Topic: Matching the levels of two sources

[StarkRavingCalm]

I taped a show the other night from two sources, SBD and AUD.
Usually I get lucky and they are both pretty close, but this not, not at all.
The SBD source is WAY low. (not enough time to check levels)

Is there a way to match the volume to the other, higher source without raising it incrementally multiple times and then judging by ear?
In other words can I:
Find max volume of AUD source, and raise the max volume of SBD source to match it?

I am using Audacity and a simple Normalize or Amplify does not do it



Thanks in advance

[Gutbucket]

There is no AUD/SBD ratio formula which works everytime.  Always do it by ear.  I can't imagine how you could determine the most appropriate blend without listening. It would be like painting while blindfolded.

Peak normalize or otherwise adjust the levels of both sources to something managable like -6dBFS then fine-tune your mix levels from there by ear. 

[page]

There is no AUD/SBD ratio formula which works everytime.  Always do it by ear.  I can't imagine how you could determine the most appropriate blend without listening. It would be like painting while blindfolded.

+1

dont be concerned with trying to make some preset (mythical) blend of "50/50" or anything like that.

[capnhook]

There is no AUD/SBD ratio formula which works everytime.  Always do it by ear.  I can't imagine how you could determine the most appropriate blend without listening. It would be like painting while blindfolded.

+1

dont be concerned with trying to make some preset (mythical) blend of "50/50" or anything like that.

+1

[Todd R]

There is no AUD/SBD ratio formula which works everytime.  Always do it by ear.  I can't imagine how you could determine the most appropriate blend without listening. It would be like painting while blindfolded.

Agreed, though if you want to mix your paints, it helps to know that blue and yellow make green.

I agree there is no magic formula, but I usually have a sense of where to start depending on what I'm doing -- 2ch aud mics + soundboard, stage mics + sbd, 2 pairs of mics like omnis and cards for a 4mic mix.  Given the situation and what I recorded and what I'm trying to accomplish, I usually have an idea of what I think would work, which I test out and listen to obviously.

On getting that head start, first off, I find 50-50 or 70-30 to be useless.  Anytime I'm asked to think about it that way, I've got to convert to dB.  Percentages to me have no meaning, maybe my brain just doesn't work that way.  Then when thinking about dB relationships, it is most useful for me to know the average RMS level of the different sources, not the max dbFS level. 

I don't need to match beforehand, just know where they are.  That way, if I think as a starting point I want my split omnis to be 6db lower in sound than my main cardioid pair, I can just work with what levels they are at (RMS) and mix higher or lower depending on the existing recording level.

Anyway, I agree that there is no magic formula and you should mix based on how it sounds, but to the extent you want to reference one source to the other in your planning, I think it is best to consider their average loudness, not the max loudness.  And of course make sure that when you mix both together you won't be clipping (trying to go above 0 dbFS) with the mixed source.

[Gutbucket]

I'm with you completely Todd,

The percentage terms baffle me too and relative RMS values between sources are far more meaningful to me.  Yet I see percentage and matrix ratios stated here at TS so often that I've assumed it reflects some specific workflow terminology of a particular software package in common use here with which I'm not familar (Wavelab maybe?) instead of simply a way of expressing the relative amounts of each source by way of words on a disussion forum.  Any idea if this is the case?

Conceptually, I tend to select a primary source as main contributor, then bring in the others enough to help reinforce the primary one without loosing focus.  That typically means the primary source makes a significantly larger contribution to what is heard, and the secondary (and sometimes tertiatry) sources contribute much less so in relative terms.  Sometimes after playing with things for a while, I'll try it the other way around and switch to using another source as the primary one to see if things work better, but the general process is the same- a clear staring role plus supporting actors.  This also affects what I do to the souces if I'm doing more than simply mixing things together, affecting decisions on EQ, compression, stereo width, panning, etc.  Keep in mind that my multiple sources are frequently multichannel surround files since that's my thing, so when mixing down to two channel I'll typically choose between one or two Left/Right pairs as the primary source, then bring up a center channel or SBD enough to help with missing detail, clarity and soundstage solidity, or to help reinforce vocals or some other under-represented instrument, and finally bring in a (usually) lesser amount of ambient room mics for depth, room ambience, bass quality, and crowd reaction.  There is no way I could guess or and get it all balanced appropriately by some rote formula, even though I have a pretty good general idea of the relative contribution of each which typically seems to work.. in the rare situations when I actually get around to mixing it all down to stereo.

I only suggested peak normalization of both sources to an arbitraty safe working value of something like -6dbFS to get both of StarkCalm's sources up to a managable level where he could then make these kinds of choices.  With the exception of avoiding clipping, I agree on RMS being more meaningful for most comparitive purposes.

[Fatah Ruark (aka MIKE B)]

One thing I've found when mixing a SBD+AUD is that I think I tend to rush things. Generally things will sound pretty good. Frequently I do a quick mix, like what I hear and then track it out.

Take your time and try different mixes.

Now I take a good long listen to my mix, then I take a break, and try something else...and repeat. Sometimes I go back to my first mix, sometimes I go with the last, sometimes one in the middle.

I've also started listening to rough mixes on multiple playback systems. I can stream music over my network to my stereo, so I'll always try that (vs. my headphones that I mix with on my computer). Occasionally I'll dump a copy on a memory card to play in my car (just to see how it sounds on a crappy playback system).