• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Why don't all speaker manufacturers design for flat on-axis and smooth off-axis?

Kvalsvoll

Addicted to Fun and Learning
Audio Company
Joined
Apr 25, 2019
Messages
878
Likes
1,643
Location
Norway
You wrote:

«I am more fascinated by your obsession with the hypothetical question related to these two particular loudspeakers...»

The reason why I chose these particular speakers was to make a practical case to apply audio science on gear that people know reasonably well. The case method is well known in situations where skill acquisition is a goal.

I am surprised you don’t think it’s intriguing to discuss the speakers’ attributes in order to put together a weight of the evidence case for a majority to prefer the one over the other. What’s the point of audio science if not to apply it to practical cases?

I also think the Salón vs M2 case is interesting because their frequency responses are similar, but people who took part in the audio society blind test describe their sound as quite different. Personally, I have only heard the M2

It would be great if ASR promoted a language - based on science - to discuss why people tend to prefer one speaker over the other even if the speakers appear similar frequency response wise.

As long as volume is kept reasonable and listening distance is not too long, most - in this case very likely the only - differences in sound are caused by linear distortions - which means frequency response.

If they sound different, those differences are also present in the measurements. But the problem is how to analyze and interpret the measurements to give useful and repetitive and accurate information about the sound from a loudspeaker. Clearly, looking at on-axis alone is not enough. You need the complete radiation pattern. You also need to study what happens in time - a speaker is not necessarily minimum phase. And diffraction and secondary sound sources can play a significant role in a loudspeakers sound. All these are linear phenomenon, and can be seen as part of frequency response.

For very strange or defective creations, it is easy to spot the defect - the frequency response (on-axis) is not flat. But for good speakers, the differences are not so obvious.

A simple in-room measurement can tell a lot of differences between the speakers, and especially if you get to listen, look at the measurements, and then listen again, you can learn and find things. The first you will notice is that it is not obvious what to look for, and the measurements often look quite similar, while sound is perceived very different.

Frequency response on-axis and the suggested additional graphs are useful for the casual buyer to see if a speaker performs reasonably accurate, kind of a does-it-work verification, but it will not tell the full story about how it sounds. It is also dubious whether the casual byer can get much good information from reading those charts, even this simplifed presentation requires expert knowledge to interpret.
 

Juhazi

Major Contributor
Joined
Sep 15, 2018
Messages
1,717
Likes
2,897
Location
Finland
Toyota Camry and Ford F-150 have been the most popular cars in USA. How then they are so different and why has neither been voted as the best car?

People have different desires, different rooms and even different spouses. This is why we need different speakers with good sound. Aiming to just one type and sound is an utopia destined to fail.

Larson published a very good test and review of 1500€ floorstanders at Audioholics. All those aim for NRC-Toole recommendations and give huge performance for the money. Alternative good designs tend to be much more expensive like Kii, ML and D&D.
 
Last edited:

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,381
Location
Seattle Area
For loudspeakers and acoustics, most of what we want to study can not be solved by distributing sound samples, the test subjects need to be on location and listen there.
There is actually a solution to this that Harman has used. You record the speaker in the room with binaural recording. You then play that back using headphones and compare it to listening tests of the actual room. If the two agree to high degree (which they have in Harman testing), then you can distribute the file to others who listen using the same headphone.

Harman used this to show that people in Asia do NOT have different taste in speakers than western listeners. Instead of replicating the entire setup, they just created the files and had the people overseas listen with headphones.

Other researchers use the same method in psychoacoustics research.
 

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,381
Location
Seattle Area
Please note that one of the authors of that article, Søren Bech, was one of the co-authors of the 2017 JAES article I brought up earlier, where the authors wrote:
Bech is indeed one of the top researchers in this field. Your casual read of his work though is leading you astray to think that there is some conflict between his views and that of the NRC/Harman research. That is not so. I have quoted Bech countless times in support of this work. Here is a great example of it from peer reviewed, Journal of AES paper by Sean Olive:

1564255710183.png


There are numerous references to Beck's research in the paper. But let me quote you the ending:

1564255776102.png


This is not a field you get to learn to criticize after doing a few google searches. You need to invest the time and effort to read it all and then see if you can find a conflict. I have read a ton of it and as we say in US, "this dog don't hunt."
 

svart-hvitt

Major Contributor
Joined
Aug 31, 2017
Messages
2,375
Likes
1,253
Bech is indeed one of the top researchers in this field. Your casual read of his work though is leading you astray to think that there is some conflict between his views and that of the NRC/Harman research. That is not so. I have quoted Bech countless times in support of this work. Here is a great example of it from peer reviewed, Journal of AES paper by Sean Olive:

View attachment 30179

There are numerous references to Beck's research in the paper. But let me quote you the ending:

View attachment 30180

This is not a field you get to learn to criticize after doing a few google searches. You need to invest the time and effort to read it all and then see if you can find a conflict. I have read a ton of it and as we say in US, "this dog don't hunt."

