Again, what is the point of a listening session if you do not consider subjective impressions to be reliable anyway, as they are subject to expectation bias and other issues ? Or does that only matter for ABX tests (= double standards) ?
Well, like I said, the point is for him to experience it himself. I would still take his subjective impressions with a grain of healthy salt, of course, and his impressions and opinions do not necessarily prove or disprove anything. Basically, it does nothing aside from gaining some experience for him.
Amplifiers can and already have been measured with real loads. Most of the time, increased distortion and other issues occur because of too high output impedance, the load having much lower than nominal impedance at some frequencies, or poor amplifier design (such as instability due to cable capacitance). All the other factors you have listed can be taken into account with more detailed measurements.
Yeah, I have seen some measured that way, too, but those measurements are typically not published to prospective buyers, and just like I demonstrated above, the published specs of the O2 do not necessarily reflect its full performance.
In any case, if you consider both measurements and ABX tests inherently unreliable, then what do you suggest for evaluating gear ?
Aside from trying the gears yourself, and making your own decision, I don't think there is any other way currently. That's why I try to base my purchase decisions less on measurements and more on my own impressions, although it has been the case that when something measures really really well (past the point of "audibility"), that I do find it subjectively "better".
So, the fact that it is possible to do something (in this case, an ABX test) incorrectly makes it entirely unreliable and irrelevant ? Interesting logic, and it could be applied to just about everything in life.
Well, but that's my entire point. An ABX test that someone did should not be a yard stick to measure whether or not a piece of gear can perform well. There are simply way too many variables.
Similarly, what someone has said about a certain piece of gear, without ABX or any scientific test, should also be taken with a healthy grain of salt.
Basically, you are the only one who can decide whether a piece of gear is for you, and that "truth" will only apply to you.
It is, too bad their claims are quite often presented as facts. For those who do not agree with them, it is also their business to disagree, and explain why.
I'd agree that some of them like to present their opinions as fact, but that's not the case for all of them. And either case, as I said above, I wouldn't trust them either way.
For those who do not agree, they can definitely come up with some measure to gauge their distrust (in this case, ABX), but that does not necessarily disprove the claim, nor does it really prove the claim if it works the other way.
Due to the nature of this "inconclusive" result, I think, again, that it is best for a person to just try whatever it is that was claimed themselves, and then if it does not apply, then that's that. The claim is invalid for that person.
And I have seen unicorns the last night, but unfortunately there were no witnesses and the pictures I have taken were accidentally deleted.
I do have another witness, though. Seems someone else has also seen this website and talked about its disappearance in 2011:
http://www.audioasylum.com/forums/MUG/messages/17/171099.html
I would just attribute it more to personal reasons (perhaps the webmaster decided to move on and do something else, or they failed to maintain the website), rather than some conspiracy, but either way, the website did exist at one point or another. Whether you would like to believe me or not is really your choice.
It depends on the definition of "miniscule". A 1 dB frequency response difference might be described as such for a headphone, but it can actually be enough to be audible. Also, since the burned-in and fresh K701 could not have been the same pair, it is a quite possible explanation that there was an audible difference due to simple random variation between the samples.
Unless Tyll somehow got a hold of materials that had contents above 10KHz, and they were of such a significant amplitude that they become audible, I'd think the differences should be more in the midrange area, where he measured the differences to be around 0.5dB or under, and that's what I meant by "minuscule".
And either way, the point is that there exist people who can hear differences, right?
No one claimed the O2 to be infallible under any conditions, so that is a straw man. The IEMs used in that measurement also have significantly lower impedance at some frequencies than the minimum recommended by the designer (16 ohms), so in this case the amplifier is basically used with a load that is outside its specifications, just like - as more extreme examples - electrostatic headphones and speakers would be.
Its author claimed and insisted that the amplifier is "infallible" for most headphones. I'm simply repeating his statements. Also, he did not recommend minimum 16 Ohm load impedance. He mentioned that his goal was to get the output impedance under 2 Ohm in order to achieve good damping factor with 16 Ohm with a "rule of thumb". Since his design ultimately ended up having an output impedance of 0.54 Ohm, he remarked that he exceeded the goal, and proceeded to post measurements at 15 Ohm load, showing the amp to have "excellent" performance there against its competition.
And either way, a 12 Ohm load should not be outside of the amp's specifications if its output impedance is truly 0.54 Ohm, and the amp was to perform like it should.