Notes · Music TechnologyUK · A-Levels

Dynamic processing

Dynamic processors control level - the difference between the loud and quiet parts of a signal. This topic builds the compressor from its transfer curve (threshold, ratio and knee), adds the timing controls (attack, release) and make-up gain, then covers limiting, gating and expansion, and finishes with de-essing, side-chaining and parallel compression. Throughout it works the numbers, because the compressor's behaviour is precisely calculable and regularly examined.

5 sections~20 min reading time3 competenciesLevel Foundation 1 · Standard 2 · Advanced 2

T·111111 / 16

Exam profile

C4 · Set and calculate compressor behaviour from threshold, ratio, knee, attack and releaseC1 · Choose and apply compression, limiting, gating and expansion to control dynamicsC2 · Use de-essing, side-chaining and parallel compression creatively and correctively

Operators:describeexplaincalculatedetermineapply

basic level

The AS foundation is what a compressor does and the meaning of threshold and ratio.

higher level

The full A-level adds the transfer curve and calculations, attack/release, limiting, gating and side-chaining.

Depth

Text

Contents · 5 sections

Dynamic processing

Dynamic range and the need for processing#

FoundationPearson 9MT0 - dynamic processing

Dynamic range of a signal

Fig. 1Dynamic range is the gap between the quietest and loudest parts. Here about 34 dB separates the quiet passage from the loud peaks - a wide range a compressor can tame.

Dynamic range is the difference, in decibels, between the loudest and quietest parts of a signal, and managing it is a core production task. A live vocal might swing from a whispered verse to a belted chorus 30 dB louder; a drum kit has sharp transient peaks far above its sustained level. Such wide dynamics can be musically expressive, but they cause practical problems in a recording and a mix, which dynamic processors are designed to solve.

The problems are concrete. A part with a very wide dynamic range is hard to balance: set the fader for the quiet passages and the loud ones jump out; set it for the loud ones and the quiet ones vanish. Loud transient peaks force you to record and mix conservatively, wasting headroom, while quiet passages sink toward the noise floor. And a part that lurches in level will not sit consistently against the rest of the mix.

Dynamic processors control level automatically and continuously, doing what a very fast, tireless hand on a fader would do. A compressor reduces the dynamic range by turning down the loud parts; a limiter stops peaks exceeding a ceiling; a gate or expander turns down or removes the quiet parts (noise, spill) to increase separation. Each acts on level according to a set of rules based on a threshold, which is what distinguishes them from a simple fader move.

Used well, dynamic processing makes a part sit steadily in a mix, controls peaks so more of the available level can be used, adds punch or sustain, and cleans up noise between notes. Used badly, it flattens the life out of music, pumps and breathes audibly, or brings up unwanted noise. The skill - as with all processing - is to apply enough control to solve the problem without destroying the musical dynamics that give the performance its shape.

For the exam, be able to define dynamic range (the loudest-to-quietest difference in dB), explain the problems a wide dynamic range causes in recording and mixing, and outline what each family of dynamic processor does about it (compress the loud, limit the peaks, gate/expand the quiet). This framing motivates every setting in the sections that follow.

Worked example

Why a static fader fails

A vocal ranges from a quiet verse at about -35 dBFS to a loud chorus at about -6 dBFS. Explain why one fader setting cannot balance it and what would help.

Compute the range
The dynamic range is about -6 - (-35) = 29 dB between the loud and quiet sections.
The fader problem
Set the fader so the quiet verse is audible and the chorus, 29 dB louder, will be far too loud; set it for the chorus and the verse disappears.
The solution
A compressor automatically turns down the loud chorus relative to the verse, narrowing the 29 dB range so a single fader setting works for both.
Result
After compression the vocal sits at a consistent level and can be balanced against the mix with a static fader.

Result: The 29 dB range is too wide for one fader; a compressor reduces it so the vocal sits consistently and can be balanced.

Typical mistakes

Confusing dynamic range (the loud-to-quiet difference) with loudness or overall level.
Thinking dynamic processors only make things louder - they change the relationship between loud and quiet parts.

