Stereo and spatial processing

Lab: The vanishing pad · Listen and compare

Starts with: a short loop of drums, bass and a very wide synth pad (Stereo Widener, Maximum Spread preset on the pad). Playing in stereo.

Reader controls: a Mono switch on the master.

Ask before switching: "When you switch to mono, will the pad get louder, quieter, or stay the same?"

Fallback: stereo and mono clips with a short description.

Owner: modules/stereo-widener on the pad; mono fold-down on the master.

Note: headphones on. This lab needs two ears.

Put on headphones and play the loop. The pad wraps around your head, much wider than the drums and bass.

Now hit Mono.

The drums and bass barely change. The pad shrinks, loses some of its sparkle, and drops back behind everything else. Phone speakers, a lot of club systems and some Bluetooth speakers play in mono or close to it, so this is what many people will hear.

Nothing was deleted. The pad was still playing. So where did it go?

Try it: split a sound into mid and side

Lab: Mid and side · Change a signal

Starts with: the pad alone through Stereo Widener at Width 0.5 (unity), Mix 100%.

Reader controls: Width (0 to 1), Mono switch, and solo buttons for Mid and Side.

Shows: two level bars (mid and side) and a correlation meter from Metergram.

Ask: "Solo the side. What's left of the pad? Now raise Width: which bar moves?"

Fallback: an annotated diagram of L/R and M/S plus four short clips (stereo, mid only, side only, wide).

Owner: modules/stereo-widener (mid/side matrix) and modules/metergram (correlation and mid/side readouts).

Solo the mid and you hear the pad's body, centred. Solo the side and you hear something strange and hollow: the shimmer, the movement, the edges. Raise Width and only the side bar grows.

What just happened

Every stereo sound is a left signal and a right signal. You can describe the same pair another way:

Mid is what the two channels have in common.
Side is how they're different.

Do the maths (optional): Mid = (L + R) ÷ 2. Side = (L − R) ÷ 2. You can rebuild the originals: L = Mid + Side, R = Mid − Side. Nothing is lost by switching descriptions.

Switching to mono adds left and right together. Look at the side signal in that sum: L + R = (Mid + Side) + (Mid − Side) = 2 × Mid. The side cancels out completely. In mono you hear only the mid.

That's the whole story of the vanishing pad. A stereo widener turns up the side. On headphones that sounds wider. In mono every bit of that extra side disappears, and if the pad got most of its energy from the side, most of the pad goes with it.

Reading the correlation meter

The correlation meter compares the two channels moment by moment:

Reading	What it means	In mono
+1	Left and right are identical (all mid)	Nothing lost
Around 0	Left and right are unrelated (very wide)	Some loss; varies with the material
Below 0	Left and right are working against each other	Parts cancel; this is the danger zone

Watch the meter as you raise Width on the pad. It slides from near +1 toward 0. A reading that stays below zero is the meter's way of saying "this will fall apart in mono".

Why it works: how our ears place sounds

Width on speakers is an illusion our brain builds from a few cues:

Level difference. A sound louder in the left ear seems to come from the left. That's what a pan knob does.
Time difference. A sound arriving a fraction of a millisecond earlier at one ear seems to come from that side.
Difference itself. When the two ears get related but not identical signals, the sound seems wide or surrounding rather than coming from a point.

A pan knob uses only level. A panned sound does have some side, but both channels carry the same signal at different volumes, so in mono they add up instead of cancelling. The correlation stays at +1, and the sound only changes level a little in the sum (how much depends on the pan law). Widening tricks work on the third cue, by making the channels genuinely different, and those differences are exactly what mono throws away.

Our take: pan to place things, widen to add atmosphere. A mix built mostly with panning will survive almost any playback system; widening is the seasoning.

Where it breaks

Lab: The widening trick that bites back · Listen and compare

Starts with: a mono synth line, copied to both channels, with the right channel delayed by 15 ms (the "Haas trick").

Reader controls: delay time (0 to 30 ms), Mono switch.

Ask: "In mono, the two copies add back together. Will that sound the same as the original?"

Fallback: clips in stereo and mono at 5, 15 and 25 ms, plus a spectrum image showing the comb-shaped notches.

Owner: delay-stage or a rendered example (labelled as rendered).

A short delay on one side is a classic widening trick. In stereo it sounds wide. In mono, the original and its delayed copy add together and cancel at a regular series of frequencies, which gives a hollow, metallic sound called comb filtering. Change the delay and the notches move.

Two more cases to watch for:

Wide bass. Low frequencies carry most of a mix's energy, and many club systems sum the sub to mono. Keep the bass and kick in the mid. Check the low end with the correlation meter: it should sit close to +1.
A mono source. If the input is the same in both channels, it has no side at all. The Stereo Widener can only scale side that already exists, so on a mono source no Width setting will make it wider. You need a source of difference first: a stereo recording, chorus, a stereo reverb, or two separate takes panned apart.

Same idea, different place: the comb filter you just heard is the same effect that makes a flanger swoosh. A flanger sweeps that short delay on purpose. See Chorus, flanger and phaser.

Make it yours

Load the bass and pad stems. Make the pair feel wide on headphones and still solid in mono.

Keep the bass centred. Check its correlation stays near +1.
Widen the pad with Stereo Widener until it feels wide on headphones.
Switch to mono. If the pad drops away, pull Width back or lower Mix until the mono version still sounds like the same pad, just narrower.
Watch the correlation meter. Aim to keep it above zero most of the time.

Save the result as your own mono-safe widener preset.

[Open this widener in the studio] (planned link: opens Stereo Widener and Metergram on the pad, with this guide's settings)

Quick check

You switch a mix to mono and the lead vocal is unchanged, but the backing vocals get much quieter. Where were the backing vocals living?
Answer
Mostly in the side: they were different between left and right, so mono cancelled much of them.
The correlation meter on your bass hovers around −0.3. What will happen on a club system with a mono sub?
Answer
Parts of the bass will cancel and it will sound weaker or thinner. Bring the low end back toward the centre.
A friend's mono synth won't get wider however far they push the widener. Why?
Answer
A mono source has no side for the widener to scale. Add some difference first, with chorus, a stereo reverb or a second take.

Next steps

Go deeper: Chorus, flanger and phaser. You met comb filtering as a problem; these effects use it on purpose.
Use it: Depth and space in a mix. Width is one dimension; front-to-back is the other.
Sideways: Low end. Why the bottom of a mix needs to stay in the middle, and how to check it on small speakers.

Go deeper

Inside the Maru Stereo Widener. It converts the input to mid and side, scales the side and leaves the mid unchanged. Width 0.5 is unity (no change) and Width 0 removes the processed side. Mix blends with the dry signal, so a partial Mix at Width 0 still keeps some of the original side. Presets range from Mono Fold to Maximum Spread. Metergram's correlation, mid/side spectra and balance views are the matching measurement tools.

Sources and examples. Pad, bass and drum loop: Maru shared source set (rights to be confirmed before publishing). The Haas example needs a rendered clip or a delay-stage configuration, to be confirmed.