The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Gear for sale  Search Gearslutz Go Advanced
conflicting 3:1 explanations
Old 20th September 2011
  #1
Gear Addict
 

conflicting 3:1 explanations

I have been getting some conflicting information in regards to the 3:1 rule. Harvey Gerst says that anything beyond 3 times the distance from the first mic to the source for the second mic is good... in other words once you pass 3 X distance, you clear of phase problems for the most part and the further you go the better.

Yet elsewhere I have read that it goes in multiples... so 3:1, 6:1, 9:1 (or 12:1 idk), and that its the in-between positions that cause phase issues.

Can anybody confirm which one is correct for me
Old 20th September 2011
  #2
Registered User
Multiples? Never heard that, and it makes no sense.

The distance thing is all about decreasing the amplitude of one signal relative to the other to a level where the notch filter phase effect isn't serious.

Actual distance doesn't matter - there is no relationship to wavelength, because we are talking about the whole audio spectrum, so no particular wavelength matters.
Old 20th September 2011
  #3
Gear Addict
 

I agree multiples don't make sense, to me at least. When the 'out-of-phase' signal is 10 dB below the original, the notch filtering effect is lessened.

10dB is more-or-less the loss when you place the second mic 3 times the distance from the 1st (double the distance is -6dB, can't remember exactly what 3 times what would be).

But yeah, it's a starting point, not necessarily the best place.
Old 20th September 2011
  #4
Here for the gear
 

So in relation to mixing, 'most' phase-incoherencies can be solved if one of the mics on a 2 mic setup is mixed 10Db lower than the other, irregardless of the actual phase problems when recorded ?
Old 20th September 2011
  #5
Registered User
Well that's one way ...

As soon as you have distance between two mics that are picking up sound from the same source, you have a time delay problem caused by the speed of sound in air.

That time delay will have the effect of boosting some wavelengths (and multiples thereof) and cutting some wavelengths (and multiples thereof). This is what we mean by a comb filter. It's like getting a graphic eq and alternatively boosting and cutting the bands ... it's a bit weird.

If you keep one of the signals fairly low, the amount of boost and cut is fairly low so it's less annoying.

Another method is to time delay one of the signals. This does not make the problem go away - but it allows you to fine tune the actual frequency bands. You might be able to get a better sound this way. Try flipping the polarity (aka the 'phase' switch) which reverses the boosts and cuts ... a different flavor of problem.

yet another way is to use an All Pass Filter ... this is basically a frequency dependent time shift. Still doesn't fix the problem, but gives you a whole different range of skewed comb filter sounds.

Or modulate one signal, so the phase effect is not static but always moving ... not always appropriate, but it's an option ...

Phase effects are not necessarily a problem. They are part of natural sounding acoustics ... just make it sound good. I think audio signals that lack any phase interaction can be sterile and boring.
Old 20th September 2011
  #6
Lives for gear
 
Ben B's Avatar
 

Quote:
Originally Posted by Purple View Post

10dB is more-or-less the loss when you place the second mic 3 times the distance from the 1st (double the distance is -6dB, can't remember exactly what 3 times what would be).
20 log 3 = 9.54 dB



-Ben B
Old 20th September 2011
  #7
Lives for gear
 
audiogeek's Avatar
 

Quote:
Originally Posted by Moerkvaerk View Post
So in relation to mixing, 'most' phase-incoherencies can be solved if one of the mics on a 2 mic setup is mixed 10Db lower than the other, irregardless of the actual phase problems when recorded ?
Well, yes sort of. But it's not quite as black-and-white as that.

In most scenarios when you're using two mics for a single sound source (think drum overheads, piano, etc), you're not going to want a 10 dB drop between the left and right channels. Generally you want the playback level to be about the same between the left and the right.

So, while that oversimplified math may work in theory, it's not really a practical solution in application.
Old 20th September 2011
  #8
you just gotta move the microphones around until the balance sounds good. I was speaking with a client the other day about this. When you follow the "rules", with mic placement, and then you have problems with your technique, the mic placement "rules" are then meant to be broken.
Old 20th September 2011
  #9
Quote:
Originally Posted by Moerkvaerk View Post
So in relation to mixing, 'most' phase-incoherencies can be solved if one of the mics on a 2 mic setup is mixed 10Db lower than the other, irregardless of the actual phase problems when recorded ?
No.

This phase problem can be solved by making one so quiet you can't hear it! ie muting one of the mics.

If you've recorded 2 mics with a phase discrepancy that causes unpleasant comb filtering, whilst some tools (eg IBP) can help, nothing will "fix" it as good as recording it properly in the first place. Time aligning can't do it (you can't time delay frequencies independently), although again it may make a subjective improvement.

