Message to golden-eared audiophiles posting at ASR for the first time...

[MattHooper]

Again…

It’s perfectly fine to say:

“ this is the type of evidence I am looking for.”

And it’s perfectly fine if you want to describe what counts as scientific evidence.

But that is ONE version of the nature or role of evidence; it does not comprise the entirety of what is often understood and rationally accepted as “ evidence.”

I already gave the Wikipedia link on this.

We reasonably accept testimony as evidence in all sorts of disciplines and situations. In fact, we are doing it all the time in our daily lives. We are making judgement calls as to whether to accept anything as evidence, other peoples testimony included, and this can be incorporated in our sliding confidence scales.

Even in science, evidence often operates on a spectrum of confidence rather than on absolute certainty. What counts as “enough” evidence is not always a purely objective matter. It often involves human judgment, prior beliefs, methodological preferences…even social or institutional norms.
That’s why scientists often dispute among one another whether any particular paper or study has met an evidential burden.

There is no reason why audio is excepted from the norms of sliding scales of evidence.
Because again evidence has to do with rationally justifying beliefs, and increasing probabilities. There is no objective cut off point for evidence - it is always a judgement call. And one can always play the “ that’s not enough evidence game.”

In normal life, if my neighbour tells me he just bought a 4K flatscreen TV, and my neighbour has a history of being truthful, then it is reasonable to take his testimony as raising the probabilities that he now owns a 4K tv.
That is how it counts as evidence. It may not be as strong evidence as actually seeing the 4K TV, but it is evidence.

So for instance, in terms of audio:

Erin (Erin’s audio corner) has shown to be quite perceptive in terms of picking up on frequency response nonlinearity, dispersion characteristics etc. during his subjective listening to loudspeakers. Most often he detects issues that show up in the measurements.

So if Erin tells me that in listening to a speaker and it seem to be lacking somewhere in the 80-150hz range, lacking some punch down there, and also had boosted high frequencies which led to exaggerated vocal sibilants (eg somewhere between 5-8k), then that to me would count as “ evidence” for the character of those speakers. It is certainly not absolute proof of any kind, but given Erin’s track record, it DOES rationally raise the probabilities for the proposition that the speaker has that character. Even before he presents me measurements.

And there are any number of other examples in audio where anecdote/personal testimony can be rationally accepted as evidence, even if it does not produce scientific levels of confidence. And as I said, the heuristic for accepting testimony is going to be along the lines “ extraordinary claims require extraordinary evidence.” The more implausible an audio claim, the less rational it is to provisionally accept testimony, as well as the inverse. In this thread we are generally addressing implausible claims, but not all audio claims are equally implausible.

It doesn’t do anybody any good to ignore that evidence works this way.

Even with audio, we have to be careful of going too far in suggesting that anything outside of a scientific or engineering standard can’t count rationally as “ evidence.” You really go down an indefensible rabbit hole with that stance.

And that’s the type of thing that is often suggested implicitly or explicitly here, and which can be also implied in saying things like “ anecdotal evidence = oxymoron.”

 

[pma]

However, what we're discussing here is music reproduction. That is something measurable, and therefore objective. Once a music reproduction system reaches a certain threshold in terms of noise, distortion, and tonality, it becomes transparent

The important word in your post is the system. Remember, this is something that is permanently avoided here. Single components are measured, not the complete systems. At today’s state of technology, the system errors dominate - ground loop noises and interference, buzz, HF EMI and it remains undisclosed in the single components tests. I do not like black & white views - “measurements” vs. “listening”. Both is equally important. Sometimes listening hints us to systematic errors.

 

[MattHooper]

tm is not the only one here trying to muddy clear waters with inappropriate interpretations of terms.

There has been no inappropriate interpretation of terms on my behalf. There has been clarification of the interpretation of terms.

I have acknowledged that a scientific version of evidence is perfectly appropriate in regards to vetting claims about audio gear.

I have simply clarified that it is not the ONLY framework that can constitute evidence, since the very nature of evidence constitutes a continuum.

This is not muddying any waters: It justifies anybody seeking scientific (or engineering) confidence levels, while being careful to not throw the baby out with the bathwater, and take scepticism too far.

Language like “ anecdotal evidence = oxymoron” is actually more apt to muddy the waters about evidence.

In some cases it is incessant and borders on mischievous. I think of that incessant type of obfuscation as trolling, and Rick is encouraging us to report trolling....hmm. I hesitate.