Active revision

Explain why a vocal that swings from a whisper to a shout is hard to balance with a static fader, and which dynamic processor would help.

Active recall

Recall the key points — then reveal.

Sources: Pearson Edexcel Level 3 Advanced GCE in Music Technology (9MT0) Specification (Pearson Edexcel)

The compressor: threshold, ratio and knee#

StandardPearson 9MT0 - dynamic processing

The compressor transfer curve

Fig. 2The transfer curve: below the -20 dB threshold, output equals input (unity); above it, the slope is 1/ratio (here 1/4). A -8 dB input maps to a -17 dB output - 9 dB of gain reduction.

A compressor reduces dynamic range by turning down the signal whenever it rises above a set level, and its behaviour is captured by its transfer curve - a graph of output level against input level, both in decibels. Below the threshold, the compressor does nothing: output equals input, a straight 45-degree line (a 1:1 relationship). Above the threshold it turns the signal down, so the line bends to a shallower slope. Reading this curve is the key to understanding compression.

Two controls define the curve. The threshold is the input level above which compression starts - the point where the curve bends. The ratio sets how hard the signal is compressed above the threshold: it is the ratio of the change in input to the change in output. A ratio of 4:1 means that for every 4 dB the input rises above the threshold, the output rises only 1 dB, so the slope above the threshold is 1/ratio. A higher ratio squashes the signal harder; a 1:1 ratio does nothing.

The maths is exact and examinable. For an input above the threshold, the output is: output = threshold + (input - threshold) / ratio. The amount the signal is turned down - the gain reduction - is the difference between input and output. So a compressor with a -20 dB threshold and a 4:1 ratio, fed a -8 dB input, gives an output of -20 + (-8 - (-20))/4 = -20 + 12/4 = -17 dB, a gain reduction of 9 dB. Below the threshold, the input passes unchanged.

The knee softens the transition at the threshold. A hard knee bends the curve sharply at the threshold - compression is either off or on - which is decisive but can sound abrupt. A soft knee rounds the corner, easing compression in gradually as the signal approaches and passes the threshold, which sounds smoother and more natural and is often preferred on vocals and mix-bus compression. The knee is thus a control over how audible the onset of compression is.

For the exam, be able to read and sketch the transfer curve (unity below threshold, slope 1/ratio above), define threshold and ratio, and calculate the output and gain reduction for a given input using output = threshold + (input - threshold)/ratio. Distinguishing a hard from a soft knee, and computing a compressed output, are among the most common and high-value questions in the whole subject.

\text{output} = T + \dfrac{\text{input} - T}{R} \quad (\text{input} > T), \qquad \text{gain reduction} = \text{input} - \text{output}

Compressor transfer function

Above the threshold T the excess is divided by the ratio R; below T the output equals the input.

Worked example

Computing compressed output

A compressor is set to a threshold of -20 dB and a ratio of 4:1. Find the output and gain reduction for (a) a -8 dB input and (b) a -30 dB input.

Check against threshold (a)
-8 dB is above the -20 dB threshold, so compression applies.
Output (a)
output = -20 + (-8 - (-20))/4 = -20 + 12/4 = -20 + 3 = -17 dB; gain reduction = -8 - (-17) = 9 dB.
$-20 + \dfrac{(-8) - (-20)}{4} = -17\,\text{dB}$
Check against threshold (b)
-30 dB is below the -20 dB threshold, so the signal passes unchanged.
Output (b)
output = -30 dB (unity), gain reduction = 0 dB, because compression acts only above the threshold.

Result: The -8 dB input becomes -17 dB (9 dB of gain reduction); the -30 dB input is unchanged at -30 dB (below threshold).

Typical mistakes

Compressing the signal below the threshold - below threshold the output equals the input (1:1); compression acts only above it.
Misapplying the ratio - a 4:1 ratio divides the amount above threshold by 4, it does not divide the whole level by 4.

