Limiter Plugins Explained: Practical Setup & LUFS Tips
Limiter Plugins Explained: Practical Setup & LUFS Tips
A limiter plugin is an extreme form of compression that enforces a hard ceiling on your audio signal, preventing any peak from crossing a set output level. It's the last plugin on your master bus, the safety net between your mix and digital clipping, and the tool that lets you push perceived loudness without destroying your track.
- Core function: Stops peaks above a defined ceiling (typically −1.0 dBTP for streaming)
- Primary placement: Master bus, final position in the signal chain
- True peak vs. sample peak: Sample-domain limiters catch individual samples; true-peak limiters also catch inter-sample peaks that appear during lossy encoding for streaming platforms
- Loudness management: Lets you increase input gain to raise LUFS without allowing the output to clip
Key Takeaways
A limiter plugin is a brickwall ceiling tool, and using it correctly means setting the true-peak ceiling first, gain-matching your AB tests, and stopping at your LUFS target rather than pushing for maximum loudness.
| Point | Details |
|---|---|
| Definition and placement | A limiter enforces a hard output ceiling; place it last on the master bus after all other processing. |
| True-peak ceiling | Set the ceiling to −1.0 dBTP and enable true-peak mode for any track headed to streaming platforms. |
| Gain-matched AB testing | Always bypass with level-matched gain to evaluate what the limiter actually does to the sound. |
| Avoid over-limiting | Streaming platforms normalize to a LUFS target, so excessive limiting reduces dynamic range without gaining loudness. |
| Vector-dsp approach | Vector-dsp plugins prioritize precise DSP, configurable release curves, and true-peak compliance for transparent mastering results. |
Table of Contents
- What is a limiter plugin, and how does it differ from a compressor?
- How a limiter actually works inside
- What types of limiters should you know about?
- Where in your session should you use a limiter?
- How to set up a limiter step by step
- How to hear what your limiter is doing
- Starting settings for common use cases
- Common limiter mistakes and how to fix them
- DSP design choices that shape how limiters sound
- The case for using less limiting than you think you need
- Vector-dsp builds limiters the way mastering engineers think about them
- Sources
What is a limiter plugin, and how does it differ from a compressor?
Think of a limiter as a bouncer at the door of your master bus. A compressor negotiates with loud signals, turning them down by a set ratio (4:1, 8:1) and letting some peaks through. A limiter doesn't negotiate. Once a signal hits the threshold, the ratio becomes effectively infinity:1, and nothing gets past the ceiling.
Compressors shape dynamics across a range of levels, adding character, controlling sustain, and sculpting the envelope of a sound. A limiter enforces a maximum output level, primarily for safety and loudness control rather than tonal shaping. As Violet Recording explains, a limiter differs from a compressor in ratio and function: compressors shape dynamics, limiters enforce a ceiling.
The attack on a limiter is typically near-instantaneous or uses lookahead to anticipate peaks before they arrive. That speed is what makes it effective as a brickwall device. A compressor with a slow attack lets transients breathe; a limiter catches them before they can cause a problem.
How a limiter actually works inside
When audio enters a limiter, the detector circuit continuously monitors the signal level. In a sample-domain limiter, the detector watches individual samples. When a sample would exceed the ceiling, the gain reduction circuit attenuates the signal just enough to keep it at or below that ceiling.
True-peak limiters go a step further. They oversample the signal (typically 4x or higher) to reconstruct what the waveform looks like between samples. Inter-sample peaks occur when a digital-to-analog converter or codec reconstructs audio and the waveform overshoots between two samples that individually appear safe. A true-peak limiter catches those overshoots. For tracks headed to streaming platforms, this matters.
Key parameters you'll encounter on any limiter:
- Ceiling / Output level: The hard maximum output level, expressed in dBFS or dBTP. Set this before anything else.
- Threshold / Input gain: How hard you're pushing the signal into the limiter. More input gain means more gain reduction and more loudness.
- Lookahead: A short delay (typically 0.1–10 ms) that lets the limiter "see" peaks before they arrive, enabling smoother gain reduction.
- Attack: How fast the limiter responds to a peak. Most brickwall limiters have this fixed or near-zero.
- Release: How quickly the limiter lets go after a peak passes. Too fast causes pumping; too slow kills transients.