If you think that my previous comments have constituted trolling… why not go ahead and report them? Go for it! I’d be happy to see if the mods agree with you in terms of what constitutes trolling.

(Because I have plenty of confidence that they are not that dogmatic, and can recognize when somebody is bringing an honest point of view to the thread)

 

[Chagall]

The important word in your post is the system. Remember, this is something that is permanently avoided here. Single components are measured, not the complete systems. At today’s state of technology, the system errors dominate - ground loop noises and interference, buzz, HF EMI and it remains undisclosed in the single components tests. I do not like black & white views - “measurements” vs. “listening”. Both is equally important. Sometimes listening hints us to systematic errors.

You’re bringing up implementation issues, and you’re right, but that’s not what I was addressing. I was speaking more generally about the distinction between music as art (subjective) and music reproduction (measurable and objective). My point wasn’t to dismiss listening but to differentiate between subjective art and objective reproduction. Does that clarify where I’m coming from?

 

[DVinylfan]

I love ASR because nowhere else on the internet can such committed debate on subjective (felt) versus objective (measured) measurements be found.
And it's important because we sense our enjoyment of music and gear ought to have some underlying theoretical basis upon which to rest.
But, a reminder: objective and subjective are false bedfellows, and many errors and misunderstandings arise from the false binary of placing them alongside each other contextlessly.
It's a longer thread, but briefly, historically, objectivity and subjectivity belong to two different philosophical traditions, and comparing them as if they were in competition and one had the possibility of outperforming the other in terms of value or accuracy is a beyond futile exercise.
This website's strength lies in its focus on scientific and empirical measurements of audio gear and the discussions arising thereof.
Its weakness lies in arguments that stray into oppositionality on false bases.

 

[tmtomh]

Again…

It’s perfectly fine to say:

“ this is the type of evidence I am looking for.”

And it’s perfectly fine if you want to describe what counts as scientific evidence.

But that is ONE version of the nature or role of evidence; it does not comprise the entirety of what is often understood and rationally accepted as “ evidence.”

I already gave the Wikipedia link on this.

We reasonably accept testimony as evidence in all sorts of disciplines and situations. In fact, we are doing it all the time in our daily lives. We are making judgement calls as to whether to accept anything as evidence, other peoples testimony included, and this can be incorporated in our sliding confidence scales.

Even in science, evidence often operates on a spectrum of confidence rather than on absolute certainty. What counts as “enough” evidence is not always a purely objective matter. It often involves human judgment, prior beliefs, methodological preferences…even social or institutional norms.
That’s why scientists often dispute among one another whether any particular paper or study has met an evidential burden.

There is no reason why audio is excepted from the norms of sliding scales of evidence.
Because again evidence has to do with rationally justifying beliefs, and increasing probabilities. There is no objective cut off point for evidence - it is always a judgement call. And one can always play the “ that’s not enough evidence game.”

In normal life, if my neighbour tells me he just bought a 4K flatscreen TV, and my neighbour has a history of being truthful, then it is reasonable to take his testimony as raising the probabilities that he now owns a 4K tv.
That is how it counts as evidence. It may not be as strong evidence as actually seeing the 4K TV, but it is evidence.

So for instance, in terms of audio:

Erin (Erin’s audio corner) has shown to be quite perceptive in terms of picking up on frequency response nonlinearity, dispersion characteristics etc. during his subjective listening to loudspeakers. Most often he detects issues that show up in the measurements.

So if Erin tells me that in listening to a speaker and it seem to be lacking somewhere in the 80-150hz range, lacking some punch down there, and also had boosted high frequencies which led to exaggerated vocal sibilants (eg somewhere between 5-8k), then that to me would count as “ evidence” for the character of those speakers. It is certainly not absolute proof of any kind, but given Erin’s track record, it DOES rationally raise the probabilities for the proposition that the speaker has that character. Even before he presents me measurements.

And there are any number of other examples in audio where anecdote/personal testimony can be rationally accepted as evidence, even if it does not produce scientific levels of confidence. And as I said, the heuristic for accepting testimony is going to be along the lines “ extraordinary claims require extraordinary evidence.” The more implausible an audio claim, the less rational it is to provisionally accept testimony, as well as the inverse. In this thread we are generally addressing implausible claims, but not all audio claims are equally implausible.

It doesn’t do anybody any good to ignore that evidence works this way.

Even with audio, we have to be careful of going too far in suggesting that anything outside of a scientific or engineering standard can’t count rationally as “ evidence.” You really go down an indefensible rabbit hole with that stance.

