Is the entire audio industry a fraud?

[ahofer]

So, the more interesting question is why is high end audio different? Why can't subjectivist just say that they think that new power cord made them perceive the music as better sounding and leave it at that? No more arguments.

Sounding science-y and cutting edge has become one of this hobby's status markers. Sadly, this is also true for the prevailing sentiment at ASR. While the evidence leans more in objectivists' favor, we still wear our dedication to it like a badge of honor, or tribal flag.

Which is an excuse to trot out my favorite bit of rationalia:

Beware Inward Apologetics

Beware oft-rehearsed but rarely-performed arguments on why-our-group-is-right: Recently I came across a quotation that expressed, with wonderful clarity, something that I kind of half-knew but had not articulated so well to myself. The historian John P. Meier …:

www.overcomingbias.com

Consider that people embrace their tribe's most ridiculous beliefs to demonstrate loyalty. What an observation!

 

[ahofer]

There is no false positive or false negative, as the outcomes are not hypotheses.

Hmm. I'm certainly not as scornful of these tests as TL, but I'm not sure I see why false negatives aren't a possibility. The listener heard no difference where there was one? TL's thesis is that this, precisely, happens due to pressure (see my response above). But it seems reasonable to say it could happen.

I suppose you could argue that the test shows "they couldn't hear the difference" with no regard for exogenous influences like test pressure or background noise.

My own confidence (Bayesian, admittedly) comes not from a single ABX test, but from the accumulated evidence, and the astonishing lack of evidence for the opposing hypothesis. Funny how little attention that gets.

 

[NTK]

Hmm. I'm certainly not as scornful of these tests as TL, but I'm not sure I see why false negatives aren't a possibility. The listener heard no difference where there was one? TL's thesis is that this, precisely, happens due to pressure (see my response above). But it seems reasonable to say it could happen.

I suppose you could argue that the test shows "they couldn't hear the difference" with no regard for exogenous influences like test pressure or background noise.

My own confidence (Bayesian, admittedly) comes not from a single ABX test, but from the accumulated evidence, and the astonishing lack of evidence for the opposing hypothesis. Funny how little attention that gets.

How do you determine that some "should have heard difference" but did not? The identification of X is either correct or incorrect in the trial, like that of a coin toss. The variability is addressed by repeating the test numerous times, and look at the variance of the results, and gives the confidence interval.

 

[ahofer]

How do you determine that some "should have heard difference" but did not?

Oh, you likely can't. But it remains a possibility (and thus suggesting a margin of error). I agree that variability is addressed by repeating, especially replicating across different tests with different potential sources of error.

 

[bodhi]

Not having a background in the sciences even I have wondered about these night and day, you have to be deaf, veil lifting differences often claimed. To me, it shouldn't even take a blind test to pick A or B , sort of like telling the difference between a firecracker and a howitzer.

I have experienced countless times that I have trouble picking out differences comparable to "firecracker and a howitzer" when not trying. I have also witnessed these kind of things from my audiophile friends.

I still remember one time 15 years ago when I accidentally unplugged my sub while hoovering and it took me a week to notice. It was because I started to wonder if that box used to have a blue led on the front panel. Another time at our card game night we brought new speakers and an amp to replace some awful computer speakers: at some point one guy switched to the old ones using AUX connection from his phone's headphone out and nobody paid any attention until the guy left. And there were audiophiles present.

I can admit that I have wooden ears and maybe my friends have as well, but those differences should be 100 times more evident than what the same guys vow to hear, for example 320kbps vs lossless.

 

[SIY]

The thing about these horrible biases and terrible ABX tests is that they fool you, show great sensitivity to things like level, frequency response, data compression, and localization, but through means never to be known by you closed-minded science types, completely and mysteriously fail when the phenomenon is something electrically not measurable and not in accord with basic engineering.

It must have taken geniuses to devise tests like that, so perfectly designed to fail when looking for effects that aren't straightforward engineering.

 

[ahofer]

The thing about these horrible biases and terrible ABX tests is that they fool you, show great sensitivity to things like level, frequency response, data compression, and localization, but through means never to be known by you closed-minded science types, completely and mysteriously fail when the phenomenon is something electrically not measurable and not in accord with basic engineering.

It must have taken geniuses to devise tests like that, so perfectly designed to fail when looking for effects that aren't straightforward engineering.

This is a good point - the tests have shown plenty of positives among measurable phenomena, and that should increase our confidence.

 

[dshreter]

Hmm. I'm certainly not as scornful of these tests as TL, but I'm not sure I see why false negatives aren't a possibility.

Because the ABX is not testing if there is a difference. It is testing if the listener can identify a difference. If you blind compare a car stereo and Genelec’s latest but fail to identify the difference, that is not a false negative. If performed correctly, the result stands on its own that a difference was not identified.

