Do We Want All Speakers To Sound The Same ?

[MattHooper]

I'm sorry I haven't read through the thread, but here's my quick thought:

We sort of do, in the sense that they should all be ready for EQ. The industry is coming around to this with auto-DSP in consumer devices. If the speaker itself exhibits flat frequency response, uniform disperson, and very low distortion at various levels, that's a good place to start. A choice of wide or narrow dispersion might be useful (I guess there's one design with variable dispersion!), and a choice of designs with different dispersion/distortion/SPL trade-offs might also be useful.

But it seems to me, with the increasing ease of DSP and pre-transducer effects, we are increasingly aiming at the same baseline capabilities.

IOW, I guess I'm saying they should all sound the same in an anechoic chamber. And this approaches the Spinorama ideal that dominates these pages.

Excellent. Thanks ahofer!

 

[MattHooper]

The research points towards good measurements correlate with preference.

Under blind conditions.

It remains a relevant question to ask how those results port over to the scenario - sighted listening - in which speakers will actually be used.

 

[Purité Audio]

Literally the entire basis of Toole’s research.
Keith

 

[trackrat888]

So, you start with speakers that sound the same from different manufacturers. How would one know what they would sound like if you add tone control or EQ? Would they respond the same way?

 

[MattHooper]

Literally the entire basis of Toole’s research.
Keith

Are you sure?

I certainly may have missed it - it's been a while since I looked in to Toole's research in more detail. But I don't remember the research being focused on being able to predict what speakers listeners will prefer under sighted conditions, but rather under blind-control conditions. I'd appreciate being pointed to the research for the former!

Further, a big reason this question is relevant: To the degree it is presumed consumers will perceive the same sonic characteristics test subjects heard under blind conditions...that would have the implication that sighted listening is not wholly unreliable.

 

[fineMen]

How do you make speakers with different directivity characteristics sound the same?

Is there a correct/better topology (or is it / should it be a matter of preference)?

I would rather ask if the relative strength of direct versus indirect sound is a parameter to be considered. It depends on directivity, room damping (curtains, obstacles, furniture) and the listerners' (plural?) positions--not only plain distance.

Alas, I cannot report on my own findings, as I, in this thread again, was already accused to lie as the stupid I am. You have to find out without my guidance. You will survive, granted.

 

[fineMen]

Under blind conditions.

It remains a relevant question to ask how those results port over to the scenario - sighted listening - in which speakers will actually be used.

Good one!

 

[ahofer]

So, you start with speakers that sound the same from different manufacturers. How would one know what they would sound like if you add tone control or EQ? Would they respond the same way?

Good question. That’s why I like the measurements of FR and distortion at higher SPLs. Helps you understand if they will respond without coloration. Or with it.

 

[Shadrach]

I'd absolutely want to at least give such a speaker a listen!

There's hope for you yet.

However, it's still possible that given all the limitations and colorations inherent in recordings themselves, I might prefer some sort of slight departure from "perfect" in a speaker (or in my amplification, as I like toobs).

Trying to include the differences in recording quality/style is rather different from a replay system.
I've heard some very good tube amplifiers that sounded just as good as good solid state amplifiers. Probably wouldn't measure as well.

What I find interesting here is your depth of subjectivism for want of a better description and why it should make you prefer a product with inferior measurements (likely case) and being in favour of listening to a system for how long? and then making your mind up on what you heard that day.

My equipment sounds different every day. I don't really know how it sounds. I'm not expecting to experience a replica of a live performance in my home.
Just a couple of saxophones playing in a confined space can be pretty teeth grinding. I think it gets overlooked that be it analogue or digital once it's been recorded it's data. There isn't any music there. Why would anyone want corrupted data? With the aid of a bit of manipulation one can make changes to that data if you don't like what the data is telling you but surely one would want the original accurate copy at least.

 

[Thomas_A]

Under blind conditions.

It remains a relevant question to ask how those results port over to the scenario - sighted listening - in which speakers will actually be used.

Well the sighted comparison in Tooles research is there as well. Affected by looks and prior knowledge of price etc

There are other areas less well researched, e.g fine tunings and loudspeaker preference in mono vs. stereo, near vs. far-field.

 

[MattHooper]

There's hope for you yet.

Ha! Fair enough.

I happily seek to hear speakers of all different types, those that measure really well (by the lights of ASR etc) and even those that don't. I really enjoy the way speakers sound different from one another, and enjoy experiencing what those differences do for the character of the listening experience. So I love the speaker auditioning experience (and hearing different set ups at audiophile's houses, at audio shows, etc).