And that’s the type of thing that is often suggested implicitly or explicitly here, and which can be also implied in saying things like “ anecdotal evidence = oxymoron.”

I take your point... to a point. Personally I think there's not really much benefit in going further down the road of arguing about whether anecdotal evidence is (a) a qualitatively different (and more limited and less robust) type of evidence or (b) not evidence at all. For the purposes of the issue being discussed at this point in the thread, it doesn't really matter, because either way "I've heard AC cords making a sonic difference" is not evidence that AC cords actually make a sonic difference. You, I, @j_j , @Newman , and most other ASR members are all in agreement about that.

By the same token, though, in your comment that I've quoted here, I do think you're doing something quite close to what Newman is complaining about: testimony is a qualitatively different type of evidence than objective empirical data, and testimony is far less reliable and far more limited in what it can be evidence of.

Yet in the parts of your comment that I've bolded above, I think you're running together two qualitatively different kinds of evidence - testimonial and empirical/objective - by claiming both are subject to probabilities and human cognition. I think that's deeply misleading and verges on a kind of unintentional hand-waving that obscures the key distinctions. The question of whether a 95% confidence interval is necessary (or, conversely, adequate) is a different kind of question than whether we should have confidence in someone's testimony.

Along those lines, I don't see Erin's subjective listening impressions as having any intrinsic value as testimony - and neither does Erin, because he always presents those impressions as part of a two-part approach with his Klippel measurements (and other associated measurements). His "testimony" is always a hypothesis about the connection between measured performance and his personal impressions. The reason we might consider him reliable - and I share your view that he is - is that his hypotheses are always plausible.

That's totally different than the claim that started this part of this thread: "I hear AC cords making a sonic difference and lots of other people do too, so it must be true." That's testimony in the religious sense: it's a statement of faith, that is "belief in things unseen"; a sonic, physical, electronic mechanism beyond measurement or rational explanation that is said to exist simply based on personal experience.

 

[MattHooper]

Nothing I’ve said undermines the scientific and engineering approach taken by anybody on ASR to audio.

It would only undermine a dogmatic overreach in terms of claims about the nature of evidence proper. And this applies to overreach talking about evidence in terms of audio as with anything else.

By the same token, though, in your comment that I've quoted here, I do think you're doing something quite close to what Newman is complaining about: testimony is a qualitatively different type of evidence than objective empirical data, and testimony is far less reliable and far more limited in what it can be evidence of.

Sure, as I wrote: Generally speaking testimony is going to make for lower confidence levels. So I’m not at odds on that.

That’s why I’ve pointed out that evidence is linked to sliding confidence levels, not some absolute. Testimonial evidence isn’t going to give scientific confidence levels.

Yet in the parts of your comment that I've bolded above, I think you're running together two qualitatively different kinds of evidence - testimonial and empirical/objective - by claiming both are subject to probabilities and human cognition.

Both empirical/objective data and testimonial evidence work along sliding scales of confidence levels, whether formally or informally. We are always making judgements as to when to rationally accept a proposition based on the evidence supplied.

And the reality is, we - certainly how we operate here on ASR - cannot always neatly separate objective evidence from testimony.

So let’s say someone reported a change in bass quality when they try a certain footer underneath their loudspeaker.

ASR members reply: “ show us the evidence we want empirical measurements not just your testimony!”

And let’s say the person produced graphs of in-room measurements showing a slight likely audible change in the bass measurements before and after impinging the footers.

The first thing to point out is that to accept the measurements in the first place, you are implicitly accepting the testimony of the person presenting them; that they really did do these measurements, they are authentic and not faked somehow, or they are aren’t measurements of something else but being misrepresented as changes in the footers.

So you’re already implicitly accepting the persons testimony - “ I took these measurements of the phenomenon I’m claiming “ - as reasonable grounds to accept the measurements.

Even though objective data is involved, there’s already an element of having to accept testimony. Further, the presence of objective data doesn’t determine the issue itself. There’s an element of judgement involved as to whether the data is sufficient evidence for the claim.

Some here may be more sceptical that a footer would change the sound in the way indicated by the measurements, and decide “ not good enough maybe they made a mistake in the measurements. I want to see more about how the measurements were done or I want to see that these can be replicated.” Other people may accept the data as evidence.