I think I will come back to my overriding point in another thread.

My easy to see point is that frequency response is an incomplete answer to the speaker question.

Your continuing defense of everything Harman is as clear as your unwillingness to measure your own Mark Levinson amplifiers (while at the same time asking for more gear to measure).
 

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,381
Location
Seattle Area
My easy to see point is that frequency response is an incomplete answer to the speaker question.
Frequency response is NOT meant to be the complete answer to speaker performance. It is however the most important with respect to its fidelity.
 

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,381
Location
Seattle Area
Your continuing defense of everything Harman is as clear as your unwillingness to measure your own Mark Levinson amplifiers (while at the same time asking for more gear to measure).
I haven't asked for more gear to measure. I have filled my lab and have to store them on the steps leading to it! I have probably 2 to 3 months worth of gear to test and more is coming.

As to your jab, I am exceptionally careful to not promote any products from Harman. I have not told anyone to buy Mark Levinson amps, or even Revel speakers unless forced.

The ML amps I have produce far more power than my dummy loads can handle. They are also so heavy I can't even slide them on wood floor let alone be willing to drag them to my test station. They have been measured by stereophile and did not get positive commentary there. And at any rate, have nothing whatsoever with the topic at hand.

I suggest you move on as you put less thought in your insults than your arguments about audio.
 

March Audio

Master Contributor
Audio Company
Joined
Mar 1, 2016
Messages
6,378
Likes
9,317
Location
Albany Western Australia
I think I will come back to my overriding point in another thread.

My easy to see point is that frequency response is an incomplete answer to the speaker question.

Your continuing defense of everything Harman is as clear as your unwillingness to measure your own Mark Levinson amplifiers (while at the same time asking for more gear to measure).

No one has ever said it is.

You are coming across with an almost wilful misinterpretation of the research as presented by Harman etc.

I think this is becoming somewhat ridiculous. You keep seeing things that aren't there.
 
Last edited:

Cosmik

Major Contributor
Joined
Apr 24, 2016
Messages
3,075
Likes
2,180
Location
UK
My easy to see point is that frequency response is an incomplete answer to the speaker question.
Every speaker has a frequency response or, more accurately, an infinite number of frequency responses depending where and how you measure them. Surely they must be a complete description of the speaker, all else being equal.
 

Sergei

Senior Member
Forum Donor
Joined
Nov 20, 2018
Messages
361
Likes
272
Location
Palo Alto, CA, USA
The study I quoted goes to the precise need we as audiophiles have: which speaker is liked better, by a statistically significant number of listeners and what metrics predict that.

After reading all the posts on this thread, I chose the phrase above as an anchor for my answer to the OP question, because it goes to the heart of the matter.

The study Amir quoted, and several other studies mentioned in this thread, do address the needs of audiophiles regarded as averaged population. They don't, however, address all needs of every audiophile. Moreover, most of such studies are funded - directly or indirectly - by large manufacturers, and thus address the needs of the manufacturers first and foremost.

Let's exclude from consideration the speakers designed by those who are not familiar with the "standard model": while such designers appear to exist, I believe that they don't tend to work for manufacturers who are successful long-term.

I believe that some smaller speaker manufacturers decided - consciously or not - to address needs of niche audiophile and professional market segments. Large manufacturers have huge advantages of scale - in marketing, R&D, components sourcing, production, distribution, and so on. However, addressing niche segments is often not financially justifiable for them.

What do such smaller speaker manufacturers have to play with? The three characteristics that make a speaker significantly statistically preferable to average population: flat frequency response on axis, smooth frequency response off-axis, low distortions.

For example, sacrificing flat frequency response on axis leads to speakers preferred by older listeners (who need high frequencies boost), budget-conscious dance music lovers (who desire low frequencies boost), and listeners preferring significantly sub-reference SPL levels (who adore smiley frequency response curves, reflecting the loudness compensation).

Sacrificing wide-angle smooth frequency response allows to inexpensively produce speakers with non-traditional radiation characteristics. This could be beneficial, for instance, to listeners who prefer or have to live or work in highly-sound-reflective environments. For example, to Dr. Geddes, whose house appears to be rich in hard wooden surfaces, according to his speaker demo videos.

