Quick question about using a Figure-8 mic for mid-side recording

[K8ch]

Hi,

I am recording an acoustic guitar using the M/S technique.
Since the mic records from both sides, should I record figure-8 mic as a stereo track?


Peace,

K8ch

[G-Sun]

Originally Posted by K8ch

Hi,

I am recording an acoustic guitar using the M/S technique.
Since the mic records from both sides, should I record figure-8 mic as a stereo track?


Peace,

K8ch

You need two mics for M/S recording.
One cardioid and one fig8.

[edwood75]

Hello,

The mic records from both sides but just outputs one mono signal.
In order to become a suitable mid side, you have to duplicate the figure of 8 track and invert the polarity of one of the channels (usually right side) and, of course, pan the signals. Or, you can just use a mid side encoder plugin like the JS (included in reaper)

[karbomusic]

You can record the mid mic on the left and the side on the right of a single stereo track and use a encoder/decoder on that track such as MSED or the JS MS effect.

[K8ch]

Originally Posted by G-Sun

You need two mics for M/S recording.
One cardioid and one fig8.

Oh yes...I'm aware of the need for the 2nd mic.
It just seemed that I should record the sides of the figure-8 as a stereo track so that it captures the left and rights sides, separately...as opposed to "combining" them into the one, single mono track.

K8ch

[Jae.Thomas]

no, you shouldn't. You have to flip the phase of one of the "side" tracks so you would PROBABLY need them on separate mono tracks.

however, the easiest way is to use a decoder plugin

[jerome_oneil]

Originally Posted by K8ch

Oh yes...I'm aware of the need for the 2nd mic.
It just seemed that I should record the sides of the figure-8 as a stereo track so that it captures the left and rights sides, separately...as opposed to "combining" them into the one, single mono track.

K8ch

No. Double the figure 8, and add one of those tracks to the cardioid signal. That's your "mid." Then take the other figure 8 and invert it's phase, and add it to the mid. That's your "side."

Edit: I'll also throw my vote in for MSED, or some other encoder. It's worth doing by hand once or twice just so you understand what is going on, though.

[Jae.Thomas]

Originally Posted by jerome_oneil

No. Double the figure 8, and add one of those tracks to the cardioid signal. That's your "mid." Then take the other figure 8 and invert it's phase, and add it to the mid. That's your "side."

Edit: I'll also throw my vote in for MSED, or some other encoder. It's worth doing by hand once or twice just so you understand what is going on, though.

I agree. I had a template for this. I still use it from time to time

[ashcat_lt]

Originally Posted by K8ch

I am recording an acoustic guitar using the M/S technique.

Why? ...

[Jae.Thomas]

Originally Posted by ashcat_lt

Why? ...

because of a psuedo-stereo sound without getting destroyed in mono?

[karbomusic]

Not sure where ashcat is going but IME, MS on acoustic really needs some distance or the stereo image won't ever seem to be quite 'right'. Which isn't a problem in a good sounding or proper room (because a bad room limits the distance you can use) but never came out the other end loving MS as much as I thought I would on a single acoustic guitar. Sounds good is good of course but something to be aware of just in case.

Would likely be more successful if recording a few musicians in a semicircle - I've done a one-man version of that before where I set the MS mic up assuming a semicircle of musicians then as I recorded each track, I just moved my chair to the next position in the circle.

It's worth doing by hand once or twice just so you understand what is going on, though.

I agree, gives you more control and a better understanding of the entire process. I don't always need the extra control but creating MS manually using three tracks really takes a lot of the mystery away.

[ashcat_lt]

Originally Posted by Jason Brian Merrill

because of a psuedo-stereo sound without getting destroyed in mono?

But it kind of will. The side signal goes away completely and the whole mix just has to change completely. Even more if you've done some different processing on each.

You could record an actual stereo image. If you do it correctly, it'll collapse to mono just fine, and maintain more of the original flavor in the process.

I'm not sure I care to argue the pros and cons, though I'd say that neither technique is likely to sound particularly natural if it's real close to the guitar, and if you're going to back it off, you really need a decent sounding room to make it worthwhile.

