Objectivist or Subjectivist? Give Me a Break

[jkeny]

Somehow in this measurements discussion we seem to be talking at cross purposes - what I & Micro are saying is that it's the complex relationships between the individual variables in the reproduced sound field that matter to our hearing perception as regards whether we find what we hear approaches something resembling an illusion of reality. The ear/brain is a superb instrument for this analysis having been honed through evolution to perfect this task. We are exquisitely aware of anomalies in this complex inter-relationship i.e we know the sound patterns from reality & anything that doesn't match this pattern is considered odd-sounding or incongruous.

The other side of the argument seems to focus on how sensitive instruments are to the measurement of individual variables in the soundfield compared to the ears sensitivity in this regard. Are we discussing two different strands here? Is this a complex Vs Simple dichotomy?

So is this the crux of the matter? With less sensitive instruments (ears) we derive more information from the sound field (because we have a processing & pattern matching engine working on it in real-time) than more sensitive instrumentation that lacks the ability to sample the same range of variables & to process them in real time! i.e with measurement instruments we can learn a lot about one or two variables but not the relationship between many variables!

If the point of measurement is to improve our sound reproduction devices to our ears, then does this statement not belie the problem "Perhaps hearing is not "replicated" by any instrument devised so far, but the measurement of the movement of air as in sound waves is far superior to the ear." Are measurements being done just to show off how sensitive our instruments are (to a small set of variables) or is there a goal to the measurements?

So, from this perspective is it not silly to say that we can't possibly hear something because it doesn't appear on our current measurements. Which implies the following, it is silly to say that we can measure everything we hear!

Again, let me refer to the video link I posted because this is from a chief engineer who's job it is to design a sound reproduction device (ESS DAC) so I presume he knows something about the engineering design but what is refreshing is that he also seems to pay attention to the sound that is being produced. At about 43:00 into the video he says something interesting - this is not a direct quote but it goes like

Despite the difference in noise levels between the regular SD to the ESS DAC, when you compare the outputs of the 2 DACs, i.e what goes to the speakers, they both look exactly the SAME.

Now this is interesting because the audiophiles he cited can hear a difference between the two DACs - I know some will choose not to believe this!

Later at at 48:55 he says:

"For me it was very interesting to make that jump to designing things that sound good even though it's somewhat uncomfortable because we can't always measure them as opposed to making things that look great in a simulation"

Despite what people might say about marketing puff, etc these are radical statements that an engineer, who values his reputation, will not make in public unless he has the back-up details to prove it.

 

[jkeny]

JackD's experiment is good evidence to suggest that our hearing is about pattern processing (as are some of our other senses). We tend to fill in the missing pieces in the sound-field & come up with a listening experience that more closely matches the full blown soundfield than it should. i.e we compensate for the missing bits. It also explains how we hear a conversation through a nosy background. It also explains Tim's issue with audiophile's reporting of the differences that they hear - to the man in the street there is not much difference if he listened to Jack's experiment, to me there was a huge difference in the songs I played with which I am familiar & the 1KHz band replay. I reported that it came from a tunnel, that it was robbed of all body, no fullness to the sound, no semblance of anything approaching realism. So this is an audiophiles description - call it hyperbole if you like & to the man in the street it is but I contend I listen for something different than just the familiarity of the song, I listen for it's realism, it's reproduction. So I guess, Tim, the reason audiophiles spout such nonsense is because they are listening in a different way to the sound. Don't get me wrong, when I am happy with the sound I abandon myself to it & let it wash over me so this is a different kind of listening, I guess?

Edit: As an aside the perception of sight is a similar analysis engine - there are separate parts of the brain that deal with just one aspect of the visual spectrum that we perceive - simplistically, one part deals with movement, a separate part deals with depth & another part deals with colour. These separate parts are amalgamated into an image plus other ancillary information. (Not to mention that the image on the retina is upside down to start with.)
So even if a very pixallated version of the Mona Lisa is presented, we know it's the Mona Lisa or if those luminous dots are put on the joints of the body & the movement of just the dots shown, we instantly recognise that it is a human body & the action they are performing. We are pattern matching machines when it comes to our sight & hearing senses.

 

[Ron Party]

Hi John. I don't have the time to watch that video to which you make reference. Is there a properly conducted DBT that supports the claims to which you refer and for which you cite the video as support? If so, where can I read about it?

