Testing audiophile claims and myths

[GearMe]

I am way too late to this discussion

But I agree with @GearMe; there is little point in arguing whether perceived differences are real or imaginary. One may get annoyed that someone is trying to make money out that perception, but ultimately, if it works for the end-user and makes them happy, then who am I to argue against their purchase decisions.

As for myself, having a background in physics and electronic engineering, I work out for myself whether a designer/manufacturer/marketeer is likely selling snake oil or possibly may be on to something, do my own cost/benefit/placebo calculations, and use my wallet accordingly...

Same here...

Talking about audio in a scientific way is not "arguing against anybody's purchase decisions." Should science refrain from saying anything about the audible differences between different audio formats just to not "argue against people's purchase decisions?" If "it works" for some end-users and makes them happy, what do these consumers need audio science for? If purchasing the highest bitrate downloads gives you the best audio experience (for whatever reason) on the planet, why even come to discussion boards like this one only to be lectured how according to science, 16/44.1 is all you need in consumer audio?

Of course science should not refrain from saying anything about the audible differences between different audio formats, but it should not do so from a hubristic perspective that science knows everything there is to know about auditory perception & processing; it doesn't. But really my point is that whether these differences are considered real or imaginary is a red herring when it comes to purchase decisions. People make these decisions based on a subjective perception, that works for them.

And you are right; end-users who are happy with their purchase decisions have zero use for a scientific slap on the wrist. They couldn't care less about the discussions that happen on forums like this.

Agreed!

 

[71 dB]

Of course science should not refrain from saying anything about the audible differences between different audio formats, but it should not do so from a hubristic perspective that science knows everything there is to know about auditory perception & processing; it doesn't. But really my point is that whether these differences are considered real or imaginary is a red herring when it comes to purchase decisions. People make these decisions based on a subjective perception, that works for them.

Science doesn't know everything, nobody does, but it is supposed to be the best knowledge we have in any given point in time. Again, audio science isn't telling anyone what to do with their money. Science provides the best available knowledge people can use the way they see fit from their subjective perspective. If you still want to spent your hard earned money on hi-res and other snake oily audio products after hearing what science has to say about them that's completely your call.

And you are right; end-users who are happy with their purchase decisions have zero use for a scientific slap on the wrist. They couldn't care less about the discussions that happen on forums like this.

So why do we have end-users like that over here? What I have witnessed here is many of those people seem to care a lot what happens here insisting (audio) science must be wrong if it contradicts their subjective experiences and feelings.

 

[bigshot]

How things sound subjectively applies to everything we can hear, not just recorded sound, and it differs from person to person. Fidelity is simple-it’s about how close the copy is to the original signal. That can be measured and tested using listening tests. No mystery there. The best case thresholds of human hearing have been studied for over a century, and they are firmly established. No mystery there either.

Quite frankly, I think the insistence on diverting discussion of fidelity into slippery slope discussions of subjective perception is just an argumentative technique used to avoid having to admit that 16/44.1 PCM covers the full spectrum of audible sound and reproduces it with accuracy that is perfect for human ears.

Subjective impressions are irrelevant when it comes to discussions of fidelity, and they can only degrade hearing when it comes to the best case thresholds of human hearing.

There is such a thing as good enough.

 

[BS5711]

“ … science knows how digital audio works, but it doesn't fully know how it sounds. How it sounds is inextricably linked to perception of sound. There is the difference between sound as a measurable physics phenomenon (progression of pressure variation through air), versus that of sound as we experience it as humans …. “

“…. there is little point in arguing whether perceived differences are real or imaginary. One may get annoyed that someone is trying to make money out that perception, but ultimately, if it works for the end-user and makes them happy, then who am I to argue against their purchase decisions …. “


The first comment is of course true but isn’t it entirely reasonable to do everything one can to differentiate measurable physical sound from perception of sound created by things that are entirely dependent on ones state of mind ?

Yes science can in some circumstances eventually be advanced by delving more deeply into certain phenomena but science certainly isn’t going to be advanced by indulging beliefs that are with 99.9% certainty completely imagined.

 

[bigshot]

Science has studied placebo effect and expectation bias extensively.

 

[2leftears]

Quite frankly, I think the insistence on diverting discussion of fidelity into slippery slope discussions of subjective perception is just an argumentative technique used to avoid having to admit that 16/44.1 PCM covers the full spectrum of audible sound and reproduces it with accuracy that is perfect for human ears.