This 3:1 "rule" is anything but. I've never seen a pro tracking engineer think about it, nor have I ever felt the need to use it myself. Just make sure that any spill is as quiet as possible, and check multi-mic setups for phase coherency, and you'll be fine.
Old 20th September 2011
  #10
Quote:
Originally Posted by Moerkvaerk View Post
So in relation to mixing, 'most' phase-incoherencies can be solved if one of the mics on a 2 mic setup is mixed 10Db lower than the other, irregardless of the actual phase problems when recorded ?
The 3-to-1 rule-of-thumb doesn't 'fix' the phase problems -- it mitigates them by helping to minimize unwanted pick up of non-target sound sources in a complex multi-mic setup.

By making sure that any given mic is at most 1/3 the distance from its target sound source as it is from the next nearest (non-target) sound source, you assure that that the target source is (something vaguely like) at least ~9 times louder* in its mic than the non-target source (assuming both sound sources start out at equal volume).

*Radiated energy decreases with the square of distance from source. Three times as far away means 1/9th the energy received, relative to the first sound source, assuming omnidirectional radiation, equal amplitudes, no reflections, etc. But since things like drums are anything but omnidirectional radiators, you don't have even radiation in all directions. By a long shot.


Think of the 1-to-3 'rule' as very general guide. It's simply there to remind you that the way to minimize phase difficulties arising from picking up a single source in multiple mics is to minimize the pick up of the source in mics that are targeting other sources.

And, of course, as psycho_monkey suggests, trying to phase align a complex drum miking is inviting the onset of insanity. You can align a single source in two mics. But when you start talking about multiple sources in multiple mics, you've got a continuously moving field. Fixing one set of relationships pulls other relationships out of kilter.
Old 20th September 2011
  #11
And just to have a Geek Out Tuesday moment.... The reason that things like sound and light decrease their intensity by the square of the distance is because they radiate outwards as a sphere. So, at any given distance from the source, the energy (originally concentrated in almost a point at the source) has now been spread out around the surface of a sphere of that radius (and the same amount of stuff spread out more means that it's becoming less concentrated, too much butter scraped over bread as Bilbo says.) Since the relationship of the surface area of a sphere relative to the radius of a sphere is based on the radius squared (4*pi*r^2), that means as the radius changes, the size of the area that the original energy is spread out around grows by that radius squared. The 4 and the pi stay the same, only the radius changes.

Or, then again, I could be comletely wrong.
Old 20th September 2011
  #12
Quote:
Originally Posted by Dean Roddey View Post
And just to have a Geek Out Tuesday moment.... The reason that things like sound and light decrease their intensity by the square of the distance is because they radiate outwards as a sphere. So, at any given distance from the source, the energy (originally concentrated in almost a point at the source) has now been spread out around the surface of a sphere of that radius (and the same amount of stuff spread out more means that it's becoming less concentrated, too much butter scraped over bread as Bilbo says.) Since the relationship of the surface area of a sphere relative to the radius of a sphere is based on the radius squared (4*pi*r^2), that means as the radius changes, the size of the area that the original energy is spread out around grows by that radius squared. The 4 and the pi stay the same, only the radius changes.

Or, then again, I could be comletely wrong.
As long as you're not a johnny comletely...


heh
Old 20th September 2011
  #13
Lives for gear
 
DigitMus's Avatar
 

The 3:1 "rule" has been mis-quoted, mis-understood, corrupted & bent out of shape more than almost any other "rule" of recording/sound reinforcement.
It was intended as a guideline for reducing bleed between 2 mics on 2 different sources. Further, it was intended to apply to close micing of sources that are approximately equal in loudness (dB SPL). It has nothing to do with 2 mics on the same source; it has nothing to do with stereo & is only related to phase anomalies with respect to the fact that they can be ONE of the negative effects of excessive bleed.

Simply stated:

When micing 2 sources with 2 microphones simultaneously(mic A-> source A; mic B -> source B), mic A should be AT LEAST 3X the distance from source B as it is from source A and vice versa.

Scott
Old 21st September 2011
  #14
Lives for gear
 

Quote:
Originally Posted by Devon8822 View Post
I have been getting some conflicting information in regards to the 3:1 rule. Harvey Gerst says that anything beyond 3 times the distance from the first mic to the source for the second mic is good... in other words once you pass 3 X distance, you clear of phase problems for the most part and the further you go the better.

Yet elsewhere I have read that it goes in multiples... so 3:1, 6:1, 9:1 (or 12:1 idk), and that its the in-between positions that cause phase issues.