- Makeup gain: Additional output gain applied after limiting, though on a master limiter this is usually handled by the ceiling/input gain relationship.
Logic Pro's built-in Limiter exposes these controls clearly and is a reliable reference for mapping general parameter names to specific GUI labels in your DAW.
Limiter vs. compressor at a glance:
| Parameter | Compressor | Limiter |
|---|---|---|
| Ratio | 2:1 to 8:1 | Effectively infinity:1 |
| Attack | Adjustable (slow to fast) | Near-instant or lookahead-based |
| Release | Adjustable | Adjustable, but critical for transparency |
| Primary purpose | Dynamic shaping, tone | Peak control, loudness ceiling |
| Typical placement | Individual tracks, buses | Master bus, final stage |
What types of limiters should you know about?
Not every limiter works the same way under the hood, and picking the wrong type for a job produces audible artifacts fast.
-
Brickwall / sample-peak limiters: The most common type. They enforce a hard ceiling at the sample level with near-zero attack. Fast and effective for preventing clipping, but they can miss inter-sample peaks. Best for quick mix-bus protection or when streaming compliance isn't a concern.
-
True-peak limiters: Oversample the signal to detect and control inter-sample peaks. Recommended for any track headed to streaming platforms like Spotify, Apple Music, or Tidal, where lossy encoding can push inter-sample peaks above 0 dBFS even when your sample-peak ceiling looks clean. The trade-off is slightly higher CPU cost.
-
Lookahead limiters: Use a short delay buffer to anticipate peaks before they arrive. This produces smoother, more transparent gain reduction because the limiter isn't reacting after the fact. The latency introduced is usually compensated automatically by your DAW, but it's worth knowing it exists.
-
Multiband limiters: Split the signal into frequency bands and limit each independently. Useful when a specific frequency range (a boomy low end, a harsh high-mid) is triggering excessive gain reduction across the whole signal. More surgical, but more complex to set up and easier to misuse.
-
Maximizers / combined limiters: These combine a soft-clipper or saturation stage with a true-peak limiter. The soft-clipper rounds off transient peaks before they hit the brickwall stage, which reduces the amount of hard limiting needed and often sounds more natural at high loudness levels. Multi-stage architectures like this, where a compressor feeds into a clipper and then a true-peak limiter, reveal exactly what each stage contributes to the final sound.
Where in your session should you use a limiter?
Master bus: This is the canonical slot. A limiter is the final plugin in a mastering chain, sitting after EQ, compression, and any stereo processing. Its job here is to set the true-peak ceiling and bring the track to its target loudness. Every mix headed for streaming, broadcast, or distribution needs a limiter in this position.
Mix buses and sub-buses: Limiters on a drum bus can tame transient spikes that would otherwise force you to pull the whole bus down in level. On a vocal bus, a limiter after a compressor acts as a safety catch for any peaks the compressor misses. Keep the gain reduction light here (1–3 dB maximum) or you'll start hearing the limiter working, which is rarely what you want on a sub-bus.
Individual tracks: Occasionally useful on a single track with unpredictable peaks (a live vocal, a snare with inconsistent hits) but generally a compressor is the better tool at this stage. A limiter on a single track can kill the transient character that makes the sound interesting.
What to avoid: Stacking multiple limiters in series on the master bus is a common mistake. Each limiter in the chain reacts to what the previous one left behind, and the cumulative effect is audible pumping and a loss of punch. One well-configured limiter at the end of the chain does the job cleaner than three mediocre ones.
How to set up a limiter step by step
Follow this workflow in your DAW and you'll avoid the most common setup errors. Good gain staging before the limiter is what separates a clean master from a squashed one.
-
Check your gain staging. Before touching the limiter, make sure your mix bus isn't already hitting 0 dBFS. A pre-limiter level of around −6 to −3 dBFS gives the limiter something to work with without forcing it into extreme gain reduction from the start.
-
Set the ceiling first. Enable true-peak mode if your limiter offers it, then set the ceiling to −1.0 dBTP. This is the standard recommendation for most streaming platforms; some engineers prefer −0.5 dBTP for extra headroom. Set this before you touch input gain.
-
Push the input gain (threshold) until you reach your loudness target. Watch the gain reduction meter. Streaming platforms normalize loudness, so target levels vary and should be set according to the platform's guidelines. Excessive gain reduction indicates the mix may need more headroom before limiting.