De-prioritizing low distortion results in designs that may sound significantly different from the "norm", yet exhibit synergies with other highly-distorting gear or older music records, which some listeners prefer for various reasons. An example would be a speaker with high level of even harmonics, compensating for high level of odd harmonics inherent in reel-to-reel tape machines.

Going in the opposite direction results in speakers with unusually low levels of specific distortions, to which some people are very sensitive. Such low distortion levels could be achieved, for instance, via electrostatic, planar, ribbon, bending wave, and other exotic transducers, which can't be economically produced, distributed, and serviced at scale by larger manufacturers.

I've seen more than one dumbfounded founder of a "normal speaker" or a "normal headphone" company at audiophile shows, seemingly exasperated by absence of foot traffic, all the while sellers of speakers and headphones that couldn't possibly be neutral and balanced, just from the first principles of acoustics, enjoyed heightened attention.
 

Thomas_A

Major Contributor
Forum Donor
Joined
Jun 20, 2019
Messages
3,421
Likes
2,406
Location
Sweden
Just one interesting question I want to ask; it is not hard to EQ a piece of a song and test two against each other and test for preference. Has anyone done the tests on this site or elsewhere? The question can eg preference be the small deviations that you get with two stereo speakers as published by Shirley Toole and others (deviations +/-1 dB in a standard room, more so in anechoic rooms). This will only deal with the center phantom image and timbral changes in the 1-8 kHz region caused by the 30° speaker positioning i stereo. It should not be hard to make the "online" test for preference.
 

MattHooper

Master Contributor
Forum Donor
Joined
Jan 27, 2019
Messages
7,194
Likes
11,806
I've seen more than one dumbfounded founder of a "normal speaker" or a "normal headphone" company at audiophile shows, seemingly exasperated by absence of foot traffic, all the while sellers of speakers and headphones that couldn't possibly be neutral and balanced, just from the first principles of acoustics, enjoyed heightened attention.

This reminds me of the somewhat puzzling relationship between the studies done by HK/NRC Sean Olive etc that, under test controlled conditions, indicate people (even teenagers) will prefer neutrally balanced speakers and headphones.

And yet....

Every time I use my car after my Rap-loving son has used it, I have to re-adjust the EQ. Left in the flat EQ positions, the system is relatively neutral sounding - music sounds balanced similarly to what I hear in my system, well-recorded voices sound even, not exaggerated, etc.

But my son can't stand the neutral settings for his music. He pushes the bass control all the way up for booming bass, the midrange ALL THE WAY DOWN, and the treble almost all the way up, for the most hideous sounding "smile" EQ I've ever encountered. My podcasts sound ridiculous on his settings.

And I'm quite sure my son is not unusual in his preferred tone settings, for people his age/into similar music.

So I have to ask, in such cases, what has the science predicted? Certainly the predictions are relevant to expectations under well controlled conditions. But has it predicted preferences out there in the real world, everything else considered? If the predictions are that my son would likely prefer more natural, neutrally balanced sound settings, those don't pan out when he has a choice between neutral and wildly unbalanced. No doubt dictated quite a bit by his preferred music.
 

LTig

Master Contributor
Forum Donor
Joined
Feb 27, 2019
Messages
5,760
Likes
9,442
Location
Europe
I've likely mentioned before about a lady taking up money for a charity at my side door. I had music playing inside. I excused myself to get a donation for her, when I returned she had stepped inside. Was looking at the speakers in something of a trance by the music. Was surprised and embarrassed when I returned. She sheepishly said, "I'm sorry, ......I shouldn't have....it sounded like someone was really playing music".

.
I had such an experience last year in our vacation. I read a book in the lobby of a small mountain hotel when someone started to play a saxophone in the billiard room close by. Went over and to my big surprise it was a tiny JBL Charge 3 blutooth speaker:eek:.

Now I own one myself.
 

MattHooper

Master Contributor
Forum Donor
Joined
Jan 27, 2019
Messages
7,194
Likes
11,806
I had such an experience last year in our vacation. I read a book in the lobby of a small mountain hotel when someone started to play a saxophone in the billiard room close by. Went over and to my big surprise it was a tiny JBL Charge 3 blutooth speaker:eek:.

Now I own one myself.


I have recordings I made of my son practicing sax, and me playing my acoustic guitar.

Playing those recordings through my MBL speakers has fooled a few people in to thinking a real instrument was being played in the house (when listened to down the hall in another room. (It was a combination of both the clarity and nature of the sound, and also the squonky "sounds like a kid practicing saxophone" nature of the recording, no doubt).
 

svart-hvitt

Major Contributor
Joined
Aug 31, 2017
Messages
2,375
Likes
1,253
I haven't asked for more gear to measure. I have filled my lab and have to store them on the steps leading to it! I have probably 2 to 3 months worth of gear to test and more is coming.

