Mid Side Miking Technique

[Frank Kruse]

Great question. I used to use coincident techniques quite frequently to mike things like drum overheads and various area miked situations, but I found the results to be only sort of OK... not obviously wrong, but not totally satisfying either.

I experimented with spaced mike tchniques such as ORTF, and found that the resulting recordings had a lot more depth and detail than my previous recordings, so I decided to abandon coincident techniques.

Yes, I think mono compatibility is a worthwhile thing, but at the same time, spaced mikes can give such a larger sense of space and dimension, so I feel that they're worth using.

So, to answer your direct question, I think that coincident techniques rob the listener of information that could have been recorded by a spaced mike technique. I know that not every engineer will hear things the way I do or strive for results that I feel are worthwhile, so I don't want to say that my way is the only way.

before this arguement gets totally rediculous, let me state that all I wanted to say in my first posting to this thread is how the MS technique works and what the advantages over the other stereo techniques are.
I just went through almost the exact same arguement over in the sound designer´s mailig list. I simply wanted to explain the theory behind MS and what is good in MS. I have used all sorts of recording setups depending on what I record so there is no need to convince me that MS is "sort of OK", "boring" or lacks depth. Especially the latter can be read in every book about micing techniques that MS offers great L,R localisation but offers less precise depth than micing setups that induce delay between the two mics like AB, ORTF, OSS, etc. etc. etc.

If you think MS is crap well so be it. But: referring to the original question of the first poster: MS is as "real" as AB, ORTF and all the others or as "unreal". You can chose. Because none of the stereo setups project reality by 100% the argument is rather useless.

[TLmix]

check out this link. It explains in detail, but to sum it up.... the center cardiod mic is recored to 2 tracks... center. the figure 8 mic is recored to L in phase.. ( make sure the front of the mic faces left) the back of the mic (facing right) is phase reversed and recored to the right channel. This is how the immaging works. a center signal is disregarded by the figure 8 mic. If a signal is on the left (front side of the fig 8 mic) is sums with the center mic on the left side... and cancels with the out of phase fig 8 on the right.. thus panning the signal to the left. If a sound comes from the right ( the back of the fig 8 mic that has been phase inverted) Since the mic has been phase inverted, the back of the mic now outputs positive phase information on the right side, thus summing with the center mic on the right/ canceling on the left, and causing the sound to pan right.

I hope that makes sense... if not, check out the link

TL

http://www.paia.com/msmicwrk.htm

[Andre Knecht]

before this arguement gets totally rediculous, […] If you think MS is crap well so be it. But: referring to the original question of the first poster: MS is as "real" as AB, ORTF and all the others or as "unreal". You can chose. Because none of the stereo setups project reality by 100% the argument is rather useless.

Frankie baby… it’s time for your chill pill.

Reading (and re-reading) the whole thread leaves me with the distinct impression that you actually might be looking for an argument. Moreover, you’re the only one having an argument. Monte is a big boy and doesn’t need others to come to his defense. Nonetheless, he never referred to MS - or anything else for that matter - as being crap, nor did he make any disparaging comments about the various solutions. He stated his personal findings in comparing stereo mic techniques. Your questions to Monte, on the other hand, struck me as being confrontational from the outset.

While I am familiar with the techniques being discussed here, and have formed my own opinions about each, I’m always intrigued to hear/read other fellow engineer’s take on the same. As such, I found all posts in this thread to be interesting and useful… until your last one.

Out of curiosity, how much love did you get “over in the sound designer´s mailing list?”

[Frank Kruse]

Andre baby,

I don´t need a chill pill at all. I just think that we are talking about two different things.
In the beginning someone was having problems with an MS decoding. I chimed in and clarified some things about MS and how it works. Then all of a sudden other people chimed in stating that MS is actually not as good as other techniques. Fine.

When I said "rediculous" I meant that I felt that we were talking about two different things that had nothing to do with each other wihtout stating that either one was wrong or right and that we were about to get into a serious arguement about two different topics one about how MS works and one about wether it´s better or worse than other stereo techniques.

Maybe the line that ended with "..so be it" was a bit harsh, I admit.

Your last question I´ll feel free to ignore if you don´t mind. I don´t understand the word "love" maybe a side effect of the pills...

frank.

[Bryan Cook]

for anyone who cares, M/S really is true stereo also because the figure 8 "S" mic gets positive and negative acoustic energy from the "front" and "back" of the mic, respectively. copying the figure 8 mic signal, and flipping one out of phase decodes this postive/negative acoustic energy from a once mixed signal. it was acoustic phase until the mic picked it up, its electrical polarity at this point.

"fake stereo" would be having a mono mid mic and a non-figure 8 mic as the S. then you would be flipping a mono signal out of phase and just getting a phasey signal around your mono mic.

a cool thing to try when doing M/S - try an omni mic as your mono/middle source!

[Stan Cotey]

Hey Monte,

Out of curiosity, how do you feel about mono buttons and spaced pairs?

Regarding MS, one of the things I like best about it is that if the thing that you are recording is in front of the mic, it's mostly represented on-axis by the M mic. With crossed figure-of-8's, any source directly in front of the recording setup is actually a bit off-axis to either of the mics.

