8 Tips for Compression in Mastering
Why compress during mastering? Is every compressor a mastering compressor? When is it best to cascade your compressors? Get tips on these questions and more.
Like equalization and limiting, compression is an important part of mastering. It’s a great way to add punch to your mix as well as a sense of overall control. Used correctly, compression can make your recordings sound richer and more energetic, plus it can help ensure that the various sections of your song flow well into one another. Perhaps most importantly, compression can act as the “glue” that ties all the sonic elements together into a cohesive whole. But, like every other process, it has its potential downfalls too. In this article, we’ll look at several important tips for using compression during mastering.
1. Know the Basics
Compression works by lowering the level of the loudest parts in your mix – the peaks you see sticking out when you view the waveform. By making them softer (a process called gain reduction), it reduces the overall dynamic range – that is, the difference between the loudest and quietest sections of a song.
You’ll find pretty much the same set of controls in every compressor plugin.
Threshold and Ratio: Threshold determines when compression begins, while ratio specifies how much gain reduction is applied. These two controls are highly interactive. A low threshold with a high ratio will cause lots of compression and result in a sound that is quite squashed and lifeless, whereas a high threshold and low ratio will yield the kind of transparency you want during mastering.
Attack and Release: Attack determines how quickly the compressor begins working once a signal exceeds the threshold setting, while release determines how quickly it stops working after the signal drops below the threshold. These two controls are so important that we’re going to devote an entire tip to them – see #5 below.
Soft vs. Hard Knee: A knee setting specifies the smoothness of the transition from no compression to full compression once the threshold is passed. A “hard” knee means that the compressor clamps down on the signal almost immediately, while a “soft” knee means the compression kicks in more gradually as the signal goes further past the threshold. Because “softer” knee settings are less aggressive, they are used far more often in mastering situations.
Make-Up Gain: This parameter allows you to boost the compressed signal, enabling you to bring your track back to its starting level, although it now ‘lives’ within a compressed, or reduced, dynamic range. The end result is that everything sounds not only louder and punchier but more even overall.
2. Compression in Mastering Is Optional, Not Essential
If you’ve already applied a fair amount of compression to the stereo buss while mixing, you may not need to use it again during mastering. Just by looking at the waveform, you should able to tell whether there are lots of peaks in the track; if not, you probably don’t need to add any compression. But as with everything audio, let your ears, not your eyes, be your judge. Don’t just use compression for the sake of it – be sure it has a purpose.
3. Not Every Compressor Is a Mastering Compressor
Not all compression plugins are able to handle the large amount of sonic information contained within a mix. Those designed for a single instrument or group of instruments may struggle when used for mastering purposes – they may color the sound in undesirable ways or even produce distortion when driven hard. Compressors designed to cope with mixes are often called buss compressors since they are usually inserted across a stereo buss, as opposed to a single channel. Two examples are the SSL G-Master Buss Compressor and the API 2500 Compressor. These types of plugins are usually the best choice for mastering, although there are times when you may want to insert a secondary compressor for multiband compression or for “color” (see tips #6 and #7 below).
In this short video, mixing and mastering engineer Yoad Nevo (Sia, Pet Shop Boys) shows how he uses the SSL G-Master at the very start of the mastering process to “glue” a song together:
4. Less Is More
As with equalization, the less compression you apply during mastering, the better the result. In fact, the quickest way to make your master sound like a demo is to overcompress it. Most mastering engineers use high thresholds and low ratios (typically 1.25:1 or 1.5:1 – rarely anything more than 2:1) in order to achieve just 1 or 2 dB of gain reduction. The idea is to feel rather than hear any compression being applied.
Engineer Yoad Nevo says, “I hardly use compression in mastering, and if I do, it’s more in terms of coloring the picture rather than trying to change the internal dynamics.” While the application of compression during mastering can help enhance a mix in some circumstances, it’s also important to remember that a recording with a relatively broad dynamic range feels musical and exciting, whereas one with minimal dynamics feels tight and fatiguing. How much dynamic range you opt to preserve is largely a judgement call based on your taste and the genre of the music.
5. It’s All in the Attack and Release
As we said, the attack and release controls are especially important. An attack time that is too short can be damaging to the music, since it will cause the compressor to grab and squash the transients – those short bursts of almost instantaneous sound (such as the beater of the kick drum or the attack of a bass note) that contain all the punch. On the other hand, if the attack time is too long, too much audio will have passed through before the compressor has time to react.