I understand others have a different approach and are more: "just give me the speaker I want, and give me the measurements that will help me ignore lots of speakers and zero in on the one I want." My own views don't contradict such an approach at all; it's just that I don't share that exact goal in terms of what I get out of this hobby.

What I find interesting here is your depth of subjectivism for want of a better description and why it should make you prefer a product with inferior measurements (likely case) and being in favour of listening to a system for how long? and then making your mind up on what you heard that day.

Well in my own case, I audition gear (speakers) when I can at a dealer, and otherwise resort to buying gear second hand to try out, and re-sell if I don't like it or when I'm done. I do not experience this as a hopeless "shooting arrows in the dark" process, but one I truly enjoy. I love being able to check out different gear.
And it's quite rare that even when I buy second hand that I'm actually disappointed. I may keep an item for a couple years, a decade, longer...but eventually selling it doesn't mean it didn't provide me with plenty of enjoyment.

When it comes to auditioning speakers at a dealer, it often happens when I'm auditioning different speakers (including at other dealers) that I'm interested in.
I have my test tracks and I listen to tons of different music I know on the speaker, as well as experimenting with speaker positioning, listener position etc. I have found that I come away with a generally reliable impression of the speaker this way: I am virtually never surprised when I end up with the speaker in my home, vs what I heard during the audition phase.

My equipment sounds different every day. I don't really know how it sounds. I'm not expecting to experience a replica of a live performance in my home.
Just a couple of saxophones playing in a confined space can be pretty teeth grinding.

Agreed! I don't actually want the sound of someone blaring a sax or trombone or trumpet full volume in my listening room. I sure as hell wouldn't want a real drummer in front of me banging away either. My ears would give out very fast. So what I like certain aspects of "the real thing" that I find engaging. My father was a jazz musician/music teacher and I grew up hearing him played sax constantly in the house (and trumpets, piano, clarinet, trombone...and more), and I played sax as well. There are certain timbral aspects that say to me "THAT is the beautiful tone I recognize of a sax. " The harmonic richness of the real thing is intoxicating. If I can get *some* of that recognizable in sound reproduction, even if not all and not reaching "perfect realism," I find it enhances my enjoyment. The same goes for other aspects. I don't need perfect replication of real world dynamics in a guitar or drum reproduction...but if one speaker seems to produce the recording with a bit more dynamic life, translating the life-energy of the performer better, that too is welcome. And on and on.

The thing is, even though I have done direct live vs reproduced comparisons (in my own home using recordings of voices, instruments we own etc), in the end I'm not going to be sitting there constantly doing live vs reproduced comparisons. What I have to work with is my own memory and impressions of what I like about the live version of a sax, voice, acoustic guitar etc. So I only have to please myself, my own impressions, in that regard. And very often when I hear a sax or acoustic guitar through speakers it just doesn't sound right - the timbral color seems "off" and not pleasing. Where sometimes I can hear it and it just strikes me as "YES! THAT is capturing something really essential, tonally, that my brain interprets as "right" and makes me want to keep listening."

I think it gets overlooked that be it analogue or digital once it's been recorded it's data. There isn't any music there. Why would anyone want corrupted data? With the aid of a bit of manipulation one can make changes to that data if you don't like what the data is telling you but surely one would want the original accurate copy at least.

My take is that, for me (depending on the gear we are talking about), the right amount of slight flavoring will tend to make almost everything sound better and more pleasing, while not being anywhere near enough to cover up the unique details of any recording. It's all an illusion; I have no qualms about manipulating it to a degree it's more pleasurable.

 

[MattHooper]

Well the sighted comparison in Tooles research is there as well. Affected by looks and prior knowledge of price etc

Yes I know about that part. You can see in many instances how speakers were rated sighted vs blind.

So you certainly do learn something about sighted listening conditions: That seeing the speakers (or having knowledge which they are listening to) can influence
the perception of the sound.

But from what I remember, there was no systematic attempt to predict which speakers listeners will prefer under sighted conditions. That was reserved for trying to determine what people will prefer under blind listening conditions. That's the difference I'm getting at. There's a perfectly good scientific rational, if you are trying to understand something, for controlling for such variables. The problem comes when you try to port the information from the tests where you've removed the variables, to predicting results when those variables are there.

As I was saying, the relevance of blind listening to sighted listening is a consequential question: the aim of the research, at least or especially when it comes to that of Harman Kardon, is not simply to come to the conclusion that "people will select differently under blind vs sighted conditions" and "this is the type of sound people prefer under blinded conditions." It's supposed to have more relevance than that: especially for a speaker company, one would presume they want to be able to predict what people will prefer under the SIGHTED conditions in which people will actually be selecting and listening to the speakers! Otherwise...the blind tests, while interesting, aren't much good at enlightening anyone on what speakers to buy or what type of speakers to make.

