Topic: M/S vs Hypers/Wides/etc for Live Recording?

[stevetoney]

In researching our passion a bit more and exploring the possibility of getting into mid-side recording, a question comes up that I suppose has been asked before. 

So...mid-side recording gives one the ability to 'dial in' the width of the stereo image by effectively mixing in the side (or reverberant) component.  Isn't that effectively what we're trying to do in a live recording environment when we decide before the show if the environment would be best suited to hypercardioids or wide cardioids? 

What I'm getting at here with this question is that, my decision to run hypers (when I had them) was largely dependent upon my assessment of how boomy the room would sound.  The boomier the room, the more apt I would be to go with hypers.  While I didn't ever have them, I've read and it follows that wide cards are the opposite...the closer to the source you are the better wides would be to use.

So, wouldn't m/s recording effectively achieve the same goals as switching capsules around, but provide for a potentially better recording scenario since m/s should later enable me to vary the overall stereo and reverb effects to my liking instead of being locked in after I make a decision and commit to a specific capsule pattern?

[dbindc]

What I'm getting at here with this question is that, my decision to run hypers (when I had them) was largely dependent upon my assessment of how boomy the room would sound.  The boomier the room, the more apt I would be to go with hypers.  While I didn't ever have them, I've read and it follows that wide cards are the opposite...the closer to the source you are the better wides would be to use.

So, wouldn't m/s recording effectively achieve the same goals as switching capsules around, but provide for a potentially better recording scenario since m/s should later enable me to vary the overall stereo and reverb effects to my liking instead of being locked in after I make a decision and commit to a specific capsule pattern?

In theory maybe yes, but what you also have to consider is that you will be picking up a lot more of the sound directly around where you are set up running MS.  The  figure 8 capsule is firing directly to your right and left and it's going to pick up a lot more chatter and reverberated sound than say two hypers at 90 degrees aimed toward the stacks.  You could I suppose narrow down the soundstage in post by mixing in a lot more mid, but from my experience so far in a few boomy rooms you are far better off using the directional pattern mics unless you can setup onstage or very FOB. 

[stevetoney]

In theory maybe yes, but what you also have to consider is that you will be picking up a lot more of the sound directly around where you are set up running MS.  The  figure 8 capsule is firing directly to your right and left and it's going to pick up a lot more chatter and reverberated sound than say two hypers at 90 degrees aimed toward the stacks.  You could I suppose narrow down the soundstage in post by mixing in a lot more mid, but from my experience so far in a few boomy rooms you are far better off using the directional pattern mics unless you can setup onstage or very FOB. 

Is what you are saying based on your experience or based on theory?

The reason I am asking this is that I thought that anything that hits the figure 8 capsule but does not hit the mid channel can be eliminated from the final mix because when you reverse the polarity on the plus and minus, the side stuff drops out and you won't hear it.  I was thinking that means that any chatter coming only from the sides can be largely filtered out of the final stereo mix because it will be phase cancelled.  Same for room reverb.

[Will_S]

Based on a mix of practice and theory:

I like m/s in a lot of situations.  But, you don't really control width and reverb independently.  If you want to cut out more side, you must reduce width and reverb simultaneously.  If you want a wider image, reverb goes up as well.

That said, you probably do have a better chance of getting things right in post, because even though you have to deal with a fixed set of width/reverb tradeoffs (set by your mic position, and initial choice of pattern for the mid capsule), at least you can try out a bunch of different combinations and pick the best compromise.

So if you know you are in a horribly boomy or chatty location, I think you're better served running hypers at a narrower angle (m/s mixes down to hypers at ~110° mixed 1:1, boosting the mid narrows the angle between virtual mics but also makes their patterns fatter).  But if you are somewhere that might range from acceptable to great sound, I like the flexibility of M/S, because in post you can go for a wide pattern if the location ended up sounding great and a narrower pattern if it turned out not so hot.

