How to Use Multiband Compression: The Complete Guide
Multiband compression splits your signal into frequency bands and compresses each band independently. Use it when your dynamic problem is frequency-specific — when the low end is too uneven but the top end is fine, for example. Don't use it as a default compressor — that's the most common multiband mistake. A single-band compressor is the right tool for most mixing situations. Multiband earns its place on the master bus for final tonal balance control, on bass instruments with specific frequency problems, and on vocal bus situations where low-mid buildup needs frequency-targeted control. If you can't articulate exactly which frequency range has the problem, reach for a standard compressor first.
Multiband compression is one of the most misused tools in music production. It's powerful, it's present in almost every DAW and mastering chain, and it's routinely applied where it doesn't belong — turning mixes that just needed a single-band compressor and some EQ into over-processed, frequency-fragmented signals that sound dense without feeling alive.
The problem is the concept is seductive: independent control over every frequency range sounds like more control, and more control sounds like a better result. In practice, more control means more ways to make something sound wrong. Multiband compression requires a clear understanding of what the tool actually does, when it solves a problem that a standard compressor can't, and — critically — what it sounds like when it's being applied where it shouldn't be.
This guide covers all of it: how multiband compression works at the technical level, the correct use case for each application context (mixing vs mastering vs individual sources), how to set every parameter, the mistakes that are most damaging to a mix, and specific settings for the instruments and contexts where multiband compression genuinely earns its place.
How Multiband Compression Works
A standard (wideband) compressor processes your entire audio signal as one unit. When the signal crosses the threshold, every frequency in the signal gets compressed equally — the low end, the midrange, and the high end all receive the same gain reduction at the same time.
A multiband compressor first splits the signal into separate frequency bands using crossover filters. Typically three to five bands, though some processors offer more. Each band then passes through its own independent compressor with its own threshold, ratio, attack, release, and makeup gain. After processing, the bands are summed back together into a single output signal.
The critical consequence: a loud transient in the low end (a kick drum hit, a bass note peak) only triggers the low band's compressor — it has no effect on the high band's compressor. A sibilant peak in the high end only triggers the high band. This frequency-isolated compression is the defining capability that makes multiband useful and also what makes it dangerous when misapplied.
Crossover Filters and Phase
The crossover filters that split the signal into bands introduce phase shift at the crossover frequencies. This is unavoidable with standard (minimum phase) crossover designs — the signal's phase response changes at the point where the bands separate. When the bands are summed back together, these phase changes can create subtle coloration and, in some cases, audible artifacts — particularly on transient material where the phase accuracy of the crossover critically affects how hits sound.
Linear phase crossover designs solve this by processing the entire frequency range in a way that preserves phase relationships, but introduce pre-ringing (a type of artifact around sharp transients) and add latency. Linear phase mode is the correct choice for mastering contexts where phase accuracy and stereo imaging are paramount. Minimum phase mode is generally preferred for mixing individual sources where lower latency and more natural transient feel matter more.
When Multiband Compression Is the Correct Tool
The question that should precede every multiband compressor insert: is the dynamic problem frequency-specific? If yes, multiband may be the right tool. If no — if the dynamics are uneven across the entire frequency range — a standard compressor handles it more cleanly and with fewer side effects.
The Correct Use Cases
Bass guitar with frequency-specific dynamics. Bass guitar often has pronounced unevenness in the low frequencies (the fundamental of the note — 40–120 Hz) compared to the upper harmonics (200–600 Hz) that define the instrument's character. A low-E string note may have significantly more low-end energy than a higher-string note at the same played volume. Compressing the low band specifically brings the low-frequency content under control without touching the harmonic character of the instrument — something a single-band compressor can't do without affecting everything simultaneously.
Vocal bus low-mid buildup. In a mix with many vocal tracks, low-mid frequencies (200–500 Hz) can accumulate into a muddy buildup when vocals get louder together. A multiband compressor on the vocal bus can specifically target and control that low-mid frequency range during louder passages, cleaning up the buildup without compressing the top end of the vocal blend where presence and air live.
Master bus tonal balance control. On the master bus, multiband compression can apply gentle leveling to frequency bands that are dynamically imbalanced — specifically controlling a bass-heavy section of a song without crushing the more balanced sections. This is the most common legitimate mastering use of multiband compression.
