Production Expert

View Original

How Many Times Should You Compress In The Mix?

In Summary

With track compression, submix compression, and treatments across the entire mix all useful tools, engineers often compress more than once, but when is it needed, and why? We shine a light on dynamic dealings for your mix.

Going Deeper

The Done Thing

We are drowning in compressors. OK, that might be a bit over-dramatic, but if anyone reading this has fewer than a small handful of them to choose from I’d like to meet them and ask them their secret. There’s no denying that some choice is good, however despite compressors’ rock-star status among many engineers, all bar none are designed to do one thing: reduce dynamic range. In the cold hard light of day, these devices are not magic but they do need one pretty miraculous ingredient: the human engineer.

When was the last time you mixed without one? With their abundance, it’s hardly surprising that compressors get a lot of use, and the DAW era does away entirely with the idea of finite devices. This has increased the use not only of track compression, but also that across track submixes, plus many engineers’ precious last coat of comp: the mix bus compressor.

With so many opportunities to get down to some dynamic deal-breaking, tracks frequently see two or even three stages of compression. Is this always necessary? If so, under what circumstances? What can go wrong and how? How do we get the best out of so much processing power? Here we take a deeper dive into navigating multi-stage compression.

Why Compress?

As mentioned, a compressor’s function is actually far less glamorous than the marketing would like us to believe. With some branches of the audio engineering family referring to even humble fader rides as “manual compression”, the kind of automatic devices that we’re all used to do exactly the same thing, only far quicker than a human hand can. Whatever the method, compression can do one or both of two things: control level, or shape sounds creatively.

All engineers approach things their own way; one way to think of reasons to compress could look like this:

1- To Create

Creative control often takes advantage of a compressor’s macro-dynamic superpowers. Often reacting on a scale of milli or even micro seconds, compressors can shape not only a sound’s consistency in the mix, but also its very character. Re-shaping envelopes is all in a day’s work when injecting punch into drums, or movement into a mix.

2 - To Control

Level control takes care of those fluctuations that can make things momentarily stick out or get drowned in the vortex of multiple sounds in a mix. Reducing these sources’ dynamic range by turning down peaks and turning up the entire result can help quiet and loud events get closer to each other to sit sounds nicely.

3 - To Cohere

In many ways, this is a combination of the first two. When deployed at the mix’s backend, cohesive compression can not only give a characterful, subtle bounce to the mix, but also further blunt any sneaky transients in danger of turning things red. With mix bus compression in this role, many refer to its effect simply as “glue”.

You can read more about why to compress and its role in level control in Dom Morley’s article Why Are You Compressing Your Mix?.

Where To Compress

Thanks to limitless audio plugin instances, compressors can live across tracks, submixes, final mix output, or any combination of the three. Perhaps one of the best ways to keep on top of what they’re achieving (or not) is to remember the flow of signal.

1 - Across Tracks And Submixes

Just like an EQ cannot boost what isn’t there, any compressor needs some dynamic variation to chew on. For example, if you’re standing hard on the drums at a track level, do they really need that drum bus compression? Lighter track treatments, on the other hand, can be the perfect prelude to some more obvious dynamic bounce across a submix.

2 - Across Mix Busses

Mix bus compression gets a lot of attention as a potential deal breaker between a mix that is stellar or something far less remarkable. It might be true that we are used to hearing the sound of mix compression, making it one of those treatments that becomes almost muscle memory, but again this can only really work its magic when it has something to do. Anyone who has tried to further compress a symphony of compressors might agree that it can be a bit like trying to polish a marble…

Getting It Wrong

1 - Too Little, Too Late

If a bus or mix compressor is being unpredictable, this could be down to track or submix compression that has been a little too forgiving. Going just that bit further on sources can mean that any mix compression will be more consistent and vice versa. Another caveat with too little compression is, well, too little compression. If the source tracks, and/or the mix as a whole sounds a little jumpy, compression is the time honoured fix where fader automation isn’t working.

2 - Too Much, Too Soon

Music needs to breathe. Over-compression is the enemy of musical changes that engage the listener, and unless you’re mixing an experimental foghorn concerto, sources need to tread the line between consistent levels and having a sense of being alive. In a worst case scenario, overcompressed music sounds dead and fatiguing to listen to. What’s more, any misgivings about making it loud will be countered by the destination when your mix gets turned back down.

3- Pulling Punch

Where a bit of slow compressor attack can let through some clout to just about anything with teeth of its own, very fast attack can blunt sounds on heavier treatments. This means trying to add bite later on could be harder to achieve. A decisive treatment with one attack time is going to work better than using multiple attack stages that don’t quite agree.

4 - Ruining Release

Heavily compressed sounds can move again when further compression is triggered by something else, be it sidechain input for effect or the more obvious case of mix bus compression. A lot of time can be spent finely honing the shape of sounds with treatments that are either subtle or less so. Slow musical envelopes especially can be easily destroyed by those brutal pumpings later on when SmashBus3000 is set to Stun at the exit.

Finding Your Own Way

Most reading this will have their own way of creating, controlling, and cohering their mix with compression. For anyone who is unsure how to do it I’m gong to make a suggestion: it doesn’t actually matter where in your mixer these three things happen. That’s after the question of whether or not it’s needed at all. What is more important is the order in which compressors happen; that may or may not involve using fewer stages of compression than we’re often led to believe.

Three ways to do it: If compression is used to create, control, and cohere sounds, these can happen using any combination of track, submix, and/or mix bus compression. The mix in the top row uses track compression to create and control sounds, leaving the mix bus compressor to cohere, or glue the results.

The Magic Number?

Going back to the question of what we’re actually trying to achieve, I think most would agree that the name of the game is a mix that possesses pleasing sounds, at consistent levels that still maintain a good impression of having some dynamic swing.

In my studio, the occasional two screen monster mix can replace some mixes that struggle to hit double figure track counts! Most of the time I am mixing into a bus compressor. If I’m compressing tracks for effect I will also use track compression for any level control and forget submix compression altogether. This keeps it simple and I find it more effective anyway. Occasionally I’ll track-compress drums but compress a submix of harmonies and BVs.

Three stages of compression can sometimes be necessary. One suggestion might be to use track compression on level control duties, and submix compression for aesthetic polish and/or some common tracks’ cohesion. As such extra compression across the entire mix might not be necessary. That is unless the engineer opts for less compression upstream before a final kiss of comp to bring the whole thing together.

Opinions vary on at what point whole-mix comp should be called up. In my experience its late arrival changes the balance too much to be useful; for that reason, for me it has to be there from the start. The only precaution is to keep one eye on how hard I’m hitting it before I get into the finer points of the mix.

So how many times should you compress in the mix? Like so many things in the craft, it depends! As mentioned above, it certainly depends for me but one thing’s for sure: it’s easy to overcompress when faced with so many opportunities to add just a little more here and there. How do we get the best out of so much processing power? With multiple stages of compression to juggle, surely the answer is to listen, listen, and listen again, and consider using fewer devices to do most of the work.

See this gallery in the original post