-
Dial in the release. Start with a medium release (around 50–100 ms) and listen for pumping. Shorten the release if the limiter is holding on too long and killing sustain; lengthen it if you hear the gain reduction cycling in a rhythmic way. Some limiters offer an auto-release mode that adapts to the program material, which is a good starting point.
-
Set the lookahead (if adjustable). A lookahead of 1–3 ms is a reasonable starting point for transparent limiting. Longer lookahead (5–10 ms) gives smoother results on dense material but introduces more latency.
-
Gain-match and AB. Bypass the limiter, then re-engage it. The bypassed version will be quieter. Use your DAW's gain trim or a utility plugin to match the bypassed level to the limited level, then compare. Gain-matched AB testing is the only reliable way to evaluate what the limiter is actually doing to the sound, because louder always sounds better to the human ear.
Metering checklist:
- Integrated LUFS meter (check against your target platform's spec)
- True-peak meter (confirm ceiling is holding at −1.0 dBTP or lower)
- Gain reduction meter (watch for excessive or erratic movement)
Pro Tip: After setting your limiter, export a short 30-second section and run it through a loudness normalizer or check it in a streaming preview tool. What looks correct on your meter can still behave differently after codec encoding.
How to hear what your limiter is doing
Metering tells you what's happening numerically. Your ears tell you whether it sounds right. These five checks cover the most common artifacts.
Transient clarity. Solo the drums or the loudest percussive element and A/B the limiter. The attack of a snare or kick should still feel immediate. If it sounds rounded or soft, the limiter is catching the transient itself, not just the peak.
Perceived loudness vs. dynamics. The limited version should feel louder and more present without sounding flat. If the track loses its sense of movement, you're over-limiting. The loudness war produced decades of records that were loud but fatiguing precisely because this balance was ignored.
Pumping and breathing. Listen to sustained sections (a pad, a long reverb tail, a held chord). If the level drops noticeably after a loud transient and then rises back up, the release is too fast or the gain reduction is too deep. Slow the release and reduce the input gain until the effect disappears.
High-frequency harshness. Aggressive limiting can introduce distortion that shows up as brittleness in the upper frequencies. Narrow-band listening (solo a high-shelf EQ band above 8 kHz) makes this easier to hear. Oversampling or true-peak mode often reduces this artifact.
Stereo image shifts. Heavy limiting can cause the stereo field to narrow or shift. Check in mono, and compare the stereo width of the limited and bypassed versions. If the image collapses, reduce the gain reduction or consider mid-side processing before the limiter.
Pro Tip: Use a spectrum analyzer alongside your limiter during the AB test. Distortion from over-limiting often shows up as added harmonic content in the 2–8 kHz range before you can clearly hear it.
Starting settings for common use cases
These ranges give you a practical starting point. Adjust from there based on what you hear.
A few notes on these ranges. Vocals and snares benefit from a faster release because the transients are short and the limiter needs to recover quickly between hits. On a full mastering chain, auto-release or a longer release time preserves the sense of dynamics better across a full song. Transparent algorithm modes (linear-phase, high oversampling) suit mastering; more aggressive or "character" modes can work on individual buses where some color is acceptable. Key parameters including threshold, ceiling, release, attack, and lookahead interact with each other, so change one at a time and listen after each adjustment.
Common limiter mistakes and how to fix them
Pumping on the master bus. The release is too fast for the program material. Lengthen the release time or switch to auto-release. If pumping persists, reduce the input gain and check whether the mix needs more headroom before the limiter.
Brittle, harsh high frequencies. Over-limiting is generating harmonic distortion. Reduce the input gain, enable oversampling or true-peak mode, or add a soft-clipper stage before the limiter to round off peaks before they hit the brickwall.
Loss of punch and transient impact. The limiter is catching the attack of percussive elements. Increase the lookahead slightly so the limiter anticipates peaks rather than reacting to them, or use a shorter release to let the gain recover faster between transients.
Stereo image collapse. Heavy gain reduction on a stereo bus can cause phase issues. Check that your limiter is operating in linked stereo mode (both channels gain-reduced equally). If the problem persists, reduce the limiting depth and address the loudness elsewhere in the chain.