Drum bus transient management. Applying light multiband compression to the drum bus allows you to control the low frequencies of the kick without changing the attack characteristics of the snare and hi-hat — instruments whose dynamics live in different frequency ranges and which need different treatment.
When to Use a Standard Compressor Instead
Use a standard compressor when the dynamic problem is consistent across the frequency spectrum — when the overall level is inconsistent regardless of what frequency is too loud. Vocal leveling, drum bus glue, instrument bus control — the vast majority of mixing compression tasks are best handled by a single-band compressor.
Use EQ rather than multiband compression when the problem is a frequency imbalance that isn't dynamic — when a frequency range is consistently too loud or too quiet regardless of the level of the signal. Multiband compression addresses dynamics. EQ addresses tonal balance. These are different problems.
Use a dynamic EQ rather than multiband when the frequency-specific problem is subtle and you want more surgical control with less coloration. Dynamic EQ introduces fewer crossover artifacts and behaves more predictably on complex program material — for gentle frequency-specific dynamic control in mastering contexts, many mastering engineers prefer dynamic EQ over multiband compression.
Setting Crossover Frequencies
Crossover frequencies determine where the bands split. Poorly set crossovers produce unnatural-sounding compression — the bands don't align with the natural frequency content of the material being processed, so gain reduction in one band audibly affects how adjacent bands sound when summed.
The goal: set crossovers at natural frequency separations in the material — points where one characteristic of the sound transitions to another. On a full mix, this means separating the sub bass from the upper bass, the upper bass from the lower midrange where most harmonic content lives, the lower midrange from the presence region, and the presence region from the air frequencies.
| Band | Frequency Range | Content | Typical Use |
|---|---|---|---|
| Sub/Low | 20–120 Hz | Sub bass, kick fundamental, bass fundamental | Control sub buildup, bass level consistency |
| Low-Mid | 120–800 Hz | Harmonic body of bass, warmth of instruments | Mud control, body consistency |
| High-Mid | 800 Hz–5 kHz | Presence, definition, vocal clarity, attack of drums | Harshness control, presence leveling |
| High | 5 kHz–20 kHz | Air, brilliance, cymbal shimmer, sibilance | De-essing, air consistency, sibilance control |
Don't place crossovers in the middle of frequency zones that you want to treat as a unit. Splitting the upper bass (200–500 Hz) between two bands means the gain reduction in the low band affects the upper bass differently than the gain reduction in the mid band — the result is unnatural-sounding compression that fragments the low-mid character of the mix.
Setting Parameters by Band
Threshold and Ratio per Band
Each band's threshold should be set independently based on the content in that band. Before touching threshold, listen to each band soloed (most multiband compressors allow band soloing) and understand what frequency content lives there. Then set the threshold for each band at the level where you want compression to engage — not globally, but specifically for what's in that band.
As a general principle: the low band often needs the lowest threshold (most compression engagement) because bass frequencies tend to be the most dynamically uneven. The high band often needs a higher threshold — high frequencies are typically more consistent and don't need as much dynamic control unless you're using the multiband for de-essing (where the high band threshold is set very specifically to engage only on sibilant peaks).
Ratio recommendations by band: Low band — 3:1 to 6:1 for mixing, 2:1 to 4:1 for mastering. Low-mid and High-mid — 2:1 to 4:1 for mixing, 1.5:1 to 3:1 for mastering. High band — 2:1 to 3:1 for mixing and mastering (higher ratios in the high band tend to introduce obvious artifacts). For all bands: less is more, especially in mastering contexts.
Attack and Release per Band
The most important parameter interaction in multiband compression: attack and release times must be set appropriately for the transient content in each frequency band, because different frequency ranges have fundamentally different transient characteristics.
Low band (20–120 Hz): Slow attack, slow release. Bass frequencies move slowly — their waveforms have long cycles. Fast attack in the low band clamps the initial transient of kick drums and bass notes, reducing punch and making the low end feel soft and weak. Start with attack 50–80ms, release 200–400ms. Adjust release to be tempo-matched — the compressor should recover between beats, not between individual waveform cycles.