My real question was why somebody who doesn't understand the fundamentals of even how to plug in the mics involved thinks he needs to do this. A learning exercise is a good enough reason, but "because that's what everybody is doing" kind of isn't.

[karbomusic]

Originally Posted by ashcat_lt

But it kind of will. The side signal goes away completely and the whole mix just has to change completely. Even more if you've done some different processing on each.

Agree... Might add that the early days of MS (if memory serves) was more about capturing a stereo image of multiple people or a roof full of people. In that respect, that mic's track *was* the mix and the concern was simply being able to collapse that. Unlike someone trying to mic instruments using MS then mix them in a mix using mono compatibility as the reason. I could be wrong there but that's seems fitting; not that the latter isn't useful but in my experience is more special purpose than other stereo micing techniques.

[G-Sun]

Originally Posted by ashcat_lt

Why? ...

Originally Posted by Jason Brian Merrill

because of a psuedo-stereo sound without getting destroyed in mono?

It's not pseudo-stereo,
it's just as true stereo as XY.
And yes, mono-compatibility is one of then great strenght.

[jerome_oneil]

Originally Posted by karbomusic

Would likely be more successful if recording a few musicians in a semicircle - I've done a one-man version of that before where I set the MS mic up assuming a semicircle of musicians then as I recorded each track, I just moved my chair to the next position in the circle.

I've found that in noisy environments (bluegrass jams surrounded by a crowd, for example), processing mid-side, and then dropping the side channel completely is a nice way of reducing the overall noise of the recording.

[Jae.Thomas]

Originally Posted by G-Sun

It's not pseudo-stereo,
it's just as true stereo as XY.
And yes, mono-compatibility is one of then great strenght.

I don't find it to be ACTUAL stereo, because how would it "see" differences in the left and right? Does that happen since one "side" is out of phase? So differences become apparent?

[jerome_oneil]

Originally Posted by Jason Brian Merrill

I don't find it to be ACTUAL stereo, because how would it "see" differences in the left and right? Does that happen since one "side" is out of phase? So differences become apparent?

Its stereo ( 2 channels), but you're no longer dealing with a "right" and "left", which sometimes takes a bit of time to get your head around.

[Jae.Thomas]

it's more like Right and not right?

or its more about .... what's NOT in the middle? Rather than left and right?

[jerome_oneil]

Originally Posted by Jason Brian Merrill

it's more like Right and not right?

It's more like "middle" and "side."

or its more about .... what's NOT in the middle? Rather than left and right?

From a strictly technical standpoint, yes. The "mid" is "Stuff from the fig-8 added with stuff in the cardioid" and the "side" is "stuff from the fig-8 subtracted from the cardioid."

That's why you invert the phase on one of the fig-8 channels. It will negate any bleed from the cardioid signal. Once it's done, you don't adjust "width" with the pan pots like you would with classic stereo signal, you control it with the faders. If you want a "wider" result, you just raise the levels of the side channel.

[clepsydrae]

Originally Posted by karbomusic

Not sure where ashcat is going but IME, MS on acoustic really needs some distance or the stereo image won't ever seem to be quite 'right'.

I love M/S and record almost everything that way, including "mono" sources like vocals, but if I'm close to the M/S pair, the tiniest rotation of my head L/R will cause a dramatic swing L/R in the resultant image, so i gotta stay right in the middle. This makes sense if you think about it... and i think it explains why close-mic'ed guitars with M/S might sound weird: it pulls the L/R too far.

M/S in general isn't as dramatic a stereo image as other techniques, of course, but up close it can get weird.

OP: you can't record a figure-8 "in stereo" even if you wanted to. I.e. you can't avoid "combining" the two sides, because there aren't two sides: in a typical two-diaphragm mic the diaphragms are combined electrically in the capsule and one signal comes out. In a figure-8 ribbon mic, there is but one ribbon that records "both sides" at once. You can record that same signal to two tracks, with one phase inverted, and it might save a little time if you were manually doing the figure 8 matrixing in the DAW, but it's basically a bad idea because you're essentially making two copies of the same thing.