There’s no definition for what difference should be identifiable. There are a number of specific things known that listeners are able to identify. And for the rest, that’s why you use experiments to test if they can be identified.

I think it’s all philosophically interesting, but practically there is much room for improvement in things we know are audible, so I’m less interested in latent mysteries of things that might be audible.

 

[SIY]

There’s no definition for what difference should be identifiable.

Positive controls perform that function and are often appropriate, depending on what specifically is being tested for.

 

[dshreter]

Positive controls perform that function and are often appropriate, depending on what specifically is being tested for.

Absolutely right, which is why I mention there are things known to be audible. It’s just to say there is not a universal definition one can apply ex ante to any two sounds, which is why you test.

 

[ahofer]

Because the ABX is not testing if there is a difference. It is testing if the listener can identify a difference. If you blind compare a car stereo and Genelec’s latest but fail to identify the difference, that is not a false negative. If performed correctly, the result stands on its own that a difference was not identified.

There’s no definition for what difference should be identifiable. There are a number of specific things known that listeners are able to identify. And for the rest, that’s why you use experiments to test if they can be identified.

I think it’s all philosophically interesting, but practically there is much room for improvement in things we know are audible, so I’m less interested in latent mysteries of things that might be audible.

Good distinction, but I think TL’s point is that the subject has failed to identify a test where he otherwise could have, due to the pressure/artificiality of the test. As I said, I reject his hypothesis due to both accumulation of evidence, and consistency with audiology threshold-of-hearing research. The burden of proof is on him or others to devise and execute a test to reject (or not) the accumulated evidence.

 

[LouB]

To the OP. of course it is & we all know it. But what industry isn't ? The Audio industry just hasn't figured out how to market leaving science out of it. Have look at Auto, Big Pharma, Food, ect. advertising all of em selling there products with smiling beautiful people living a life were envious of.
If I buy that Nesspresso machine I'll be cool as George Clooney, maybe buy that Lincoln and when I take my shirt off I'll look just like Matthew McConaughey. Now there's no "fraud" by definition in those ads and thats the beauty of it, Madision ave. at it's finest.
I guess the Audio industry isn't big enough or have a wide enough consumer base to warrant those kind of million dollar ad campaigns so there fraud is blatant and easily disproved.

 

[Mr. Widget]

I think this is a much more complex topic, more appropriate to psychology and social science, and outside the expertise of the bulk of the forum. I've been able to add a little bit from my knowledge of behavioral finance.

Thanks for the interesting observations and the link.

Regarding psychology etc., I think you are spot on. I think there is much to study outside the confines of measuring and researching gear. To me the aesthetic design and our aural perception is more interesting than SINAD scores and the like. (Sure, like the rest here I am a geek for this stuff too, but...)

For example why are so few women drawn to this hobby. I personally only know one female audiophile... and thinking about it only a very small group of men are really interested in this hobby. Many men have a passing interest, but outside of groups like this one, standing in line at the grocery store the likelihood of meeting someone who actually knows the difference between a woofer and a subwoofer is pretty low. Finding one who cares is even lower.

 

[ahofer]

Thanks for the interesting observations and the link.

Regarding psychology etc., I think you are spot on. I think there is much to study outside the confines of measuring and researching gear. To me the aesthetic design and our aural perception is more interesting than SINAD scores and the like. (Sure, like the rest here I am a geek for this stuff too, but...)

For example why are so few women drawn to this hobby. I personally only know one female audiophile... and thinking about it only a very small group of men are really interested in this hobby. Many men have a passing interest, but outside of groups like this one, standing in line at the grocery store the likelihood of meeting someone who actually knows the difference between a woofer and a subwoofer is pretty low.

My family finds my fascination amusing and they often make fun of me. Although when my oldest son bought a house, I pointed him here and he put Genelec 8030s + Genelec Sub, and an SMSL Dac in his living room.

 

[atmasphere]

I'm highly familiar with Earl's work. And why any amp that's not severely broken passes his metric. Earl uses a cheap old solid state receiver.

But once again, you've made assertions about relative audibility of low levels of distortion with ZERO evidence to back it up, just more assertions, hand waves, and misapplied references. This is a continuing issue and you really need to do better.

What Earl uses for an amplifier does not detract from his point. Actually I'm pointing at a disconnect between the camps of subjectivist and objectivist. Both are composed of humans and since that is true, both capable of making mistakes. IMO/IME they both make the mistake of discounting the other and both are guilty of confirmation bias. And both dig in their heels at any challenge, larger or small, aqusing each other of hand waving and the like, getting nowhere. That's simply human nature and we are witnessing it on a massive scale in the political world.

I don't ignore measurement and I don't ignore what I hear and the feedback of thousands of customers.