I didn't say anything re. 16/44.1 PCM.

16/44.1 PCM can be implemented well, and it can be implemented poorly. Well-implemented 16/44.1 PCM can sound very good indeed.

But let's be clear; 16/44.1 PCM done in accordance with Nyquist/Shannon theory (including all its strict idealised sharp bandwidth filtering and idealised reconstruction conditions) is NOT how it is implemented in practical 16/44.1 PCM audio.

 

[2leftears]

The first comment is of course true but isn’t it entirely reasonable to do everything one can to differentiate measurable physical sound from perception of sound created by things that are entirely dependent on ones state of mind ?

To some extent yes, but surely that differentiation is not really possible until science knows how to quantify (or even qualify) how we perceive sound. They still struggle how to come up with relevant metrics for sound signals.

 

[BS5711]

To some extent yes, but surely that differentiation is not really possible until science knows how to quantify (or even qualify) how we perceive sound. They still struggle how to come up with relevant metrics for sound signals.

If we can measure the physical sound isn’t the rest of the audio experience down to perception and obviously just preferences ?

That perception will vary from person to person with the listening experience for some people being dramatically more influenced by stimuli beyond the sound pressure waves entering their ears than others ?

 

[Brahmsian]

Firstly, Levinson was specifically talking about PCM rather than digital audio in general.

He wasn’t talking only about hi-res content either, so I don’t understand your references to inaudible content “that doesn’t exist.”

In addition, the difference between LP/cassette and digital audio recording was very well studied. Digital audio recorders were invented around 20 years before digital audio was released to consumers. I’m not sure about physiological effects but clearly the loss of high frequency content with analogue recordings was not generally preferred by consumers, who routinely applied the “V shaped” EQ curve to try and counter the HF loss with analogue recordings, while digital audio did not suffer that loss. Lastly, a “more soothing effect” doesn’t really make sense with music products because music is not always supposed to be soothing, audiences have also demonstrated a preference for exciting, grandiose, surprising, shocking and even harsh and dissonant/unpleasant elements in music.

When I was a kid, I used to listen to punk. But guess what? I still found low-level tape hiss and the sound of the needle in the groove soothing. We all know about some people’s preference for coloration/distortion, so even though I said goodbye to analog (tape and LP) a long time ago and don’t particularly miss it, I understand how some people might react negatively to the squeaky-clean sound of a DDD CD.

Science exists to discover the truth/facts about the reality of the universe we inhabit, not to address every crackpot idea anyone decides to invent and that is particularly true when that crackpot idea is not based in reality because science has no way of testing it. There are no scientific studies on the physiological/psychological response to unicorns, fairies, flying pigs, yeti’s, mermaids, etc., because there is no scientific way of testing the response to something that doesn’t exist. In such cases, the only thing science can do is state there is no reliable evidence to support the assertion, only reliable evidence that contradicts it and leave it at that, until there is something falsifiable that science can test.

G

Archimago wasn’t writing a science paper; it was a blog post in which he was literally addressing crackpot marketing claims. As I see it, if you’re going to address the crackpot theories, you have to address the main claim/premise (about the physiological effects) head-on. Was Archimago merely preaching to the choir or does he want to persuade others?

 

[bigshot]

But let's be clear; 16/44.1 PCM done in accordance with Nyquist/Shannon theory (including all its strict idealised sharp bandwidth filtering and idealised reconstruction conditions) is NOT how it is implemented in practical 16/44.1 PCM audio.

Oversampling DACs do a fine job of it, and most DACs fit that description. The importance of dithering is usually overstated in audiophile circles. In my sig file is a seminar for the AES that compares dithered and undithered audio and there is very little difference between them. In typical use in the home, you would never notice a difference.

 

[Elegiac]

The thing about subjective audio quality, is that exploitation of it is wrong. Even if the person selling snake oil believes in what they are selling, it is still unhealthy.

Audio is so ephemeral. It's not a cupcake or a comfy chair... hell, even a pleasant smell has more physical presence than audio. The mind cannot rest upon a solid, consistent perception of it easily. Tweaking your system with legitimate electronic parts can get stressful enough, very quickly. The mind in 'tweaking mode', is a horrible, uncomfortable thing. And it is exactly this demented, unhappy monomania that snake oil peddlers have as the basis of their business.

 

[bigshot]

To some extent yes, but surely that differentiation is not really possible until science knows how to quantify (or even qualify) how we perceive sound. They still struggle how to come up with relevant metrics for sound signals.

