MS Stereo combines two different microphone characteristics and a special matrix to create the stereo image and localization. M is an abbreviation for Mid, and S for Side.
In the basic setup, a cardioid microphone capsule pointing forward provides the M signal. A bi-directional microphone (figure-of-eight) positioned at the same point but angled 90° provides the Side-signal. The positive lobe of the bi-directional microphone is pointing left.
It is not possible to monitor the stereo signal correctly without applying an MS-matrix.
To obtain the left / right stereo image, the M- and the S-signals are processed by simple addition and subtraction:
L = M + S
R = M – S
There are various ways to carry out this calculation: The illustration shows two ways to reach the LR-format: Using a mixer or using a transformer.
If using a mixer, feed the M-signal (the cardioid microphone) to one input and set the pan-pot to the center-position (providing an equal amount of the M-signal on the left and right channel). The S-signal from the bi-directional microphone is split equally into two parts. One part is fed into a channel and panned fully left. The other part is fed into another channel – phase inverted – and panned fully right. If this channel does not contain a phase-reversing button, use an external phase reversing cable (swap of pin 2/3 at one end).
Notice that most DAWs can perform the process directly, perhaps by applying plug-ins.
The interesting thing about MS recording is the fact that the stereo width can be adjusted, during – or even after – the recording takes place. It is all about the balance between M and S. Gaining M and reducing S makes a more narrow recording. In the extreme, full M and zero S provides clean mono. Gaining S and reducing M provides over width.
This principle works due to the convention that the in-phase part of the bi-directional lobe is pointing to the left. Providing the left channel (M+S) most left-side acoustic information adds up, whereas as right-side information more or less cancel out. Obtaining the right side channel by subtracting the signals cancels out left-side information and add up right-side information. (Remember: minus minus, equals plus.)