25 Tips for Compression
Compression is a vital part of everyday music-making and production – and recently the subject of much intense debate as the Loudness Wars continue to rage... So how and when is it best to deploy it? Hollin Jones gives you the lowdown.
Use with caution
Compression should not be seen as a way of squeezing elements of a track too hard, but rather as a means of keeping things controlled and smooth within a mix. Settings are very much project and genre-dependent, but on a typical guitar track you might want to look at using a ratio of around 4:1 for moderate compression or perhaps 8:1 for heavier stuff like lead guitars. Compress drums individually if possible rather than as a whole, as different parts of the kit require different treatments. Bass often requires a little more compression than other tracks because of the large difference in volume and energy between low and high bass notes.
Split the bands
Consider using a multi-band compressor in situations where you are dealing with audio material that is more complex, such as an entire mix. You can have three or four bands, and each has its own set of gain controls, enabling you to shape each band separately. The bottom end of the track, for example, could have slow attack and release times and a smooth amount of gain reduction. The top end, on the other hand, could require faster attack and release settings so that it is more reactive to peaks. A benefit of multi-band compressors is that no single frequency controls the amount of gain reduction applied to the whole track, so you can push each one harder (or, indeed, leave it more natural-sounding) but treat other parts of the frequency spectrum differently.
Round the side
An interesting use for compression can be found if your compressor supports sidechaining. This enables you to use one signal to control how much compression is applied to another signal. For example, you could feed a compressor’s external sidechain input with a copy of a kick drum track and then feed a bass track to the compressor’s processing input. If you use a very low attack alongside a low or medium release value on the compressor, the result should be that the volume of the bass part is pulled down more or less instantly every time the kick drum sounds and then released just as quickly. This can avoid the overwhelming volume that can result from two very bass-heavy sounds occurring at the same time, but leave the bass louder when it isn’t coinciding with the kick drum part.
Get 3 issues of Music Tech for just £5 saving 72% or USA 12 issues for $99
Every issue has 25 tips, software tutorials, production features, news & reviews
PLUS a free DVD
Signal path
When compressing bass guitar, consider placing the compressor before any EQ in your signal chain. If you do this the other way round, the frequencies you boost with EQ will be compressed more and you might lose, for example, highs and mids. If you compress the signal prior to EQ’ing it and then use EQ to add bottom end, you will end up with a more even, weighty and accurate sound. You can use compression pedals but it’s sometimes better to add the effect as a plug-in so that it can be freely modified after recording.
Take control
Sidechaining your compressor – if this feature is supported – is a great way to achieve advanced dynamics control without the tedious business of automating compressor settings repeatedly over the duration of a track. You don’t even have to use an actual element from your track as the sidechain input source – rather, you could use some kind of rhythmic source (such as a MIDI-triggered instrument or a sample) to apply a rhythmic feel to the way that another part is compressed.
Blend it in
You can achieve some interesting effects by blending compressed and uncompressed versions of the same signal. For example, you could try taking the lowest things in a track, typically the bass and the kick drum, and bussing them to a stereo compressor. Apply a relatively large amount of compression to these so that they are squeezed – the exact amount will be up to you – then route them back to a new channel and blend that with the uncompressed signal. This can help the bottom end sound controlled while still leaving space for other sound sources, such as guitars or vocals.
Make-up gain
Remember to pay attention to the make-up gain control, which your compressor will almost certainly have. This is especially important in the case of multi-band compressors, since compression effectively turns down the signal, so you could end up with certain frequencies being controlled but too quiet in relation to the rest of the track. Bands that have been more heavily compressed may therefore need more make-up gain applied to bring them back up. Be careful, however, since what you are doing is almost re-mixing or re-EQ’ing the track by doing this. You should be using only a few dB of difference when making up gain. Many compressors have an auto control for this, which can be helpful.