Then imagine someone arrives on the forum claiming they hear big differences between different USB cables in their digital set up and they produce what they claim to be in-room measurements of speaker response with a different USB cables. The measurements show very obvious audible deviations in the frequency response.

Should those measurements automatically be accepted as sufficient evidence?

In this case, I would think: no. it’s going to be adjudicated against the background knowledge and beliefs of the person looking at the measurements. Given how highly implausible it is for USB cables to affect the sound in that way, plenty would find it reasonable to say “ It’s more likely a measurement/ experimentor error.” They may demand more information, a different test, replication of the test by others, even being able to replicate the test themselves, perhaps.

So what counts as “ evidence” is generally going to be a judgement call as to when any
“evidence” is sufficient to warrant belief in the proposition. In deciding which data we accept as evidence , there can still be an element of extraordinary claims require extraordinary evidence, of the type we use for less formal evidence like testimony.

In crime cases, there’s all sorts of objective evidence presented. But very often just one bit of objective evidence is not seen to Justify a conviction. One bit of evidence can simply start to raise confidence levels or probabilities, and additional evidence keeps raising them until there’s a point where judgement is made by a jury, whether it constitutes sufficient evidence or not.
Testimony is mixed in there as well as evidence raising probabilities.

Nothing about what I’m arguing undermines the approach taken by anybody at ASR; eschewing the weaker form of testimonial evidence for more objective evidence is perfectly reasonable.

Where you get into trouble is in the move to “Therefore, it is unreasonable for anyone to think anecdote/testimony constitutes evidence.”

That just becomes an untenable stance on the nature of evidence, including as it applies to audio gear. And yet there are those who make those type of moves on ASR.

Along those lines, I don't see Erin's subjective listening impressions as having any intrinsic value as testimony - and neither does Erin,

That’s fine and defensible.

However, it depends what you are saying in terms of no intrinsic value. If you mean that Erin thinks that subjective impressions aren’t always reliable, and also it’s better to have objective measurements to correlate what somebody is hearing to the measurements, sure. In that sense Erin ideally wants measurements involved.

But if you mean something stronger - that Erin doesn’t think his subjective impressions have any evidential value at all, then that doesn’t seem to be the case.

Take a look at plenty of his reviews, his Borresen x3 review for instance, and it’s very clear that during his subjective listening process he is taking some of what he perceives to be veridical - evidence of dips in the midrange and highs, boost in the midbass, playing with speaker positions to try and ameliorate the issues, phoning the manufacturer to ask about positioning, etc. There’s a difference between Erin being an engineer, who wants to correlate what he hears to measurements, and the suggestion that Erin does not see his subjective impressions as evidential at all.

I mean, after all it’s common even on ASR to take our subjective listening impressions as evidence to initiate some work on our system: If someone is noticing an obvious annoying emphasis in the bass, that can initiate bringing in room treatment or investigating the problem with measurements room correction, etc.

The reason we might consider him reliable - and I share your view that he is - is that his hypotheses are always plausible.

Yes, as I’ve argued plausibility plays a very big part in accepting testimony, but in this case so does a track record of listening accuracy on Erin’s part.

That’s why I find it reasonable to accept Erin’s subjective reports as a form of evidence - it plays the role of raising the probabilities that the speakers in question DOES exhibit that behaviour, and that we will likely find that behaviour in the objective measurements.

That’s the nature of evidential reasoning. One piece of evidence can start to raise probabilities - Erin reporting an impression. Add some more - in this case Erin’s measurements - and your confidence levels can scale up. You can always decide for yourself to put no stock in Erin’s subjective reports. But in the big picture, evidential reasoning is not usually some absolute all or nothing scenario.

Cheers

 

[Mart68]

The strength of the evidence required surely depends on the nature of the claim it is supposed to support?

The claim 'I listened to two loudspeakers and they sound different' probably does not require any supporting evidence.

The claim 'I listened to two power cables and they sound different' would require a high level of supporting evidence.

And then you have everything in between those extremes.

.

 

[Newman]

The strength of the evidence required surely depends on the nature of the claim it is supposed to support?

The claim 'I listened to two loudspeakers and they sound different' probably does not require any supporting evidence.

I disagree. It is trivially easy (and widely confirmed via reports on personal experiences) that people can say how different two loudspeakers sound, then discover that it was actually the same loudspeaker both times, and they had not switched over.

The opposite also happens: Toole reports on sighted listening tests where people give the same ratings to two loudspeakers that look identical (same badge and model ID) but are internally different to a clearly audible degree.