 

[jkeny]

Hi John. I don't have the time to watch that video to which you make reference. Is there a properly conducted DBT that supports the claims to which you refer and for which you cite the video as support? If so, where can I read about it?

I've given the exact video timings for where I quoted - takes 2 mins & I also gave the slides.
I see the old card being played here "Peer reviewed, DBT, otherwise it's just anecdotal" Why not say what Winer says which is even more direct - I paraphrase "unless somebody personally shows me this then I don't believe it"

BTW, am I allowed to ask questions of Tim, you never did clarify what your warning post to me meant or Tim's interpretation of it? I'm confused.

 

[FrantzM]

Good Morning John

Perception has to do with processing and if that is what you are talking about I don't how anyone can discuss about this... Principally , perceptions require a stimulus to exist, else it is a phantom perception. I hope we are not dealing with that. These are usually called illusions, dreams or worse... There must be a stimulus or a group of stimuli. Replicating the stimuli allow with some serious processing by the brain involving processes and phenomenon I (or us all) fail to understand in many cases or thoroughly unexplained up to know but likely measured, cataloged and studied or being (or to be) studied by Science in ways that would allow us to comprehend them ... When presented with these stimuli our organs and our brains make of these stimuli perceptions but the perception begins with the stimuli. No stimuli= no Perception. When we perceive without stimulus then it is called a dream or worse ..

SO by corrrectly replicating the stimuli, we can approximate the perception and yes be fooled. So the soundscape painted by two speakers seems to put sound where there is actually no speaker at all , in particular strongly in the middle for example and elsewhere too thus the imaging we re so fond of, the stimuli had to come first, we had to replicate them first and the best way possible by (err) measuring (which is what capturing is after all) whatever signals caused these perceptions and re-producing the signals if and as well as possible and the only way to do this is to measure with the utmost precision, something our senses are very poor at despite the super computer behind them. So we use instruments to measure the signals and from that we sometimes construct a set of stimuli to be sent to our senses to be processed into our perception by the supercomputer we were born with to process the signal and make of these the concerti we love, the music we hear, the movies that we watch ... In other words, Our dear multi-media modern world.

Not much else I can say. I will be polite and retreat in the background for now ... I have things to deliver by tomorrow ... So ...

 

[jkeny]

Yes, Frantz, measuring/studying the stimulus with the intention of recreating it exactly is a fine goal. We know that it hasn't been achieved perfectly yet. So the divide between Subj & Obj is an argument over how far wrong this recreation/reproduction is, right? Now the subj says I can hear a problem & the obj says my measurements tell me that there is no audible problem. We immediately have a dilemma because it is admitted that the reproduction is not perfect, it is admitted that the measurement equipment does not have the processing power (& experience) that the ear/brain system has, it is admitted that measurements cannot yet be correlated to what sounds best & yet we are being asked to believe a finite set of "exact" measurements over our hearing.

I know our hearing has many biases & variables but these can be neutralised/dealt with. So we are still left with the vista that incomplete measurements are being touted as more accurate than our hearing. If we just admit, like Martin Mallinson does in the video that he does not know why certain factors (phase modulation) are audible then we would all get over this divide.

 

[FrantzM]

John

I don't need the measuring equipment to process much . I want it to measure, I,we process measurements or the captured signs , only we do that. And if we manage to replicate the signals/stimuli then we accomplish the goal the illusion the reproduction ....
That is all. I want my microphone to add nothing. I and only I may need more from this signal after processing by my own brain, I and only I make the decision to use a microphone that is not flat toand pass its signal (measurement after all) into a processor to make Britney Spears voice something vaguely akin to a person singing intune ... The microphone doesn't know that and it is all right .. Maria Callas or Britney Spears same thing to the mike: bunch of vibrations.. We processes these into music or in the case of Britney... Noise

