Hey all, very interesting and intelligent discussion here so I thought I'd fill in some of the details that get cut when your word count for an article is 500 words. There are a lot of nuances and it sounds like this is the kind of crowd that would appreciate knowing more.
First, I don't always use the single mic technique. I did Nestor Ausqui's most recent album using two mics and stereo, for the same general reasons mentioned here - capturing more ambience, more control over the guitar spectrum, etc. I did have to give up some of the single mic advantages - recording's always a tradeoff - but it was definitely the right technique for this performer and guitar.
I used the single mic technique for the first time on the late Linda Cohen's last album, "Naked Under the Moon." Linda was a delightful person and an excellent classical guitarist and teacher, but was always interested in pushing the boundaries musically. When she did "Leda" (released 1971), the idea of combining acoustic guitar and ambient electronics was quite ahead of its time and due to my reputation as being involved with electronics, I was often credited with "inventing" modern new age music long before it started taking off in the 80s. However, if that album did in fact invent modern new age music, she gets all the credit as it was entirely her idea.
Anyway, Linda played a Contreras guitar very softly and with much dynamics. The single mic thing was born out of frustration. In 1971 it was still a tape-based world, and to capture her dynamics was a problem. Whoever pointed out that one mic over multiple tracks doesn't mean less noise is technically correct, but when using two mics the mics weren't matched and neither was the noise. Spreading it out in stereo created a "soundstage" for the noise.
We couldn't do anything about it back then but "Naked Under the Moon" was done digitally with an ADAT for capture and a PARIS system for editing, her first (and only) album to use a DAW. I noticed that using one mic
appeared to give less noise because it was "up the middle" and masked by the centered material. The low frequency channel nuked the noise anyway, and the channel with the highs was panned slightly to the right; while it contributed to the noise, you at least didn't perceive stereo noise.
The Contreras guitar was also a problem because it had a HUGE bump at, IIRC, 220Hz or so. It was great for concerts, because that bass would hit the back rows but it was a bear to record. I usually ended up having to print with a bit of a notch just to set levels that could accommodate her dynamics without having too much noise.
Regarding phase, I've never had any luck with solving problems by using delay unless the point sources are fixed, for example, bass DI and miked amp are easy to fix because they don't move. Even if the mic does move a bit because of a kick drum or cabinet shaking the floor, the wavelength of a bass is so low it has to move a LOT to make a difference.
But guitarists move around when they record, Linda perhaps more than some. The body tends to stay anchored more than the neck, which made things worse because the period of high frequencies is so much shorter. So if she moved about an inch, anything at 5kHz was out of phase. Two inches, and it started affecting the guitar's harmonics. There's not a lot of energy from a nylon string guitar up there except for the fingernail sounds and articulations, but those contribute to the "in the room" feeling. Furthermore, if she got up to listen in the control room and sat down again, it was often way more than a few inches off but the "boom" would still be super-prominent because it didn't move as much, had longer wavelengths, and was loud. So this wasn't a polarity issue, it really was about phase with different peaks and valleys at different frequencies that varied over time...and you all know what comb filters do.
In addition to minimizing the phase issues, part of the reason for the single mic technique was to better re-create what Linda sounded like sitting in front of you and playing. The "stereo" aspect with separation via EQ had nothing to do with capturing the stereo interaction with the room, because it can't do that. What it does do is create a stereo image of the guitar itself that is very "up close and personal" because psycho-acoustically, the finger scrapes are on the right, and the body is on the left, differentiated by EQ. If you get the cutoffs and levels right, it sounds like the guitar is as close to you as your speakers are.
The final benefit was that by essentially putting the boom in its own little tonal ghetto, I could control it much better in relation to the rest of the guitar.
As to sounding tinny, artificial, etc., the technique is like a combination lock. If all the tumblers aren't in place, it won't open. There are 100 ways to make this technique sound bad and only a few that sound good. As mentioned in the article, it's crucial to experiment with the levels and cutoff frequencies. There is no "one size fits all" setting or I would have saved it as a preset
In fact sometimes it just plain doesn't work, period, depending on the player, the guitar, and the playing style. But when it DOES work, it works really well. People remarked on "Naked Under the Moon" having by far the best guitar sound of all her albums and while the single mic technique was certainly not the only reason why, I do feel it made a significant contribution to giving a stronger, more defined sound that supported her playing better.
As always, you have to be sensitive to the artist and do whatever you can to represent their music. I've done classical harpsichord recordings with Kathleen Mcintosh and for her album of Antonio Soler pieces, was a good boy and recorded harpsichord the way it's "supposed" to be recorded. But when we did her Bach CD, I said to hell with it and tried close-miking it...very rock and roll. I played her both the traditional and close-miked versions, and said I'd do it whichever way she wanted. She FLIPPED OUT over the close-miked. "That's what it sounds like when I'm playing harpsichord!! Do that!!" I really thought the purists would get upset, but no one even noticed, other than to say she sounded great. And you know what? She did, because the close-miking removed a "layer" between you and her playing, and you could realize just how well she played that music.
Anyway, thought you'd find the above interesting. Carry on