M/S recording

[Saudade]

Don't know if this is the right place to post....anyways

Lets say I'm recording a vocalist who is singing her own multiple backing vocal parts.

I set up a M/S recording, and get her to stand a little to the left for those parts that are intended to appear left, and vice-versa for right.

My question is, after I have recorded and decoded the tracks, can I still CONTROL how much to the left I want those "left" tracks and how much to the right I want the "right" ones? Can it still be done via traditional panning in a DAW as I understand I should end up with 3 separate mono tracks after decoding????

Or do I have to decide on the spot during recording how far left or right physically she needs to be based on what I hear in the monitors?

Thanks for any info...

[zilver]

I don't know if i understand you completely... but :

If you record three times a M/S signal it's six tracks.
After decoding the M/S to L/R you have six tracks (or three stereo tracks).
These stereo tracks can be panned more left or right.

But why would you want to record the back vox in stereo ? Do you want to record the acoustics of the studio also ? Like it's an old fashioned one mic recording of an ensemble ?
If not, it's easier to record three mono tracks and pan them.

Good Luck, Hans

[mosrite]

Don't know if this is the right place to post....anyways

Lets say I'm recording a vocalist who is singing her own multiple backing vocal parts.

I set up a M/S recording, and get her to stand a little to the left for those parts that are intended to appear left, and vice-versa for right.

My question is, after I have recorded and decoded the tracks, can I still CONTROL how much to the left I want those "left" tracks and how much to the right I want the "right" ones? Can it still be done via traditional panning in a DAW as I understand I should end up with 3 separate mono tracks after decoding????

Or do I have to decide on the spot during recording how far left or right physically she needs to be based on what I hear in the monitors?

Thanks for any info...

Forgive me if Im not understanding you correctly but here goes:
If you are making a stereo recording using an M/S array then why would you then pan any of the elements after the fact? If you are overdubbing the backing vocals and wish them to be situated in a different position within the stereo field then simply keep the array static and re-position the singer.

[tINY]

Here is an instance where flexability later will require that you use a dead vocal booth and artificial revberberation.

Once you have several stereo pairs, the only thing you can do is adjust relative intensities between the pairs - if you want to keep the origninal stereo queues in place.

If you are just using M/S because you like something other than the stereo field it provides, do whatever sounds good.



-tINY

[pkautzsch]

Convert it to XY and feel free to pan it wherever you like. You might notice that even by changing the ratio between M and S you will pan her in a way.
Though, as others already have written, I don't see the point in recording one singer in stereo outside classical or other "natural" music.

[zilver]

Originally Posted by pkautzsch

Convert it to XY and feel free to pan it wherever you like .

Saudade, this might confuse, therefore :
XY is the more exact name for the "normal stereo" you get when you insert a M/S converter plugin.

Hans

[Jim vanBergen]

traditionally, the way to decode the M-S is to use the SIDE figure eight, mult its signal to two tracks, and throw one side of the figure eight out of phase. That leaves you with THREE tracks from the two mic signals: L-Mid-Right are the result. Each track is able to be panned as you desire, though you can muck up a nice recording if you try.

Does that help?

[sssaudio]

Originally Posted by zilver

Saudade, this might confuse, therefore :
XY is the more exact name for the "normal stereo" you get when you insert a M/S converter plugin.

Hans

Actually, that's not true. In the truest sense of the word, and in Concert Recording terminology, XY is just the term for a stereo pair whose capsules are coincident, set to cardioid, and create a 90 degree angle. The stereo track that one gets after decoding an M/S setup is simply M/S decoded. People tend to call almost every kind of stereo setup "XY." Especially drummers when they set up their overheads. But there are complete differences between XY, AB, MS, Blumlein, ORTF, and a whole slew of others. Just trying to clarify.

[JazzYoda]

Logic will allow you to change the stereo positioning information in m/s, using the "direction mixer". I don't know about other programs.

[Saudade]

woah...now I am more confused

ok see Jane Doe is doing all her BV parts after she's done the lead.

So I was thinking if i can mic her in some stereo method she can do (overdubbing)BV part 1 by standing far left of the array, bv part 2 mid-left, part 3 mid right, part 4 far right etc. So after that playing back the stereo tracks I should have an stereo image of 4 Jane Does singing togther.