Low-mid band (120–800 Hz): Moderate attack, moderate release. This band contains the harmonic body of most instruments — it needs enough attack time to let initial transients through but shouldn't be as slow as the low band. Attack 20–50ms, release 100–200ms.
High-mid band (800 Hz–5 kHz): Moderate-to-fast attack, fast-to-moderate release. The presence and attack characteristics of drums, vocals, and melodic instruments live here. Attack 10–25ms, release 80–150ms. Faster attack in this band can reduce perceived clarity and definition — be careful about clamping high-mid transients.
High band (5 kHz–20 kHz): Variable depending on purpose. For de-essing or sibilance control: fast attack (2–10ms) and fast release (30–80ms) to catch and release sibilant peaks quickly. For general air and brilliance control: slower attack (15–30ms) and moderate release (100–200ms) to avoid a choppy, unnatural character in the high frequencies.
Multiband Compression on the Master Bus
Mastering is the most common professional use of multiband compression — and also the context where it's most likely to do damage when applied incorrectly. On the master bus, every decision affects the entire mix simultaneously. Multiband compression applied too aggressively on the master bus can unmix a well-balanced track — artificially separating the frequency ranges in a way that makes the mix feel disconnected, or reducing dynamics so heavily that the music loses its sense of movement and energy.
Mastering Approach — Gentle Balance Control
The mastering application where multiband compression is most justified: a mix where one frequency range is dynamically inconsistent compared to the others. The bass gets loud in the chorus but the rest of the mix is balanced; or the high end gets harsh on the verse but is fine on the chorus. Standard compression would address these sections but affect the entire frequency spectrum — multiband compression addresses only the problem range.
Settings for mastering multiband: 4-band configuration with the crossovers set around 120 Hz, 800 Hz, and 5 kHz. Ratio 1.5:1 to 2:1 across all bands — anything more aggressive on the master bus starts to damage the mix. Threshold set for a maximum of 1–2 dB gain reduction per band in the most active sections. Attack slow enough to preserve transients in every band. If you're seeing more than 3 dB of gain reduction on any band, the problem is too significant for multiband to solve transparently — it needs to be addressed in the mix, not the master.
Mastering Approach — Final Tonal Shaping
Some mastering engineers use multiband compression at the end of the chain as a final frequency balance tool — applying very light compression (0.5–1 dB GR maximum) to nudge the tonal balance of the master without adding static EQ. The advantage is that the frequency balance adjustment is dynamic — it only activates when the frequency range gets louder, leaving the quieter sections untouched.
This approach requires extremely refined hearing and significant experience — the difference between this working well and making the master sound worse is a matter of fractions of decibels. If you're not sure whether you're achieving this level of control, default to a static EQ curve instead.
Multiband Compression on Individual Sources
Bass Guitar
Bass guitar is one of the most productive sources for multiband compression because its dynamic problems are frequently frequency-specific. A common problem: the low-fundamental notes (E and A string open) have significantly more low-frequency energy than the same notes played up the neck, creating a lopsided frequency balance that makes the bass sound inconsistent across the range of the instrument.
A targeted approach: 3-band multiband with crossovers at approximately 120 Hz and 500 Hz. Set the low band (sub to 120 Hz) with a moderate threshold and ratio (4:1) to control the excessive sub energy on the open strings. Set the mid band (120–500 Hz) with a gentler threshold — this range is more consistent and needs less control. Set the high band (500 Hz+) to bypass or minimal compression — the upper harmonics of the bass are typically consistent and don't need frequency-specific dynamic control.
Drum Bus
Drum bus multiband works well when you want to add density to the low end of the kick without crushing the top end of the snare and hi-hat — a combination that standard compression can't achieve because compressing the kick's low-end punch also compresses the snare's transient snap.
A functional drum bus multiband approach: split at 120 Hz and 3 kHz. Low band: ratio 4:1, attack 30ms, release matched to tempo — this compresses the kick's sub energy and thickens the low end. Mid band: light compression, ratio 2:1, moderate attack and release — gentle glue on the body of the kit. High band: minimal compression or bypass — preserve the transient character of the snare crack and hi-hat shimmer. Use parallel dry/wet blending to control the amount of compression character entering the mix.
