Do you default to throwing a compressor on every track in your mix? Do you automatically reach for the compressor whenever you mix certain instruments?

Maybe you should re-think that.

Last month I posted about using compression on your master fader. I wrote about how compression can be a great way to glue a mix together, but that you need to be really careful, or you might compress the life out of the music.

What about compression on individual tracks? Is it okay to squash everything separately? Well, there’s no right or wrong answer, but I’ve heard plenty of recordings where the engineer heavily compressed each track in the session. The songs end up sounding very flat and lifeless. When asked if they used compression on the master fader, the engineer will proudly say, “Nope! I wanted to preserve the dynamic range of the song.”

That’s a great goal, but over-compressing the individual tracks essentially does the same thing as over-compressing the master bus. I was recently listening to an older episode from the Project Studio Network podcast, where a mix engineer (I believe it was Charles Dye) was talking about his approach to a particular record. He wanted the record to have around 6 dB more dynamic range than the music he normally mixed. This involved using less compression, obviously, but it affected how he compressed the individual tracks just as much as it did the master bus compression.

Listen Before You Compress

I know this seems like an obvious piece of advice, but do you really listen to a track before slapping a compressor on it? I admit, I’m guilty of this. I just assume the track needs compression before I really critically listen to it.

Your goal should be to make the music sound as amazing as possible with as little compression as possible. Do I think you shouldn’t use compression at all? No way! The problem, though, is that unnecessary compression can create a whole bucket of new problems for your mix. Rather than making things harder for yourself, listen to the track first, then decide what changes need to be made to make it sit better in the mix.

Only then should you reach for a compressor, and even then, you should have a very specific goal in mind for what you want to accomplish sonically with that compressor. For example, you may want to bring out the attack of a kick drum part. Or maybe you want to tame some of the louder bass notes.

What I Do

With that said, there are a handful of tracks I almost always compress. Kick drum, snare, bass, and lead vocals usually get some compression. However, it really depends on how the parts were tracked. For example, there are a few songs on my album where I didn’t compress the lead vocal much at all. Why? Because the tube mic I used had a lot of natural compression to it, and I sang through a little bit of compression going in.

Things I don’t like to compress? Percussion, acoustic instruments (acoustic guitar, mandolin, piano, etc.), pads. That’s not to say I won’t compress them if they need it, but I find that compression on an acoustic instrument can change the tone WAY too much. A piano, for example, has so much harmonic content. When you compress a piano track, these harmonic frequencies get louder, potentially making it sound a bit unnatural.

Just like everything with mixing, use your ears. Know what you want it to sound like, and work hard to get it to sound that way.

What do you think? I’ll need 10 comments on this post.

[Photo by foxspain]

  • Terrell Harper

    Wow Joe you out did yourself again I’ve watched almost every video for studio one 3 that you did I’m still figuring things out I got a few more things to learn but its the only part of my game so far that I’m lacking b4 I cam execute my ep thanks man as always your a life saver

    • You’re very welcome. I want to challenge you, though. It sounds like you are waiting to acquire enough knowledge before you make your EP. Make your EP right now. Learn on the job. Acquire knowledge as you go. The only REAL learning happens when you’re in the thick of it, making art, releasing it to the world. Go do that. Today.

  • Dudly Ulysse

    damn this is really gonna help me out, this feels like an eye opener for me

  • amin

    hail, this article is so useful to me. at least i knew that why my piano sound become unnatural after mastering. could you please tell me a brief about compressing the grand piano track which plays the main role on my songs?

    • While it’s hard to give you one piece of advice, I would suggest considering NOT using compression at all if possible. If it sounds right. Otherwise go with a slow attack so as not to lose the fullness of the piano.

  • Great article! I’ve been struggling with compression in that I am either over compressing or under compressing. Finding that right balance has been tough but I am new to this.

  • Rafa

    Try riding the wave instead of compressing… You get the same results without the nasty squashing. You can even slap Wave’s Vocal Rider so you don’t even have to do it manually anymore.

  • Carlos.A

    Hi , Can you help me with some tips when Compressing Bass Drums? , lets say if i want to have more Punch or have more bass etc /’im learning so any help will be execlent

  • Alex

    Is it a good idea to compress going in? like on my vocals

    • I do a lot of the time. But just a LITTLE compression. If you overcompress, that’s something you can’t undo later.

  • John Judge

    I very rarely use compression anymore. I would rather go into the wave form & manually drop the loudest peaks in the editor.

    Does it take a lot of time & work? Yes it does…I have spent up to 5 hours doing this one one song! BUT the end result was a cleaner track that does not need a 5th of compression – limiting that the original needed, IF it needs any at all.

