My mix feels narrow. Would mid/side processing help, and how?

Measure before widening

Read the correlation of the mix. Close to +1, the two channels are nearly identical and the mix is close to mono. Turning up the side signal then raises very little, and what it raises is reverb tails and noise. That mix needs width put into it: pan parts apart, double-track a guitar or a synth and pan the two takes left and right, or use a stereo reverb.

Between about +0.3 and +0.7 there is side content to work with, and mid/side processing is the right tool.

Mid/side moves that work

• A high shelf on the side only, 1 to 2 dB above 8 kHz. The mix sounds wider and more open, and the centre does not get brighter.
• A high-pass on the side at 100 to 150 Hz. Low frequencies in the side signal add width nobody hears and take level away in mono.
• A small cut in the mid where the centre is crowded, such as the low-mids under a vocal.
• A width control at 110 to 130 percent on the instrument bus or the master.

The limit

Every decibel added to the side is a decibel that disappears on a mono speaker. Check the mix in mono after each move, and keep the correlation above zero. A mix that is wide on headphones and hollow on a phone has been pushed too far.

In Zammoth

The engineer measures correlation for the whole mix and for each band, so it can say whether there is side content to raise before it changes anything:

Each EQ band can be set to act on the mid or the side. The Compressor can run on the mid or the side as well. Stereo Width scales the side from 0 to 2, the Stereo Imager does the same in three bands, and Band Width does it in one band you place. The Stereo Width device has keys for listening to the mid or the side alone. See the Stereo Imager reference and the EQ reference.