Active revision

A compressor has a threshold of -20 dB and a ratio of 4:1. Calculate the output level and the gain reduction for an input of -8 dB, and for an input of -30 dB.

Active recall

Recall the key points — then reveal.

Sources: Pearson Edexcel Level 3 Advanced GCE in Music Technology (9MT0) Specification (Pearson Edexcel)

Attack, release, knee and make-up gain#

StandardPearson 9MT0 - dynamic processing

Gain reduction over time: attack and release

Fig. 3Gain reduction responds over time: the attack ramps into full reduction after the threshold is crossed, and the release ramps back to zero after the signal falls below it.

The transfer curve says how much a compressor turns the signal down; the timing controls say how quickly it responds. The attack time is how long the compressor takes to reach full gain reduction after the signal crosses the threshold. A fast attack clamps down almost instantly, catching and taming transients (but softening the punch of a drum); a slow attack lets the initial transient through before compressing, preserving the punch and attack of percussive sounds while still controlling the sustain.

The release time is how long the compressor takes to stop compressing after the signal falls back below the threshold. A fast release recovers quickly, which can add energy and density but may cause audible pumping (the level breathing up and down); a slow release is smoother and less obtrusive but can leave the compressor holding down the signal into quieter passages. Attack and release together shape the feel of the compression as much as the amount does.

Setting attack and release is a musical decision tied to the source. On drums, a slow attack and moderate release keeps the transient punch while controlling sustain; on a vocal, a moderate attack and release smooth the level transparently; on a mix bus, a slow attack and a release timed to the tempo let the mix breathe with the groove. Getting these wrong - a fast attack that kills all the punch, a fast release that pumps - is a common cause of lifeless or distracting compression.

Because compression turns loud parts down, the overall level of the output is reduced, so a make-up gain (output gain) control boosts the whole signal back up to compensate. This is what makes compression sound louder and denser: the peaks are pulled down and then the whole thing is raised, so the average level rises and the quiet detail comes up relative to the (now-controlled) peaks. Make-up gain is set to match the compressed output back to the input level (or louder) for a fair comparison.

For the exam, be able to explain attack (speed to full gain reduction - fast catches transients, slow preserves punch) and release (speed of recovery - fast can pump, slow is smoother), relate them to sources such as drums and vocals, and explain make-up gain as the boost that compensates for the level lost to compression. Understanding that these timing controls shape the character, not the amount, of compression is the key insight.

Worked example

Compressing a snare with punch

A snare drum needs its sustain controlled but must keep its punchy transient. Describe the attack, release and make-up gain settings.

Slow attack for punch
Use a relatively slow attack so the initial transient (the crack of the stick) passes through before the compressor clamps down, preserving the punch.
Moderate release
Set a moderate release so the compressor recovers between hits without pumping, controlling the sustain and body after the transient.
Set threshold and ratio
Lower the threshold until the body of the snare is compressed by a few decibels at a ratio of around 4:1, leaving the transient largely untouched.
Make-up gain
Add make-up gain to bring the compressed snare back up to (or above) its original level, so it sounds punchier and denser rather than quieter.

Result: A slow attack keeps the transient punch, a moderate release controls the sustain, and make-up gain restores the level - a punchy, controlled snare.

Typical mistakes

Using a fast attack on drums and killing the transient punch when a slower attack would keep it.
Forgetting make-up gain, so the compressed signal is quieter and the compression seems only to reduce level.

Active revision

Describe the attack and release settings you would choose to keep the punch of a snare drum while controlling its sustain, and explain the role of make-up gain.

Active recall

Recall the key points — then reveal.

Sources: Pearson Edexcel Level 3 Advanced GCE in Music Technology (9MT0) Specification (Pearson Edexcel)

Limiting, gating and expansion#

AdvancedPearson 9MT0 - dynamic processing

Compressor versus limiter transfer curves

Fig. 4Both bend at the threshold, but the limiter (very high ratio) flattens to a brick-wall ceiling above it, while the compressor keeps a shallow upward slope.