Vocal Bus
In a dense vocal arrangement (lead + multiple harmonies), low-mid buildup (200–500 Hz) during louder sections is a common problem. A multiband compressor on the vocal bus with a low-mid band set to engage only when the vocal group gets loud cleans up this buildup dynamically — leaving quiet sections open and warm while controlling the mud when the voices stack up in the choruses.
The Most Common Multiband Mistakes
Using Multiband as a Default Compressor
Inserting a multiband compressor on every track because "it gives you more control" is the most common mistake at the intermediate level. Multiband is a specialist tool for frequency-specific dynamic problems. Using it everywhere creates crossover artifacts across every track in your session, introduces unnecessary phase shift at multiple frequency points, and adds processing overhead — all with no benefit over a standard compressor for the majority of compression tasks.
Too Much Gain Reduction Per Band
Pulling the threshold down until you're seeing 6–8+ dB of gain reduction per band in a mastering context is one of the fastest ways to ruin a mix. The frequency ranges are now being compressed at dramatically different rates, the inter-band relationships that make the mix feel coherent are destroyed, and the result sounds fragmented — the low end, midrange, and high end no longer behave as though they're coming from the same piece of music. In mastering, 1–2 dB per band is the working limit for transparent control.
Mismatched Attack Times Across Bands
Using the same attack time for all bands is a sign that the multiband settings haven't been customized for the material. The low band needs slow attack to preserve kick transients; the high band may need fast attack for sibilance control. Same attack across all bands means the low band is either attacking too fast (killing kick punch) or too slow (inadequately controlling bass peaks), while the high band isn't responding appropriately to transients.
Using Multiband to Fix Problems That Are Better Fixed at Source
If the bass guitar recording has a specific frequency that's boomy because of the room or the amp position, fix it with EQ at that specific frequency — don't compress an entire frequency band to manage it. If the vocal has inconsistent low-mid resonance because the singer moved position during tracking, fix the performance or comp takes. Multiband compression applied to mask a source problem typically creates a new sound problem while partially masking the original one.
Practical Exercises
Beginner Exercise — Hear the Difference Between Bands
Open any multiband compressor on a stereo mix reference track. Solo the low band (below ~120 Hz) — you should hear only the sub bass and kick fundamental. Now solo the low-mid band (120–800 Hz) — this is the body of the mix: bass harmonics, the warmth of guitars and keys, the body of the vocal. Solo the high-mid band (800 Hz–5 kHz) — this is presence and definition: vocal clarity, snare attack, midrange detail. Solo the high band (5 kHz+) — air, shimmer, cymbal.
Listen to each band individually on different musical genres: a hip-hop track, a rock mix, an acoustic recording. Notice how the energy distribution between bands differs significantly. A hip-hop mix has enormous energy in the low band; an acoustic guitar recording has almost none. This exercise trains your ability to hear frequency-specific dynamics — the prerequisite skill for using multiband compression intentionally.
Intermediate Exercise — Frequency-Specific Problem Identification
Take a bass guitar recording and insert a multiband compressor. Before touching the threshold on any band, play the recording and watch the gain reduction meters on each band independently. You'll likely see the low band's meter moving significantly more than the mid band's — the sub bass of the open strings is causing the low band to work harder. The mid band shows more consistent activity. The high band shows minimal activity.
Now set the threshold on the low band until you're getting 3–4 dB of gain reduction on the loudest low-frequency peaks. Leave the other bands uncompressed (set thresholds very high so they never engage). Compare the bass with multiband on vs off. The low end should be more consistent across the range of the instrument — open string notes should sit closer in level to notes played higher on the neck. This is multiband compression solving an actual frequency-specific problem.
Advanced Exercise — Mastering Multiband on a Full Mix
Take a finished mix that passes your standard checks (good dynamic range, good tonal balance) and insert a 4-band multiband compressor after your master bus limiter (or before, depending on your mastering chain philosophy). Set all four bands with ratio 1.5:1 and thresholds so high that no band is compressing at all — confirmed by looking at all gain reduction meters at zero.
Now play the loudest section of the mix. Slowly lower the low band threshold until you see 0.5 dB of gain reduction on the loudest moments. Listen critically — does the bass feel more controlled? Less controlled? Does the change feel natural or does it introduce obvious artifacts? Now try the same on the low-mid band: 0.5 dB maximum, listen. Then the high-mid, then the high band.
