Don't think about what effects are used in music. Our hearing doesn't detect reverb plugins. It detects ILD/ITD + spectral cues in time. Only care about what kind of ILD/ITD -information reaches your ears. Is it within "human hearing space" or not?
This is patently incorrect. If it were correct, reverb would be undetectable in mono (single speaker placed centrally), where there is effectively little or no ILD/ITD, which is clearly NOT the case. I'm not sure what you mean by "human hearing space" so I cannot respond to your question.
Our everyday sound world is more monophonic than people realise, closer to mono than ping pong stereo.
Again, no it is not. Our everyday world is NEVER monophonic, unless your "everyday world" is living in an anechoic chamber!
[1] What a crossfeeder does to sound is what our hearing expects sounds coming to our ear having. [2] That's why the "messing-up" is beneficial. That's why people don't say the stereo image with speakers is messed up because your left ear hears right speaker and vice versa. [2a] Spatial information in "stereo space" MUST BE messed with to map it into "human hearing space". [3] You can live your life without a crossfeeder, but that makes you spatially ignorant.
1. Again, NO it is not! Our hearing does not expect just the simple crossfeeding of the sound, it would only expect that if normal, everyday life were living in an anechoic chamber. An anechoic chamber is in fact so alien to what human hearing "expects" that the first time inside one can be a very strange experience for many people, as the brain cannot accept the reality of what it's hearing and can start to play all kinds of weird tricks to make sense of it, even to the point of causing severe nausea in some people. What the brain actually expects is not just crossfeed but acoustic reflections. This is where it gets complicated because we have reflections in the recording itself and reflections of those reflections from the listening environment.
2. That all depends on the messing-up! You appear to have a typical audiophile black and white approach, much the same as the "all distortion is bad" opinion. In fact, distortion is not "bad", much/most distortion is not only "good" but absolutely essential, it depends on the type of distortion, the amount and it's context! Same with "messing-up", it ALL depends on how it's messed-up! What we hear in say the left ear from the right speaker is not just the right speaker signal delayed by some small amount (you quoted 250um) to compensate for ITD, what we hear (and absolutely expect to hear!) in our left ear is the reflections off the left wall of the room from the right speaker output, which is a delay in the several milliseconds range, plus a freq colouration of those reflections (which is completely unrelated to any head transfer functions). Messing-up with listening room reflections is beneficial because the recording has been created in an environment with room reflections in the first place, the delays and reverbs chosen/programmed according to how they interact with those room reflections. "Messing-up" the recording playback using ONLY a 250um delayed crossfeed is an entirely different sort of messing-up! It's just as likely, if not MORE likely to be the opposite of beneficial. There CANNOT be a black or white answer to this because it ENTIRELY depends on what reverbs/delays have been employed in a mix and how they have been employed.
2a. True but, HOW that spatial information is messed-up is vital, not just any messing-up will do!
3. The argument can be made for exactly the opposite to your statement! Yes, in some cases crossfeeding might provide an improvement, particularly with early stereo mixes which often contained very rudimentary stereo information but you are stating that nearly all recordings benefit from crossfeed and the most likely, logical explanation for such a statement is that you are relatively insensitive to some of the spatial information on recordings and are therefore not bothered by the inappropriate "messing-up" which often occurs with crossfeeding. If this is the case, that would make you the one who is "spatially ignorant"!
I know your answer to the above, something along the lines of: "With no crossfeed you still haven't got any room interaction but at least with crossfeed you've got one of the elements the brain expects." True and in some cases that's enough for crossfeeding work reasonably well but in other cases it can do more harm than good. In some cases, taking away that room interaction does not have much of a detrimental effect on the spatial information, it all depends on the amount of reverb applied and the parameters of that reverb, such as diffusion, stereo spread, stereo width and the timing, positioning and relative balance of the early reflections, almost all reverb algorithms already contain a fair amount of crossfeed to start with, although it's user configurable! Secondly, almost all mix engineers will check their mix using headphones, depending on the target media. Obviously the mix doesn't work how we intend on HPs and while we can't do much as far as the left/right positioning of the dry source sound/instruments are concerned without compromising the reproduction on speakers, there are typically tweaks we can apply to the spatial information (reverbs, delays, compression, EQ, etc.) which may have little impact on the speaker reproduction but significantly improves the HP reproduction. Again, how much we can do depends on the mix in the first place, the types and amounts of spatial information employed and of course, the time/effort dedicated to making such adjustments, which can vary from almost none at all to a considerable amount. Obviously, anything above "none at all" and you're going to damage/destroy it with your crossfeeding!
It helps that crossfeeders simulate what happens in reality so the decoding is easy for our brain, much much easier than decoding the un-crossfed signals with excessive spatial information.
This is so typical of many audiophiles; take a perception/preference and either just invent some complete nonsense to explain why it's the truth/real or, take some actual facts but ignore other vital facts to arrive at a more plausible/factual explanation which due to the omitted facts is still nonsense! This quoted statement falls into the latter category. Crossfeeding does indeed "simulate what happens in reality", the reality it simulates is what would happen if we listened to a recording on a stereo speaker system in an anechoic chamber and typically, nothing is more unreal or difficult for "our brain to decode" than listening in an anechoic chamber. This is the complete opposite of your quoted statement!!! AGAIN, in practice it all depends on the recording, our personal sensitivity to all the various types of spatial information and our personal preferences. Some audiophiles for example prefer and actively seek out the most "excessive spatial information" they can find, regardless of the meaning of the word "fidelity".
I've no objection to your preference for crossfeed, what I object to is you taking what is simply your personal preference and trying to redefine it as objective fact which applies to everyone and that me and everyone else who does not share your preference is "spatially ignorant". In all likelihood, I'm far less spatially ignorant than you, because for the last 25 years creating and manipulating spatial information is a significant part of what I do for a living. Additionally, your objective facts are not objective, they're clearly subjective and last but not least, you're completely ignoring/omitting other, vital facts! You have a strongly held preference and have developed a fairly elaborate explanation to turn that preference into an absolute factual belief, so there's virtually no chance I'm going to have even the tiniest influence on your views. This post is mainly aimed at others, who might find some of what I've said interesting or thought provoking.
G
