Testing audiophile claims and myths

[bigshot]

I still haven't made up my mind whether some posters in this group are sophisticated bots or not... It's clear that the intended audience for all these words is themselves.

 

[Dogmatrix]

Excellent technical questions have been raised and answered , it would seem all that remains is the unknown so I ask

What is the significance of the unknown if it remains unknown ?

And

Where could the unknown be hiding ?

Not a Bot

 

[craiglester]

I'm still confused how anyone could say that audible differences could be beyond measurement. I was under the impression measuring equipment was comfortably capable of measuring anything audible. If you can't measure it, is it really there?

 

[bigshot]

Where could the unknown be hiding ?

WHERE YOU LEAST EXPECT IT! THE SHADOW KNOWS!

 

[Dogmatrix]

WHERE YOU LEAST EXPECT IT! THE SHADOW KNOWS!

I know , but the Shadow is busy

 

[gregorio]

Does anyone here think that is relevant to the improvement of the art of audio engineering?

I don’t know about anyone else, but I don’t! And that seems to be the issue, you are coming at this from the point of view of “improving the art of audio engineering” (audio equipment design) but I’m not, I’m just answering the question/myth.

Isn't it just a step forward in the art of pedantry?

No, it’s refuting a common audiophile myth, but maybe you see that as pedantry because you are only thinking in terms of what is useful/interesting to improve the art of equipment design?

You usually do better than this...

Do I?

Now, the next step is: Can audio subjectivism find differences in audio that objective analysis cannot yet qualify or quantify?

That’s not the next step or rather, it’s potentially the next step after refuting the myth but it’s exactly the same “step” you appear to have been making all along. The answer is already in my previous responses to you: We can measure all audible differences at least by virtue of being able to record them digitally but we cannot (yet) un-mix a recording and therefore isolate all the individual components so they can be analysed, for example isolate the 2nd violins in a typical orchestral recording or separate certain distortions from other distortions. However, human perception does allow this to a degree in certain cases, as demonstrated by my 2nd violins example and the “cocktail party effect”, which has been well studied.

G

 

[jagwap]

I don’t know about anyone else, but I don’t! And that seems to be the issue, you are coming at this from the point of view of “improving the art of audio engineering” (audio equipment design) but I’m not, I’m just answering the question/myth.

No, it’s refuting a common audiophile myth, but maybe you see that as pedantry because you are only thinking in terms of what is useful/interesting to improve the art of equipment design?

Do I?

That’s not the next step or rather, it’s potentially the next step after refuting the myth but it’s exactly the same “step” you appear to have been making all along. The answer is already in my previous responses to you: We can measure all audible differences at least by virtue of being able to record them digitally but we cannot (yet) un-mix a recording and therefore isolate all the individual components so they can be analysed, for example isolate the 2nd violins in a typical orchestral recording or separate certain distortions from other distortions. However, human perception does allow this to a degree in certain cases, as demonstrated by my 2nd violins example and the “cocktail party effect”, which has been well studied.

G

It is the "myth" the objectivist have been fighting here for years: that subjective listening can not hear things that cannot be scientifically analysed. It was always phrased as cannot be measured because EVERYONE commonly thinks of that as SINAD, Intermodulation, FFT etc measurement, except you. You decided to define a recording as a measurement. So we rephrase, to avoid, well, whatever.

So now, here it is. One of the key objectivists, agreeing that listening can reveal what analysis cannot (yet), after all these years of confronting those who suggested anything like it.

I kind of expected music, trumpets, fireworks.

A window opens:
"Sound Science Forum Level 2:
pedants corner"


castleofargh Can we pin this as a sticky?

 

[PhonoPhi]

It is the "myth" the objectivist have been fighting here for years: that subjective listening can not hear things that cannot be scientifically analysed. It was always phrased as cannot be measured because EVERYONE commonly thinks of that as SINAD, Intermodulation, FFT etc measurement, except you. You decided to define a recording as a measurement. So we rephrase, to avoid, well, whatever.

So now, here it is. One of the key objectivists, agreeing that listening can reveal what analysis cannot (yet), after all these years of confronting those who suggested anything like it.

I kind of expected music, trumpets, fireworks.

A window opens:
"Sound Science Forum Level 2:
pedants corner"


castleofargh Can we pin this as a sticky?

Recordings can be perfectly classified as measurements.
One must have a detector(s) to record the sound.

How recordings and their possible limitations including artefacts relate to hearing with all its psychoacoustic complexity is another story.

Can recordings (at least in their current form) capture all what can be heard - no, but it is a different duscussion.

Yet, the very nature of recordings makes them perfectly quantifyable, leaving no room for people listening to reproductions of these recording (however diversely imperfect in their turn these reproductions are and however creatively hyped to audiophiles) for "hearing things", pretending hearing things and debating their pretenses any meaningfully, let alone scientifically.