Accordingly, start with an attack time in the 30–40 ms range (an ms, short for millisecond, is a thousandth of a second) and be prepared to dial it up as far as 100 ms. When you’ve got it just right – and bear in mind that you may not be able to ascertain the optimum attack time until you set the release time correctly – the percussive elements of the mix will pass through unscathed, despite the compressor reining in the overall level.
The release time is a little trickier since the goal is to get the action of the compressor to blend in with the musicality of the track. Too short of a release time will result in audible pumping; too long and the compressor will rarely if ever return to its zero state, resulting in nearly constant gain reduction. The sweet spot occurs when the release time complements the attack time. In mastering applications, you’ll generally want to set the release time so that it is slightly longer than a beat. Simply divide 60,000 – the number of milliseconds in a minute – by your track’s beat-per-minute (BPM) tempo to get the number of milliseconds in each beat. This way, the compressor is still working somewhat – though at a reduced intensity – when the next transient (usually the kick or snare) comes along and exceeds the threshold. The result is smooth gain reduction which follows the music, as opposed to it kicking in and out randomly. Depending upon the tempo, typical mastering compressor release times might range from 300 to 800 ms or more.
As you can see, there are no attack and release values that are ideal for every track. When you get them right, your compressor works almost invisibly, but when those settings are too fast or too slow for the music, compression becomes obvious and intrusive.
6. Understand Multiband Compression
While a standard mastering compressor processes the entire track as one sound, a multiband compressor breaks up the frequency spectrum into several bands and allows you to compress or expand each differently. (Expansion is the opposite of compression; by lowering the level of soft signals, it increases rather than decreases dynamic range.)
Because it is more inconspicuous and subtle than traditional compression, multiband compression can be a powerful mastering tool, used for everything from taming overly loud cymbal crashes to tightening up the low end without affecting the rest of the track. Plugins such as the Linear Phase Multiband Compressor (which offers five user-definable frequency bands and advanced mastering features such as adaptive thresholds, automatic makeup gain, and finite response filters) let you focus on specific instruments or areas of a mix and boost or attenuate them with a great degree of accuracy.
As with every other kind of mastering process, multiband compression can yield tremendous results when done properly, but it can also negatively impact your mix if overdone or applied incorrectly. For example, be sure to set the same (or at least similar) ratio in all the bands being affected or you run the risk of adding an imbalance to the sound.
7. Cascade Your Compressors
Sometimes even the best mastering compressor can’t do the job alone. Adding a second compression plugin can be much more effective than simply increasing the intensity of the first one.
To get more gain transparently:
If you’re looking to keep your compression transparent yet still achieve lots of gain, it’s often better to use two similar compressors in series (i.e., one after the other), thus cutting in half the workload on each. For example, the first compressor could be used to begin delicately taming some transients – maybe a dB or so of gain reduction every now and then – by setting it to a very low ratio (1.2:1 or 1.25:1) with a relatively slow attack. The second compressor can then be set to a marginally higher ratio and slightly faster attack in order to go after any remaining peaks a little more aggressively.
To deal with specific problem areas, try using a multiband plugin like the Linear Phase Multiband Compressor as the second compressor; see tip #6 above for more information.
To add color to the mix:
To add color and character to your mix, consider adding as your second compressor a plugin modeled on a tube or transistor-based hardware device. These impart what some call a “fatness” or “warmth” – in reality, an emulation of a signal that is saturated, modulated or slightly out of phase. Analog circuitry also tends to add random anomalies that lead us to believe that we are hearing an improvement over the original signal. A good example of this kind of plugin is the CLA-2A Compressor/Limiter – a particular favorite of Yoad Nevo, who says “even though it’s not really a mastering compressor, it can sound cool because its processing is kind of slow due to the nature of the optical circuitry of the original unit it’s modeling. Using it gives you the best of two different worlds, because the main master buss compressor makes everything punchy and brings the sound closer, whereas the CLA-2A adds some coloring and does something nice to the stereo image, making it slightly wider.”
In this webinar, Nevo shows how to use both CLA-2A and the SSL G-Master plugins when mastering a song:
8. Compare Constantly
When it comes to compression, the Bypass button can be your best friend. Every time you tweak a parameter, click on the button to ensure that you’re actually making the mix sound better. Keep flipping back and forth so you can hear the processed signal versus the original. If your plugin has a “Compare” or "A/B" function that toggles between two sets of parameters (as is provided by most Waves plugins), use it frequently to test different settings in quick succession. This video shows how to use the A/B function in Waves plugins:
Hopefully these tips have given you a better understanding of how to apply compression during mastering. For more mastering tips, check out this article on how to EQ effectively during mastering.