in my short experience out of phase audio yields a suffered low end a weird sounding mid range.
But in this scenario, as I understand it, out of phase relative to the other audio involved in recording/playback is fundamentally different than inverted but relative phase. Not sure I understand the parallel you're drawing...?
I really don't know how that could have went three years without being discovered.
I think it took 3 years to identify precisely because it is not audible (at least not to me). When Grace first reported that the first 100 V3's did not have a DC offset of 0, some people completely freaked out. One of the explicitly stated reasons by Mike Grace that they didn't catch it is because
the DC offset simply wasn't audible. (Still, of course, they offered to fix all the boxes affected under warranty.)
Anyway...I've been thinking about phase within the context of 2 channel recording, and my initial thoughts suggest that it doesn't (or shouldn't) matter, that what's important is relative phase across the two channels - and that's why I don't hear a difference when ABX-ing a 2-ch recording with a particular phase v. inverted phase. Lemme test my thoughts on everyone...
We have a speaker. A + voltage pushes the cone outwards, a - voltage pulls the cone inwards. The result, in a perfect world, is a single-cycle sine wave of given amplitude. In reality, the waveform propagates in all three dimensions and is much more complex, but for the sake of simplicity in discussion, let's look at this sine wave in two dimensions. For this two dimensional waveform, relative to the horizontal plane on which my speaker rests the + side of the sine wave arcs upwards and the - side of the sine wave arcs downwards. At some defined point, X, from the speaker, the waveform reaches my ears. Let's assume that at this point, the + arc reaches my ears first, followed by the - arc. Make sense?
Now, two modifications to the above scenario:
<01> Change my listening orientation at point X
For example, hanging upside down with my head in precisely the same location, X, as above. Now, the waveform reaching my ears is effectively inverted relative to to the previous scenario. The + arc still reaches my ears first, followed by the - arc, but the orientation of the waveform relative to my ears is 180º inverted. Does this fundamentally change the way I hear the waveform? Since our listening position is never precisely the same across any period of listening time (unless you're strapped to a body board, with a neck brace, restricting all movement), our listening orientation impacts the phase throughout our listening experience.
<02> Change my listening position from point X to point Y
Let's assume point Y is forward of my current listening position by exactly 1/2 of the waveform cycle. The result is that now the - arc of the waveform reaches my ears first, followed by the + arc. In the more complex real world, different frequencies will reach our ears with different phase - some frequencies' - arc will reach our ears first, other frequencies' + arc will reach our ears first, and still others somewhere in between. So does this phase offset fundamentally change the way I hear the waveform? Maybe in the most absolutely controlled of playback environments, but even for most very high end playback systems I don't see how phase inversion would matter to the listener.
It seems to me that if we extrapolate this very two dimensional model into three dimensional reality, phase inversion will occur as a result of either or both of the above scenarios, and that the phase inversion (and time arrival offset, for that matter) makes no difference, or at least is no more or less important than one's ears' orientation (in all three dimensions) relative to the speaker.
Thoughts?