The goal is to determine which bands, if any, actually benefit from this light touch of frequency-specific dynamic control. In many well-mixed tracks, the answer is none — no band benefits from multiband compression at the master bus level. In those cases, the correct answer is to not use multiband. This exercise trains the discipline of using the tool only when it genuinely improves the result.
Frequently Asked Questions
What is multiband compression?
Multiband compression splits an audio signal into multiple frequency bands (typically 3–5) and applies independent compression to each band. This allows you to compress the low end without affecting the high end, or control a problem frequency range without touching the rest of the signal. Each band operates as a completely separate compressor with its own threshold, ratio, attack, release, and makeup gain.
When should I use multiband compression instead of a regular compressor?
Use multiband compression when the dynamic problem is frequency-specific — when one frequency range is too dynamic while others are fine. Common scenarios: a bass guitar with a boomy low end but controlled mids, a vocal mix where the low-mid resonance is inconsistent, or a master bus where the bass is causing pumping while the rest of the mix is controlled. If the dynamic problem affects all frequencies equally, a standard compressor is the better tool.
What crossover frequencies should I use for multiband compression?
A common starting point for a 4-band multiband on a full mix: Low band 20–120 Hz, Low-Mid band 120–800 Hz, High-Mid band 800 Hz–5 kHz, High band 5 kHz–20 kHz. These are starting points — the optimal crossover points depend on the specific material. Set crossovers at natural frequency separations in the content, not at arbitrary round numbers.
Can multiband compression make a mix sound worse?
Yes, and this is one of the most common problems in intermediate mixing. Multiband compression can introduce phase artifacts at crossover points, create unnatural-sounding dynamics when individual bands compress at different rates, and make a mix feel fragmented and lifeless if overused. A mix that's been over-multiband-compressed often sounds dense and controlled but lacking life and movement. The correct amount is typically much less than beginners expect.
How much gain reduction should I aim for on each band?
For mixing: 1–3 dB per band is typical for transparent control. For mastering: 1–2 dB per band maximum for gentle glue. On individual sources like bass guitar: 3–6 dB on the problematic band is reasonable. If you're seeing 8+ dB of gain reduction on any band in a mastering context, the problem is more fundamental than multiband can fix without causing damage.
What is the difference between multiband compression and dynamic EQ?
Multiband compression applies compression to a frequency band — threshold, ratio, attack, release all apply, and the band is compressed relative to a fixed threshold. Dynamic EQ applies EQ gain changes that are triggered by the signal level. Dynamic EQ is generally more surgical and less colored; multiband compression is more aggressive and can shape character as well as dynamics. For subtle frequency-specific dynamic control, dynamic EQ is often the cleaner choice.
Should beginners use multiband compression?
No — not until you can hear what a standard compressor does and use it intentionally. Most beginner and intermediate mixes that reach for multiband compression don't need it — the problem is better addressed with EQ, single-band compression, or fixing the source recording. Multiband is a specialist tool, not a default option.
What are the best multiband compressor plugins?
FabFilter Pro-MB is widely considered the most transparent and flexible multiband compressor available. Waves C6 is an industry-standard mixing multiband. iZotope Ozone's Dynamics module is excellent for mastering contexts. The built-in multiband in most DAWs (Logic's Multipressor, Ableton's Multiband Dynamics) are competent starting points.
What is linear phase mode in a multiband compressor?
Linear phase crossover filters split the signal into bands without introducing phase shift at the crossover frequencies. This preserves the phase relationships of the original signal. Use linear phase mode for mastering and mix bus applications where phase accuracy matters most; use minimum phase for mixing individual sources where lower latency and more natural transient feel are important.
How do I set multiband compression attack and release times?
Set each band's attack and release independently based on the content in that frequency range. Low band: slow attack (30–80ms), slow release. Mid bands: moderate attack (10–30ms), release 100–200ms. High band: faster attack (5–15ms), shorter release (50–100ms). The key rule: faster transients in a band need faster attack times.