Can anybody confirm which one is correct for me
3:1 rule is a rule of thumb that keeps comb filtering to less than 1 dB so you have a usable recording

anything larger should work better
anything closer would have bigger comb filter peaks
Old 21st September 2011
  #15
Comb filtering and phase incoherence can be used to your advantage. You might find that by ignoring the 3:1 rule on, say, a bass drum, that you can dial out an ugly resonance in the low 100s, reinforcing the fundamental. Or maybe you can change the tonality of a guitar sound by adjusting two mics' volumes to EQ the sound.

I'd say you'd be better off listening to what you might come up with, rather than to strictly adhere to any "rule" just for the sake of it. Phase is a powerful tool!

Furthermore, the 3:1 rule is far from infallible. Take an overhead mic against a snare close mic. In pretty much every conceivable scenario, you'll be obeying the rule (snare drum close mic is a couple inches or so off the head, while the OH is many feet over the drum). And yet, you can still get HUGE amounts of cancellation to the point of the snare practically disappearing, or you can still get 90 degree cancellation (which is, IMO, a very bad thing).
Old 21st September 2011
  #16
Lives for gear
 
DigitMus's Avatar
 

Apparently no one even read my post...
Old 21st September 2011
  #17
Lives for gear
 
Bristol_Jonesey's Avatar
I did, and it made total sense
Old 21st September 2011
  #18
Lives for gear
 
rob S's Avatar
I read it.
Spot on.
Old 21st September 2011
  #19
Lives for gear
 

The situation is usually very messy. You'll need to experiment, listen and adjust.

Another factor to consider is that the level falloff with distance only extends to the critical distance, at which point indirect sound will have the same level. You don't see much level falloff beyond the critical distance. Sabine's approximation is d = .057 SQRT(V/RT60) in SI units. Of course, a real instrument will probably have a very complex projection pattern which will make the real behavior more complicated and messy, but in any direction there will still be a distance beyond which indirect sound rules, and it may be closer than you think.

Also, the behavior can vary radically by frequency. In the low frequencies, the room modes may be strong and reverberate long. For those frequencies, there is essentially no "close." The response may be dominated not so much by relative distance, but rather by the room response at each mike's location and, to a degree, the projection pattern of the instruments. Hence the behavior noted above, where a "close" mike and a "distant" mike both have competing levels of sound at certain frequencies but with different phase.

Cheers,

Otto
Old 21st September 2011
  #20
Gear Maniac
 

Quote:
Originally Posted by Kiwi View Post
Another method is to time delay one of the signals. This does not make the problem go away ...
eh? Why not? Once they're aligned and in phase, aren't they aligned and in phase?

Wikipedia tells me that the speed of sound through air is only slightly dependent on frequency, so what's going on here?

(If you've lined up a distant and a close mic to the sample in a DAW and heard phasing, then I'll accept that a reasonable reason to have you opinion, but I would like to know if it's just your subjective experience rather then a fact.)

Thanks Kiwi.

Edit: Sweet post DigitMus.
Old 21st September 2011
  #21
Gear Addict
 

These replies are very helpful! Digimus, I would give u rep points if i could
Old 21st September 2011
  #22
Lives for gear
 

Quote:
Originally Posted by TwardyLikeTank View Post
eh? Why not? Once they're aligned and in phase, aren't they aligned and in phase? ...
Just include that it implies a single source in rather clean direct paths to the mics, or +/- any other paths/reflections that might be in there.

I'd bet we should add that those two mics ought to be in a fairly direct line as well. If one is to the side lots of our 'single sources' get a bit more complicated. ..Re; '3:1 in stereo micing an ac guitar..

It is fascinating how a ratio' can get touted at times as some kind alignment method though. That continues to be one totally bizarre thing in audio circles.
Old 21st September 2011
  #23
Quote:
Originally Posted by DigitMus View Post
Apparently no one even read my post...
I did. My only 'quarrel' would be that this single bit, taken out of context:
Quote:
It has nothing to do with 2 mics on the same source
I know what you mean, but, of course, the problem is when, in a complex multi-sound source, multi-mic setup, a given mic that is targeting one source picks up enough of another source to cause 'phase' issues (resulting in comb filtering) if enough of that unwanted pickup is combined with the (wanted) pickup from the mic for that second sound source.

Talking about this stuff is actually more complicated than doing it, in many ways. heh
Old 21st September 2011
  #24
Gear Addict
 
nicpope's Avatar
 

If the 3:1 rule actually applied to multiple microphones on a single source it would suggest that placing one microphone 1 inch from a guitar speaker and a second microphone 3 inches from the same guitar speaker would produce a phase coherent output....nope.

It is however a decent rule of thumb when recording multiple performers in a live space as detailed above, and ALSO for the same reasons can be a decent guidline for setting up spaced omni stereo pairs in smaller rooms, not that this was the original intention but the concept is the same.
Old 21st September 2011
  #25
Lives for gear
 

Quote:
Originally Posted by TwardyLikeTank View Post
eh? Why not? Once they're aligned and in phase, aren't they aligned and in phase?
For a single, point source with a spherical pattern for all frequencies and no reflections or room interactions, yes.