A limiter is a compressor with a very high ratio - effectively infinity-to-one - so that once the signal reaches the threshold, the output cannot exceed it: the transfer curve becomes flat (a brick wall) above the threshold. Limiting is used to catch and stop peaks absolutely, protecting against clipping and, in mastering, to raise the overall loudness by shaving the peaks so the whole track can be turned up. Because it acts only on the highest peaks with a fast attack, transparent limiting can add loudness with little audible change.

Gating works at the opposite end of the level scale: a gate attenuates or silences the signal when it falls below a threshold, passing it only when it is loud enough. It is used to remove noise, hiss or spill in the gaps between wanted sounds - silencing a guitar amp's hiss between phrases, or cleaning the bleed from a tom microphone so it opens only when that tom is hit. A gate's controls include the threshold, the range (how much it attenuates), and attack/hold/release times to shape how it opens and closes.

Expansion is the gentler relative of gating: instead of hard silencing, a downward expander increases the dynamic range below the threshold, turning quiet parts down proportionally so the difference between loud and quiet grows. It reduces low-level noise and spill more subtly than a gate, useful where a hard gate would sound unnatural. Gating and expansion are thus the mirror image of compression - acting on the quiet end to increase dynamic range or remove low-level noise, rather than on the loud end to reduce it.

These processors complete the dynamic toolkit. Compression reduces the range from the top; limiting caps the peaks absolutely; gating and expansion clean up or reduce the bottom. On a drum kit, for instance, gates tighten the individual mics by removing spill, compression controls the dynamics, and a limiter on the mix bus catches stray peaks - each processor doing its distinct job on the level.

For the exam, be able to describe a limiter (very high ratio, a brick-wall ceiling, for peak control and loudness), a gate (silences below threshold, to remove noise/spill, with range and timing controls) and expansion (gently increases range below threshold), and to place each against compression on the level scale. Recognising a limiter as extreme compression, and a gate/expander as acting on the quiet end, ties the family together.

Worked example

Choosing the right dynamic tool

Select the correct dynamic processor for (a) preventing any mastering peak going over -1 dBFS, (b) removing amp hiss between guitar phrases, and (c) smoothing a vocal's level, and justify each.

The mastering peaks
Use a limiter set to a -1 dBFS ceiling: its brick-wall, very-high-ratio action guarantees no peak exceeds the ceiling, allowing safe loudness maximisation.
The amp hiss
Use a gate: set the threshold above the hiss but below the guitar so the gate opens for the phrases and closes (silencing the hiss) in the gaps.
The vocal
Use a compressor with a moderate ratio to reduce the vocal's dynamic range so it sits consistently, with make-up gain to restore level.
Why not swap
A compressor would not stop peaks absolutely (a limiter does), and a gate acts on the quiet end where the hiss is, not on the loud vocal peaks.

Result: Limiter for the peak ceiling, gate for the hiss between phrases, compressor for the vocal - each processor matched to the end of the level scale it controls.

Typical mistakes

Confusing a gate (acts on quiet signals below threshold) with a compressor (acts on loud signals above threshold).
Thinking a limiter is a separate process from compression - it is compression at a very high ratio with a brick-wall ceiling.

Active revision

State whether a limiter, a gate or a compressor is the right tool for (a) stopping mastering peaks exceeding a ceiling, (b) removing hiss between guitar phrases, and (c) evening out a vocal, and justify each.

Active recall

Recall the key points — then reveal.

Sources: Pearson Edexcel Level 3 Advanced GCE in Music Technology (9MT0) Specification (Pearson Edexcel)

De-essing, side-chaining and parallel compression#

AdvancedPearson 9MT0 - dynamic processing

Side-chain ducking routing

Fig. 5Side-chaining: the kick drum feeds the compressor's key input, so each kick hit triggers gain reduction on the bass - the bass ducks under the kick.