Practical Exercises
Identifying When Multiband Is Needed
Load a vocal recording into your DAW and listen critically for 2 minutes. Identify the specific frequency range that feels problematic — is the low end boomy? Are the mids thick and uncontrolled? Write down the frequency range (e.g., "100–300 Hz feels muddy"). Now insert a standard compressor on the same track and try to fix the problem. Notice how it affects all frequencies equally, even the ones that sound fine. This experience shows why multiband compression exists: when ONE frequency range needs control and others don't, standard compression can't help — it will compress everything. Save your notes. You've identified a legitimate multiband use case.
Multiband vs. Single-Band Comparison
Record or import a bass guitar that has uneven low-end energy and some mid-range harshness. Duplicate the track twice. On the first, insert a standard compressor set to 4:1 ratio, 10ms attack, 100ms release, threshold adjusted until you hear visible gain reduction. On the second, insert a multiband compressor. Isolate the low-end band (80–250 Hz) and compress only that section with the same ratio/attack/release. Leave the mid and high bands untouched. A/B switch between both tracks. Which sounds more natural? Which one kept the character of the higher frequencies while controlling the problem? Document which frequency range needed the fix and why the multiband version preserved more of the original tone. This teaches you the actual advantage: frequency-specific control.
Master Bus Multiband for Tonal Balance
Export a full stereo mix and create a new session. Import the mix onto a stereo track and insert a multiband compressor. Listen for tonal imbalances across the frequency spectrum — does the mix feel low-heavy? Does the midrange crowd the vocals? Set up 4 bands: sub-bass (20–80 Hz), low-mids (80–400 Hz), mids (400–2kHz), and highs (2kHz+). Apply subtle compression (2:1 ratio, 20–30ms attack, 150–300ms release) only to bands that feel unbalanced. For example, if the low-mids mask clarity, compress at -6dB threshold. If the highs feel harsh, reduce gain on that band instead of compressing. Make surgical adjustments — each band should compress less than 3dB. A/B your mix before and after. The goal isn't obvious compression; it's tonal polish that makes the mix feel cohesive. Document your settings and listen on multiple playback systems.
Frequently Asked Questions
Use multiband compression only when you have a frequency-specific dynamic problem — for example, when your low end is too uneven but the midrange and high end are fine. If you can't articulate exactly which frequency range has the problem, reach for a standard compressor first. Multiband compression is best suited for mastering, bass instruments with specific issues, and vocal bus situations with low-mid buildup.
A multiband compressor uses crossover filters to divide the signal into separate frequency bands, typically three to five bands though some offer more. Each band then passes through its own independent compressor with separate threshold, ratio, attack, release, and makeup gain settings. After individual processing, all bands are summed back together into a single output signal.
The key advantage is that loud transients in one frequency range only trigger that specific band's compressor, leaving other frequencies unaffected. For instance, a kick drum peak in the low end won't trigger compression in the high band, allowing you to maintain dynamic control independently across frequency ranges. This provides frequency-targeted dynamic control that a single-band compressor cannot achieve.
The most common mistake is using multiband compression as a default compressor on every track, rather than as a specialized tool for specific frequency problems. This leads to over-processed, frequency-fragmented signals that sound dense without feeling alive. Multiband compression should only be applied when you have a clear, frequency-specific dynamic issue that standard compression cannot solve.
Multiband compression is most effective on the master bus for final tonal balance control, on bass instruments with specific frequency-related dynamic issues, and on vocal bus situations where low-mid buildup needs frequency-targeted control. These are the contexts where the tool solves problems that standard compression cannot address, rather than being applied indiscriminately across your mix.
Most multiband compressors use three to five bands, though some processors offer more options. The typical three to five band approach provides a good balance between granular frequency control and the risk of over-processing. More bands allow finer frequency control but increase the complexity and potential for creating artifacts if not used carefully.
Each band in a multiband compressor has its own independent threshold, ratio, attack, release, and makeup gain settings. This independent control over each frequency band is what allows you to apply different compression characteristics to the low end, midrange, and high end simultaneously based on each band's specific dynamic needs.
Multiband compression is seductive because independent control over every frequency range sounds like more control, and more control intuitively sounds like it should produce better results. However, in practice, more control means more ways to make something sound wrong, leading to over-processing and frequency fragmentation. The tool requires clear understanding of what problem you're solving rather than applying it for perceived control.