Mid/Side Mixing: Part One – Decoding M/S recordings
If you’ve never recorded using Mid/Side (M/S) techniques you should definitely give it a go. The basic premise is that the forward facing cardioid mic captures the direct sound whilst the figure-8 mic positioned at a 90-degree angle to the sound source captures side reflections from the room. Due to the deep ‘nulls’ in the polar pattern of a figure-8 mic, the direct sound will be all but excluded leaving only the room sound.
In the next section we will be talking about using a mono signal as well as a stereo signal to achieve the mid/side process, so let’s quickly clear up where these will come from. The forward facing cardioid microphone will, as expected, give you the first mono signal we talk about, the stereo signal we then discuss comes from your side microphone recording in a figure-8 pattern, this again will produce a mono signal so to produce the required stereo signal, first duplicate the mono side signal, and then invert the phase of one of them, these two can now be combined to create the hard panned stereo signal to be used alongside the mono signal from the front facing cardioid, leaving you with a mono and stereo file.
After you have recorded in M/S you need to decode the signal so it is useable. To do this, you can either use one of the many M/S decoder plug-ins available, or just set up some simple routing in Pro Tools – we’ll get on to this a little later.
M/S is a useful mixing tool, and the best thing is you don’t need to have recorded in M/S to take advantage of the technique. This tutorial will show you how to decode an M/S recording so that you can mix in M/S, and offer a few introductory hints and tips for mixing in M/S.
Decoding an M/S recording
You should be starting out with the on-axis mono recording and the figure-8 recording which will either be on two mono audio tracks or one stereo track depending on your recording setup. It is best to work with a mono and a stereo track as opposed to three mono tracks. So if you need to, now add a stereo audio track and by selecting the two figure-8 audio file regions from the regions list and dragging onto the new track you can create your stereo file.
Now we need to duplicate the stereo (figure-8 recorded) audio track and pan one track hard left and the other hard right.
Quick Tip: To duplicate a track, select it and use shortcut Shift_Option_D (mac) Shift_Alt_D (windows) or use the duplicate function in the track menu
The whole point of mixing in M/S is to be able to independently control the middle and wide elements of the stereo image. Right now, the stereo image you hear sounds OK, but it’s about to get a whole lot wider!
We need to flip the phase (invert) of the duplicated stereo audio track; there are a couple of ways of doing this so choose for yourself:
- Select the region in the edit window and click Audiosuite>Other>Invert – this will apply the phase inversion to the file itself – this is different to using a realtime plug-in as below
- Insert the ‘trim’ (or the1 band EQ) plug-in onto the duplicated stereo track and click the invert button
It’s Just a Phase…
At this point, you may wondering about phase and mono compatibility. Since we have identical material exactly 180° out of phase panned hard left and right respectively, if you switch your outputs to mono you shouldn’t hear anything at all!
This is why we should take care with the extent of M/S processing used – all that stereo width might sound awesome when played back over your lovely monitors (hopefully perfectly positioned in a 45° triangle to your listening position) but when your mix makes it big and people listen on their mono portable radio’s or over shop’s PA systems, all of that phase cancellation will ruin your hard work!
A lot of engineers use a phase correlation meter or graph inserted on their master fader so they can keep an eye on the the phase interactions and mono compatibility of their mixes. Pro Tools comes pre-loaded with the Phasescope plug-in which is definitely worth your while using; we produced a guide to the plug-in which you can read here.
A Word On Mixing in M/S
About now It would be wise to either group the two stereo audio tracks or route them to an auxiliary as I have; this will let you control the Mid and Side elements more easily.
It is best to drop the fader on the side element all together and slowly bring the fader up until you like what you are hearing; don’t forget to keep an eye on your mono compatibility too as the more side you use the less compatible your mix will become.
A fair few modern plug-ins come with a M/S option; Izotope’s excellent Ozone 4 mastering suite allows you to control the compressor, limiter, EQ, harmonic exciter and reverb units in either stereo or M/S – but why would you want to use this feature? Here’s a couple of examples to get the creative juices flowing:
- A final mix you are mastering is lacking brightness – you could boost a little high’s in the side element to increase the perceived width and shimmer.
- A final mix has a bassline that needs some extra compression – you could add some frequency specific compression over just the mid element – the same principle applies to vocals!
- You want your mix or indeed one track to have a little more ambience without detracting from the performance – you could add a little reverb over the side element.
Watch out for parts two and three where we will discuss converting a stereo recording into a useable M/S format and we will explore the various M/S plug-ins available for Pro Tools.
We would love to hear about your experiences with M/S processing or your thoughts on the tutorial and the suggestions above – please leave a message for us below or on our Twitter page.
Why use two stereo tracks for the sides? My understanding of mid/side is that you take the side mic, duplicate it, flip the phase on one, and then drag both to a stereo track. The sides are now on ONE stereo track. Making two stereo tracks is redundant, is it not?
Just a question, would this NOT work without a mid mic. Could you just use say a ribbon mic, duplicate the tracks, invert the phase and have a stereo image and control it with the amount of pan?