I'm always wonder a bit when folks state specific mix ratios between sources here at TS. How do you know what the ratio is? I can't imagine folks are doing RMS ratio calculations to determine what the ratio might be. Is there some "combine" function of specific mixing software that specifies this?
If so, unless the separate sources had exactly the same levels prior to mixing, any level difference between the files would be invisible to the mixing software (unless the software was pretty advanced). It's basically the same situation as the Mid/Side ratio thing I described earlier in the thread - the Mid/Side controller assumes the Mid and Side channels were recorded with equal sensitivity microphones and recording gains, and if that's not the case the ratio displayed on the controller display is not the actual ratio.
A dB difference between sources would seem to me easier to determine and more useful (being a logarithmic rather than a linear relationship) if one wanted to describe the relationship between sources to someone else, just get a numeric "feel" for a specific level difference, or determine how much difference is in play before one source effectively renders the other inaudible.
"I record them as if separate sources.
Mix them by ear.
Adjust output level by eye.
In other words-
I record optimizing recording gains for each. Import the files and adjust the relative levels of those sources in the mixing software by ear, listening for whatever sounds best and works right (sometimes going back and forth between a few different "good" choices before making a decision, but never by any specific ratio or dB difference), and adjust the resulting level in the mix bus by eye, checking the peak-hold meter so that the highest peak ends up around -.5 dBfs or so."
OK, I am clearly not articulating myself well on this topic. My apologies for all the confusion.
I essentially do the same thing you do...as stated above.
My original question was only relating to your original recording levels, which you did answer. As you are aware, when recording multi tracks, and then combining them, they increase which can then create a handful of peaks. If, when recording the live source, you lower your levels based on what you'd normally record at, then you'd be raising your levels in post if they are not already the way you like them once you mix tracks. When recording as you normally would with two tracks, levels will have to be lowered before mixing tracks. My question was based on that. After getting the swing of multi-track recording, I lowered my original recording levels by about 3dB in anticipation of post production. I've never had a problem with this (yet). In other cases, when I want the different microphone sources to also shine on their own, I will record as in they are pairs only, and end up having to lower the levels in post.
The ratios a I sighted were bullshit and I shouldn't have stated them that way, I just thought for the sake of simplification, it would best describe what I was trying to say. I was obviously incorrect. My sources usually differ from being the same to a 3-4 dB difference (before mixing, in order to attain the sound I like best). That is to say, If one pair of mics are Naks and the other pair are Gefells, when mixing, I will determine what I need more in the mix (by ear), and adjust accordingly. If I remember correctly, my most recent 4-track recording ended up at a 2dB difference between the two source pair of mics. Naks are warmer and provide a more resonating bottom end, while the Gefells offer a crisp, more precise high end. They blend nicely, but it's a matter of preference what your ear may prefer.
I hope that helps clarify what I meant to say.