Do We Want All Speakers To Sound The Same ?

[Holmz]

I am thinking more along the lines of analysing the frequencies up to circa 2KHz (although in a perfect world we want 20kHz)- similar to what Dirac is proposing with SRC - but doing it in real time. (I am setting the 2khz limitation based on what Dirac has published with regards to their SRC system, showing that the benefits - as the system currently stand, mostly are achieved below 2kHz)

Real time analysis of not just the speakers direct signal, but the reflections as well, and then real time correction of those sound flaws both direct and reflected through active cancellation - so the end result is closer to the theoretically targeted sound field.

The only thing that can be corrected with certainty is the direct path response.
We cannot correct the box or baffle or resonances.

It is only the motor/cone/driver non-linearity that we can make linear.

That would ultimately be the path to Nirvana - the ability to simulate the soundfield we want within any listening space (given sufficient speakers, and processing power).

That is a pipe dream, that IMO is a step too far.

A more linearity system can be achieved, but what you are suggesting would take more speakers than 2.

 

[dlaloum]

The only thing that can be corrected with certainty is the direct path response.
We cannot correct the box or baffle or resonances.

It is only the motor/cone/driver non-linearity that we can make linear.

Yes - at speaker level, building a better speaker requires that we build/design out the box/baffle resonances... (ball speakers, like the Gallo designs, do a good job!... panels do too.)

That is a pipe dream, that IMO is a step too far.

A more linearity system can be achieved, but what you are suggesting would take more speakers than 2.

Yes - it will take more speakers - which is why it will initially appear in home theatre / surround style setups - but there is no reason for it to be limited to that type of software.

 

[Holmz]

…
Yes - it will take more speakers - which is why it will initially appear in home theatre / surround style setups -

The S/W to make a more linear response, for a direct path sound, is not the same S/W as a some larger realtime room bounce thing.
I know that the former is achievabl, but the later sounds like something I would have on a list for Santa.

but there is no reason for it to be limited to that type of software.

The main “reason to be limited”, is that S/W generally builds upon the easy, and evolves towards the more difficult.

 

[prerich]

However, you're not in their room, you're probably not listening on their specific speakers (there will be variations even between the same model), your ears are different from theirs, stereo recording is very limited it what it captures of an unamplified event, and so on.

It isn't truly possible to hear what the artist created, as it was created - it is a fantasy scenario. Too much distortion, of one sort or another, has happened at many steps along the way in the recording process and happen upon reproduction too.

I'd like to hear what the artist created too, but all I got was a crummy stereo recording!

So in reality...I'd like that crummy stereo recording to serve me. I'd like it to be something that my flawed receptors (my ears) will find pleasant to my particular taste. The final question is this - do you enjoy it? As I read this thread - your response is right on. Also, having known people in the recording industry and actual singers....many of the performers stay away from the mastering process...they actually hate it...they "leave that up to the engineers". I'd say 3 of 5 performers feel this way (some have performed so many takes that they no longer want to hear the song again). Also many events although live, are amplified. Enjoy the music ..... your way.

 

[Digby]

So in reality...I'd like that crummy stereo recording to serve me. I'd like it to be something that my flawed receptors (my ears) will find pleasant to my particular taste. The final question is this - do you enjoy it? As I read this thread - your response is right on. Also, having known people in the recording industry and actual singers....many of the performers stay away from the mastering process...they actually hate it...they "leave that up to the engineers". I'd say 3 of 5 performers feel this way (some have performed so many takes that they no longer want to hear the song again). Also many events although live, are amplified. Enjoy the music ..... your way.

Precisely.

 

[prerich]

The fi in hi-fi: if the fidelity is high, the transparency is high, the realism is high, the similarity is high, the sameness is high.

The tone of a hi-fi component, superimposed over the tone of a musical instrument, is no longer the tone of a musical instrument. I want to experience the tone of musical instruments in my home.