True-peak violations after export. Your sample-peak ceiling looks fine but the exported file clips. Enable true-peak mode and set the ceiling to −1.0 dBTP, not −1.0 dBFS. These are different measurements.
Stacking multiple limiters. Each limiter reacts to the residual peaks left by the previous one, creating unpredictable interaction. Use one limiter at the end of the chain and address loudness upstream with compression and gain staging.
Pro Tip: To preserve transients without sacrificing loudness, try placing a clipper (soft or hard) before your limiter. The clipper shaves the very tip of transient peaks with minimal audible effect, which means the limiter sees a less aggressive signal and needs to work less hard. The result is more loudness with less pumping.
DSP design choices that shape how limiters sound
The difference between a limiter that sounds transparent and one that audibly squashes your mix often comes down to three design decisions: release curve shape, lookahead implementation, and detection mode.
Release curve shape determines how the gain reduction recovers after a peak. A linear release ramps gain back up at a constant rate, which can sound mechanical on program material. A logarithmic or program-dependent release curve follows the natural decay of audio more closely, which is why "auto-release" modes in well-designed limiters tend to sound more natural than a fixed setting. The curve shape also affects how the limiter handles back-to-back peaks: a poorly designed release can cause the gain reduction to stack up on dense material.
Lookahead latency vs. transparency is a genuine trade-off. Longer lookahead gives the gain reduction circuit more time to act smoothly, which reduces distortion artifacts. But every millisecond of lookahead adds latency to the signal path. In a mastering context where the DAW compensates for plugin delay automatically, this is rarely a problem. In a live or near-real-time context, it matters. DSP algorithm choices directly affect this balance, and the best implementations minimize latency while still achieving smooth gain reduction.
Peak detection vs. RMS detection changes what the limiter responds to. Peak detection catches instantaneous sample peaks, which is what you need for a brickwall ceiling. RMS detection averages the signal over a short window, which is more representative of perceived loudness but slower to react. Most brickwall limiters use peak detection for the ceiling and may use RMS or a hybrid approach for program-dependent release behavior.
True-peak detection adds another layer: oversampling the signal to reconstruct inter-sample content before applying gain reduction. This is computationally more expensive than sample-peak detection, but it's the only reliable way to prevent inter-sample peaks from causing clipping after encoding. Vector-dsp's design priorities center on precision, low-latency processing, and true-peak compliance, which reflects the practical reality that a limiter that misses inter-sample peaks is not actually protecting your master.
Multi-stage architectures, where a compressor feeds into a soft-clipper and then a true-peak limiter, distribute the workload across stages. Each stage handles a different aspect of the signal, which reduces the amount of hard limiting any single stage needs to do. The result is typically more loudness with fewer audible artifacts than a single-stage brickwall limiter pushed hard.
The case for using less limiting than you think you need
Most producers over-limit. The instinct is understandable: louder sounds better in the moment, and a limiter makes it easy to push the level up. But the tracks that hold up over repeated listening are almost always the ones where the limiter was used as a safety net, not a loudness engine.
Streaming platforms normalize loudness to a target LUFS level, which means an over-limited track at −8 LUFS will be turned down to match a well-mastered track at −14 LUFS. The over-limited version doesn't win on loudness; it just arrives at the listener with less dynamic range and more distortion. The loudness war taught the industry this lesson the hard way, and the shift to LUFS-based normalization was the direct response.
The practical implication: set your ceiling, hit your LUFS target, and stop. If the mix sounds flat at the target loudness, the problem is in the mix, not the limiter. More limiting won't fix a mix that lacks energy; it will just make the flatness louder.
Vector-dsp builds limiters the way mastering engineers think about them
The gap between a limiter that technically prevents clipping and one that sounds transparent at high loudness comes down to the DSP decisions made before you ever touch a control. Vector-dsp approaches plugin design from that starting point: precise detection, configurable release curves, and true-peak compliance built into the architecture rather than bolted on.
For producers and engineers who want meticulous control over peak management without the guesswork, Vector-dsp plugins are available in VST3, AU, and AAX formats, compatible with every major DAW on Windows and macOS. A free demo is available before any purchase commitment. Visit Vector-dsp to explore the current lineup and download a demo.
Sources
- Limiter in Logic Pro for Mac
- How To Use A Limiter in Mastering (Parameters Explained)
- What Is a Limiter? Mixing & Mastering Guide