2 mics, recorded tracks offset.

[bouldersoundguy]

Well either...because if you're distance from mic2 to mic1 is increased three fold the distance of mic1 to source....then the distance of mic2 to source is automatically increased proportionally, and ends up being 4x what mic1 is to source.

You could move one mic laterally to change the distance between the mics without changing the distance to the source or the phase and level relationships (ignoring source radiation patterns). The distance between the mics is incidental.

The part in the end of the video shows how "phasey" the combination of both mics sounds when mic2 is less than 3:1 of mic1...there's just that point in the distance/time arrival where the phase issue is not as much of an issue.

It's the ratio of the two source-to-mic distances and the resulting signal levels that achieves separation. The distance between the mics is incidental.

Anyway...I have not to this day ever bothered to take out a ruler to measure when placing mics/mulitple mics....it's more of a by-eye and then what-sounds-best technique.

I don't think I've ever actually used a ruler. Distance ratios on that order seem self evident to me.

 

[miroslav]

You could move one mic laterally to change the distance between the mics without changing the distance to the source or the phase and level relationships (ignoring source radiation patterns).

Mmmmmm.... in very few cases.

If you have a mic 1 foot away from a guitar cab speaker....at 3:1 the other mic (even laterally) is still about 4 feet away from the source....and would be the case with most sources, unless it was a really wide *single-point* source, of which there are few.
With very large sound producing things like drum kits, choirs, grand pianos, etc....there is technically no single-point source to measure from as a whole...
...so then it's the measurement between the two mics.

Anyway...I think if you follow 3:1 mic-to-mic or mics-to-source....you end up in about the same place, so maybe that's why the 3:1 rule is often mentoined in both cases, and not such a big deal if applied either way.