COMPRESSION IN THE STUDIO - PART TWO
LAST ISSUE, WE LOOKED BROADLY AT THE VARIOUS TYPES OF COMPRESSORS AND THEIR BASIC FUNCTIONS AND CONTROLS. IN THIS ONE, WE’LL TAKE A LOOK AT VARIOUS WAYS OF USING COMPRESSORS IN THE STUDIO AS CORRECTIVE AND CREATIVE TOOLS.
LIMITING VERSUS COMPRESSION
A good starting point is to grasp the difference between compression and limiting – quite simply, it’s all about the ratio. A compressor has a low ratio that gently attenuates the level of the source material when it goes beyond the threshold, whereas a limiter has a high ratio that limits all of the source material once it hits the threshold. In a nutshell, a limiter is a compressor with a high ratio and, generally, a fast attack time. Compression with ratio of ten‑to‑one or more is generally considered limiting. Brickwall limiting (discussed later in detail) has a very high ratio and a very fast attack time.
GLUING YOUR MIX TOGETHER
Many tracks these days are put together piece by piece, multi‑tracked at different times in different spaces and even different studios. Regardless of the genre, at the end of the day it’s imperative to have your track sounding like a coherent and cohesive entity, where everything sounds as though it belongs together. There are many factors contributing to this, and many ways in which to enhance the ‘togetherness’ of a final mix – none as powerful as compression.
The ‘glue’ effect happens when you are treating either the entire mix together, or subgroups of tracks – for example, rhythm sections, guitars and backing vocals. The general approach is to use a very low ratio, somewhere between 1.5‑to‑one to two‑to‑one, since your goal here is not to compress the signal too much. Use a long attack time, to avoid deteriorating the peaks. The release time ought to follow the groove, so the compressor breathes naturally, following the tempo and rhythm of the music. Use a relatively soft knee setting, so the compressor acts progressively, but not too much, otherwise the compression will be ‘too audible’.
The threshold should be set so that you get around ‑1dB (decibels) of gain reduction, so that all you get is a very gentle effect. When choosing the ‘glue’ compressor, it’s usually best to go for something with character that may introduce some harmonic distortion, like a vintage analog compressor or a serious virtual emulation of one. The distortion acts as a final layer of varnish on your track.
TWO-STAGE COMPRESSION
This technique involves using two compressors on one source in a chain. Sometimes it’s preferable to treat dynamics in two stages, with two different devices. Rather than try to tame peaks and give the track a nice, warm audio massage with one unit, why not use a separate device to deal with the stray spikes, and use a second unit to deal with more general compression duties? It makes perfect sense, as each task requires very different settings, and ironically, compressors work best when being fed with a limited dynamic range – it allows for a smoother transition between contrasting sections.
For example, think of how dynamic the human voice can be. It needs to be treated with care in order to retain some of its natural nuances, while at the same time keep things within an audible range in the mix, whilst using compression to enhance the vibe and sonic flavour. Two compression units in tandem will give the engineer much more control over how the voice sits in the mix. Another great source to try this on is bass.
SIDECHAIN COMPRESSION
‘Sidechain compression’ is the process of using one sound source to trigger compression on another sound source. Originally, this technique was mainly used to automatically ‘duck’ a music track to allow for a voiceover in radio or TV. Many software compressors have a sidechain input built in, and some hardware units will have a physical input jack.
A classic application of this technique (especially in dance music) is to place a compressor over a bass line, then trigger its sidechain input from the kick drum track. This effectively drops the bass level on every kick hit, which helps the kick punch through, whilst ‘sucking’ the bass back momentarily.
Extending the idea, it can be applied to a sound source that needs to drop in level to allow room for another part to take precedence momentarily, such as triggering a string pad via the melody instrument to allow it to poke through the mix. Check out some EDM tracks to hear the effect – or better still, try it yourself!
MULTIBAND COMPRESSION
One small, yet critical piece of information for those working
with audio compressors is that low frequencies have more energy or power for the same perceived loudness. The upshot of this is that lower frequencies will trigger more compression than you'll have at higher frequencies.
What this means is that if you slap a compressor over any source which encompasses a broad range of frequencies – from very low to very high (a full mix, a drum kit sub mix, some piano or synth parts...) – the lower frequencies will trigger the compressor disproportionately to the mids and highs. Thus, you can end up with a mix which virtually pulses in time with the low-end parts. This has its place, of course, but is often an unnatural, ugly sound.
Normally when you place a compressor or limiter over an audio source, the entire frequency range is treated together. A multiband compressor allows the user to select specifically which frequencies or ranges are compressed. This is a powerful tool for treating issues across entire mixes, or on instruments which cover a broad frequency range.
It’s possible to surgically zoom in on frequencies that are creating problems, and tame them without needing to rely solely on EQ. Using compression can be a smoother approach than simply dipping a frequency with EQ. Typical uses include master buss for clearing up problem areas like low-mid buildup, or on lead vocals to tame harshness in the 5-10k range.
DE-ESSER
An even more surgical way of compressing source material at specifically chosen frequencies can be achieved by using a de-esser. This is usually used across a vocal track to reduce the sibilant ‘ess’ sounds that often occur on many vocal recordings with words emphasising the letter S.
There are a number of factors which can lead to overly sibilant vocals – over-compressing a vocal; recording a vocal in a poor acoustic environment, over-boosting frequencies that cause sibilance, or simply recording a singer who has a naturally sibilant voice. Treating sibilance with a broad, EQ-cutting top-end is usually far too extreme, and results in a lifeless, dull vocal with little presence.
Any compressor with a sidechain input can be turned into a de-esser by inserting an EQ and boosting the offending frequencies – many modern de-essers, however, have very sophisticated circuitry and controls that are optimised for achieving results beyond what would be easy with a simple compressor with an EQ in the sidechain.
LOOK-AHEAD COMPRESSION
‘Look-ahead’ compression analyses an input signal ahead of time, and essentially applies compression (based on your desired settings) before the transient actually occurs. It’s most useful in ‘limiting’ applications, as it prevents the ugly crushing effect that results from a peak being unnaturally slammed by a limiter. Other practical applications include anything with prominent, fast transients – for example, snare drums.
An interesting way to explore the concept using a regular compressor is by duplicating the original signal onto another track in your DAW, shifting the file back a few milliseconds in time, then placing a compressor on the original signal, and using the duplicated audio as the sidechain input to trigger the compressor.
Parallel compression refers to feeding signal from a source you’d like to compress into a parallel channel, FX or auxiliary buss which is heavily compressed. Essentially, you end up with a dry fader, and a compressed fader side-by-side. The idea is simply that the original audio remains natural, unaltered and unprocessed, whilst the compressed signal adds colour and can be blended at will. This is especially good for drums, as you retain all the original punch and fatness, whilst dialling in the room and vibe of the compression track as desired.
BRICKWALL LIMITING
This technique simply means setting an absolute ceiling on the level of a track by ensuring it cannot pass a certain threshold. This is generally used to control peaks and clipping on a mix, and is the last thing in the chain. It has technical functionality as system ‘protection', particularly in live applications and broadcast.
In the studio, this style of limiting is intended to make the overall mix as loud as possible without clipping – the quiet passages are raised in level to effectively minimise dynamics in the track. For some genres this is desirable, whereas for others, the ‘volume war’ does little to enhance the integrity of the work and only crushes dynamics.
Compressors are fun, and can be some of the most creative tools in the studio at any level. Whether you use software, hardware, or a combination or both, the technical and creative variables that can be obtained are... Dare I say it... Limitless.