Originally Posted by ashcat_lt

But it kind of will. The side signal goes away completely and the whole mix just has to change completely.

I do disagree on this point in terms of M/S weaknesses: this is true of any stereo image: the antiphase components cancel and the signal changes. It's only a problem with M/S when you use too much of the side channel, which, admittedly, lots of people do, because it's easy to go overboard with it (ask me how i know :-) ). If properly balanced, though, it's no worse than any other stereo technique, IME. (Also, since M/S isn't as "spread-y" as other stereo techniques, people sometimes overdo it on the side channel to compensate.)

You could record an actual stereo image. If you do it correctly, it'll collapse to mono just fine, and maintain more of the original flavor in the process.

I'll throw in with those saying that M/S is just as "actual" as any other stereo method. It's not a trick, just a different way for a pair of mics to make a rough representation to our ears of 3D waves bouncing around in space.

Part of the value is that when/if the image collapses (as it largely does when played out of a boombox, for example), the resultant signal is a known quantity -- the mid mic -- as opposed to some less predictable consequence of combining two mics; it's a signal that was recorded on-axis. (You could close-mic with X-Y but now you're at 45 degrees -- may or may not be important, given the mic.)

I like to use M/S for giving "mono" signals a little body; e.g. with vocals or an amp or a flute, just a touch of the side gives it dimensionality, and when it collapses I have a close, on-axis, "perfect" signal. Equal results may well be achievable with other stereo techniques, but I like the predictability of M/S.

I also like the fact that M/S setups can be more compact and less intrusive for live recording.

Like X/Y, M/S is also coherent for arrival time, which can be cool, depending on your needs.

And note that any EQ you do on M/S signals really should be done with linear-phase EQ! It's not a subtle difference.

In terms of visualization, I made this image a while back to help folks understand how M/S encoding works:



...you can see how the "mid" signal (in orange) tracks the mid point of the left and right signals, and the "side" signal (pea green) tracks the difference between the left and right signals (e.g. all the places where L and R cross, "side" is at zero) -- it's indicated in two spots how the polarity of the side is determined; you can think of it as the distance from the mid signal to the left signal.

[clepsydrae]

Originally Posted by jerome_oneil

If you want a "wider" result, you just raise the levels of the side channel.

...and just beware that if you go too far, you get excessively anti-phase and your mix will suffer when it collapses to mono.

[karbomusic]

I love M/S and record almost everything that way, including "mono" sources like vocals, but if I'm close to the M/S pair, the tiniest rotation of my head L/R will cause a dramatic swing L/R in the resultant image, so i gotta stay right in the middle. This makes sense if you think about it... and i think it explains why close-mic'ed guitars with M/S might sound weird: it pulls the L/R too far.

That's pretty much the thing I've noticed - I call it sounding "crooked" where you just can't seem to make it sound balanced. I don't like it on acoustics if one has to mic too closely (aka defensively) to avoid room issues. I'm including the assumption that I'm recording stereo because I don't want to throw away the sides and go mono.

I have other spaced pair methods that work better for me because it creates a stereo acoustic track with just enough of a hole in the middle to nicely fit a vocal - it works as well because I can just drop a mic or collapse to mono with success because the mics are far enough from each other in relation to the source that phase is not a concern, and at least one of the mics usually produces a fine mono result if I actually had to do that.

Edit: Another thing I have done, which is quite fun, is to put the mid on the source and the side across the room like a room mic, adjust timing for phase alignment if needed, it's a creative what does this sound like thing though.

[clepsydrae]

Originally Posted by karbomusic

That's pretty much the thing I've noticed - I call it sounding "crooked" where you just can't seem to make it sound balanced. I don't like it on acoustics if one has to mic too closely (aka defensively) to avoid room issues.

Yeah, same here. Anything where the acoustic instrument is large and different tonalities are coming from different directions. I like close m/s on cello, though, because the alignment can put it in a symmetrical center of the sound (as opposed to the guitar). Come to think of it, i wonder how m/s would sound on guitar if you tilted it so that the fig-8 was pointing to either side of the neck, with the mid in the middle, over the 14th fret... of course if you move at all when you play it's a bummer. :-)

For me, another argument for m/s is mic'ing cabinets: given how precisely people position mics over speakers, you can line your mid mic up just right and still get a stereo recording.