Now measurements do correlate with many things we hear... Oh Yes!! 30 dB down in the midrange and Britney Spears and Maria Callas may not even sound like a person singing ... Phase rotation in the Highs and The cymblas sound bleachy .. Too high even harmonic distortion and the amp sound bleachy and dry ...
We don't know it all but we know quite a bit... On that we also know that THD , at least as measured most often doesn't tell us much. We have a glimpse why often SET sound so glorious...
Our protocols will have to be reviewed, refined and atuned to certain perception people tells us about. We have to make sure we are NOT chasing shadows when the perception cannot be ascertain with any degree of repeatability or accuracy... exemplean audiophile says that knowing which cables was playing, the difference he perceived were "night and day". Pretty stark even considering the imprecision of human language. One is LIGHT the other NO light... Now let me get you blind in other words, I remove the knowledge of which cable is in the system ... Why isn't it so clear then??? ... You can do it with speakers with an accuracy close to 100% why is it that with cables you can't .. Is it because, maybe, the "Night and Day" difference is one of those perceptions without stimulus aka as dreams, illusions, phantasms??? Just wondering .... and to me that is the crux of the matter...

 

[microstrip]

What I find amusing, sorry that is the only epithet I can find to this rationale is the brain interpretation theory ... Tomelex please letus live with our 2-channel for now

(...) Measurements don't have to be direct in Science no one has yet gone around the earth with a ruler to measure its circumference we have other means, other measurements to help us do that , same with Audio .. By measuring using the very basic component of a sound namely its level, Phase, frequency and duration we know everything about that sound signal. We can dream there is other things we need to measure but for a sound I think that all there is to it .. When those sounds are combined into something we call music then our brain start making all kind of things about them including liking them or not and even that is not that reliable we find ourselves not liking things we would have killed for, while the stimulus stays its stupid same.. I used to loooove Electric Light Orchestra now I have several of their albums and find myself hardly listening to any of them ... then again they are on CD so it could be the medium

Frantz,

Do you really believe it is as simple as that? That all the new types of measurements Audio Precision have developed (and may be many of them are still being developed) for the new generation of audio measuring instruments are just marketing?

 

[FrantzM]

Microstrip

For sounds, yes. The interrelation between the variables make things more complex or difficult but for sound those are the parameters we need to know.

Frequency
Phase (in relation to each other)
Intensity
Duration
Direction (Where it comes from in a 3-D space)

If there are more I would be pleased to learn. The very reason of me discussing here ,although there are others amongst them the music and the friendship that develops from these exchange ... I know for example where I can go to listen to a SF Aida in Portugal ...

 

[Whatmore]

Hi John. I don't have the time to watch that video to which you make reference. Is there a properly conducted DBT that supports the claims to which you refer and for which you cite the video as support? If so, where can I read about it?

I've given the exact video timings for where I quoted - takes 2 mins & I also gave the slides.
I see the old card being played here "Peer reviewed, DBT, otherwise it's just anecdotal" Why not say what Winer says which is even more direct - I paraphrase "unless somebody personally shows me this then I don't believe it"

BTW, am I allowed to ask questions of Tim, you never did clarify what your warning post to me meant or Tim's interpretation of it? I'm confused.

So was there a properly conducted DBT or not?

 

[Mosin]

This whole business is about flavors. Whether it is digital vs. analog, solid state vs. tube, moving coil vs. strain gauge, belt vs. idler, there will always be various camps; just name your poison. All those camps use science in some way to get there, but at the end of the day it is still about flavors, right or wrong.

 

[jkeny]

John

I don't need the measuring equipment to process much .
I want it to measure, I,we process measurements or the captured signs , only we do that. And if we manage to replicate the signals/stimuli then we accomplish the goal the illusion the reproduction ....
That is all. I want my microphone to add nothing. I and only I may need more from this signal after processing by my own brain, I and only I make the decision to use a microphone that is not flat toand pass its signal (measurement after all) into a processor to make Britney Spears voice something vaguely akin to a person singing intune ... The microphone doesn't know that and it is all right .. Maria Callas or Britney Spears same thing to the mike: bunch of vibrations.. We processes these into music or in the case of Britney... Noise

Now measurements do correlate with many things we hear... Oh Yes!! 30 dB down in the midrange and Britney Spears and Maria Callas may not even sound like a person singing ... Phase rotation in the Highs and The cymblas sound bleachy .. Too high even harmonic distortion and the amp sound bleachy and dry ...
We don't know it all but we know quite a bit... On that we also know that THD , at least as measured most often doesn't tell us much. We have a glimpse why often SET sound so glorious...
Our protocols will have to be reviewed, refined and atuned to certain perception people tells us about. We have to make sure we are NOT chasing shadows when the perception cannot be ascertain with any degree of repeatability or accuracy... exemplean audiophile says that knowing which cables was playing, the difference he perceived were "night and day". Pretty stark even considering the imprecision of human language. One is LIGHT the other NO light... Now let me get you blind in other words, I remove the knowledge of which cable is in the system ... Why isn't it so clear then??? ... You can do it with speakers with an accuracy close to 100% why is it that with cables you can't .. Is it because, maybe, the "Night and Day" difference is one of those perceptions without stimulus aka as dreams, illusions, phantasms??? Just wondering .... and to me that is the crux of the matter...