Nope. The metrics have been defined, and those definitions go back over a century to the Bell Labs... Sound reproduction involves frequency, amplitude, distortion and time. Those four core concepts measure everything we can hear. Each one has its own set of concepts and rules, but those are just details.

If you can tame frequency response, amplitude and dynamics, distortion of various types and time error to the degree that the reproduction exceeds the human ear's ability to hear it, you are golden. You don't need to worry about it any more. Regular old CDs do that. The only problem that might exist is some sort of design or manufacturing error that degrades sound so that it is performing below spec. That isn't the fault of the format. Regular old CDs should be perfect. You have to mess it up to get it wrong.

 

[bigshot]

I still found low-level tape hiss and the sound of the needle in the groove soothing.

Some people use their home audio systems to listen to music. I guess some listen to noise. That's fine, but that isn't what audiophiles claim to do. It's easy to introduce noise. Just fine the most soothing noise you can find... rain sounds? waterfalls? birds tweeting? wind?... make a loop of it and play it 24/7 on your home stereo. It will faithfully reproduce your favorite noise as well as it reproduces your favorite music.

Archimago has done some great tests of home audio fidelity. You should read his blog. You'll certainly learn a lot. I've learned a lot from it myself.

 

[castleofargh]

To some extent yes, but surely that differentiation is not really possible until science knows how to quantify (or even qualify) how we perceive sound. They still struggle how to come up with relevant metrics for sound signals.

1/ different people do get different sound at their eardrums. If they got the same in an experiment controlling what arrives at the eardrum, they would interpret it differently because their brain isn't calibrated with/used to the same sounds. It's even more obvious and significant for headphone use. Frequency response alone is involved in nearly all(all?) sound interpretations, but we're getting a different FR.

2/ Do you ask for a graph to quantify your love? Same thing here, remove all the stuff that is clearly emotional and subjected to change over time or from non audio variables, and we probably have more data than most people are willing to ingest already.
Every other post about gears involves soundstage. Most of the time, it's already the wrong word for what they talk about. But also we(science, and those who care to read some of it) know a lot about sound localization. What comes out is that we're not nearly as good at sensing as we like to think we are. And because of that, the brain is even more likely to go pick up cues outside the realm of sound, to help in his "low cpu" job of fast prediction. The obvious conclusion is to not expect a graph to justify our half made up and almost never accurate impression of, say, distance. But that tends to be true for most perception of space, whatever it is that someone meant by "soundstage". There is no hope or point in quantifying anything for guys who think of themselves as an infallible sensory machine with infinite range. No amount of research will validate that delusion.

3/ if all someone has, is a sighted impression, he needs to **** about sound claims because sound is not the exclusive source of his experience.
And again, what audio measurement is going to quantify preconceptions and visual cues? Telling people that their feelings of sound involve more than sound, just does not work in this hobby. It's not about knowledge or quantifying it, it's about ego and people with exceptional level of denial when it comes to their flaws and mistakes.

Have people understand and truly accept those 3 points, and all audio forums will move between 50 and 4000 years into the future in terms of scientific knowledge. Meanwhile, we'll continue navigating in forums driven by textbook propaganda. You're the best for belonging to some group, it's us VS them, there is a lot to fix, are you sure you're not having that problem, if you don't say you're hearing it then you have cardboard ears or peasant rig, many people agree with me, my claims are as valid as yours, no u...

 

[castleofargh]

Archimago wasn’t writing a science paper; it was a blog post in which he was literally addressing crackpot marketing claims. As I see it, if you’re going to address the crackpot theories, you have to address the main claim/premise (about the physiological effects) head-on. Was Archimago merely preaching to the choir or does he want to persuade others?

What could he possibly do that he didn't do? There is very little concrete data about what the alleged PCM fix is actually doing. There is even less data about what problem PCM has in the first place. And the best demonstration, from the snake oil poster guy, refers to pseudo-science tools.
Yes, Archi takes cheap shots, like I did, because there isn't much else to do about such an obvious nothingburger, beside listing the few known facts around it and mocking the concept for standing on nothing.
Even if I believed in something needing a fix with PCM, I would still think Levinson is banking on snake oil here. That's how little credibility, or evidence, I see. From what is advertised, I don't even feel it's nothing like our friend selling the word "quantum". My limited assumption, because the guy is really using many words to say nothing of substance about C Wave, is that it will lower the fidelity(maybe in a euphonic way, maybe in an inaudible way, that IDK).