 

[gregorio]

It is the "myth" the objectivist have been fighting here for years: that subjective listening can not hear things that cannot be scientifically analysed.

No, that is neither the stated myth nor the myth we’ve been fighting! The actual myth we’ve been fighting is that there are audible differences that cannot be measured. “Audible differences” is not “subjective listening” and “measurable” is not “scientifically analysable”.

So, you’ve invented a statement that’s true but significantly different to the myth and then fallaciously claimed the myth is therefore true! That’s a non-sequitur.

Incidentally, your statement (not the myth!) has been demonstrated true numerous times. Take for example the McGurk Effect, we can subjectively listen and hear a difference between “baa” and “faa” but we cannot “scientifically analyse” that difference in the audio signal (because there isn’t one).

It was always phrased as cannot be measured because EVERYONE commonly thinks of that as SINAD, Intermodulation, FFT etc measurement, except you.

Are you really claiming that NO ONE thinks the metered/measured amplitude of an audio signal is a measurement?

You decided to define a recording as a measurement.

Of course I didn’t, Shannon, Nyquist and others did, long before I was even born!

Your arguments are just getting ridiculous now!

G

 

[jagwap]

No, that is neither the stated myth nor the myth we’ve been fighting! The actual myth we’ve been fighting is that there are audible differences that cannot be measured. “Audible differences” is not “subjective listening” and “measurable” is not “scientifically analysable”.

Then put that as a sticky at the top of this forum. Because no one else separates them in common discussion. "There was an audible difference but I couldn't measure it" is the common discussion here, and while you have wilfully decided to take that as not true, because you can record it in order to deny this, without fully explanation that is what you mean until recently. People mean the type of measurement that analyses the signal, and if you didn't know that you are obtuse. If you knew that you are purposely obstinate and argumentative for your own amusement.

So, you’ve invented a statement that’s true but significantly different to the myth and then fallaciously claimed the myth is therefore true! That’s a non-sequitur.

So what is the myth? Define it clearly, because it has been taken as what I have said by all those you argue with. Your definitions do not always align with teh rest of the profession.

Incidentally, your statement (not the myth!) has been demonstrated true numerous times. Take for example the McGurk Effect, we can subjectively listen and hear a difference between “baa” and “faa” but we cannot “scientifically analyse” that difference in the audio signal (because there isn’t one).

So you having been fighting people here all these years, knowing you agree with them, but you are just taking them for a ride on your version of semantics.

Are you really claiming that NO ONE thinks the metered/measured amplitude of an audio signal is a measurement?

Of course not. Now you are putting words in to my mouth. That is covered by "etc."

Of course I didn’t, Shannon, Nyquist and others did, long before I was even born!

Your arguments are just getting ridiculous now!

G

All of this has been ridiculous. You are a hinderance.

I think I get where you are coming from on this forum, and it no better than Big Shot despite all the knowledge.

 

[castleofargh]

It is the "myth" the objectivist have been fighting here for years: that subjective listening can not hear things that cannot be scientifically analysed. It was always phrased as cannot be measured because EVERYONE commonly thinks of that as SINAD, Intermodulation, FFT etc measurement, except you. You decided to define a recording as a measurement. So we rephrase, to avoid, well, whatever.

So now, here it is. One of the key objectivists, agreeing that listening can reveal what analysis cannot (yet), after all these years of confronting those who suggested anything like it.

I kind of expected music, trumpets, fireworks.

A window opens:
"Sound Science Forum Level 2:
pedants corner"


castleofargh Can we pin this as a sticky?

There is a tendency to react strongly to ideas like ”we can’t measure everything” even though it’s just a fact, because we overwhelmingly read it from people who wish to discredit and disregard objective approaches, and doing so, remove accountability when they make empty claims.
If we’re discussing quantum physics and particles, nobody knowing anything would oppose the claim that we can’t measure everything.
For all the wrong reasons, context matters.

Because I’m slightly sadistic, I propose another direction:
Audibility is defined by controlled listening, not measurement. But isn’t a controlled test a measurement?

 

[gregorio]

Because no one else separates them in common discussion.

Again, that’s nonsense! A large percentage of the posts on this subforum are all about separating “audible differences” from “subjective listening”. On other forums on head-fi your assertion is more true, not least because any mention of the tests used to separate the two is forbidden. Your statement is false because many do separate them. So, your logic appears to rely on: I am wrong because a subset of audiophiles are ignorant (and wish to remain that way). How is that not ridiculous?

"There was an audible difference but I couldn't measure it" is the common discussion here,

No it’s not! It might be a common discussion but it’s not the common discussion or the myth being addressed. Why on earth would I disagree with yet another DIFFERENT made up statement that is (or could be) true? How do I know what that individual could or couldn’t measure? Maybe they don’t have a voltmeter or any sort of measuring device and therefore “couldn’t measure” anything at all. Obviously though, just because they don’t own a voltmeter (or don’t know how to use it) doesn’t mean that voltage differences are not measurable!

and while you have wilfully decided to take that as not true,

Of course I’ve wilfully decided to take the myth as not true. You think maybe I should willingly decide that the reliable, objective evidence is false while a myth without supporting reliable evidence is true?

because you can record it in order to deny this, without fully explanation that is what you mean until recently.