Master buss compression
Stereo buss compression can be useful, but remember that it affects the whole mix. Producers sometimes add a dB or two of compression when mixing down to add some punch or clarity to the mix. This is fine, but how much you add depends on how the track is going to be mastered. Usually, mastering engineers will add compression and limiting to a track during the mastering process, so it may not be necessary to add it at the mixdown stage (at least, not across the whole mix). Radio and TV broadcasters also add their own form of volume boosting during transmission, so you can see that it’s easy to get a little carried away with compression.
Parallel lines
To compress during mastering, consider using parallel compression. Take one version of the track and apply moderate mastering processing (compression, limiting and EQ). Duplicate the audio file to a new audio track – perfectly in sync with the first – and dial in some more extreme settings. Use the mixer faders to slowly blend in a small amount of the heavily processed track with the original and see how this affects the overall sound.
Hardware compression
If you’re recording live sources such as guitars, drums or vocals, it’s possible to apply a little compression between the source and your DAW to control any unexpected peaks in the signal. This works best, however, when the effect is inserted between the two end points and as such is rather easier to achieve with hardware effects. It’s also common to record a bass, for example, through a compressor pedal to give a more consistent sound. Remember that any processing applied during recording can’t be removed later.
Let it breathe
The best way to get compression to add punch to a track is to allow the attack of the sound through and experiment with release times to tweak the effect. Faster attack times will reduce the amount of punch that the signal has; slower release times can cause an unwanted ‘pumping’ effect as the compressor takes too long to ‘let go’. The ideal setup, which you can find by experimentation, is to make the compressor seem to breathe at the same pace as the track. For hardcore dance music you might actually want to encourage the pumping effect.
Mastering compression
If you are compressing at the mastering stage, it’s helpful to know if compression has already been applied across the stereo buss (and if so, how much). Compression settings in mastering range from around 1.5:1 and 3:1 and thresholds can be around -20 to -10. If you’re unsure about how these kinds of settings will affect your masters, try loading up a preset in your mastering plug-in; look at its settings and how these are changing the sound. Try different presets and you’ll quickly start to understand how different settings will affect that particular track.
Be selective
The amount of compression you apply depends greatly on what you are working on. 6dB or less of compression is considered moderate and is useful for controlling peaks and dynamics rather than imparting any specific colour to the sound. Higher levels, such as 15–20dB of compression, can be used on elements such as drums and electric guitars. In the case of snare drums, for example, you can set the threshold of your compressor to either compress every hit or to compress only those hits above a certain level, which will maintain a more natural feel.
Compare and contrast
If you compare, for example, an old jazz record with a modern pop song you will hear a tremendous difference in the dynamic range. The old record will almost certainly sound much quieter and more natural, while the new one will be consistently very loud. This is thanks to compression – and also limiting – being applied to squeeze the pop song very hard and make the ‘quiet’ parts as loud as the loud parts. In general there’s a balance to be struck between these extremes for many genres of music. Compress to give it lots of body, but not to crush or destroy the dynamics.
Mix control
Some compressors have a Mix control, which enables you to feed a certain amount of the dry signal back into the wet signal path, giving you more control over the way in which the compression is working. This is not dissimilar to the way that parallel compression works, only it’s rather easier to set up. Also remember to periodically mute effects such as compressors to remind yourself how they are affecting the sound – it’s easy to forget after prolonged periods of listening.
Split the frequencies
Percussive sounds contain more energy than sounds that occur continuously over time. If you are using a multi-band compressor, consider applying more gentle compression to the higher frequencies – higher threshold and lower ratio settings – which should have the effect of letting the transients of the percussive sounds through while retaining a punchy effect on the sustain of the continuous sounds. Experiment with the crossover points of your multi-band compressor to suit the makeup of the track you are working on.
Choose your weapons
Compression can be used to bring out elements of a mix, but should not necessarily be applied to every track (at least, not heavily). Some elements need to be in the background for the track to work, while others are best kept in the foreground. It’s easy to over-compress and suddenly find that everything’s fighting with everything else. A good rule of thumb is to decide what needs to live where in the mix and in the soundstage, then use light compression more generally and heavy compression more sparingly.