[Todd R]

It doesn't really matter that the fig8 is firing directly right and left, since you would be mixing it so that it picks up the same amount that the hypers or cards or whatever that you would be mimicing, so if you had those actual mics you would get that same amount of side information.

But the issue that this really gets at, the side firing fig8, is that with MS the included angle between the virtual mics is set depending on whether you are mimicing hypers vs cards, etc.  So if you want to mix down to hypers, you will have a smaller included angle, and if you want to mix down to subcards you will have a larger included angle.  This is one of the biggest limitations of MS recording, as far as I can tell.

All that said, schoeps (I think) has developed the technique of double-MS recording that uses a card forward and a card rearword, together with the fig8, that allows you to independently choose mic pattern and included angle as you mix down.  There's been a lot of discussion on ts.com on the double MS technique.

[Will_S]

So if you want to mix down to hypers, you will have a smaller included angle, and if you want to mix down to subcards you will have a larger included angle.

I think this is backwards.  With a cardioid mid, the smaller the included angle, the more cardioid-like (fatter) the virtual mics, and the larger the included angle, the more figure 8-like (directional).  You can't get anything fatter than a virtual cardioid out with a cardioid mid.

With an omni mid, again the widest angle (all side) will be virtual cardioids, and as you add more mid you get more of a subcardioid, with a narrower included angle.

[stevetoney]

It doesn't really matter that the fig8 is firing directly right and left, since you would be mixing it so that it picks up the same amount that the hypers or cards or whatever that you would be mimicing, so if you had those actual mics you would get that same amount of side information.

But the issue that this really gets at, the side firing fig8, is that with MS the included angle between the virtual mics is set depending on whether you are mimicing hypers vs cards, etc.  So if you want to mix down to hypers, you will have a smaller included angle, and if you want to mix down to subcards you will have a larger included angle.  This is one of the biggest limitations of MS recording, as far as I can tell.
All that said, schoeps (I think) has developed the technique of double-MS recording that uses a card forward and a card rearword, together with the fig8, that allows you to independently choose mic pattern and included angle as you mix down.  There's been a lot of discussion on ts.com on the double MS technique.

Todd, thanks for the response.  It's logical and I understand everything you said, but I don't understand the conclusion that I've highlighted in bold.  Why is that considered a limitation to M/S recording.  Wouldn't having flexibility to mix to virtual hyper sound or virtual subcard sound be a desirable flexibility rather than a limitation?

[Brian Skalinder]

When I'm reasonably close, I like M/S because of the option to retain some control over the stereo width / ratio of direct to indirect sound when decoding M/S to stereo.  IME, the problem, as I move farther back from the source -- especially in crappy sounding rooms -- is that the more I dial back the Side to remove indirect sound and "room boom", the more mono the recording becomes.  I generally don't prefer mono-ish recordings.  Even if the PA is mono, a stereo environment (room, crowd, etc.) is a big part of what makes a recording "come alive" for me.  So I generally try to avoid mono-ish recordings, if possible.

In cruddy sounding rooms -- at a certain distance -- I generally prefer spaced cards (hypers would work, too) because it provides better stereo imaging while still minimizing indirect sound and "room boom".  The trade-off, of course:  do I want better stereo imaging (spaced) or better indirect sound and "room boom" rejection (M/S).  At a certain distance, in a cruddy sounding room, it becomes increasingly difficult to achieve both.  In these cases, I'll run the cards with wider spacing and smaller included angle.  The greater spacing provides difference-in-time stereo (which M/S does not offer) while the smaller included angle provides greater direct sound relative to indirect sound (though not as much as reducing or removing the Side in M/S, unless the included angle approaches 0º).  Now, running spaced cards, I've lost the flexibility of decoding in post, but in these cases I basically know that the only way I'll get a suitable recording M/S is to dial down the Side so much that the recording becomes very mono.