Sadly - this is an impossible task, unless you have the band in the space where they played...if they played unamplified - which limits the actual instruments allowed (Fender Rhodes, Hammond B3, Electric Bass and guitar along with other amplified instruments - no go). The tone of musical instruments also changes with the venue they're played in. The tone changes with the age of your ears. The variables are almost infinite. The engineered mix is generally not the artist interpretation, but the engineer's interpretation of the performance ...of how he/she thinks it should sound, and what will make it sell. Enjoy the music....I've decided I'm growing too old to seek perfection in an imperfect body...let's have some fun.

 

[fpitas]

I wish this thread had a poll so I could vote for Soul-Crushing Conformity.

 

[ahofer]

There seems to be a deep emotional attachment to the idea of a bespoke speaker match to listeners. Which is fine, of course. But my working hypothesis is that this could be replicated with EQ and *perhaps* some additional distortion, all of which can be added cheaply in the digital domain. Which takes me back to the idea that what you want is the best and most flexible starting point - flat response, even dispersion, high transient capability without altered FR, and low distortion. Weirdly enough, that combination is what sounded best to the majority in Toole's research, without EQ or effects.

 

[symphara]

There seems to be a deep emotional attachment of a bespoke speaker match to listeners. Which is fine, of course. But my working hypothesis is that this could be replicated with EQ and *perhaps* some additional distortion, all of which can be added cheaply in the digital domain. Which takes me back to the idea that what you want is the best and most flexible starting point - flat response, even dispersion, high transient capability without altered FR, and low distortion. Weirdly enough, that combination is what sounded best to the majority in Toole's research, without EQ or effects.

Not being fastidious, what's "high transient capability without altered FR" and where does it show in measurements (e.g. those done by @amirm)?

 

[MarkS]

What's amazing to me is that speakers that measure badly, even extremely badly, can still "sound good" to lots of people.

 

[Purité Audio]

Almost certainly because they have never heard a really good loudspeaker or they have and found it a bit dull, compared to the system at home.
Keith

 

[Mr. Widget]

What's amazing to me is that speakers that measure badly, even extremely badly, can still "sound good" to lots of people.

Do you mean that they produce "good sound" or do you mean they sound good to lots of people?

I believe the work of Sean Olive and his team at Harman showed that a significant majority of untrained listeners preferred what we would consider speakers that measure well, but that it wasn't unanimous. There were still lots of people who didn't prefer the speakers that measure well. I don't know if follow up work has been done to determine what the outliers preferred, was it random/apathy, or were they drawn to other specific types of sound.

I am thinking Altec A7 vs. a modern state of the art speaker. There are fairly large numbers of people who actually prefer the A7 type of speaker even though they are easily shown to be technically inferior. My theory on this is the dynamic nature of the speakers. There is a liveliness in those high efficiency designs that modern speaker lack.

 

[prerich]

What's amazing to me is that speakers that measure badly, even extremely badly, can still "sound good" to lots of people.

Doesn't surprise me, as by the time we are old enough to buy gear - we have already developed a hearing paradigm from our youth. This paradigm is subject to change, but we have our own ideals as to what constitutes good sound .... to us. Yes measurements matter - they tell us if something is fundamentally broken - but have you ever seen that older person that can't let go of their stove top coffee pot? There are ways that are scientifically better to brew coffee, but they want that old pot. Navy chiefs many times have mold growing in their coffee cups...don't you dare wash it. Some pick the mold out being careful not to disturb the "mocha" built up on the cup ....for years!!! It's what they love...same will happen for speakers, music types, etc. I applaud Amir for bringing objective test to gear, we can tell that they work well - but a person's ears...may not work well - so what sounds good to them may be distorted but it makes them happy.

 

[prerich]

There seems to be a deep emotional attachment of a bespoke speaker match to listeners. Which is fine, of course. But my working hypothesis is that this could be replicated with EQ and *perhaps* some additional distortion, all of which can be added cheaply in the digital domain. Which takes me back to the idea that what you want is the best and most flexible starting point - flat response, even dispersion, high transient capability without altered FR, and low distortion. Weirdly enough, that combination is what sounded best to the majority in Toole's research, without EQ or effects.