[alanofoz]

Sorry, but I have to say that a fair bit of the above is technically wrong or misleading, a notable exception being clepsydrae's post #20. You could just read that post and be done.

Although I will add that converting to stereo, i.e. left & right needs to be carried out; MSED is as good a method as any.

A disadvantage of X/Y recording is that the two mikes need to be well matched or the stereo image can be blurred. No such problem with M/S.

I really like the results I get using M/S on an acoustic guitar, but that could just as easily be a lucky combination of mikes.

[karbomusic]

Originally Posted by clepsydrae

Yeah, same here. Anything where the acoustic instrument is large and different tonalities are coming from different directions. I like close m/s on cello, though, because the alignment can put it in a symmetrical center of the sound (as opposed to the guitar). Come to think of it, i wonder how m/s would sound on guitar if you tilted it so that the fig-8 was pointing to either side of the neck, with the mid in the middle, over the 14th fret... of course if you move at all when you play it's a bummer. :-)

For me, another argument for m/s is mic'ing cabinets: given how precisely people position mics over speakers, you can line your mid mic up just right and still get a stereo recording.

I know that the majority of the M/S experiments I carried out were in a less than ideal room, long before I treated it, which forced me to mic closer than I would like to - which probably gave me a worse taste in my mouth than it should have. Meaning, I've used it less since then while at the same time, recommending using a good room so I should revisit it some.

I performed probably a hundred different micing tests in that room, each as a Reaper project documenting mics/configuration/instrument/positon/amp etc. However the majority were done before the room was up to snuff and laziness ensued on doing it again. Found some interesting stuff either way.

I don't think I added previously that I record band rehearsal with my Zoom Q8 and the detachable M/S mic 'seems' to do a bit better than the X/Y based on how we are setup and distance from the sources. I actually prefer the M/S in that regard. Would it be wrong to consider the side being on axis to the far left/right components or does the process acoustically result in the same amount of off-axis penalty in the final result if there is such a thing?

[clepsydrae]

Would it be wrong to consider the side being on axis to the far left/right components or does the process acoustically result in the same amount of off-axis penalty in the final result if there is such a thing?

Hmmm... dunno... I'm having trouble thinking through that one :-), what with the antiphase lobe, orientation of the mics, common recording scnearios...

Wes Dooley/Ronald Streicher's AES paper on M/S seems to agree with you:

"Another related benefit is that with the M microphone aimed directly at the centerline of the sound source and substantially on axis to it, the midsection of the source is not subjected to the off-axis coloration of image ambiguity common to many other stereo techniques (assuming of course, good frequency and polar characteristics of the M microphone). Similarly, since the reverberant sound field is largely on axis to the S microphone, the coloration of the reverberant field is also reduced."

Incidentally, I like this tool for visualizing stereo pickup patterns:

http://www.sengpielaudio.com/Visualization-AB60-E.htm

(no mid/side unfortunately)

[karbomusic]

Originally Posted by clepsydrae

.....Similarly, since the reverberant sound field is largely on axis to the S microphone, the coloration of the reverberant field is also reduced."

Cool, I was hesitant to vocalize such a conclusion but ^that is pretty much what I thought I had noticed and why I kept using it. We rehearse in a bit of a circle with everyone facing in which places a couple of amps near the -90 and +90 degree points assuming mid is 0 degrees and had at least thought those seemed a little less colored than I usually get with X/Y.

Incidentally, I like this tool for visualizing stereo pickup patterns:

http://www.sengpielaudio.com/Visualization-AB60-E.htm

(no mid/side unfortunately)

Handy link, I'll bookmark it and save myself some time next time.

[clepsydrae]

Originally Posted by karbomusic

Handy link, I'll bookmark it and save myself some time next time.

I like how it shows the distribution of the virtual sources in the perceived vs. actual fields. Handy to get an idea of what it might sound like. (Not as handy as trying in reality, of course. :-) )

[EpicSounds]

record M/S as a single stereo file or two mono. Either way you need to decode if your preamp doesn't do that for you.