Digital audio data is a measurement (of amplitude over time). So:
1. How can people spend hours, days or even years discussing every aspect of reproducing digital audio data if they don’t even know what it is? And,
2. I’ve explained it countless times anyway and more than once in this thread!

So what is the myth?

What do you mean what is the myth, you’ve just quoted it and responded to it. Don’t you read what you’re responding to?

Of course not. Now you are putting words in to my mouth. That is covered by "etc."

Right, so you “of course” agree that a measurement of amplitude is a measurement and as digital audio is a measurement of amplitude, ergo recording something as digital audio MUST be a measurement. Therefore:

People mean the type of measurement that analyses the signal, and if you didn't know that you are obtuse.

Hang on, you can’t have it both ways. Do you and others, as you’ve just claimed, “of course” know that a measurement of amplitude is a “measurement” OR do you/they think it’s ONLY a “measurement” if it separates/isolates phenomena so they can be analysed? Which is it, it can’t be both! And if you didn’t know that, then you are deliberately obtuse!

All of this has been ridiculous. You are a hinderance.

I thought for a while you were trying to make a reasonable (though misguided) point but your last few ridiculous posts indicate you were probably just trolling, shame.

But thanks, yes, my aim is indeed to be a hindrance! A hindrance to the attempted justifications of false audiophile claims and myths!!

G

 

[Corti]

Audibility is defined by controlled listening, not measurement. But isn’t a controlled test a measurement?

i didnt read the discussion beforehand, just addressing your sadistic question

i would say a controlled test is a controlled test and it can include measurements but not necessarily needs to.

e.g. an apple falling to the ground. you can measure the physical impact it has on the ground.
but you can also do this as a controlled test where you dont measure, but you put different people under the falling apple and they report if it hurts when the apple falls onto their head.
there you dont measure anything but it would still be a controlled test because you control the parameters of the test. however the results do not rely on measurements but on the fact that the people have pain receptors and their ability to communicate what happens.

isnt this the main difference between empirical science like psychology and hard science like physics?
both are of important value and reasonable approaches to understand the world.

a thought about audibility vs. measurements in audio when someone claims to hear something, that could not be measured:
between the heard sound and the claim lies a long chain of different signal transformation (i.e. a physical wave impact on the eardrum, which will be transformed into action potentials and runs through long neuronal networks which will be processed by even larger and complex filter systems, which then leads to a certain behaviour that is controlled somewhere else in the brain.).
i think there are a lot of influences/manipulations along this transformation chain that may lead to a claim someone makes about hearing something even though it can not be physically measured (because there never was a sound-difference in the first place).
i dont think you can ever practically create two tests with the exact same conditions either (even though science relies on reproducibility).
but creating the same condition again for a machine that measures is way easier/more reliable than for a humanbeing with all these neuronal signal transformations and internal filters.

so yea, i think what im trying to say is that people can actually hear differences when detectors do not measure differences, just because the neuronal signal-transformation-chain is so long and blackbox-like and can change dramatically even within seconds, leading to additional/altered perceived information that was not inherent to the sound-wave.
what someone 'hears' is not the same what a device measures (i go by the definition that we dont hear sound-waves but we hear what our brains filtered and transformed so we can make sense of it. if the registration of sound-waves itself was already called 'hearing' then any machine that detects sound would also 'hear').
to create a device that measures what a human would perceive, the device would need to have a human neuronal circuit included, which is kinda absurd. but without this i dont see this discussion ever coming to an end.

oof, this post has gotten longer than initially intended.

 

[Prog Rock Man]

I had the same two headphones for over a decade, till both literally fell apart within in months of each other, resulting in the purchase of two new headphones in quick succession.

One purchase I am disappointed with, the other I am happy with.

One set of new headphones gives me goosebumps when I listen to certain tracks, the other does not. Both of my previous sets of headphones were capable of giving me goosebumps.

I don't care how well they measure, or how they are made. I now know that sound quality is measured by pleasure.

 

[bfreedma]

I had the same two headphones for over a decade, till both literally fell apart within in months of each other, resulting in the purchase of two new headphones in quick succession.

One purchase I am disappointed with, the other I am happy with.

One set of new headphones gives me goosebumps when I listen to certain tracks, the other does not. Both of my previous sets of headphones were capable of giving me goosebumps.

I don't care how well they measure, or how they are made. I now know that sound quality is measured by pleasure.

Sound preference can certainly be "measured by pleasure", but sound quality connotes a more objective set of measurable targets.

Or said another way, preference is variable by the individual listener, quality is a static target across all listeners