Yes, Frantz, all you say is reasonable but what I & many before are saying is that it appears hearing can currently reveal more than measurements are showing. Now the reasons for why this is are endless but it seems to me to be a falsehood to state that all things that are heard can be measured & then not be able to demonstrate this. The time to make that statement is when it can be demonstrated. Otherwise it is just a bit of self-opinionated boasting & like any such claim requires testing.

It's like saying that because we know how DNA works & have mapped the genome we know how it works. Well I don't think you will find anybody making that claim. There is more to it than just the individual mapping of the building blocks - the order of when they become active, the inter-gene sections, etc. all have to be worked out & until they are claiming that we know the human genome & it's operation is an empty boast. It seems to me that the stage that measurements are at is in mapping the building blocks but some way from understanding how they all interrelate

 

[Steve Williams]

This whole business is about flavors. Whether it is digital vs. analog, solid state vs. tube, moving coil vs. strain gauge, belt vs. idler, there will always be various camps; just name your poison. All those camps use science in some way to get there, but at the end of the day it is still about flavors, right or wrong.

I absolutely concur

 

[Bill Hart]

I'm a little surprised by the extreme divergence of views thinking, perhaps naively, that circa 2012 we had achieved a level of enlightenment that went beyond 'measurements are the final arbiter of sound' or that all 'tweaks' or component matching on a system-wide basis are simply imaginary.
So, a few questions and observations.
1. If measurements are the final arbiter, for say, an amplifier, what measurements are we talking about?
2. Pick 3 or 5 amplifiers that all have comparable measurements- do they sound the same? If not, why not? And why is the difference not revealed by the relevant measurements? Or, is it the 'objective' view that these amplfiers really do sound the same, and that any perceived differences can be eliminated in double blind testing?
3. What does synergy with related components, including wire, have to do with any difference in sound between the amplfiers in point 2, above?
4. I have heard differences in my system as a result of changes in power cords on the amplifiers, or changes in the interconnect or speaker cable, or how a power supply is coupled or decoupled from its mounting platform. Are these simply imaginary on my part? If not, are these differences measurable?
5.How can the overall result, using measurements, be predicted on a system-wide basis? Is it simply a matter of testing for frequency response and the like, of the system as a whole?


I guess I'm advocating the intelligent use of measurements and gathering as much useful information as possible but I still see a role for subjective judgment in sound quality, at least at the level of component matching to assemble a system, which is something, I gather, that meaurement of individual components cannot determine.
Admittedly, such component matching may be reduced to trying to find compatable colorations that, overall, produce a desirable result. But if we accept that every component has some sonic character, they are all colored or inaccurate to a greater or lesser degree. (Or those differences are simpy imaginary and are not revealed by appropriate measurements).
Finally, 'accurate' to what? The original performance or recording, which is impossible as a benchmark? To the chosen set of measurements? Isn't that circular, i.e, a particular measurement is given credence, and then the component is judged based on how closely it hews to that select measurement?
I am not suggesting that we just operate on a 'feel goods- sounds good' basis. I want data, to the extent science or engineering can explain it, on why one component sounds better than another. (Yes, i know 'better' is entirely subjective but I'd like to hear a cymbal that sounds like a cymbal does in real life, or a cello, or a human voice- it is something that can fool me into believing that the sound being re-created in my room is 'alive,' rather than reproduced- not something, admittedly, that happens consistently or very often). But, my sense, reading the hard line objectivist position* is that the ultimate sound quality produced by a component may be subordinate to measured performance, i.e., if it measures well, that's the end of the inquiry. And that's where I have difficulty- the fact that we can't measure it doesn't mean it doesn't exist or can't be heard. I would think that the true value of measurements is to serve a predictive function, and not simply to serve as ends to themselves.