Sub sonics
Very low frequencies can cause problems for compressors (and especially limiters) as they force such effects to work harder to process them, potentially resulting in too much gain reduction being applied across the whole mix and leaving your master quieter than it could otherwise have been (and with less dynamic range). Try using a high-pass filter to filter out any sounds below 30 or 35Hz and see if this frees up more headroom for the track as a whole.
De-essing
A de-esser is basically a compressor with a very specific frequency target. You can use the Threshold setting to determine how much of the sibilance is removed through compression; sometimes there’s also a Frequency control, which enables you to target a precise area. De-essing is almost always best used with caution, since over-application very quickly starts to sound unnatural and squashed. It’s better to take precautions during recording – by using a good popshield – than try to fix problems afterwards.
Background noise
Compression can amplify the noise in the sound being treated because when a signal is below the threshold, the compressor raises the gain. It’s possible, therefore, to accidentally raise the volume of hiss and other background noise during compression. It’s important, however, not to obsess about removing all of this, as in doing so you will almost certainly lose part of the music that you wanted to keep. A little hiss is a natural part of recording, so be gentle in your efforts to eliminate it.
Compression vs limiting
Although compressors and limiters both affect volume, they must be used in different ways. Compression changes a sound more than limiting does – compression is best thought of as a tool for changing the dynamics of the music in a way that is more subtle and musical. Limiters are best when they are more transparent, not colouring the sound too much but adding loudness without significantly altering the character of the sound. A side effect of limiting can be to add punch and clarity, which is often desirable.
Simple compression
One of the most gentle forms of compression is to use it simply for controlling the peaks in recorded material, leaving the majority of the recording untouched and unaffected. To achieve this you would usually set a threshold level that is slightly above the average level of the track. By using this technique only the peaks will be compressed; you could compress them more heavily by raising the ratio setting, up as far as 8:1. Consider an upper limit of around 10dB of gain reduction as a rule of thumb.
Total effects
Another way in which to use a compressor is to process an entire track. In this case you might want to set a low ratio of between 1.1:1 and 1.5:1 and a threshold of around 30–35dB below the peak level of the track. You will need to play with the levels a little until your specific track sounds right to you, but this is a good way to help sounds blend into a mix better, gently evening out the differences between quiet and loud passages without dramatically reducing dynamic range.
Less is more
If you are applying master buss compression when mixing down whole tracks or stems, err on the side of caution. The idea during mixdown isn’t to squeeze loads of volume out of the track – that can be done with a limiter at the mastering stage. Instead, aim to make the track gel together properly; to sound like it’s musically finished, if not really loud as yet. Also bear in mind that any tracks in the mix that have heavy compression applied to their channel might sound over-compressed when further compression is applied down the line, so take care.
Crossing over
When using a multi-band compressor, be aware of the crossover points – many models let you set these manually (within certain limits). Listen to the track and identify where the different frequencies intersect. Admittedly, this is easier in dub than in rock, for example, but you can use multi-band compression to gain better control over the overall dynamics of a track (providing that the crossover points are correctly set). Remember that your mix should be correct prior to mixdown – don’t rely on multi-band compression or mastering to fix it.
Subscribe for just £5 for 3 issues or USA 12 issues for $99
Read more Production Features
Reverb Masterclass
This feature first appeared in Music Tech Magazine issue 97
Filed under
Home,
General Features,
Features
Sign in to download this article
New users, please register here
See also...
 MTM 104 |
Steinberg RND Portico Plug-Ins Review |
 MTM 104 |
Audio-Technica AT2022 Review |
|
 MTM 103 |
25 Tips for Cubase 6 |
MTM 102 |
Ultimate Guide to Filtering |
|
MTM 103 |
Harness the power of additive synthesis |
MTM 103 |
Bob Katz Interviewed |
|
MTM 103 |
Drum Production Masterclass |
 MTM 105 |
Bettermaker EQ 230P Review |
