Typically, corrective EQ, where you pull down problem frequencies or dramatically reshape a sound, is better performed before compression — that way the compressor is acting on something like the wanted signal — and arguably most of this could best be achieved long before you actually start mixing!
I’ve seen lots of different advice lately about whether it’s best to put an EQ before a compressor or after it. Some EQ before, while other people say that means the compressor will just be fighting your EQ moves. Some say you should do both, but that sounds like a lot of work! I realise there’s not necessarily a right or wrong way, but is there an approach you’d recommend?
Jake Martin
SOS Reviews Editor Matt Houghton replies: It depends what you’re trying to achieve, but the first thing I’ll say is that EQ isn’t always necessary. If the performance is good, the recording space quiet and the recording technique appropriate, it’s perfectly possible that you won’t need to use EQ at all, or that you need no more than the high‑pass filter you’ll find built into many preamps. A second point is that as well as EQ’ing pre/post compressor, a third possibility is EQ’ing the compressor’s side‑chain (control signal): create a duplicate of the pre‑compressor signal, and you can EQ that to your heart’s content en route to the compressor’s external side‑chain input (if it has one).
Let’s assume we’re working in a DAW (the signal path isn’t as rigid as on some mixers) and we’re dealing with one stage of compression. I usually perform corrective equalisation (dipping undesirable components, such as background noise or over‑prominent resonances) before compression. Conventional (static) EQ isn’t the only tool for that sort of correction, though; I might tackle resonances with dynamic EQ or de‑essers, and treat hiss or rumble with fingerprint or AI‑based de‑noisers; I’d use those before a compressor too (compression modulates the noise floor, making some of those tools less effective).
Also, I’d typically attend to most of that sort of corrective work either at the end of the recording session, or in a separate ‘pre‑mix’ session — the sort of work an assistant might do when prepping a session for a top‑level mix engineer, along with other housekeeping tasks like colour‑coding and labelling tracks, creating folders, and working towards a basic, initial, static ‘faders up’ balance. This way, the often time‑consuming and detailed corrective work doesn’t burn time, tire ears, and distract from the big musical decisions I need to make in the earliest stages of the mix, and mixing remains a creative process rather than a ‘left brain chore’!
When I come to mixing, my EQ moves are all about the tonality and musical balance, typically with EQs placed after the compressor, for a couple of reasons. First, when you EQ into compression, the compressor tends to iron out your EQ moves, so if I decide as the mix progresses that further EQ tweaks are necessary or I want to automate the EQ to, say, boost a sound’s top end in the chorus, I’ll likely need more radical settings. With most EQs that means more phase‑shift too and, depending on what I’m EQ’ing, that can interact unhelpfully with other tracks. With some analogue‑modelling EQs it may mean a more saturated sound too, which may or may not be desirable. Second, if broader tonal shaping before the compressor also changes the overall signal level or shifts significant energy to a different part of the frequency spectrum, the compressor may react differently.
Finally, I usually take a ‘top‑down’ approach to EQ, starting with broad‑brush tonal EQ moves at the mix‑bus level, and working back from there through subgroups and down to the individual sources only if required. Again, the aim is to mix quickly in the early stages so my ears are ‘fresh’ when making all the important musical decisions. I don’t always have the bus compressor in from the outset, and will often dial in things like a high‑shelf boost first and add the compressor ‘downstream’. But in this context the two processors won’t fight in the same way as they can on individual sources, because the compressor’s ratio is fairly low, I’ll be looking for gain reduction of no more than a couple of dB, and I’m very much making broad‑brush EQ changes with gentle shelves or very wide bell bands, rather than pushing or pulling specific components in the signal.
You really don’t want to be overthinking or overcomplicating this stuff when mixing, so I’d encourage you to experiment and gain experience...
That’s the theory — which might all seem a bit much to take in! It’s great to get this sort of thing clear in your head, but you really don’t want to be overthinking or overcomplicating this stuff when mixing, so I’d encourage you to experiment and gain experience of the interaction between EQ and compression, and the relative merits of EQ and other ‘problem solvers’ you might use. Your decisions when mixing will then soon become second nature, which is really important because, as I alluded to above, you really don’t want these sorts of considerations to distract you from focusing on the big musical picture when you’re mixing.