*Disclaimer: I am not attributing the 'hardline' objectivist view to any one person or posting here, nor am I trying to set up straw arguments or distort the position any one person may have taken. Instead, I have tried to summarize what I gathered were the basic positions being taken here for the purpose of better understanding those views. And if I have it wrong, my apologies in advance.

 

[jkeny]

This whole business is about flavors. Whether it is digital vs. analog, solid state vs. tube, moving coil vs. strain gauge, belt vs. idler, there will always be various camps; just name your poison. All those camps use science in some way to get there, but at the end of the day it is still about flavors, right or wrong.

-1, I don't agree. I believe that this is a cop-out when faced with complexity that can't be easily measured/processed/analysed!

BTW, it is interesting to note through this thread just some of the views expressed in the original article & how true they are despite the protestations of many here. Just look through the thread & you will smile at the prescience of the statements in that article. So do people really think she is now absurd?

 

[mep]

I'm a little surprised by the extreme divergence of views thinking, perhaps naively, that circa 2012 we had achieved a level of enlightenment that went beyond 'measurements are the final arbiter of sound' or that all 'tweaks' or component matching on a system-wide basis are simply imaginary.
So, a few questions and observations.
1. If measurements are the final arbiter, for say, an amplifier, what measurements are we talking about?
2. Pick 3 or 5 amplifiers that all have comparable measurements- do they sound the same? If not, why not? And why is the difference not revealed by the relevant measurements? Or, is it the 'objective' view that these amplfiers really do sound the same, and that any perceived differences can be eliminated in double blind testing?
3. What does synergy with related components, including wire, have to do with any difference in sound between the amplfiers in point 2, above?
4. I have heard differences in my system as a result of changes in power cords on the amplifiers, or changes in the interconnect or speaker cable, or how a power supply is coupled or decoupled from its mounting platform. Are these simply imaginary on my part? If not, are these differences measurable?
5.How can the overall result, using measurements, be predicted on a system-wide basis? Is it simply a matter of testing for frequency response and the like, of the system as a whole?


I guess I'm advocating the intelligent use of measurements and gathering as much useful information as possible but I still see a role for subjective judgment in sound quality, at least at the level of component matching to assemble a system, which is something, I gather, that meaurement of individual components cannot determine.
Admittedly, such component matching may be reduced to trying to find compatable colorations that, overall, produce a desirable result. But if we accept that every component has some sonic character, they are all colored or inaccurate to a greater or lesser degree. (Or those differences are simpy imaginary and are not revealed by appropriate measurements).
Finally, 'accurate' to what? The original performance or recording, which is impossible as a benchmark? To the chosen set of measurements? Isn't that circular, i.e, a particular measurement is given credence, and then the component is judged based on how closely it hews to that select measurement?
I am not suggesting that we just operate on a 'feel goods- sounds good' basis. I want data, to the extent science or engineering can explain it, on why one component sounds better than another. (Yes, i know 'better' is entirely subjective but I'd like to hear a cymbal that sounds like a cymbal does in real life, or a cello, or a human voice- it is something that can fool me into believing that the sound being re-created in my room is 'alive,' rather than reproduced- not something, admittedly, that happens consistently or very often). But, my sense, reading the hard line objectivist position* is that the ultimate sound quality produced by a component may be subordinate to measured performance, i.e., if it measures well, that's the end of the inquiry. And that's where I have difficulty- the fact that we can't measure it doesn't mean it doesn't exist or can't be heard. I would think that the true value of measurements is to serve a predictive function, and not simply to serve as ends to themselves.

*Disclaimer: I am not attributing the 'hardline' objectivist view to any one person or posting here, nor am I trying to set up straw arguments or distort the position any one person may have taken. Instead, I have tried to summarize what I gathered were the basic positions being taken here for the purpose of better understanding those views. And if I have it wrong, my apologies in advance.

Hey Whart! Can you pass the popcorn please?

 

[jkeny]

+1 Whart

 

[microstrip]

Microstrip

For sounds, yes. The interrelation between the variables make things more complex or difficult but for sound those are the parameters we need to know.

Frequency
Phase (in relation to each other)
Intensity
Duration
Direction (Where it comes from in a 3-D space)

If there are more I would be pleased to learn. The very reason of me discussing here ,although there are others amongst them the music and the friendship that develops from these exchange ... I know for example where I can go to listen to a SF Aida in Portugal ...