But, as I say, IF it turns out that the results of blind speaker tests port to what people will also prefer under the normal sighted listening which we all use, THEN that has to imply some significant (not "perfect/infallible" but significant) reliability in sighted listening. Otherwise "what people preferred under blind conditions" is useless for the purposes of telling us what people will perceive as "good/better" in real world listening conditions.

 

[Thomas_A]

Yes I know about that part. You can see in many instances how speakers were rated sighted vs blind.

So you certainly do learn something about sighted listening conditions: That seeing the speakers (or having knowledge which they are listening to) can influence
the perception of the sound.

But from what I remember, there was no systematic attempt to predict which speakers listeners will prefer under sighted conditions. That was reserved for trying to determine what people will prefer under blind listening conditions. That's the difference I'm getting at. There's a perfectly good scientific rational, if you are trying to understand something, for controlling for such variables. The problem comes when you try to port the information from the tests where you've removed the variables, to predicting results when those variables are there.

As I was saying, the relevance of blind listening to sighted listening is a consequential question: the aim of the research, at least or especially when it comes to that of Harman Kardon, is not simply to come to the conclusion that "people will select differently under blind vs sighted conditions" and "this is the type of sound people prefer under blinded conditions." It's supposed to have more relevance than that: especially for a speaker company, one would presume they want to be able to predict what people will prefer under the SIGHTED conditions in which people will actually be selecting and listening to the speakers! Otherwise...the blind tests, while interesting, aren't much good at enlightening anyone on what speakers to buy or what type of speakers to make.

But, as I say, IF it turns out that the results of blind speaker tests port to what people will also prefer under the normal sighted listening which we all use, THEN that has to imply some significant (not "perfect/infallible" but significant) reliability in sighted listening. Otherwise "what people preferred under blind conditions" is useless for the purposes of telling us what people will perceive as "good/better" in real world listening conditions.

Well, to me such experiments have small value. I would rather see more blind experiments in stereo setups in near-field and in far-field. Some of them could be done using dummy head recordings for evaluations. Even if they are far from ideal some experiments carried out at ASR show interesting results IMO.

 

[HeadDoc12]

It's been a busy couple of days for me (in the US) so bear with me if this is repeating what others have said. To me, the ideal state of affairs is that all speakers be CAPABLE of being indistinguishable from a live performance. By which I mean, under ideal double-blind test conditions, a statistically significant majority of listeners would be unable to tell the real from the playback; all speaker flaws or differences would be inaudible. Obviously, once we get into the size and space limitations of different listening spaces, this would fall apart to a large extent, but even then, speakers could all aspire to that ideal once it is achieved. Then, music-lovers would know they are hearing what the recording engineer heard, and the artist intended (I also listen to a fair amount of electronic music that was never "played live," but I think you all know what I mean). Then, we could all still have fun with DSP: not just adapting the speakers to our rooms (which active speakers would be able to do automatically), but messing around with the sound to suit the recording, the setting, and our mood. Back in the 1990s, I owned the very first Dolby Digital capable receiver (I believe it was briefly called Dolby AC-3 before the marketing guys set things right), a giant black Yamaha beast whose model number I can't remember. It had DSP that supposedly mimicked famous performance spaces around the world, and a whole bunch of other possible locations. Was it a silly gimmick that put bells and whistles above sound quality? 100% yes. But I still have fond memories of listening to "You Can't Always Get What You Want" modified to sound like it was in a cathedral - the boys choir echoed around like crazy. So, once speaker "perfection" is achieved, I would imagine DSP boxes advancing into a whole new category of ways to manipulate the music. Or, more likely, software-based systems that would let you do all kinds of modifications to the music: pitch, speed, key, remixing, replace Neil Young's vocals with Lizzo's, stuff like that. So there could still be plenty of variety even listening to the same recording.

 

[MAB]

Alas, I cannot report on my own findings, as I, in this thread again, was already accused to lie as the stupid I am. You have to find out without my guidance. You will survive, granted.

Where? I didn’t see those accusations towards you here. I did see you oddly misstate the OP’s name. He very kindly responded to you, on topic. What else are you looking for?

 

[Shadrach]

Ha! Fair enough.

I happily seek to hear speakers of all different types, those that measure really well (by the lights of ASR etc) and even those that don't. I really enjoy the way speakers sound different from one another, and enjoy experiencing what those differences do for the character of the listening experience. So I love the speaker auditioning experience (and hearing different set ups at audiophile's houses, at audio shows, etc).