    This works for me!

    • HipSlack

      Nice! I agree, this is sometimes the best tactic!

  • Dave Chick

    I like what Sparqee was saying – kind of the way I see compression. It really all depends on what style you’re mixing and what you’re trying to achieve. I don’t really see why anyone needs to apologize for using compression – if it’s working for you and/or your clients, then good on ya!

  • I’m just starting out and learning, but I’ve really come to appreciate very mildly-applied buss compressor – it really does seem to “glue” things together when correctly applied (e.g. the bounced waveform shouldn’t be a block that makes the mastering engineer laugh at you)
    Everything else I just kind of listen and see if other tracks may individually benefit – I almost always do a little compression on the vocal if there wasn’t any during tracking, kick & bass as well. Been learning how/where sidechain compression can be handy there. It’s all subjective but and at this early stage I know I’ve over-compressed some things, but that’s learnin’ for ya.

  • I’m guilty of over compressing as well but unfortunetly the choice is not mine. In hip hop/rap I have to go with the compressed hard sound of what’s current in my genre. Sure it kills me but at the end of the day I just want like most of us to have my musical skills and talents accepted as good as a commercial release. I’ve since found that anything that sounds good can be a commercial release but you know what I mean lol the way I see it is like all music we either evolve and conform or stay the same and be left behind in what was.

  • Rob

    I hate to admit it but i compress nearly every track. I think this is down to lack of confidence when mixing. I’ve always approached each track thinking if i compress it may sound better. :S

  • Like randy mentioned I actually use compression (more often than not) to increase dynamic range (e.g. bring out the attack on the kick or snare. For vocal & bass I do use it reduce dynamic range but for my own vocals I use it more as an envelope & tonal shaper. On the drum bus and master bus it’s all about finding that ever so subtle pump that gets the song thrumming along to the rhythm of the kick & snare. Now granted that last bit really only applies to driving upbeat tunes but then most of my songs are that way.

    There are all kinds of examples of over the top limiting on commercial albums that people like to whine about and you can certainly compress the life out of individual instrument tracks as well but compression is just a tool. It’s just a sound. Back in the day when guitarists first started overdriving their amps there were people that liked it and “purists” that hated it. Over the top compression and limiting (to my mind) is just another flavor of distortion (and once again some like it, some hate it). There are plenty of guitar sounds that I think are too distorted but others like it. That’s cool. If you like hammering a track or a mix with a compressor, I say go for it. Maybe you’ll change your mind after you hear it. Maybe you’ll learn something by going too far. The worst that will happen is that you’ll hear the mix in a couple of years and say “what was I thinking?”. So what, it’s art. It’s temporal. Write a new song and make a new mix. Learn and move on. For most people it’s way easier to learn from their own mistakes than to learn from other people’s mistakes. I say go for it make the mistake, take it too far and then move it back into your personal realm of artistic sanity. 🙂

    p.s. I’m finding that if I want to enhance attack without loosing too much top end a Transient Designer is often a better solution than compression.

  • I try not to use compression on guitar as I like the normal dynamic range, but I will use if needed in a certain situation. I use automation to change volumes. I look at compression like using reverb less is more. I will use compression on vocals and percussion. I will use a touch on bass guitar. I will compress on the master buss with Ozone 4 or T-Racks Deluxe. T-Racks Deluxe is still available on Ebay for $39.99 for mac or windows, great plugins try the demo. That’s my take on compression.

  • famouspatrick

    I tend to not use a compressor at all unless I think it is absolutely necessary. A lot of what I record is based around me playing acoustic guitar and singing, with a fairly low track count. I have a fairly even guitar style and very good mic technique, so I don’t need much, if any, there. The rest of my stuff is MIDI or audio loops, which are usually recorded pretty well to begin with so I just use a touch where needed. Perhaps if I had better engineering skills with compression I would use it more, but for right now, I usually find I like the sound better with a minimum amount or none at all.

  • Lukas

    Great points, Joe.
    When it comes to compressing individual tracks I tend to go with only as much compression as a given track needs. It is so easy to overdo it, though. Sometimes I just prefer to go in and adjust the regions’ levels with audiosuite gain plugin before compressing – for more natural effect. The same goes for vocals, I prefer to automate levels a bit before applying compressor. With the 2buss compression, mhm… it really depends what your goal is and what genre your dealing with. Somewhere around 1-2dB of gain reduction is enough to glue the mix together. The rest is mastering engineer’s job…:)
    Good post, Joe!

  • christopher [chrisw92]

    I tend to use compression more on percussion (for example a snare drum) but only EQ is the only insert what I always use on tracks.

    My “philosophy” is if it sounds very good without an insert, don’t automatically think that it will sound better with particular inserts. for instance a few of my friends have been taught to use specific plugins on specific instrument tracks and when I question them why they done it, 9 times out of 10 they say they have no idea its just what they learnt to do… I’m not saying that its particularly bad, nor am I saying its good, I just believe that if you don’t question yourself why you are using a plugin you don’t really fully understand what it does and if it will make the track sound better or worse compared to using either none or a different plugin.

    oh and also IKEA rocks! (just read the newsletter) its a good thing I’m just about 3 miles from one… with a bit of imagination you can make really nice/cheap studio furniture, here is my favourite “modifications”; http://ikeahacker.blogspot.com/2007/09/hack-bedside-tables-into-music-studio.html

    http://ikeahacker.blogspot.com/2010/05/ikea-for-music.html

  • Its true that compression can change the tone of acoustic instruments, but sometimes it can be a nice effect. If anyone’s heard the song “Piano Lessons” by Porcupine Tree, this is a perfect example – the most heavily compressed piano I’ve ever heard, but it works and it fits.

    But I agree with you, for a natural sound its best to avoid over compressing. I’m sure thats why a lot of my mixes are still suffering, after EQ I tend to always go for the compressor, even if just to tame some peaks. I’ve started using Tapehead to lightly compress a lot of tracks though, which helps.

  • I almost always compress the kick drum, but lately I’ve tried to not compress the snare as much. You can really kill a great snare pop with fast attack times. I’ve used a transient shaper. Then I’ll compress the drum buss to glue the kit and catch some of those “wild” hits. I’m also trying to not compress vocals as much. Spending more time riding the vocal to preserve dynamics.

  • Mark B.

    i guess i tend to use compression alot, but i am trying to learn to be aware of what the compression is DOING rather than just blindly turning knobs. I do use it alot, but i DO try to use it lightly on acoustic guitar. i tend to squash bass and “strummy” acoustic. i learned about parallel compression from this blog, so i often leave comp off the main drum track, and throw some on an aux. honestly, i have yet to muck about with comp on the master. i’ll get there.

  • Great article, first of all. I might be in a unique situation with this:
    Most of the music I’ve been recording lately is solo acoustic guitar. And not multitracked or double-taked acoustic guitars…just one guitar recorded with two or three condensor mics. So, preserving the entire dynamic range of the acoustic guitar in THIS case doesn’t quite hold up (unless I want to keep my hand on the volume knob while I listen…) so some sneaky amounts of compression are required.

    But there’s an additional problem. The guitarists I’ve been recording (including myself) often use the body of the guitar like a drum (Tommy Emmanuel anyone?), and come up with all kinds of creative ways to use the instrument percussively. As you can imagine, this creates even more dynamic range in the waveform than if the player were just strumming, and the preamp gains obviously need to be set so the percussive hits don’t clip. This obviously lowers the levels of the ‘regular’ playing.

    The first problem I encountered as an engineer is that the gap between the spike created by the percussive hits and the levels of the ‘regular’ playing is pretty darn big. I started out using two passes of compression on the track…one compressor (fast attack, fast release) to catch only the percussive hits to bring them down closer to the string sounds, and then another compressor (slower attack, longer release) to lightly compress the total track. But I still had the problem where my levels on the strumming were just too low for my liking…not enough gain. And as you know, you can’t really do much about that.

    So lately, I’ve actually taken to setting up a compressor to engage LIVE while the track is being recorded (can’t remember the technical term for this). Usually I’d never do that, but having that compressor there from the beginning, just to catch the percussive spikes, helps me raise the gain on the preamps so I can get decent levels on the strings. I still find I need some compression during the mix, but definitely not multiple compressors layered over each other.

    And on another topic, I’m still experimenting with using bus compression only, or not using bus compression and compressing each track (typically I have 3 tracks) individually.

    Love to hear more thoughts on this!

  • I think the key to this discussion is what you hope to accomplish by using a compressor. Most people assume compression means lowering the dynamic range. But you can also use a compressor to INCREASE dynamic range. A distorted electric guitar already has very little dynamic range, but if you put a compressor on it with a slow enough attack, you can actually bring up the volume of the pick action relative to the sustain. The same way a claw hammer both drives and removes nails, a compressor can be a versatile tool for manipulating dynamic range. The application of the tool has as much to do with the results as the choice of tool. Or as Michelangelo put it, “A hammer and chisel do not a sculptor make.”

  • dave

    That’s good food for thought! If you double a bass track (with the copied track having a distortion and a ‘wacky’ eq) would you compress the original only? I’ve tried both ways and can’t really distinguish which is the best!