As to your jab, I am exceptionally careful to not promote any products from Harman. I have not told anyone to buy Mark Levinson amps, or even Revel speakers unless forced.

The ML amps I have produce far more power than my dummy loads can handle. They are also so heavy I can't even slide them on wood floor let alone be willing to drag them to my test station. They have been measured by stereophile and did not get positive commentary there. And at any rate, have nothing whatsoever with the topic at hand.

I suggest you move on as you put less thought in your insults than your arguments about audio.

@amirm, what you write on testing Mark Levinson amplifiers doesn’t make sense.

You ask for donations to an elaborate speaker test setup, which means speakers need to be sent from wide and afar to your place. Yet, you cannot carry the Mark Levinson (70 kg) amplifier to your measurement desk, or the Audio Precision to your amplifiers?

You write that Stereophile have measured the ML amplifiers, as if there’s no need to do an independent review. Stereophile have tested the Benchmark amplifier too (in 2015), years before you tested the very same amplifier.

Your note on not promoting products from Harman sounds nice. Yet, every time someone questions the Harman research method, you come to Harman’s defense.

I think measurments of the ML amplifiers would make sense to illustrate once and again how far electronics design has come. Modern amplifiers weigh a fraction of legacy gear, yet have similar or even better performance. Measurements of the heavy ML amplifiers could possibly be a very strong signal for high-enders to embrace new ideas and reject the old myth that good electronics must be heavy.

It would be nice to see ML’s amp pitted against say @March Audio ’s newest and “biggest”.
 

Sergei

Senior Member
Forum Donor
Joined
Nov 20, 2018
Messages
361
Likes
272
Location
Palo Alto, CA, USA
But my son can't stand the neutral settings for his music. He pushes the bass control all the way up for booming bass, the midrange ALL THE WAY DOWN, and the treble almost all the way up, for the most hideous sounding "smile" EQ I've ever encountered.

So I have to ask, in such cases, what has the science predicted? Certainly the predictions are relevant to expectations under well controlled conditions. But has it predicted preferences out there in the real world, everything else considered? If the predictions are that my son would likely prefer more natural, neutrally balanced sound settings, those don't pan out when he has a choice between neutral and wildly unbalanced. No doubt dictated quite a bit by his preferred music.

Maybe I could answer, given more information. What is the model of the car, including year? What is the full range of EQ in decibels? Who is his favorite singer?
 

MattHooper

Master Contributor
Forum Donor
Joined
Jan 27, 2019
Messages
7,194
Likes
11,806
Maybe I could answer, given more information. What is the model of the car, including year? What is the full range of EQ in decibels? Who is his favorite singer?

Sorry, not going to be much help. For one thing I'm not a car guy. I think of our cars as "the blue car, the gray car" etc. Second, I don't have that technical information on the sound system - I'm not one of you engineer-types who would keep track of that stuff. Third: "singer?" My boy likes Rap music ;-)

Not going to be helpful, but FWIW we have a Volkswagen Golf (I dunno...6 years old?), Volvo (I dunno...older...but with very nice Dynaudio soundsystem 'the upgraded package.') Also my son's Honda Civic 2018. In all cars my son will EQ to the "smiley" settings.


(Car information truly does not stick in my headl, it's not something that interests me at all. Whenever someone asks me which model car I drive, I'm honestly at a loss to remember. Can only really give the brand).

What were you getting at, with those questions?
 

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,381
Location
Seattle Area
The study Amir quoted, and several other studies mentioned in this thread, do address the needs of audiophiles regarded as averaged population. They don't, however, address all needs of every audiophile. Moreover, most of such studies are funded - directly or indirectly - by large manufacturers, and thus address the needs of the manufacturers first and foremost.
Once again no. See this recent post on why this research was funded by non-profit government funded Canadian research: https://www.audiosciencereview.com/...-on-axis-and-smooth-off-axis.8090/post-203005

That is on top of the research being peer reviewed, winning numerous awards for its authors and the papers themselves.

Harman's belief is that by building what customers prefer in controlled testing, then their business will expand. That business by the way, has been the multi-billion dollar automotive business, no hi-fi audio although that benefits too.

As to your general comment, the assumption is that most people like ice cream. Can you find people who do not, sure. But if you are going to make something tasty for people, ice cream would be a much better bet than Durian fruit. :)

Warning: R-rated language:

P.S. I actually like the taste of Durian fruit. It is the smell that is the problem...
 
Top Bottom