OTOH, just picture two mikes, one by hi-hat, one by floor tom. The hi-hat sound shows up first in the hi-hat mike, the floor tom sound first in the floor tom mike and let's pretend the snare is exactly in the middle and arrives simultaneously at both mikes.

Shift the floor tom mike signal ahead to align the hi-hat in both signals and you've created arrival time differences on the snare in the two signals and increased the existing difference for the floor tom.

Hope that helps.

Cheers,

Otto
Old 21st September 2011
  #26
Lives for gear
 
DigitMus's Avatar
 

Gee, I guess my posts aren't invisible, then. heh

theblue1: I see what you're saying, but you're still talking about multiple mics on multiple sources in your example, not the use of 2 mics (intentionally) on a single source, which the 3:1 rule is commonly mistaken to apply to.
The rule, as I stated it above, is very simple - implementing it, as in your example, isn't always. What remains is that the 3:1 rule, followed correctly, will eliminate or significantly reduce MOST phase problems when micing multiple sources individually. More complex situations and factors like room reflections were not ever intended to be addressed by it. It is NOT, nor was it ever (well...before the internet, anyway heh ) a cure-all for every phase anomaly caused by multiple mics. Drum micing alone is a very complex stew of bleed, phase, arrival time and other problems - that's why there are so many threads, and even books, written about drum micing.
I don't know why I ramble on about this, except that the mis-understanding of 3:1 causes it to be invoked and discussed in all kinds of situations that it simply doesn't apply to.

Scott
Old 21st September 2011
  #27
Quote:
Originally Posted by DigitMus View Post
Gee, I guess my posts aren't invisible, then. heh

theblue1: I see what you're saying, but you're still talking about multiple mics on multiple sources in your example, not the use of 2 mics (intentionally) on a single source, which the 3:1 rule is commonly mistaken to apply to.
The rule, as I stated it above, is very simple - implementing it, as in your example, isn't always. What remains is that the 3:1 rule, followed correctly, will eliminate or significantly reduce MOST phase problems when micing multiple sources individually. More complex situations and factors like room reflections were not ever intended to be addressed by it. It is NOT, nor was it ever (well...before the internet, anyway heh ) a cure-all for every phase anomaly caused by multiple mics. Drum micing alone is a very complex stew of bleed, phase, arrival time and other problems - that's why there are so many threads, and even books, written about drum micing.
I don't know why I ramble on about this, except that the mis-understanding of 3:1 causes it to be invoked and discussed in all kinds of situations that it simply doesn't apply to.

Scott
Not at all (on the invisibility thing).

In fact, I'd thought about addressing that exceedingly minor point (only because I was afraid someone who didn't quite get it might grab onto that detail and run off the edge of the field with it) but I didn't want to seem like I was just reaching in and cherry picking a wee bit that might be misinterpreted by others -- or, worse, offering some sort of discourse of my own that might, itself, then be misinterpreted.

Certainly, as I think most all of us here would readily agree, the '3-to-1 rule' is not pertinent to, say, using multiple mics on a guitar amp, where one might actually like the sound of some phase-related cancellation/comb filtering -- or, conversely, where time-alignment of multiple mics might actually make sense.


BTW, one of the best drum sounds I got in my early days -- and something of a practical lesson -- was a four track location recording with really minimal mic availability. I think I used a 421 for the kick, an SM57 for the snare/hat, another '57 for the toms, and a '58 (!) for the overhead. Sure, the cymbals were less than silky, but it was a really solid sound that was super easy to balance out. Of course, crucial to all that was the fact that it was a good kit played by a good drummer in a pretty decent room.
Old 22nd September 2011
  #28
Gear Maniac
 

Thanks ofajen and Wayne,cool replies.
Old 22nd September 2011
  #29
Lives for gear
 
DigitMus's Avatar
 

Quote:
Originally Posted by theblue1 View Post
... Of course, crucial to all that was the fact that it was a good kit played by a good drummer in a pretty decent room.
Ain't it funny how when that happens, all the minutia about drum sounds that people argue about on the internet just falls completely off the radar?

Scott
Old 22nd September 2011
  #30
Lives for gear
 

Quote:
Originally Posted by DigitMus View Post
Ain't it funny how when that happens, all the minutia about drum sounds that people argue about on the internet just falls completely off the radar?

Scott
This would be why I track mono to one track on tape with an overhead mike, an out-front mike, or both. Quick Sound Field, good sound on the drums and me playing stuff right, that's all it takes. That last one is the tricky part...

Cheers,

Otto
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump