Which came first, the chicken or the egg?

Which should come first, EQ or compression?

Feel free to state your opinion for both questions in the comments section, but this article will deal mainly with the second. (Although, you never know…)

Esse, one of my Production Club members, emailed me a question over the weekend. (Side-note: Esse lives in Italy. I’ve also got Production Club members from England, New Zealand, Brazil…crazy, right?!)

Anyway, Esse’s question was about EQ and compression. What order should you use them? Is there one way you should do things? What are some reasons you would try it one way over the other?

First off, “grazie mille” to Esse for the great question. As I was emailing him back, I thought this would make a great blog post, so here’s my email to Esse (with a few edits):

Here’s the basic idea behind how I mix. While a compressor technically decreases the volume of the audio, in reality, our ears hear it as also INCREASING the quieter frequencies. If you have a guitar that has some soft rumble down around 60 Hz, compressing the guitar will make that 60 Hz rumble LOUDER. Once the signal has been compressed and made louder, it can become hard to EQ it out.

In that scenario, I would use an EQ before the compressor to get rid of that rumble. That way the rumble is gone BEFORE it has a chance to be turned up by the compressor.

Compressors are kind of like glue. You should only send things to the compressor that you WANT to be glued together. Otherwise, it’s really hard to “unglue” something once it’s glued together.

On the other hand, compression DOES still turn things down…it essentially turns loud things down and soft things up. So, if you’re trying to do an EQ BOOST, you probably want to do it AFTER the compressor. Let’s say you want to do a high-frequency boost on a vocal track, to make it sound more “airy.” If you boost 10k before the compressor, the compressor will essentially turn that boost back down. So, if you want to boost something, it’s a good idea to boost it AFTER the compressor. That way your boost won’t get turned down by a compressor.

It’s not uncommon to have one EQ before the compressor (for cuts), then another EQ after the compressor (for boosts).

I don’t do this very much because I rarely boost anything. I like to use EQ to cut certain frequencies, so I usually EQ before the compressor.

Keep in mind that my first answer to any question like this is to use your ears. There are no hard and fast rules. Sometimes it might sound better to boost before compression and cut after. I’m simply sharing with you how I approach EQ and compression. It’s worked fairly well for me, but different engineers will have completely different approaches to EQ and compression.

The important thing is that you understand how EQ and compression affect the sound. Experiment with different settings. Listen to the changes in the audio. Any time you invest into messing around with EQ and compressor settings is time well spent.

Over time, you’ll find that you’re doing less and less guess-work when it comes to EQ and compression. You’ll know what sound you’re wanting to create, and you’ll know how to use EQ and/or compression to get there.

Agree? Disagree? I’d like to hear about it. Do me a favor, and leave a comment below.

[Photo by jonny.hunter]

  • Phillip Simmons

    From Phillip
    The right and accredited way while avoiding plagiarism.

    I simply put my friend cd in the cd drive then up load it to my wavlab. I then look at the equalization the oscilloscope and jot down all those frequencies his cd and sound lows mids and highs are at on the parametric equalizer. I then put the adjustment on my sound equalization
    And presto. The sound sounds good but needs some work but it not that much work. For example the eq enthusiast noticed that 384 hz is for the boxy low frequency note. when its turned up it makes it sound hollow. When the frequency is turned down the boxy sound does not sound likes hollow. I thought the was a neet way it effect sound veses hollow or boxy filled up wave form.

    The rest i just research ed them frequency an chose a couple more and put the adjustments on the parametric equalizer.

    I use studio one three by presonus.
    My friend makes techno music. Hes a cridic at the sound sample playing it low s mids and highs. I mean all most all of the bands are on assignment. Lowend and high end.

    Thanks
    Phillip Simmons

    Jazz man

  • Nathan

    what i do is – Dynamics first (Comp,Limiter,Gate) and some EQ if needed (like cutting obvious abnormal frequencies) but after that its “post production” time – you can’t go back and start filming again (sort da say) because you already created the dynamic “picture” of that sound, so now comes – EQing the sound with its companions in the mix (other sounds).. so to me its: Dynamic-EQ-EQ *Please Note :the first EQ does not have to necessarily be after the comp or after any other dynamic,its a matter of judgement and taste – remember -we mix with our ears ,not with mathematic formulas

    • Why a limiter?

      • Nathan

        don’t seem to understand

        • I’m asking why you use a limiter on your individual tracks.

          • Nathan

            well , not always, it depends, sometimes i use it to get a different sound. for instance if i want a clear and dynamic fast attack on a plucked instrument, but once again ,i hate to generalize

  • Thanks. This was highly informative. Love the glue analogy and makes me rethink how much I boost eq.

  • Wow, thanks! I never thought about it before. Yes, compression reduces the louder frequencies the the most since they have the largest amplitudes in the fourier breakdown of the signal into sinusoides. Since the louder frequencies are quiter, the quieter frequencies now sound louder. I will now eq before compression for cuts and after compression for boosts.

  • First a compress from the side with a simple gate at 60 db. Then reach for a 7band eq on lead. I use 7 band cause a 4 band dont give me a high pass filter. I dont use the middle frequecy. Only use middle frequencys on dubs and adlibs. On lead I swept the frequencys then cut 3 dbs.I only add in the highs on lead.I always leave the banwitch at 100. Then I grab for the compresser with a 10 ms attack with a 3.0.1 ratio. I compress 3 dbs to even out my eqing. Depending on style of music I will grab a another compresser(vocal leveler) then compress another 3 db. This results is best for what I have. Eq and compression all depends on your signal and your mic and acoustics.

  • Carlos.A

    Hey , Thanks to all of you , Ive learn a lot in this page with all the coments , cheers from Colombia 😀

  • Matt

    weird… just the other day I was having trouble getting the right sound on a track (i think it was a bass track but I can’t even remember). I always thought logically to EQ first because why would I want to compress frequencies that I wanted to eliminate in the first place?

    I got frustrated and for some reason it struck me to try compression before EQ – wow! what a difference. Not only that but to my surprise it got me a lot closer to what I was aiming for… so I rolled with it.

    Be wary of habits – especially when you’re learning (as I am). Don’t be afraid to break them. I found that you can learn a lot more lessons by breaking habits (or what you think is “right”) than you can by blindly sticking to the same routines that somebody else says is correct. Regardless of the outcome of your ‘experiment’… you’ll gain some valuable experience firsthand.
    My 2 cents.
    – Matt

  • aLf

    “It’s not uncommon to have one EQ before the compressor (for cuts), then another EQ after the compressor (for boosts).”

    Yeah, that´s how I do it too. I love the “sentence”:

    “Cut to correct, boost for effect!” 🙂

    Thanks!

  • MichaelHe

    I also EQ before comp, but I also do some EQing after comp…
    The reason is, that I create the sound with the first EQ. And when it’s how I like it, I put in a comp to make it sound more clued together, and it’s sounds better in the final mix…. But the comp do change something in the tone, so the sound I created with the first EQ is now changes.. Therefore I do some miner EQing after comp, to get bakc the same sound… It could be some air on a vocal ex…

    It is also a good idea to have an EQ plug-in after comp, even though you change very little… because.. When you are listening to the final mix, and here some EQ problem with ex. the lead vocal, it’s very easy to correct it by changing the second EQ a little… If you try to correct this on the first EQ the comp will change the sound in other ways than just the slight EQ change you want… ex. if you hear the final mix and you can hear that the lead vocal needs a litlle more top, you can raise the top in the first EQ, but the comp will then maybe be more activated by the higher top, or make some changes in the tone because of the increased top in the comp input…
    This is very much easier to do with the second EQ because it ONLY raises the top and nothing else…

    • GREAT points, Michael. You’re right; it’s really important to remember that any level changes you make to the signal before the compressor will change how much compression occurs.

      Boost the bass, now there’s obvious pumping in the high end. Boost the highs, now the bass sounds too choked.

      Awesome insight, Michael.

      • MichaelHe

        Thanks :o)

        Another ploblem is your automation…
        If you have levelmixed your lead vocal, and you think that the vocal need a little more midtone maybe.. If you raise some midtone in the first comp you will not raise the midtone at all. You will rather lower the top and bottom because the raised midtone will raise the overall level of the input of the comp, and it will kick in…
        This will course that instead of raising midtone you will lower the importen frequences like pronoucing and s and t sound and so on….
        So therefore I always do this small changes after the comp….

  • famouspatrick

    I have to say that I have found the post and the comments very enlightening, and I thank all of you for sharing your knowledge and experience with those of us with less experience.

  • Wow, it’s very nice to come across a thoughtful mixing discussion, thanks Joe.

    I usually start with my usual cuts then add the compression. I’ve become a big fan lately of multiple layers of comp, I find a touch here and a touch there leads to big results with lots of individual control along the way. I will EQ after only if there is a unique tone in there that was created by the compression, otherwise the lollipop out of toffee analogy above is bang on.

    And of course, as all my old instructors said would happen I’ve learned that if it sounds good, do it. Happy accidents are a wonderful thing!

    Thx for the post!

  • I find that more often than not, what needs to be done is not necessarily what your approach was when you started mixing. I’ve often put compression on something pre-eq and then found that it brought out something nasty that I didn’t notice before. At that point, the thing to do is re-arrange the signal flow or put some corrective EQ pre comp. In short, I don’t think that there is an answer that works for every instance. How is that for non-commital?

  • This tends to correlate with my approach too. I almost always eq before compression because it sounds more natural – the compressor is responding to the sound I’m hearing.

    Sometimes I’ll EQ after compression, where the compression is a fundamental part of the sound (such as pumping), and pre-EQ would change the compressor’s behaviour.

    Your comment at the end about cutting EQ before compression and boosting after compression is an interesting one. It would work well if you *really* want to bring out a particular aspect of the tone. On the other hand, I’d imagine it could sometimes cause problems too – such as boosting sibilance on a vocal part without allowing the compressor to control it (although if your processing is that heavy to cause a problem, you probably need a de-esser anyway!).

    I’ve written more about this on my blog here:

    http://kimlajoie.wordpress.com/2007/06/26/ordering-of-eq-and-compression/

    And:

    http://kimlajoie.wordpress.com/2009/09/07/eq-cutting-vs-boosting/

    -Kim.

  • Steven

    I tend to eq before the compression. If I boost after, I tend to dislike the sonic character (plugin probs maybe?) of the result. It sounds – er… ‘sticky’. It kind of loosens up the audio somehow. Pulling round a 4dB boost after the compressor feels like trying to trying to pull a lollypop out of a vat of toffee….

    I realise how random this is, but I just don’t like the sound of post compression eq.

    Cheers

  • Great article and relevant topic. It does the change the sound if you switch the order. Asking these types of questions forces you to analyze why you are using certain plugins in the first place. What are you trying accomplish in your mix with this effect?

    Thanks Joe.

    • EXACTLY!! I catch myself reaching for a compressor automatically on a track, then I have to slap myself and ask what it is I don’t like about the tone before I just go into default “I must compress everything” mode.

  • Joe R.

    Very interesting,
    I usally EQ then compress.
    I found the Cut before/boost after very interesting.
    I dont normally boost, but that concept helped me get my head around compression and what it actauly does.

    Time to experiment!

  • Hi Joe,

    How I wish I had a community and forums in which to discuss things like this when I was starting out but, unfortunately, it was long before the days of the inner tubes. Anyway, I’m glad that people like you are around to initiate discussions for folks today.

    When I was first getting into recording and pondering this question of EQ and compression, I happened to once be sitting at a table at a wedding with an award-winning engineer by the name of Ed Rak. Ed owns Clinton Recording, a world renowned studio in NYC. Ed’s advice to me, or at least his method of doing things which I eagerly took as advice, was to use EQ before compression. The concept made sense in terms of keeping frequencies in check. As usual, there are no rules and there are certainly ramifications to any approach but, I have to say, I’ve never (not once) ever felt the need to do it any other way. That doesn’t mean I couldn’t come up with a scenario where, presumably, it might work the other way around. I’ve actually never really thought about it for about 20 years…

    Who knows, maybe I’ll just try it the other way just to exercise my brain and shake up the prejudicial plug-in muscle memory a bit.

    🙂

    Slau

    • I’m the same way, I almost never EQ after compression. However, the other day I was EQ-ing a vocal that I had sung through an AKG D5 dynamic mic. There wasn’t much top end at all, and boosting the highs before the compressor made it sound way too harsh, so I EQ’d the vocal, removed the low-end rumble, compressed it like I normally would, then used a 2nd EQ to add a high shelf. The highs sounded much tighter, since the whole track had been compressed, but it came across as tight rather than harsh.

      Of course…this was just a scratch vocal… HA!

      • From a mix it while you’re recording it standpoint, choosing a dynamic microphone and them compressing it was counterintuitive in trying to achieve a vocal with top end. It should be noted that compression has a habit of making things dark, so when you compress after you’ve put in all sorts of top end with EQ you’re going to be losing some of it.

        Understandably, it was a scratch vocal and the dynamic served it’s primary purpose in that vein, but it’s food for thought to take into consideration the electrical components in microphones as well as compressors(and plugins that emulate compressors) and what they’re doing to affect the other things in the chain.

        I know too many engineers that decided compression on the way in was a good idea only to find themselves trying ti dial in copious amounts of high end in the mix to try and get back what the compressors ate up.

  • Al

    Well, I am absolutely a beginner, in other word a self taught.So expect to hear a noob’s idea =)
    so what I do is, I use EQ to polish the sound of each piece (eg. brighter kick or snare, removing boxy guitar sound or muddy bass, etc) , that’s before Compression.
    I also use EQ on the final mixed song, for instance making left channel sound a bit fatter or mid-ranged than right channel, that’s after all Compressions.

    Consequesntly, I use EQ mostly before the compression, depends on what I’d like to hear on final output.