In Stereo - Mid/Side Processing Explained

When we think about mixing, and more specifically element positioning, we often think in terms of volume - where do the guitars sit in the mix compared to the drums, is the snare cutting through the vocals, and so on. Even with EQing elements - all we're really doing is taking specific frequencies and turning them up or down.

Sometimes (or maybe a lot of the time if you're dealing with fancy surround sound Dolby Atmos business) we think about panning elements left and right.

But another incredibly useful, and often mandatory, way of thinking about your mix is from the perspective of mid and side processing - what we do to the sound in the center, compared to what we do on the sides.

Beyond just being a useful mixing trick, understanding M/S is absolutely crucial if you're expecting your track to be played on large sound systems. If you're writing a club banger, and you don't mono your low end, then your track will completely fall apart when played live.

This post outlines what M/S processing gives you, where it's useful, and where it's mandatory.

Quick Maths

The mid channel contains everything that's identical in both L and R - a centered kick, a mono bass or a centered vocal all sit here at full level.

The side channel on the other hand contains the differences - stereo reverb tails, hard panned guitars and their subtle differences, stereo widening effects and so on.

Important, and unintuitive nuance: hard-panned elements technically aren't present only on the sides. A guitar panned 100% to the left shows up in the mid and the side.

"But that doesn't make sense, my DAW meter only shows audio coming out of the left channel when pan hard-left?"

The mid channel is the sum of L and R. The side channel is the difference between them.

The practical consequence is what matters: any processing you apply to the side channel will affect that hard-panned instrument. Boost the side, the instrument gets louder (in a weirder way - L goes up, R goes negative). EQ the side, the guitar gets EQ’d.

This is why M/S moves should usually be smaller than L/R moves of the same apparent weight - you’re touching more of the mix than the channel label suggests, and aggressive moves on the sides can introduce artifacts on things that weren’t meant to be stereo-processed at all.

The most important M/S move: mono your low end

The low end of your mix - kick, bass, sub, anything below roughly 120hz - should be mono.

Not because it sounds better in some abstract sense - if it isn't, your mix will fall apart on large sound systems.

Low frequencies carry a huge amount of energy, and when sub bass information is different in the left and right channels - a stereo bass synth, a wide-panned kick or a reverb tail that dips into the sub-100hz range - those two signals interfere with each other the moment they're played back on anything that isn't a pair of studio monitors.

The nerdy explanation - if you have two versions of a low-frequency signal in L and R that aren't identical, they have some phase relationship to each other.

If they're perfectly in phase (identical), they sum to double the amplitude - this is the exact same process as what happens when you take two identical tracks and perfectly play both copies together, they simply double in volume.

If they're perfectly out of phase (one is the inverse of the other), they sum to silence. Most stereo low end lives somewhere in between - partially correlated, partially not - which means that when those two channels get added together, some frequencies reinforce and others cancel.

In practical terms, any playback device that sums L and R - like a club sound system, vinyl, or car stereo - will expose whatever phase problems are lurking in your low-end, potentially cancelling out your most impactful frequencies. Vinyl in particular is very unforgiving - you might end up with needle skips or damaged lacquer during cutting as a result of not mono'ing your sub.

The fix is one M/S move: a high-pass filter on the side channel, cutting everything below somewhere between 100 and 300 Hz. This doesn’t touch the mid channel, so the kick and bass keep their level - it just forces the low frequencies to be identical in L and R, which is the definition of mono. Anything in the stereo bass that was going to phase-cancel now can’t - there’s no stereo information left down there to cancel with.

Other useful areas to look at

Widening without smearing the centre.

The most common use. A gentle high-shelf boost on the side channel, or a gain increase on the sides relative to the mid, pushes width into the top end where the ear perceives it as air and space. Because you’re not touching the mid, the vocal and kick stay where they are. Compare this with an L/R stereo widener, which often messes with the centre as a side effect.

De-harshing without killing guitars.

If the vocal is sibilant or the 3-5 kHz region in the centre is aggressive, an M/S EQ lets you dip that region on the mid channel only. Hard-panned guitars and cymbals keep their bite because they live mostly on the sides.

Controlling side-channel dynamics.

An M/S compressor with a lighter ratio on the sides than the mid keeps the centre punchy without squashing the width. This is particularly useful on mix-bus or mastering compression, where a single stereo-linked compressor often pulls the sides down every time the kick hits - the image narrows on every transient. M/S compression avoids that.

Rebalancing a finished mix.

This is where M/S earns its place in mastering specifically. If you’ve been sent a stereo bounce and the vocal is slightly buried, a small boost to the mid channel in the vocal’s fundamental range will lift it. It’s not a substitute for a proper remix, but it’s often enough to save a master that would otherwise need to go back.

A sensible workflow

When reaching for M/S, the question to ask first is whether the problem is actually a centre-versus-sides problem. If the vocal is too loud across the whole mix, that’s not an M/S issue - it’s a level issue. If the mix feels narrow in the top end but the centre is where you want it, that’s M/S territory.

If the kick and bass are fine in stereo but lose punch in mono, that’s also M/S territory.

Once you’re sure you need it: make smaller moves than feel right, check mono often, and watch correlation. A good M/S move usually sounds like almost nothing in isolation - half a dB here, little shelf there - but the mix as a whole opens up or tightens up in a way that a single stereo-linked move couldn’t achieve. If you’re applying 4 dB boosts to the side channel and widening the stereo image dramatically, you’ve probably overshot, and the mono sum will tell you so.

The technique is powerful precisely because it operates on a dimension that L/R processing can’t reach. That’s also why it punishes heavy-handedness faster than normal processing does.

Used surgically, M/S is one of the highest-leverage tools in mastering. Used as a widener cranked to 11, it’s one of the fastest ways to produce a mix that falls apart the moment someone plays it on a phone.

As we mentioned in our audio metrics article - good metering goes a long way, so make sure to grab a good stereo imager instead of relying on your ears all the time.