Frantz,

All this quantities are permanently changing - the key issue is the variation. However variable signals such as music can generate enormous amounts of data that can not be analyzed and as such we say can not be measured. So, instrumentalists develop sequences of test signals that have a limited variation, but try to test the errors that they assume that the ear can perceive in real life. However these tests have limited scope. They are improving because the knowledge and instrumental capabilities of electronic instruments have (and are) increasing fast and a better understanding of the perception in the sound reproduction. Two decades ago any amateur could understand the specifications of an audio analyzer, today you have to have a degree in mathematics and signal processing to fully understand and use the full power of the new analyzers for development.

"Complex test sequences can be created within the UI, or the APx API can be accessed for full control from any .NET application or LabVIEW. " This sentence was taken from the Audio Precision page. Together with SOTA hardware this is one the greatest values of these instruments - the user can develop his own tests.

But by industry audio standards we are still in this situation: (text taken from the Burmester site)


Measurements are able to make a statement about the technical and mechanic quality of the piece of equipment. They cannot, however, predict the actual sound. Proof is given by the fact that it is possible to build two devices, which have exactly the same technical data but a completely different sound.

The musicality Dieter Burmester is demanding from his components is that they process the unaltered signal from the source to the speakers. This results in the unequalled reproduction, which is characteristic for all Burmester appliances.

In order to achieve the extraordinary Burmester sound, we use the latest technologies and finest electronic components in combination with the results of longstanding research and developments added by our uncompromising quality requirements.

Finally, apart from technological perfection, the human ear is still the most crucial criteria. Only when a newly developed device has passed the last authority – the skilled ears of our staff – the undiluted listening pleasure is guaranteed.

 

[Bill Hart]

Hey Whart! Can you pass the popcorn please?

I am not trying to be unnecessarily provocative here, it is not my style. But, I would like some help to understand because I'm not an extremist of any flavor and while i don't have the engineering expertise some do, that is precisely why I am asking these questions. If this is viewed as inflammatory, I apologize. That is not my intention.

 

[FrantzM]

Yes, Frantz, all you say is reasonable but what I & many before are saying is that it appears hearing can currently reveal more than measurements are showing. Now the reasons for why this is are endless but it seems to me to be a falsehood to state that all things that are heard can be measured & then not be able to demonstrate this. The time to make that statement is when it can be demonstrated. Otherwise it is just a bit of self-opinionated boasting & like any such claim requires testing.

It's like saying that because we know how DNA works & have mapped the genome we know how it works. Well I don't think you will find anybody making that claim. There is more to it than just the individual mapping of the building blocks - the order of when they become active, the inter-gene sections, etc. all have to be worked out & until they are claiming that we know the human genome & it's operation is an empty boast. It seems to me that the stage that measurements are at is in mapping the building blocks but some way from understanding how they all interrelate

Jkeny

That is not the argument here... I keep throughout my posts repeating that we don't know it all. Simply measuring is gathering data .. Interpretation is needed to make of it knowledge. You can have a large depository of data, it is a start but no the end, you need to interpret it and for that you need brains. I am not discussing that. never did nor did I see anyone mention that siple data collecting will lead to anything .. As a matter of fact the interpretation of data is what we are discussing about We need reliable data and our senses are unreliable, yet they can be useful since after all we are gathering data for them in this regard in Audio or video where we know that we can play some trick s and can have senses fooled and repeatably (Depth of soundstage, spaciousness, etc). We know that by observing measuring and listening or watching) we refine the instruments our senses have limitations and so far we have been pretty successful in making instruments more sensitive than our senses.. Microphones and Video sensors routinely see and hear more and better than our ears .. If we then use our brains to interpret these results , we know more... that 's all .. nd knowing more about what and how we hear produce beter results i-e reproduction ..

We are not there yet but we are forging ahead .. manufacturers are measuring, you better believe it and making sure they can replicate the results .. They must know at least why the basic electrical parameters of their circuitry translates into what their market likes and for that they must measure. If the mystique/marketing calls for them to say they do it all by ears and that increase their sales so they will say.. They will in the meantime, measure and measure more and measure better ..all the time continuously even if it is to produce a flavor far from the original signal as long as they seel that flavor they will measure and say they listen only ..Marketing ... You think for example that anyone canmatch two speaker within 0.5 dB by ear???