(FWIW, I've used Stereophonic Zoom as a rough guideline for getting started with my spacing / angles, and adjusted accordingly to my tastes and environment.  I've long seen tapers running wide spaced cards with a very small included angle, and I'd never really had anyone explain effectively why (think "dude, it just works"), until I read Stereophonic Zoom.)

Another key question to consider:  does one prefer difference-in-intensity stereo (coincident configs, like M/S, XY) or difference-in-time stereo (near-coincident or spaced configs).  Some people simply prefer difference-in-time stereo, while others prefer difference-in-intensity stereo.  My personal preference:  the closer I am to the sound source, the more I prefer difference-in-intensity;  the farther I am from the sound source, the more I prefer difference-in-time.

Not sure if that addresses your questions or not, or even helps, but...$0.02.

[stevetoney]

^ Of course it helps.  Awesome response.  Thanks alot Brian.

[dbindc]

Any conclusions I've made are purely based on my rather limited experience with MS rather than any sound theory knowledge.  I only got the figure 8 cap a few months ago and have done maybe 7-8 recordings with it.  But as far as recording MS in less than stellar sounding rooms, Brian's post pretty much sums up my perceptions.

[stevetoney]

Any conclusions I've made are purely based on my rather limited experience with MS rather than any sound theory knowledge.  I only got the figure 8 cap a few months ago and have done maybe 7-8 recordings with it.  But as far as recording MS in less than stellar sounding rooms, Brian's post pretty much sums up my perceptions.

Thanks dbindc and everyone else.  Being new to this technique, I had done some studying the last two days and the answers all of you provided really helped me to understand how best to apply it.  I know at some point I'll pick up that ak20, it's just a matter of when. 

[JD]

Rather than start a new thread I'll throw this M/S question out here. Sorry for the thread-jacking

Has anyone ever tried to use a mono SBD feed for the Mid channel on a M/S recording?

Would this be a good way of adding some ambiance to a mono recording?

I keep toying around with this idea for a festival I record, where I must stay near the board and there is too much chatter for a straight audience recording. Plugging into the board is no problem, but it's a mono feed and sounds pretty lifeless. Also, I'd rather not deal with running more than 2 tracks.

[Gutbucket]

Brian nailed it while I was typing, so I've deleted most of it..

But also keep in mind that you can manipulate the mid/side ratio of any stereo recording using any mic technique after the fact.  If the recording was not mic'd as mid & side, you only need to encode it to M/S first.  Technically, its foolproof with any coincident mic technique, but I often find it useful for spaced mics to perfect the soundstage, theory be damned, use your ears.

As for crowd pickup- IME, if the mics are on a stand above the crowd, the figure-8 will be picking up the side reverberant sound of the room strongly, but can actually pickup less nearby chatter because it has its null pointing straight down, backwards and forward and increases in sensitivity as the input angle approaches horizontal towards the sides (above eveyone's heads).  It would be somewhat similar in that aspect to a cardioid pointed at the ceiling, but a cardioid would also pick up chatter from in front and behind.

Disclaimer- I rarely use hypers, I've only used M/S up close & if forced to choose, I have a general time-difference preference.

[Gutbucket]

Has anyone ever tried to use a mono SBD feed for the Mid channel on a M/S recording?

Hmmm. Maybe. You'd have to use a figure 8 as the single mic and closely time-align the two tracks. You might be better off just running stereo mics but I'd like to hear it.  Interesting idea.

[fotoralf.be]

You'd have to use a figure 8 as the single mic and closely time-align the two tracks.

I see another prob there. This would only work as long as the fig-8 is behind or in front of the source of the M signal. Shift it, say, to the let and you'd have to delay the S signal components coming from the left and advance those coming from the right. I mean, if you have a recipe for doing this, I'd be all too glad to hear about that. Might help to solve a bunch of related probs. ;-)

Pending that solution, I guess we all would be well advised to stick to coincident MS arrangements.

Ralf