The problem is that amps don't amplify sine waves. The measurement community likes to say that the distortion is so many dB down (perhaps 90dB) and therefore is masked; therefore inaudible. If you look at a signal on an oscilloscope you see something different. The distortion is integrated into the waveform. If we take my prior example of a
clarinet

we can see that the waveform clearly has higher ordered harmonic content which is how we tell it from a trumpet. It is not conjecture on my part that the ear uses higher orders to sense sound pressure and so is keenly sensitive to them (this has been known for well over 80 years and isn't hand waving )- you can prove it to yourself if you've not already done so. What you seem to be suggesting is that by altering the higher ordered content of the signal shown above that it will not alter the sound of the instrument (in this case by adding more harmonics) since the distortion is masked. Obviously it isn't- it modifies the sound of the clarinet. You can hear the difference between a Stradivarius and a middle of the road instrument on a car radio. But if you really want to hear what the Strad sounds like you will need to have less distortion because the distortion of the car radio (some of which are quite good these days) is changing the sound of the instrument in the same way a sine wave is changed passing thru a circuit.

If I were to embody the disconnect, the measurement side seems to take the position that distortion is masked by the signal itself; therefore all amps sound the same... The subjective side takes the position that you can hear things and ascribes them the magic of the design (preferring that over admiting that the meaurements show what they hear) and are willing to pay dearly for it, even though they might be buying something a country mile away from neutral. IME the measurements are saying exactly what they are hearing and oddly, the measurement people don't want to admit that either, clinging to the 'all amps sound the same as long as the distortion is low enough' (this being due to the idea that the distortion is masked, when in fact it sculpts the sound of each instrument). (What I contend is the the amps will sound the same regardless of technology if they have the same distortion spectra and if FR is consistent. The more you reduce the distortion, the more it will sound neutral as long as attention is paid to how the ear interprets higher ordered harmonics.)

This nonsense will continue. I doubt either party has the veracity to see something in the other's argument, IMO because both are wrong about what's happening.

IME the measurements tell you how the circuit will sound if you have enough of them. I don't think this was true in the 1980s and a lot of the subjectivist crowd seems to live their lives as if the measurement tech has not advanced past 1985 and so can't be trusted. This ignores what you see in the measurement pages of Stereophile of course, but many people run their lives according to decisions or observations they made about life at a much younger age and suffer from it as a result. In my work I've found that the lower you can get the distortion the more neutral the circuit becomes- IOW its on a scale rather than a yes/no thing that is commonly purported. For it to really not color the sound it needs to be a lot lower IMO than most say. I'm of the opinion that there are amps out there now that are operating in this realm; none of them existed or were even possible 20 years ago.

You can resort to calling this hand waving and the like but in doing so you'll be doing the same thing. Rather than resulting to insult, what you need to refute my comments is to show that somehow the clarinet waveform can exist perfectly while the distortion is acting as a noise floor so it could be masked. But if the clarinet signal itself is distorted then it can't be masked. Do you get my point?

So, the more interesting question is why is high end audio different? Why can't subjectivist just say that they think that new power cord made them perceive the music as better sounding and leave it at that? No more arguments.

FWIW dept.: power cords behave according to Ohm's Law like everything else. What you might not get is that the voltage drop across the cord can affect the performance of an amplifier that draws a bit of current. To see what the power cord is doing you therefore measure the performance of the amplifier: full power, distortion and output impedance. The more feedback the amp has, the less power it draws, the less the power cord will have an influence. So a powerful zero feedback amplifier running class A will be affected most, particularly if its a tube amp with an unregulated filament circuit. I've seen power cords affect such amps by dropping the output power by 40 Watts so this is very real.

But the solution isn't a megabuck power cord, just one that can handle the current without heating up. If you have such an amp you might need to go to Menard's and get some electric stove power cable and make a power cord out of that. The IEC and AC power plugs will be the lion's share of the cost of the cord.

 

[ahofer]

You can hear the difference between a Stradivarius and a middle of the road instrument on a car radio.

Not sure about that, people have had trouble with this live.

somehow the clarinet waveform can exist perfectly while the distortion is acting as a noise floor so it could be masked. But if the clarinet signal itself is distorted then it can't be masked. Do you get my point?

Isn't all harmonic distortion part of the combined waveform? I'm not sure this contradicts masking theory, but not my area of expertise.

 

[Killingbeans]

Obviously it isn't- it modifies the sound of the clarinet.

It modifies the waveform. But our hearing doesn't function as an oscilloscope. It's more more akin to a spectrum analyzer?