The claim 'I listened to two power cables and they sound different' would require a high level of supporting evidence.

And then you have everything in between those extremes.

Except they aren't actually all that extremely different.

The conclusion I draw is that proper listening controls are more or less essential for any report on the sound of the sound waves themselves.

cheers

(PS I do get the generic point you wanted to make, but your first example leans towards an implicit acceptance of sighted listening reports on loudspeakers, which is IMHO a problematic direction to take. I mean yes, in extremis, "I listened to a subwoofer and an iphone and they sounded different" doesn't require support, and that's your point, but we can't rationally extend it to any meaningful comparison where the difference may or may not be clear, and still waive the need for supporting evidence. It's not even worth trying to find the point where evidence requirements start to escalate, because the sighted listening effect is strong enough that, for practical purposes of evaluating sonic attributes of gear, we really need to see the evidence that the listening test was controlled.)

 

[sergeauckland]

The strength of the evidence required surely depends on the nature of the claim it is supposed to support?

The claim 'I listened to two loudspeakers and they sound different' probably does not require any supporting evidence.

The claim 'I listened to two power cables and they sound different' would require a high level of supporting evidence.

And then you have everything in between those extremes.

.

This comes down yet again to Carl Sagan' s aphorism that Extraordinary claims require extraordinary evidence. That loudspeakers sound different isn't extraordinary, that power cables do is.

S

 

[Mart68]

I disagree. It is trivially easy (and widely confirmed via reports on personal experiences) that people can say how different two loudspeakers sound, then discover that it was actually the same loudspeaker both times, and they had not switched over.

The opposite also happens: Toole reports on sighted listening tests where people give the same ratings to two loudspeakers that look identical (same badge and model ID) but are internally different to a clearly audible degree.


Except they aren't actually all that extremely different.

The conclusion I draw is that proper listening controls are more or less essential for any report on the sound of the sound waves themselves.

cheers

(PS I do get the generic point you wanted to make, but your first example leans towards an implicit acceptance of sighted listening reports on loudspeakers, which is IMHO a problematic direction to take. I mean yes, in extremis, "I listened to a subwoofer and an iphone and they sounded different" doesn't require support, and that's your point, but we can't rationally extend it to any meaningful comparison where the difference may or may not be clear, and still waive the need for supporting evidence. It's not even worth trying to find the point where evidence requirements start to escalate, because the sighted listening effect is strong enough that, for practical purposes of evaluating sonic attributes of gear, we really need to see the evidence that the listening test was controlled.)

Yes I agree. For sake of brevity I did not forensically examine the various permutations. I just wanted to make the point that the degree of evidence required is a sliding scale that is directly related to the claim that's being made.

 

[GabrielPhoto]

I honestly dont understand how this thread is so long because this is rather simple.
If you CLAIM to be able to hear it, then you should be able to easily prove it with measurements or a proper blind test. If not, then is all in your head and no amount of sugar coating can change the fact.
I see some attempts at making up excuses and alternatives but that is just absurd.
It is that simple...measurements or blind tests...
Anything else is just a cute anecdote

 

[antcollinet]

I take it you know exactly how uncommon this mythical "common sense" really is.

As I keep saying:

The problem with humanity** is that half of the people are below average intelligence.




** to be honest I normally say "democracy" rather than "humanity" But not here.

 

[MattHooper]

The strength of the evidence required surely depends on the nature of the claim it is supposed to support?

The claim 'I listened to two loudspeakers and they sound different' probably does not require any supporting evidence.

The claim 'I listened to two power cables and they sound different' would require a high level of supporting evidence.

And then you have everything in between those extremes.

.

Exactly!

Though I would add the caveat:

When it comes to the example of a testimony of hearing differences loudspeakers, to say that “ it doesn’t require any more supporting evidence”… it’s going to depend on the exact nature of the claim, and the type of evidence being sought.

Depending on the claim, it can be acceptable in the informal lower confidence levels version - of the type many on ASR clearly accept. It can be a very plausible claim.

On the other hand, it’s not going to satisfy the search for scientific confidence levels, because it’s not controlling for the variables of sighted bias.

And as I say, when we don’t have scientific confidence controls available to us for a particular claim, we can navigate the reasonableness of a testimony using a heuristic like “ extraordinary claims require extraordinary evidence.”

This comes down yet again to Carl Sagan' s aphorism that Extraordinary claims require extraordinary evidence. That loudspeakers sound different isn't extraordinary, that power cables do is.

