What is Mid/Side processing?
Mid/Side processing is a special way to process stereo signals. This stereo signal consists of 2 mono signals, one for the left and one for the right channel. However, when we create this stereo signal, it is likely that some sounds will occur on the left or on the right side. Nevertheless, we can pan some signals to the center. The concept of Mid/Side processing is to process the center or mid sounds separately from the sounds on the sides. For example, we might want to apply an equalizer to the sounds that are in the center of the mix and a compressor to the sounds that are positioned on the sides of our mix.
Use it with other processing tools
On its own, Mid/Side processing does not actually change the signal. I recommend using it with some other processing tool like equalization or compression. When using Mid/Side processing, a conventional stereo signal with a left and right channels is separated into a mid and sides channel. This way it is possible to equalize or compress the sounds in the center of a mix independently of the sounds on the sides.
Inside Mid/Side processing
To furthermore understand the process of Mid/Side processing, we first need to take a look at the three principal ways a mastering engineer can deselect a given track. The first and the most immediate deselection tools are, of course, the left and right channels of a conventional audio signal. Mastering equalization, for example, might differentiate between the two channels of an L/R stereo mix. Perhaps applying slightly more highs to a left-side than to right. Another dissection tool is multi-band processing. Here, a two-track stereo mix is divided up into three or four different frequency bands. With multi-band processing, an audio engineer can decide to apply more compression to one frequency band than another. He can shape both the timbre and dynamics of the end master.
Dissecting a track thru Mid/Side processing is essentially a “third dimension” of signal processing. However, to understand the Mid/Side we need to understand a new way of looking at the stereo. As we know, a stereo playback is a neat audio trick, using two speakers to create a 180 degrees soundstage. What’s particularly interesting about this two-way speaker setup is that the center of the soundstage is created by the two speakers having a degree of “shared” information. In short, a signal fed at equal amplitude to the left and right speaker creates a “phantom center” image. Almost as if there’s a third speaker positioned equidistant between the two speakers. In effect, the mid signal is the sum of the left and right channels.
Controlling the master with M/S
Let’s look at the middle of the soundstage as the sum of the left and the right channels. It logically follows that the two “extreme” sides are formed by the difference between the left and right channels. Now we are looking at the stereo image in a new way-formed from the sum. The difference between the two channels with discrete controls for the mid and side of the master. By controlling the master using mid and side controls (rather the left and right), we yield new ways of flexibility. We can differentiate between EQ and compression applied to the instruments in the center of the mix (such as vocals) and signals that exist on the sides (such as cymbals, overheads, reverbs and so on).