04-05-2015, 07:15 AM
[Pauli posted a general question about this multitrack in a separate thread, but I've decided to answer it here so all this kind of information stays in one place -- Mike S.]
Well, you can get very heavily in-depth with tweaking phase-relationships these days, what with sample delays and phase-rotators both being pretty common currency in most DAWs. However, I don't personally think you need do anything more than experiment with your polarity buttons in this instance. It's also worth mentioning that I always check polarities while recording, so the default polarity relationship of the tracks is the one we used for the rough mixes during tracking -- and which survived to my final mix as well, I seem to remember.
A few additional bits of advice, though:
This is a really interesting question, and once I've not had before. You're right that fighting loudness bias is a constant challenge while mixing, but this is one case where increased loudness is often an indicator of a better polarity-match, so I wouldn't worry too much about it here, beyond keeping the question in your mind. If you suspect that loudness bias is affecting your decision making, then duplicate the track in question, invert the duplicate, level-match the duplicate, and then switch between the inverted and uninverted tracks.
One word of caution, though. One of the reasons you have close mics in an ensemble recording is so that you get some independent level control over different instruments. If you choose a close-mic polarity setting that achieves a timbre for its instrument which relies on cancelling (rather than reinforcing) the sound of that instrument's spill on other mics, then its usefulness for balancing will likely be reduced -- you might even find that fading the mic up will actually push the instrument backwards in mix! I'm not about to cast scorn on that approach, because the end justifies the means, but it's as well to be aware that it's possible to make a rod for your own back in this way if you're not careful.
Hope that makes sense! Talking about phase and spill in multimiked ensemble recordings can be quite complicated...
pauli Wrote:Anyone have any tips on handling phasing in a fully multi-miked mix like this? I'm wondering what your priorities are in settling on phase relationships and where you're willing to make sonic sacrifices for benefits elsewhere.
Well, you can get very heavily in-depth with tweaking phase-relationships these days, what with sample delays and phase-rotators both being pretty common currency in most DAWs. However, I don't personally think you need do anything more than experiment with your polarity buttons in this instance. It's also worth mentioning that I always check polarities while recording, so the default polarity relationship of the tracks is the one we used for the rough mixes during tracking -- and which survived to my final mix as well, I seem to remember.
A few additional bits of advice, though:
- Start your mix with the tracks which have the most spill, because that'll make the effect of the polarity switches more readily audible.
- Don't just listen to the instrument whose mic you're polarity-switching. Because of all the spill, changing the polarity/phase-relationship between the mics has the potential to change any/all sounds in the mix.
- Choose you polarity setting with the mix balance roughly where you want it, because the relative levels of mics affect the strength of any comb-filtering between the mics.
- Remember that some mix processing (EQ in particular) will affect a sound's phase response, so it's sometimes worth double-checking your polarity settings after adding any other processing.
pauli Wrote:Big snag I keep hitting is loudness bias... lots of the time I'll invert polarity or fiddle with a phase rotator, and the phase summation makes it louder, but not always more appealing in retrospect. Seems like a lot of the time the louder phase correlations also sound really "loose" or uncontrolled with fresh ears... full tone, but also a little unmanageable. Anyone have any tips on that score?
This is a really interesting question, and once I've not had before. You're right that fighting loudness bias is a constant challenge while mixing, but this is one case where increased loudness is often an indicator of a better polarity-match, so I wouldn't worry too much about it here, beyond keeping the question in your mind. If you suspect that loudness bias is affecting your decision making, then duplicate the track in question, invert the duplicate, level-match the duplicate, and then switch between the inverted and uninverted tracks.
One word of caution, though. One of the reasons you have close mics in an ensemble recording is so that you get some independent level control over different instruments. If you choose a close-mic polarity setting that achieves a timbre for its instrument which relies on cancelling (rather than reinforcing) the sound of that instrument's spill on other mics, then its usefulness for balancing will likely be reduced -- you might even find that fading the mic up will actually push the instrument backwards in mix! I'm not about to cast scorn on that approach, because the end justifies the means, but it's as well to be aware that it's possible to make a rod for your own back in this way if you're not careful.
Hope that makes sense! Talking about phase and spill in multimiked ensemble recordings can be quite complicated...