A de-esser is a frequency-conscious compressor that reduces harsh sibilance - the 's', 'sh' and 't' sounds that spike around 5-8 kHz on close-miked vocals. It listens only to that high band and compresses the signal when the sibilance exceeds a threshold, ducking the harshness without dulling the rest of the voice. It is effectively a compressor whose detector is filtered to the sibilant frequencies, and it is a routine corrective tool on vocals in Component 4.

Side-chaining feeds the compressor's detector from a different signal than the one being compressed, so one sound controls the level of another. The classic musical use is 'ducking': a compressor on the bass (or on a music bed) is triggered by the kick drum (or a voiceover), so the bass dips every time the kick hits, carving space and producing the pumping, rhythmic effect heard across dance and pop. In broadcast, a presenter's microphone side-chains a compressor on the music so the music ducks under speech automatically.

The side-chain is what makes this possible: the signal that triggers compression (the key input) is separated from the signal being processed. Understanding the routing - the trigger sets the gain reduction, but the reduction is applied to the target - is the key to using it. The amount, attack and release of the side-chain compressor set how deep and how rhythmic the ducking is: a fast release gives an obvious pump, a slow one a gentle dip.

Parallel (or New York) compression blends a heavily compressed copy of a signal with the untouched original, rather than compressing the whole signal. The dry signal keeps the natural transients and dynamics, while the crushed parallel copy adds density, sustain and body underneath, and the balance of the two sets the effect. It is popular on drums and vocals because it adds power and consistency without sacrificing the punch that heavy compression alone would flatten.

For the exam, be able to describe de-essing (band-conscious compression of sibilance), side-chaining (one signal triggers the compression of another, for ducking and pumping effects) and parallel compression (blending a crushed copy with the dry signal for density without losing transients). These are the creative and advanced applications of the compressor that recur in both Component 2 and Component 4.

Worked example

Setting up kick-to-bass ducking

You want the bass to duck rhythmically each time the kick hits, for a pumping dance feel. Set up and tune the side-chain.

Route the key input
Place a compressor on the bass and set its side-chain (key) input to the kick drum, so the kick triggers the compression while the bass is the signal being compressed.
Set threshold and ratio
Lower the threshold until each kick produces several decibels of gain reduction on the bass at a moderate-to-high ratio, so the bass dips clearly on every kick.
Tune the release
Set the release to control the pump: a fast release makes the bass spring back quickly for an obvious pumping effect; a slower release gives a gentler dip - the release is the main feel control.
Check in tempo
Play the groove: the bass should duck and recover in time with the kick, opening space for the kick and creating the rhythmic pump.

Result: Feed the kick to the bass compressor's key input; threshold/ratio set the depth and the release sets how pumping the ducking feels.

Typical mistakes

Thinking a de-esser is an EQ - it is a compressor triggered by the sibilant band, ducking only when sibilance spikes.
Confusing which side-chain signal does what - the trigger (key input) sets the gain reduction applied to the target.

Active revision

Describe how you would set up a side-chain so that a bass line ducks each time the kick drum hits, and state one setting that controls how pumping the effect sounds.

Active recall

Recall the key points — then reveal.

Sources: Pearson Edexcel Level 3 Advanced GCE in Music Technology (9MT0) Specification (Pearson Edexcel) · Ofqual - GCE AS and A level qualifications (subject-level conditions and requirements) (Ofqual)

References & sources

Sources

Pearson Edexcel

Pearson Edexcel Level 3 Advanced GCE in Music Technology (9MT0) Specification

Ofqual

Ofqual - GCE AS and A level qualifications (subject-level conditions and requirements)

Dynamic processing

5 sections~20 min reading time3 competenciesLevel Foundation 1 · Standard 2 · Advanced 2

Dynamic processing

Key points

Why a static fader fails

Key points

Computing compressed output

Key points

Compressing a snare with punch

Key points

Choosing the right dynamic tool

Key points

Setting up kick-to-bass ducking

Sources

Dynamic processing

Key points

Why a static fader fails

Key points

Computing compressed output

Key points

Compressing a snare with punch

Key points

Choosing the right dynamic tool

Key points

Setting up kick-to-bass ducking

Sources