than a neutral speaker would be in service of that goal.

How does a designer choose to deviate from neutrality in such a way as to make the sound "more live"? Please explain that design process to me.

There isn’t of course, you can try and persuade someone that added distortion somehow sounds more ‘live’ just marketing.
Having a coloured loudspeaker amongst neutral designs almost certainly helps at the point of sale, actually living with them another story.
Keith

 

[Duke]

You prefer (it seems) as neutral a speaker as possible. Other people may not, and that includes the variety of goals and approaches among speaker designers.
Personally, as someone who enjoys comparing the qualities of live sound to reproduced as part of my criteria, I'm very glad there are some speaker designers who use a similar criteria. I've enjoyed designs by such companies, and I'm glad there are different strokes being produced for different folks.

OK, I'm completely mystified by this comment. As someone who attends a LOT of live classical performances, my goal is to have my system sound as much as possible like the live sound that I hear at these performances. And I don't see how anything other than a neutral speaker would be in service of that goal.

How does a designer choose to deviate from neutrality in such a way as to make the sound "more live"? Please explain that design process to me.

Given that tradeoffs are virtually inevitable and that finances are seldom unlimited, a person may choose to trade off some objective neutrality for better spatial quality (better image separation and precision and/or a more realistic sense of immersion and envelopment). A person may choose to trade off some objective neutrality for better dynamics. A person may choose to trade off some objective neutrality for more bass extension. A person may choose to trade off some objective neutrality for a more relaxing and forgiving presentation. A person may choose to trade off some objective neutrality for the sake of practicality given the constraints they must work with (room, speaker positioning, WAF, whatever).

And other people may prioritize objective neutrality over all other considerations.

So I agree with Matt Hooper. I think having a wide range of choices in loudpeakers is a feature, not a bug.

 

[MattHooper]

Given that tradeoffs are virtually inevitable and that finances are seldom unlimited, a person may choose to trade off some objective neutrality for better spatial quality (better image separation and precision and/or a more realistic sense of immersion and envelopment). A person may choose to trade off some objective neutrality for better dynamics. A person may choose to trade off some objective neutrality for more bass extension. A person may choose to trade off some objective neutrality for a more relaxing and forgiving presentation. A person may choose to trade off some objective neutrality for the sake of practicality given the constraints they must work with (room, speaker positioning, WAF, whatever).

And other people may prioritize objective neutrality over all other considerations.

So I agree with Matt Hooper. I think having a wide range of choices in loudpeakers is a feature, not a bug.

Yes, thanks Duke. Since I’ve been down the road of this discussion exhaustively before, I’m not sure I’m up for it again at the moment. But you captured some of what I would say.

 

[MarkS]

Given that tradeoffs are virtually inevitable and that finances are seldom unlimited, a person may choose to trade off some objective neutrality for better spatial quality (better image separation and precision and/or a more realistic sense of immersion and envelopment).

OK, great. Now tell me: how do you design a speaker with "better spatial quality"? Does this correlate with any measurements? Or do you just tinker and listen, tinker and listen, until you happen to hit upon a design with this feature?

A person may choose to trade off some objective neutrality for better dynamics.

This one I sort of understand: I would guess that lower distortion at higher volume is what "better dynamics" means in practice (but please correct me if you think it's something else).

A person may choose to trade off some objective neutrality for more bass extension.

This one I get, in principle, but much easier and better to add a sub, at almost any budget.

A person may choose to trade off some objective neutrality for a more relaxing and forgiving presentation.

OK, again: how do you design a speaker with "a more relaxing and forgiving presentation"? Is there any measurement that this correlates with?

A person may choose to trade off some objective neutrality for the sake of practicality given the constraints they must work with (room, speaker positioning, WAF, whatever).

In principle, but again in practice small monitors and a small sub like my SVS Micro 3000 will fit in any room.

 

[Duke]

OK, great. Now tell me: how do you design a speaker with "better spatial quality"? Does this correlate with any measurements? Or do you just tinker and listen, tinker and listen, until you happen to hit upon a design with this feature?

A very narrow radiation pattern (like the Sanders Sound electrostat) results in extremely precise image localization, albeit in a small sweet spot. A lot of sidewall interaction from having a wide pattern can result in a widened soundstage that extends beyond the speakers. Interaural crosstalk cancellation like that incorported into the Polk L800 loudspeaker, or derived by the BACCH processor, can expand the soundstage and improve both sound image localization and the sense of envelopment. Multidirectional and omnidirectional systems can offer an increased sense of immersion and envelopment when set up correctly. And multichannel systems offer even more of an increase in the sense of immersion and envelopment.

This one I sort of understand: I would guess that lower distortion at higher volume is what "better dynamics" means in practice (but please correct me if you think it's something else).

Briefly (and brevity is not my forte!), imo "better dynamics" implies the ability to play louder and with less compression, and in particular less compression of the peaks. My understanding is that sufficiently loud music activates the limbic system and is perceived as "more exciting". Of course we don't want distortion to become increasingly audible and objectionable as the SPL goes up.

OK, again: how do you design a speaker with "a more relaxing and forgiving presentation"? Is there any measurement that this correlates with?

My understanding is that the ear's sensitivity to some types of distortion peaks in the 3-4 kHz region, which is also where the ear is most sensitive to sound pressure level, as indicated by equal-loudness curves. In particular, the ear's sensitivity to a type of distortion which often occurs in horns peaks in this region. Often there is a crossover in or near this region as well, and drivers are seldom at their best at the top and bottom of their passbands. Ime erring on the side of a bit of dippage in this region can make a speaker more forgiving. Ime there are two other frequency regions where a bit of dippage can arguably make a speaker more relaxing to listen to, but one of them varies from one speaker to the next, and the other is related to a recording artifact, so there is a lot of "it depends" involved in whether or not introducing a bit of gentle dippage in these other two regions is worth considering.

 

[garyrc]

the other is related to a recording artifact

About 1 in 5 of my CDs sound a bit anomalously "hard" in the upper midrange. I have never heard this on a SACD or a reel to reel tape. It's worth it to me in order to enjoy the clear, clean, dynamic sound on the other 80%.