Sound - Hearing, Frequency, Wavelength

Sound - Hearing, Frequency, Wavelength: The ear actually functions as a type of Fourier analysis device, with the mechanism of the inner ear converting mechanical waves into electrical impulses that describe the intensity of the sound as a function of frequency. Ohm’s law of hearing is a...

www.britannica.com

 

[SIY]

What Earl uses for an amplifier does not detract from his point. Actually I'm pointing at a disconnect between the camps of subjectivist and objectivist. Both are composed of humans and since that is true, both capable of making mistakes. IMO/IME they both make the mistake of discounting the other and both are guilty of confirmation bias. And both dig in their heels at any challenge, larger or small, aqusing each other of hand waving and the like, getting nowhere. That's simply human nature and we are witnessing it on a massive scale in the political world.

I don't ignore measurement and I don't ignore what I hear and the feedback of thousands of customers.

The problem is that amps don't amplify sine waves. The measurement community likes to say that the distortion is so many dB down (perhaps 90dB) and therefore is masked; therefore inaudible. If you look at a signal on an oscilloscope you see something different. The distortion is integrated into the waveform. If we take my prior example of a
clarinet

we can see that the waveform clearly has higher ordered harmonic content which is how we tell it from a trumpet. It is not conjecture on my part that the ear uses higher orders to sense sound pressure and so is keenly sensitive to them (this has been known for well over 80 years and isn't hand waving )- you can prove it to yourself if you've not already done so. What you seem to be suggesting is that by altering the higher ordered content of the signal shown above that it will not alter the sound of the instrument (in this case by adding more harmonics) since the distortion is masked. Obviously it isn't- it modifies the sound of the clarinet. You can hear the difference between a Stradivarius and a middle of the road instrument on a car radio. But if you really want to hear what the Strad sounds like you will need to have less distortion because the distortion of the car radio (some of which are quite good these days) is changing the sound of the instrument in the same way a sine wave is changed passing thru a circuit.

If I were to embody the disconnect, the measurement side seems to take the position that distortion is masked by the signal itself; therefore all amps sound the same... The subjective side takes the position that you can hear things and ascribes them the magic of the design (preferring that over admiting that the meaurements show what they hear) and are willing to pay dearly for it, even though they might be buying something a country mile away from neutral. IME the measurements are saying exactly what they are hearing and oddly, the measurement people don't want to admit that either, clinging to the 'all amps sound the same as long as the distortion is low enough' (this being due to the idea that the distortion is masked, when in fact it sculpts the sound of each instrument). (What I contend is the the amps will sound the same regardless of technology if they have the same distortion spectra and if FR is consistent. The more you reduce the distortion, the more it will sound neutral as long as attention is paid to how the ear interprets higher ordered harmonics.)

This nonsense will continue. I doubt either party has the veracity to see something in the other's argument, IMO because both are wrong about what's happening.

IME the measurements tell you how the circuit will sound if you have enough of them. I don't think this was true in the 1980s and a lot of the subjectivist crowd seems to live their lives as if the measurement tech has not advanced past 1985 and so can't be trusted. This ignores what you see in the measurement pages of Stereophile of course, but many people run their lives according to decisions or observations they made about life at a much younger age and suffer from it as a result. In my work I've found that the lower you can get the distortion the more neutral the circuit becomes- IOW its on a scale rather than a yes/no thing that is commonly purported. For it to really not color the sound it needs to be a lot lower IMO than most say. I'm of the opinion that there are amps out there now that are operating in this realm; none of them existed or were even possible 20 years ago.

You can resort to calling this hand waving and the like but in doing so you'll be doing the same thing. Rather than resulting to insult, what you need to refute my comments is to show that somehow the clarinet waveform can exist perfectly while the distortion is acting as a noise floor so it could be masked. But if the clarinet signal itself is distorted then it can't be masked. Do you get my point?

FWIW dept.: power cords behave according to Ohm's Law like everything else. What you might not get is that the voltage drop across the cord can affect the performance of an amplifier that draws a bit of current. To see what the power cord is doing you therefore measure the performance of the amplifier: full power, distortion and output impedance. The more feedback the amp has, the less power it draws, the less the power cord will have an influence. So a powerful zero feedback amplifier running class A will be affected most, particularly if its a tube amp with an unregulated filament circuit. I've seen power cords affect such amps by dropping the output power by 40 Watts so this is very real.

But the solution isn't a megabuck power cord, just one that can handle the current without heating up. If you have such an amp you might need to go to Menard's and get some electric stove power cable and make a power cord out of that. The IEC and AC power plugs will be the lion's share of the cost of the cord.

Wall of words notwithstanding, you still haven’t provided any data or evidence to back up your claims. Yet again.

 

[Dan]

the slightly older guy who all but admits he hates audio science simply because it spoils his fun

Love this. So true.

 

[MarkS]

I don't ignore measurement and I don't ignore what I hear

If you are not listening blind, then you are using your eyes, not your ears, to judge.

No need for fancy ABX tests and statistics. Just listen blind. Do whatever you usually do to decide that Amp A sounds better than Amp B, but without knowing which is which.

The number of audiophiles who have actually tried such an experiment is vanishlngly small.