What would a sighted, subjective, non-level-matched, and otherwise uncontrolled observation tell us? Nothing.
Also, what does it matter that someone disagrees?
Basically, I remain unimpressed and do not see any evidence being offered.
I'll be the last one here on headfi to justified expensive headphones ring /equipment...but after 4-5 years of searching the best amp to match for my 702's I've got to the same result and conclusion as I start with. Some of as that have the chance to hear this ring can claim that they prefer more bass, or blabla,bla...but I guess nobody will argue about the K-702 and Phonitor combo in terms sound of resolution and transparency.
EDIT: Without getting into TMI maybe you should check this combo for yourself a report back...
This is actually interesting!
I've read some differing opinions on phase shift, from "absolute phase does not matter" to "we need to stop everything to fix this" - I tend to subscribe to the former school of thought, although I don't think it's nearly that cut and dry. Mostly because I agree with what you're saying about bandwidth.
The "point source" argument for speakers gets a bit more interesting - even a single driver will have phase shift (hence "absolute phase does not matter, because it can never be recreated" - remember that recording microphones don't have absolute phase either!), and I've never seen anything substantial on 'stats and phase (I assume they should behave linearly enough though; they're more or less universally championed by their designers as "linear and perfect" (this is more or less lifted from STAX, Koss, MartinLogan, Quad, etc marketing - it all sort of blurs together)). So I'm curious, is basically my point. See here: http://sound.westhost.com/amp-sound.htm Again, I'm getting a lot of the same "wonky exotic audiophile gear" as the problem child, and modern and competent devices as being comparable/interchangeable. 0.5 dB of shift is not a problem imho, but I'm sure some people see that on a spec and have a heart attack (again, there's more deviation from the output device!).
Noise is a bit more contentious - a lot of people like Bryston (I don't know hardly anything about Arcam), and the noise floor is always one of the first arguments made. Where I have an issue is when you're talking about a noise floor that's either too low to be audible, too low to be audible with conventional transducers, too low to be audible due to environmental noise, or too low to not be stepped on by the signal. In other words, we all dream about high bandwidth and low noise devices that reject more noise than a mom on Friday night, but that doesn't mean it would absolutely help us. And just because we can recognize this device to exist, at least in theory, doesn't mean the expensive machines are that. Finally, listening tests do not agree that more expensive or more fancy/overbuilt/better designed/etc amplifiers are inherently superior to to cheap ones; that's a commercial fallacy. Granted, all of those tests are done with loudspeakers, but it's not like we're jumping off the deep end into another realm when talking headphones. It's like uh, I have a fan in my listening room (which cannot be removed), and I always hear that fan; that fan is environmental noise above any device's noise floor - if it occurs in the same frequency range, it will mask it. Period. So who cares below that? Same for white noise on recordings - it will mask any white noise from the amplifier, decoder, whatever. Can't do a thing about it. Even if I killed the fan, my listening area is not a "quiet room" - the noise floor is not down there at 0 dB. Even if it was, now find me a recording that has a super duper low noise floor (say, 90 dB down). Now find me a few more, and make them something I want to listen to. I know, it's a bit glib and pragmatic, but it's to the point.
Sure, there's always going to be junk out there, but unless the thing is bleeding out noise or clipping, there's bigger fish to fry.
Years ago I read a scientific paper stating that the human ear can detect phase shift as low as +/- 15 degrees. Bandwitdth and phase shift are tied together. If you want less than +/- 15 degrees of phase shift from an audio amp then you want a very wide bandwidth amplifier.
It wasn't my research, I am only paraphrasing it. If I get a spare moment, I might try and track this paper, or a similar paper, down on the web.
Apparently one of the advantages of a properly designed eletrostatic loudspeaker is the reduction in phase shift from the speaker as it is a one way speaker, i.e. no separate woofer and tweeter with crossover components.
When I am referring to noise I am partially referring to noise that sees to be buried within the signal that the amp amy or may not add. Supposedly this is one of the secrets of the Bryston and Arcam sound, they have done a lot to reduce various sources of noise but neither company does much to properly explain what they are doing. One reason might be that it would take a 50 page scientific paper to explain it all. How do you explain radiated and conducted noise immunity and emissions to a layman?