Version: Firmware 13.1.3 + DOS 2.2.4 Date: 2019-02-12

[David A]

@David A , with respect, I think you're missing my point.  At the risk of taking this thread further off topic...

It's possible that terminology is the problem here.  I thought we were all using the word "difference" to mean "a noticeable change in sound quality"; apologies for seeming obtuse if that's not what you meant.

Since sound quality is purely subjective, any report of sound quality difference is unique to the individual(s) who did the listening.  Therefore if they hear a difference, or do not, then the (subjective) difference is unarguably there or not, as the case may be.  (Unless they are deliberately mis-reporting, hence the caveat in my previous post -- not that I think there's any of that going on here!)

One person's subjective experience doesn't give any significant information about whether anyone else may or may not hear a difference, or indeed whether there is an objective -- that is, measurable in principle, even if perhaps not in practice -- difference in output from the system (say, at the point where sound waves enters the listener's ear).

Of course if you have enough listeners independently and repeatably reporting a sound quality difference, that starts to build evidence that there may be an objective change in output of the system, but in principle listening tests can never answer that question with certainty.  Conversely there may be an objective change in output that no-one ever reports hearing.

In any case, there is no "right" or "wrong" and nothing to "prove" either way about whether or not any given listener hears a difference, since the experience is personal and subjective.  In particular there is no conflict or mystery to be resolved if some listeners do and some do not hear a difference, whether or not there is an objective change in output

We're definitely seeing things differently.
 
I make a distinction between a difference in sound and a difference in sound quality. No one would argue that we can't hear differences between vinyl and CD playback (surface noise, anyone?) but someone can listen to both, hear the difference, and say that though they are different they think the sound quality is identical/similar. A difference in sound doesn't mean that the listener will hear a difference in sound quality but if someone says they hear a difference in sound quality then they're definitely hearing a difference in sound because there can't be a difference in sound quality if there is no difference in sound while 2 different sounds can be regarded as equally good or bad if their different strengths and weaknesses balance out.

If we're talking tests then we have to be talking about tests for differences in the sound. Sound quality is subjective as you say and we can't test for subjective differences. We can test for differences in sound and we can try to relate any differences we find to people's subjective reports of sound quality but we can't test for differences in sound quality, we can just gather and analyse the reports people make.

And there is something to prove in proving that people can hear a difference. First, it shows that an actual measurable difference is present and then, since not all measurable differences are audible, proving that people can hear it shows that the difference is not only actual but audible and that stops all of the arguments over whether or not people can really hear what they're claiming to hear when they say there is a difference. Proving the existence of an actual difference proves which statement made about whether or not there is a difference is right and which is wrong, and proving that people can  hear a difference means that the subjective impressions of listeners have value when it comes to deciding which you want to use.

And if you don't think that analysis of subjective impressions and of what people prefer have value, then just look at any electronic room correction system using DSP to tailor the sound to a particular target curve because almost without exception the target curve being used will be based in part on studies of listener preferences. Preferences are certainly individual and they aren't right or wrong, but they do have value and knowledge of the preferences of a large enough group of people can be and is used to make products work better so data on preferences definitely has value.

I definitely agree we're both looking at this from very different perspectives.