I understand others have a different approach and are more: "just give me the speaker I want, and give me the measurements that will help me ignore lots of speakers and zero in on the one I want." My own views don't contradict such an approach at all; it's just that I don't share that exact goal in terms of what I get out of this hobby.



Well in my own case, I audition gear (speakers) when I can at a dealer, and otherwise resort to buying gear second hand to try out, and re-sell if I don't like it or when I'm done. I do not experience this as a hopeless "shooting arrows in the dark" process, but one I truly enjoy. I love being able to check out different gear.
And it's quite rare that even when I buy second hand that I'm actually disappointed. I may keep an item for a couple years, a decade, longer...but eventually selling it doesn't mean it didn't provide me with plenty of enjoyment.

When it comes to auditioning speakers at a dealer, it often happens when I'm auditioning different speakers (including at other dealers) that I'm interested in.
I have my test tracks and I listen to tons of different music I know on the speaker, as well as experimenting with speaker positioning, listener position etc. I have found that I come away with a generally reliable impression of the speaker this way: I am virtually never surprised when I end up with the speaker in my home, vs what I heard during the audition phase.



Agreed! I don't actually want the sound of someone blaring a sax or trombone or trumpet full volume in my listening room. I sure as hell wouldn't want a real drummer in front of me banging away either. My ears would give out very fast. So what I like certain aspects of "the real thing" that I find engaging. My father was a jazz musician/music teacher and I grew up hearing him played sax constantly in the house (and trumpets, piano, clarinet, trombone...and more), and I played sax as well. There are certain timbral aspects that say to me "THAT is the beautiful tone I recognize of a sax. " The harmonic richness of the real thing is intoxicating. If I can get *some* of that recognizable in sound reproduction, even if not all and not reaching "perfect realism," I find it enhances my enjoyment. The same goes for other aspects. I don't need perfect replication of real world dynamics in a guitar or drum reproduction...but if one speaker seems to produce the recording with a bit more dynamic life, translating the life-energy of the performer better, that too is welcome. And on and on.

The thing is, even though I have done direct live vs reproduced comparisons (in my own home using recordings of voices, instruments we own etc), in the end I'm not going to be sitting there constantly doing live vs reproduced comparisons. What I have to work with is my own memory and impressions of what I like about the live version of a sax, voice, acoustic guitar etc. So I only have to please myself, my own impressions, in that regard. And very often when I hear a sax or acoustic guitar through speakers it just doesn't sound right - the timbral color seems "off" and not pleasing. Where sometimes I can hear it and it just strikes me as "YES! THAT is capturing something really essential, tonally, that my brain interprets as "right" and makes me want to keep listening."



My take is that, for me (depending on the gear we are talking about), the right amount of slight flavoring will tend to make almost everything sound better and more pleasing, while not being anywhere near enough to cover up the unique details of any recording. It's all an illusion; I have no qualms about manipulating it to a degree it's more pleasurable.

Thanks for the replies. I was just curious. I did a lot of blind testing and some double blind many years ago for an avionics company. I guess it's spoilt the fun for me you have listening to lots of different system and swaping componants around.

 

[phoenixdogfan]

Personally I would discount any poor measuring designs long before the evaluate at home stage.
Better measurements equals more transparency which is what I value.
Keith

Measurements are a filter for me as well. Not too many Zu's are getting an in home audition from me.

 

[MattHooper]

Thanks for the replies. I was just curious. I did a lot of blind testing and some double blind many years ago for an avionics company. I guess it's spoilt the fun for me you have listening to lots of different system and swaping componants around.

Understandable. I've done a fair bit of blind testing over the years and, while I know some can find it unsettling to see their subjective impressions melt away under rigorous controls, I've always found it fascinating and like I've yet again learned something significant. There really is nothing like an actual experience to make something sink in to one's bones.

On the other hand, I don't feel I have to blind test everything I'm doing (that would be a drag and a hassle at some point). So I don't mind some ambiguity, being in a spot where it could be a bias effect on my part or it could be a 'real sonic difference' I'm enjoying. I may just go for the ride. I don't have to make my life a science experiment but I DO want to stay consistent on whatever claims I'd make on a subject. Maybe I'm experiencing something subjectively pleasing but if I don't have good enough evidence for why, I wouldn't want to pretend I "know" more than I know.

 

[MattHooper]

Measurements are a filter for me as well. Not too many Zu's are getting an in home audition from me.

And that is why measurements can be so useful for some people's goals!

 

[krabapple]

Yes, ideally, of course. If I wanted 'colors', I'd add them with EQ. Andin that imaginary techtopia no doubt there would be a market for 'emulator' software/boxes , like what guitar plays have now, for those strange souls who think speakers should be like musical instruments and really *want* to hear what Magneplanars and Klipschorns sounded like.