However - I would like Harman to do this and to state demographics, men or women, culture, types of music tested etc. I remember doing the lossless vs lossy test on Apple a few years ago. On songs that I enjoyed - I was 100% ...on songs that I didn't I was barely 50%. Apple took that to say that Lossless vs. lossy didn't matter to me. However my listening paradigm kicked in - if I didn't like the song - I found myself saying when will it end? So I hurried through those selections. There are many psychological things going on when one listens to music ...too many to count. Too many missing variables, and I'm wondering would that change any of results....and I'm a huge fan of Toole. I'm getting old now and thinking outside of the box a bit...when Father Time shows you your mortality - you tend to try and enjoy things a bit more.

 

[MattHooper]

There seems to be a deep emotional attachment of a bespoke speaker match to listeners. Which is fine, of course. But my working hypothesis is that this could be replicated with EQ and *perhaps* some additional distortion, all of which can be added cheaply in the digital domain. Which takes me back to the idea that what you want is the best and most flexible starting point - flat response, even dispersion, high transient capability without altered FR, and low distortion. Weirdly enough, that combination is what sounded best to the majority in Toole's research, without EQ or effects.

I fully agree that you have expressed a very reasonable approach! And I think it would work for many people.

I do have doubts it would cover for everyone's tastes and goals though (and hence wouldn't work as a universal recommendation).

Of course there are non-audio considerations like a pleasing speaker design. Right now the options for speakers that we know meet that technical criteria is a truly teeny slice of the available speakers, and at least in my case (and I know others) I'm not a big fan of how most of them look. (I sure as heck dislike the studio monitor look for home audio, and the small number of consumer speaker brands we know to meet the criteria above don't really appeal to me). I've chosen speakers whose sound I really like, but also meet my taste aesthetically.

Also, as I've expressed, I like how speakers sound different and I've owned a wide range of speakers all of which sounded different, all of which I truly enjoyed and I'm very glad to have owned them. Those include narrow dispersion dipoles like Quads to Omnis like MBL, to first order crossover speakers (Thiel, Meadowlark), high order crossover speakers, big speakers, small speakers, some with cabinet resonances others without...and on and on. I did indeed have a
parametric digital EQ in my system for many years and occasionally gave a go at manipulating the sound of one speaker to sound like a previous, but never really achieved it. Perhaps with more extensive attempts (and more sophisticated software) I could have gotten closer, but I have doubts I could have replicated all the speakers I've enjoyed simply by manipulating the sound of a single neutral speaker.

 

[Digby]

Doesn't surprise me, as by the time we are old enough to buy gear - we have already developed a hearing paradigm from our youth. This paradigm is subject to change, but we have our own ideals as to what constitutes good sound .... to us.

I wonder though....do we really have a separate part of brain allocated to 'stereo reproduction memory' or do we just reference our knowledge of everyday sounds/instruments when listening to loudspeakers?

Probably most people listen to amplified (rock/pop/electronic) music, more so than unamplified (folk/classical), so I suppose there is no 'true north' to reference in such cases.

 

[prerich]

I wonder though....do we really have a separate part of brain allocated to 'stereo reproduction memory' or do we just reference our knowledge of everyday sounds/instruments when listening to loudspeakers?

Probably most people listen to amplified (rock/pop/electronic) music, more so than unamplified (folk/classical), so I suppose there is no 'true north' to reference in such cases.

That detective.....is the right question.

 

[Sgt. Ear Ache]

Do I want all speakers to sound the same? Yes. I do - In the sense that when I feed known signal X into my system I want (as much as is possible) known signal X to be what reaches my ears. If, for instance, I feed pink noise into my system...why would I want what reaches my ears to be white noise, or brown noise? So I want a system that is neutral (not altering the signal in any unwanted way) and speakers that are not going to be too difficult to EQ for the room in order to (again, as much as possible) maintain that neutrality.

 

[Digby]

What's amazing to me is that speakers that measure badly, even extremely badly, can still "sound good" to lots of people.

That's because your ear is not a measuring device and can hide/mask/adapt to a great variety of sins.

 

[symphara]

Almost certainly because they have never heard a really good loudspeaker or they have and found it a bit dull, compared to the system at home.
Keith

Dull is best. Particularly when fed with a signal from a Roon Nucleus server. Nothing like spending a fortune on a pretty PC case to improve your sound.