So my question is, if I choose the M/S method (which I read somewhere somehow allows me to manipulate the stereo image further during mixing), will I be able to lets say, do stuff like, switch the positions of Far Left Jane Doe with Mid-Left Jane Doe, or make Far left Jane Doe even "further" left (if the decoded stereo image is not wide enough)?? In other words can I still pan the signals to manipulate the "position" of each BV part in the stereo field without artifacts or phase cancellation? Or will other stereo methods be preferred over M/S for this purpose???

Or should I just screw it and record all the BVs with a single mic?

Thanks for the time...

[tINY]

You probably could do what you say, But....

..normally you would use a coincident pair (M/S or X/Y) on a recording where you want the room sound recorded as well. When you manipulate the signals to "reposition" the takes, you will also manipulate (and probaby screw up) the room sound.





-tINY

[JazzYoda]

*******************************/Mid-Sid...hone_Technique

[sssaudio]

Originally Posted by tINY

You probably could do what you say, But....

..normally you would use a coincident pair (M/S or X/Y) on a recording where you want the room sound recorded as well. When you manipulate the signals to "reposition" the takes, you will also manipulate (and probaby screw up) the room sound.





-tINY

I agree. M/S ialso widely used because of it's perfect mono compatibility. Since the two side components are 180 degrees out of phase with each other if they're both panned center, they cancel out, leaving you with only the forward facing cardioid component (or omni if you chose that). Once you go and start messing with the panning you can definitely get some phasing issues.
It's not a bad idea to do what you want to do as M/S provides an amazing stereo image. Having your singer standing in different spots each time (assuming that the m/s system doesn't change positions) is definitely better than recording in mono and panning. Panning is just a manipulation of gain. M/S is true stereo. But if you're going to be messing with the position of it afterwards you're probably better off doing them in mono. Use the M/S only if you get the positions right and want to stick with them.

[Gilliland]

Originally Posted by sssaudio

Actually, that's not true. In the truest sense of the word, and in Concert Recording terminology, XY is just the term for a stereo pair whose capsules are coincident, set to cardioid, and create a 90 degree angle. The stereo track that one gets after decoding an M/S setup is simply M/S decoded. People tend to call almost every kind of stereo setup "XY." Especially drummers when they set up their overheads. But there are complete differences between XY, AB, MS, Blumlein, ORTF, and a whole slew of others. Just trying to clarify.

However, MS and XY are interchangeable, at least in theory. With "perfect" microphones in "perfect" coincident postion, MS and XY can be shown to be mathematically identical - once you decode MS into normal stereo, the resulting stereo is exactly what you'd get if you had used XY in the first place. And it would be equally possible to convert XY to MS (though I can't see why you'd want to).

Of course, there are no perfect microphones, and it is not possible to place their diaphragms in perfect coincidence, so there will always be some difference. Nonetheless, in most cases that difference is negligible. For all intents and purposes, MS and XY will produce identical results.

[JazzYoda]

Originally Posted by sssaudio

I agree. M/S ialso widely used because of it's perfect mono compatibility. Since the two side components are 180 degrees out of phase with each other if they're both panned center, they cancel out, leaving you with only the forward facing cardioid component (or omni if you chose that). Once you go and start messing with the panning you can definitely get some phasing issues.
It's not a bad idea to do what you want to do as M/S provides an amazing stereo image. Having your singer standing in different spots each time (assuming that the m/s system doesn't change positions) is definitely better than recording in mono and panning. Panning is just a manipulation of gain. M/S is true stereo. But if you're going to be messing with the position of it afterwards you're probably better off doing them in mono. Use the M/S only if you get the positions right and want to stick with them.

If you do it in M/S you can simply discard the side mic track if it doesn't work out and end up with a perfectly useable cardioid mono recording of each :)

[zilver]

Originally Posted by Gilliland

However, MS and XY are interchangeable, at least in theory. With "perfect" microphones in "perfect" coincident postion, MS and XY can be shown to be mathematically identical - once you decode MS into normal stereo, the resulting stereo is exactly what you'd get if you had used XY in the first place. And it would be equally possible to convert XY to MS (though I can't see why you'd want to).
.

Thanks Gilliland, Hans

[zilver]

Originally Posted by sssaudio

= Panning is just a manipulation of gain. M/S is true stereo.

M/S is also manipulation of gain. It's how much of the side signal (S) is added or substracted from the M signal.

Cheers, Hans

PS hope this doesn't end in the confusion of the ORTF thread........

