Mid/side processing: EQ, compression and mono checks
Last updated 6 min readTags:MixingMasteringStereo
Learn what mid and side contain, when to use M/S EQ or compression, and how to check changes to tone, width and mono playback.
Mid/side processing lets you EQ, compress or adjust the level of a stereo signal’s mid and side components separately. It is useful when a tonal or dynamic problem needs a different adjustment in each component; for example, harshness in wide percussion while a centred vocal already sounds right.
The important limitation is that mid and side do not isolate instruments. A side EQ boost may leave a dry, centred vocal untouched, but it can also brighten guitars, cymbals and vocal reverb.
What do mid and side contain?
Mid/side, often shortened to M/S, represents the same stereo signal in a different form:
- Mid is the sum of left and right. It contains centred sounds and contributions from sounds panned elsewhere.
- Side is the difference between left and right. Identical audio at the same level and polarity in both channels cancels here.
A dry mono vocal panned to the centre appears only in mid. A guitar panned fully left appears in both mid and side. A stereo reverb will usually contribute to both. Side is one difference signal, rather than separate signals for the left and right edges. [1]
For a gain-compensated example:
M = (L + R) / 2
S = (L - R) / 2
L = M + S
R = M - S
The final two equations reconstruct left and right. With no processing between encoding and decoding, the result is the original signal. Plug-ins may distribute the gain compensation differently, but a correctly configured M/S chain should pass audio unchanged with neutral settings.
When is mid/side processing useful?
M/S is worth trying when ordinary stereo processing improves one part of the image but spoils another. If the whole mix is too bright, a stereo EQ may be sufficient. If one guitar is too bright and you have its track, adjust that track first.
On a stereo bus or finished mix, these are reasonable starting points:
| What you hear | Possible adjustment | What to listen for |
|---|---|---|
| Harsh wide percussion, with a well-balanced centred vocal | A small side EQ cut in the harsh range | Softer percussion without dulling guitars or ambience |
| Excess low-frequency stereo information | Reduce low frequencies in side | A narrower low end without losing useful bass texture |
| Wide layers competing with the lead | Lower side slightly | A clearer lead without an unnecessarily narrow mix |
| A centred vocal sounds harsh | A small mid EQ cut | Less harshness without weakening snare, guitars or other material in mid |
| The stereo ambience swells too much on loud passages | Gentle side compression | More consistent ambience without obvious changes in width |
These are listening tests, not instrument-selection rules. Solo mid and side briefly to hear their contents, then judge each adjustment in the full mix.
M/S EQ changes tone and width together
A high-shelf boost in side changes the tone of that component and can make the top end seem wider. A low-shelf cut in side reduces low-frequency width. Mid EQ also affects panned sounds, because they contribute to mid as well. [2]
For a broad tonal adjustment, try about 1 dB first. Increase it only if the improvement remains clear at matched loudness. The size of the necessary move depends on the recording.
A side high-pass filter can reduce stereo bass content, but it cannot restore bass that already cancels in mono. See mono bass for the distinction between narrowing the low end and fixing cancellation.
M/S compression can make width change with the music
With side level unchanged, gain reduction in mid raises the relative prominence of side during compression. Gain reduction in side does the reverse. Make-up gain and other settings also affect the result. [1]
Listen through transitions, drum hits and sustained notes. If the image seems to expand and contract distractingly, reduce the compression or reconsider whether independent M/S dynamics suit the problem.
How to set it up
A plug-in with built-in M/S processing handles the conversion internally. Insert it on the stereo track or bus, select mid or side for the relevant band or processor, and leave the other component unchanged. Some EQs offer this choice per band. [2]
For a manual chain, use an M/S encoder, separate processing for M and S, then a matching decoder. Check channel assignments, gain and delay compensation before making adjustments. With neutral settings, the chain should match bypass in level, tone and stereo position.
Width controls are not interchangeable M/S encoders. For example, Ableton Utility has specific Mid/Side settings; its ordinary Width control also changes the stereo balance, but does not simply output raw M and S on separate channels. Follow the device’s routing instructions. [3]
Does M/S EQ need linear phase?
Minimum-phase EQ changes phase as well as level around its filters. If you EQ mid and side differently, their relationship changes, which can alter the apparent position of panned sounds after decoding. Steep side high-pass filters deserve particular attention. [1]
Linear-phase EQ avoids that frequency-dependent phase rotation, provided the paths remain correctly time-aligned. It still changes the M/S level balance, so it does not preserve every aspect of stereo placement. It also adds latency and can produce pre-ringing, an audible sound before a transient. [4]
Try a gentler filter or smaller adjustment first. Compare phase modes if placement changes are audible, and listen for softened or smeared attacks as well as width. Linear phase is an option to assess, rather than a requirement for every M/S move.
What should you check in stereo and mono?
In stereo, compare the processed and bypassed versions at similar perceived loudness. Check whether the original problem has improved, whether the lead still has the right prominence, and whether panned sounds have shifted unexpectedly.
Then use a proper mono sum. With the equations above:
Mono = (L + R) / 2 = M
Processing confined to side cancels out of that mono sum. If mid, overall gain and downstream processing remain unchanged, the mono signal is unchanged too. This follows directly from the decoding equations.
The mono check therefore answers two different questions:
- After a side-only change: does the existing mono mix still carry the musical parts you need? An improvement heard only in stereo will disappear in mono.
- After a mid change: has the adjustment improved or damaged the mono balance? Any mid EQ or compression is present in mono too.
An output trim changes mono level, and a limiter after the M/S processor may react differently to the altered stereo signal. Check the complete chain as well as the individual plug-in.
If bass, a vocal layer or a musical hook is already weak in mono, extra side level will not repair it. Find the track or stereo effect responsible; the mono compatibility guide explains what to check.
Sources
- iZotope: What is mid/side processing?, for signal contents, phase behaviour and dynamic width changes.
- FabFilter Pro-Q: Stereo options, for independent mid/side EQ and per-band routing.
- Ableton Live manual: Audio effect reference, Utility, for device-specific width and Mid/Side controls.
- FabFilter Pro-Q: Processing mode, for linear-phase latency and pre-ringing.