I do like M/S for acoustic guitar but not too close. 3-6ft away works well.

[G-Sun]

Originally Posted by Jason Brian Merrill

it's more like Right and not right?

or its more about .... what's NOT in the middle? Rather than left and right?

One could argue that normal mono, played back on a regular stereo-system is pseudo mono,
as there is an illusion of sound coming from the center.

[Jae.Thomas]

Originally Posted by G-Sun

One could argue that normal mono, played back on a regular stereo-system is pseudo mono,
as there is an illusion of sound coming from the center.

man this is getting all philosophical

[jerome_oneil]

Originally Posted by Jason Brian Merrill

man this is getting all philosophical

There is no middle.

[James HE]

In general I do not mind the stereo field being a little lopsided when using a M/S pair for acoustic guitar, I almost prefer it not so balanced. When you sit with an acoustic guitar (in a non classical position at least) the sound hole is much closer to one ear than the other, and you also feel the vibration in your body a bit stronger on one side as well.

In a dense mix with other instruments, that lopsidedness is mostly an asset anyways, helps the guitar sit in the mix really well.

The best thing to me about setting up a M/S pair for acoustic is that you get to just set it up and forget it. You can can just vary the performers position relative to the mic to get a different sound and you don't have to worry about repositioning mics so much.

[K8ch]

Originally Posted by ashcat_lt

My real question was why somebody who doesn't understand the fundamentals of even how to plug in the mics involved thinks he needs to do this. A learning exercise is a good enough reason, but "because that's what everybody is doing" kind of isn't.

Why would my inquiry compel you to engage in name-calling?
It was not an unreasonable question, and it was posted in the "newbieland" forum, where it belonged.
Apparently you're not burdened with well-developed social skills.

[G-Sun]

Originally Posted by K8ch

It was not an unreasonable question, and it was posted in the "newbieland" forum, where it belonged.

No question is wrong in the newbie-forum,
and using m/s is valid for stereo-recording
(and has the benefit of superior mono-compatibility)

That said, it's not the most common technique to record a guitar.

[K8ch]

Thanks for your response. I appreciate it.

Yes, I understand it's not the typical way to record an acoustic guitar.
I also understand it can be a very effective way to record, so, I thought I'd give it a try to see what kind of results I am able to get because that's how I learn.

In addition to all that, I would prefer my recordings to collapse to mono well...and since the guitar usually plays a very prominent role in the music I write, it seemed like a worthwhile thing to check out.


Anyway..thank you for responding in a civil manner.
I'll just keep "ashcat IL" in my prayers.




Peace,

Keith

[G-Sun]

Originally Posted by K8ch

Thanks for your response. I appreciate it.

Yes, I understand it's not the typical way to record an acoustic guitar.
I also understand it can be a very effective way to record, so, I thought I'd give it a try to see what kind of results I am able to get because that's how I learn.

In addition to all that, I would prefer my recordings to collapse to mono well...and since the guitar usually plays a very prominent role in the music I write, it seemed like a worthwhile thing to check out.

Yes, indeed check it out.
That's how we learn.
Now, experiment with spaced pair. That's fun

[ashcat_lt]

Originally Posted by K8ch

Why would my inquiry compel you to engage in name-calling?
It was not an unreasonable question, and it was posted in the "newbieland" forum, where it belonged.
Apparently you're not burdened with well-developed social skills.

Wasn't calling names. It was a legitimate question. Like I said "That's how I learn" is a reasonable answer. Getting all defensive and questioning my social skills really isn't.

[Lucian]

I love M/S recording classical guitar and have run into the "crooked" issue where one side channel is louder than the other. Rather than boosting the low side I experimented using phasebug on the mid channel and usually +/- 6 degrees of phase correction evens everything out for me. I hope this is not blasphemy haha. I asked about this in the recording forum but never got a reply.
Anyway, it works for me and makes sense but I am no pro.

[clepsydrae]

Neat! -- my brain is melting a little trying to reckon with the effects of intentional phase adjustment of the mid -- i'll have to spend some time with a scope plugin and play around. Thanks for the idea.