Exactly.

We can still be wrong even about what we hear between loudspeakers. If we want to raise our confidence levels we ought to employ controls for sighted bias.
But in the absence of such controls, we can still navigate the evidential quality of testimony based on how reasonable or plausible the report is, and we can accept such testimony sometimes with the obvious caveats.

 

[Mort]

I honestly dont understand how this thread is so long because this is rather simple.
If you CLAIM to be able to hear it, then you should be able to easily prove it with measurements or a proper blind test. If not, then is all in your head and no amount of sugar coating can change the fact.

Many many individual new threads are moved here by admins so we don't have a thousand threads basically with the same content.

 

[Newman]

I honestly dont understand how this thread is so long because this is rather simple.
If you CLAIM to be able to hear it, then you should be able to easily prove it with measurements ...

Although many measured differences are inaudible and don't support a claim that one can hear it. "Proving it with measurements" is a two-step process: show a measured difference, and show that such a difference is above commonly-agreed hearing thresholds.

...or a proper blind test. If not, then is all in your head and no amount of sugar coating can change the fact.
I see some attempts at making up excuses and alternatives but that is just absurd.
It is that simple...measurements or blind tests...
Anything else is just a cute anecdote

Some anecdotes in the audio press are so cute they should be limericks.

cheers

 

[GabrielPhoto]

Although many measured differences are inaudible and don't support a claim that one can hear it. "Proving it with measurements" is a two-step process: show a measured difference, and show that such a difference is above commonly-agreed hearing thresholds.


Some anecdotes in the audio press are so cute they should be limericks.

cheers

Indeed for me nothing beats a blind test because of that very same reason. Just because say, a DAC measures SINAD 115 and the other does 120, means little if either is way above human capabilities

 

[MattHooper]

Indeed for me nothing beats a blind test because of that very same reason. Just because say, a DAC measures SINAD 115 and the other does 120, means little if either is way above human capabilities

Yup. This is why I did my own blind tests when I had several CDPs/DACs.

Right now, I am in dialogue with a golden eared Positive Feedback Magazine reviewer about his rave review of the mind bendingly expensive dCS Varese digital front end. “ new level reached in digital playback, uncovers musical details previously buried in the noise floor by other DACs” and all that.

I’m among those in the same thread pointing out these are dubious claims on any number of fronts, including what we know of already state of the art DAC performance, the limits of human perception, the fact no measurements have been presented that indicate any likely audible new information is being uncovered, the fact that nobody has presented any listening tests, controlling for bias demonstrating the claims etc.

The reviewer has retorted (not uncommonly among Golden ears) that he is a scientist himself in the life sciences, and that the differences he was hearing were so obvious they don’t require blind testing.

As you can imagine, I’ve been explaining that’s not how things work:

HERE

Among other things, I asked him to imagine a researcher in a study for a new medical treatment opining “ this medicine has such obvious effects we don’t need a placebo control.”

It’s always demonstrative of how pseudoscience can flourish even under the nose of scientists, as soon as they have stepped out of their lab and abandoned strict controls in order to believe what they want for their pet hobby.

 

[GabrielPhoto]

that the differences he was hearing were so obvious they don’t require blind testing.

That is so hilarious

Gotta love them. LOL

 

[MattHooper]

Good Gosh, I am compelled to post this here because of the subject of this thread, Golden Eared audiophiles and their attitude towards measurements and controlled, listening tests.

I’m trying to impress the relevance of measurements and blind testing on the “ scientist” who did the positive feedback review of the Varèse. Especially when as in this case the very claim about a large audible advantage for the Varèse over a lower priced competently designed DAC is under dispute.

And this is the type of reply I’m getting from him:

“Double blind from life sciences, where it originated, refers to both the clinical staff and the patient being blind as to whether the test article is a real drug or a placebo--the pills appear identical. This is readily transferable when we talk about something like cables, which have subtle effects and could go either way. But when the listener is experienced, and we're talking about big things such as tubes vs SS, OG vinyl vs modern digital, or Varese vs a normal, it's silly to think the listener remains blinded during the study--there's no way that the "pills" will appear identical during the test..

So not saying I can test perfectly well sighted, I'm saying that real blinding is impossible if a person has experience with one of the components being tested and the difference is significant.”

So…if the person claims sonic difference is really obvious, then blind testing is no longer applicable. Because it would be so obvious, they would know what they are listening to.
How to deal with this level of question-begging?