[pkautzsch]

Hoping not to confuse things:
As the theoretical (ideal) XY does the whole thing only by level differences, it is the same as panning, and an MS signal that has been converted to LR can be handled the same way as an XY signal. Any coincident setup (except Blumlein because of the negative phase lobe of the fig-8) can be panned and summed to mono freely without phase issues, in contrast to setups involving time differences.

What IS different between stereo "in place" recording and panned mono is that the acoustics of the room will sound more natural when recorded in stereo. The lack of naturalness doesn't matter too much, though, when applying pop-typical artificial reverb.

I think we've learned a lot about confusion in that ORTF thread...

[David Spearritt]

Any coincident setup (except Blumlein because of the negative phase lobe of the fig-8) can be panned and summed to mono freely without phase issues

Blumlein is also completely fine summed to mono, there are no phase issues. When summing to mono, phase issues arise from left/right phase anomolies not front back.

[sssaudio]

Originally Posted by Gilliland

However, MS and XY are interchangeable, at least in theory. With "perfect" microphones in "perfect" coincident postion, MS and XY can be shown to be mathematically identical - once you decode MS into normal stereo, the resulting stereo is exactly what you'd get if you had used XY in the first place. And it would be equally possible to convert XY to MS (though I can't see why you'd want to).

Yeah you're right about that, in theory. Though that's only assuming that the M mic is cardioid. It can also be omni or figure 8 if one was so inclined.
The cool thing though is that with DAWs you don't have to commit to bouncing the track to a stereo track. You can just send the input of hte M mic to one track and send the input of the S mic to two tracks with one out of phase. If you put those two S tracks into a group there's no reason really to bounce them down to stereo and commit to how you wanted it mixed as you would if you were recording M/S directly into a hardware M/S matrix and then into your recording media of choice. What I was getting at is that by doing it this way you really always maintain the control of S component whereas with XY it's already in it's stereo form.

[edwinhurwitz]

Originally Posted by Gilliland

However, MS and XY are interchangeable, at least in theory. With "perfect" microphones in "perfect" coincident postion, MS and XY can be shown to be mathematically identical - once you decode MS into normal stereo, the resulting stereo is exactly what you'd get if you had used XY in the first place. And it would be equally possible to convert XY to MS (though I can't see why you'd want to).

.

The reason to convert XY to MS is to be able to apply processing to either the mid or side of a stereo track separately. For instance, if you want to bring up a vocal that's in the middle, but you don't want to affect the guitar that's in the right channel, by coding into m/s, you can eq the m, then decode back into xy. All kinds of things can be done like this.

Edwin

[pkautzsch]

Originally Posted by David Spearritt

Blumlein is also completely fine summed to mono, there are no phase issues. When summing to mono, phase issues arise from left/right phase anomolies not front back.

I should have been more precise. I remember reading a discussion (don't know if it was here or on the Neumann forum) in which someone had set up a vocal group in a circle around a Blumlein pair. The gist of that discussion was:
As long as the source is in the intended angle of +-45° (ie at the max exactly on axis of one of the mics in the Blumlein setup) everything is fine. Problems can, however, arise when sound comes quite exactly from the side (+- 90°; some singers in that circle setup, but also room reflections). Then it's picked up by the positive lobe of one mic and equally strong by the negative lobe of the other mic, thus eliminating when mono-ed.

[David Spearritt]

Originally Posted by pkautzsch

As long as the source is in the intended angle of +-45° (ie at the max exactly on axis of one of the mics in the Blumlein setup) everything is fine.

But summing to mono, gives one single forward facing fig 8, no phase problems there.

Problems can, however, arise when sound comes quite exactly from the side (+- 90°; some singers in that circle setup, but also room reflections). Then it's picked up by the positive lobe of one mic and equally strong by the negative lobe of the other mic, thus eliminating when mono-ed.

But this is out of bounds for a Blumlein pair anyway, so its not relevant. There should be no sources in those areas. Room sounds coming in on the side quadrants don't matter either as they are incoherant.

[pkautzsch]

Thanks for pointing out the "single fig-8 when summing" thing.

Originally Posted by David Spearritt

There should be no sources in those areas

Exactly.
Should. Seemingly there are people who have singers in a circle around Blumlein, and who then encounter problems. If everyone knew about the setups they use, many threads in gearslutz wouldn't exist, and people wouldn't have all those experiments that, though seldom technically "correct", often sound interesting.
But we're getting philosophic here.