I think each technique (ORTF, M-S, Blumlein, Decca Trees, etc.) has its applications, plusses and minuses and I'd choose different techniques depending on what I was recording, the room I was in, how much ambience I was trying to capture, the availability of mics, etc.

Regarding decoding M-S in Pro Tools, I do what was described ealier (duplicating the S track, inverting the polarity of one and panning them hard left and right). I then make a mix group of the two S tracks to link the faders.

When working with M-S decoding, I'll start by setting the M level first, and then bringin up the linked S tracks until I hear the stereo width I'm looking for. I think it's pretty important to hit the mono button from time to time while doing this; you can create a greatly enhanced stereo image with an elevated S level but the price of it will be low volume in mono.

Regarding compression, I use stereo compressors with M-S signals with no problem, as long as the compressor has truly linked side chains. I don't like compressing the M and S signals independently, due the the size of the stereo image changing with varying levels of gain reduction.

Nice thread.

- Stan

[dBgogogo]

Having a mono button in PT would be very helpful.

[Andre Knecht]

Having a mono button in PT would be very helpful.

True.

As a workaround, I’ve been inserting a bypassed Waves S1* with the image control collapsed to mono. When I need to check mono, I simply defeat the plug’s bypass. All my stereo templates feature the above on my “monitoring out” master fader.

* But any stereo plug in featuring panning controls for its two channels will do just a swell. If you run out of DSP, deactivate the plug instead of bypassing it.

IHTH.

[Monte McGuire]

Hey Monte,

Out of curiosity, how do you feel about mono buttons and spaced pairs?

With a spaced pair and sounds recorded in in the center of the image, there's little time difference between the two mikes, so summing to mono doesn't cause many problems. Of course, sounds to the sides of the image will get a slight time delay into the two mikes and will yield a comb filter effect when summed to mono, but hopefully, that isn't the bulk of the signal.

In practice, it doesn't seem to be a huge problem to sum a spaced pair like an ORTF array to mono.

Regarding MS, one of the things I like best about it is that if the thing that you are recording is in front of the mic, it's mostly represented on-axis by the M mic.

MS does give you this advantage over XY, and it means that you can use a mike with a less than ideal polar pattern and still end up with the center informattion sounding good. However, to do any of these array techniques well, you really need a mike that has a smooth polar pattern and an off axis response that's essentially flat, so it's not always easy to do in practice.

A mike like the KM-84 has a really nice polar pattern in this respect. The front half hemisphere has pretty much the same frequency response, making it work really well for all of these sorts of array techniques. Most small capsule mikes are only a little worse, especially at high frequencies, but many large capsule mikes have very significant colorations at 90 degrees, making them less useful for arrays.

With crossed figure-of-8's, any source directly in front of the recording setup is actually a bit off-axis to either of the mics.

Yes, but the nice thing is that many true figure eight mikes (like quality ribbons) have very good off axis response, so it's not a huge problem. A figure eight made with a pair of back to back large capsule cardioids (like the AKG C414 or Neumann U87 style capsules) won't have this advantage, since you still have the inherent "slop" in the polar pattern. A mike like the KM-86 that uses back to back cardioids that actually have a nice polar pattern and uniform response seems to solve this problem.

I think each technique (ORTF, M-S, Blumlein, Decca Trees, etc.) has its applications, plusses and minuses and I'd choose different techniques depending on what I was recording, the room I was in, how much ambience I was trying to capture, the availability of mics, etc.

Right... as always, your tastes, the room in question, the music being recorded and your available equipment always do matter. For example, if you don't have two of the same kind of mikes, then MS can solve that problem pretty well whereas XY is going to work pretty poorly using a different mike for each channel.

When working with M-S decoding, I'll start by setting the M level first, and then bringin up the linked S tracks until I hear the stereo width I'm looking for. I think it's pretty important to hit the mono button from time to time while doing this; you can create a greatly enhanced stereo image with an elevated S level but the price of it will be low volume in mono.

If you're stuck without a mono button, you can just mute the S channels and as long as your decoding gains are accurate, you'll end up with a pretty true picture of what mono sounds like.

Regarding compression, I use stereo compressors with M-S signals with no problem, as long as the compressor has truly linked side chains. I don't like compressing the M and S signals independently, due the the size of the stereo image changing with varying levels of gain reduction.

It's important to use a linked compressor, if you don't want any image shiift, but then again, the image shift you get from an unlinked compressor working on the M and S signals separately is an interesting sound too. Usually, there's more dynamics in the M channel, so compressing them separately causes the image to widen and the relative amount of "room" to increase during signal peaks, so that might be useful in some situations. It really depends whether accuracy or just "good sound" is important. If the latter, than you don't have to worry about the rules if what you get sounds good...

Nice thread.

Indeed!

-monte-

[punk]

I built a MS matrix using the schematic at the jensen xformer site as a guide:

http://www.jensen-transformers.com/as/as065.pdf

I put the mics through the decoder before the preamp.

One advantage of MS is that you can use different mics. So, if you're on a small budget and your best mics happen to be your matched pair, you don't have to commit them both to overheads.

just my 2 cents. YMMV.

byron