If you ever get into mixing music in the studio or in a live venue, you’ll need to know what a compressor is and how to wield it with precision.
But when you first start out, compressors can seem a bit scary. They have lots of knobs with weird names (“soft knee” anyone??) and it’s hard to know if you’re making the track better or worse with each virtual tweak.
This week I want to hit the topic of compression head on by breaking down a typical compressor so you see how truly simple (and powerful) of a device it is.
This Is All Your Compressor Does
Before we dive in, let’s just be clear about what a compressor does in the first place.
At it’s most basic (and I love basic) a compressor is simply an automatic volume fader control device. It can literally turn down loud stuff automatically and give you a more even volume throughout. That’s it!
Why is that helpful?
Well most music is very dynamic, meaning it gets loud and soft with great variety throughout a song. This can make it hard for an instrument or vocal to be consistently heard and felt throughout in a final mix. Compression, can help you get that nice, consistent, up front sound you’re after.
To be fair, compressors are used for more than just volume control. They can give something more energy or more sustain. They also can act as tone shaping devices.
But for today, let’s look at the knobs and tweakable settings on a typical digital compressor plugin so you can feel right at home and in charge!
Threshold
The most important knob on a compressor in my opinion is the threshold. It’s the knob that actually allows compression to take place or not (and how much).
It makes absolutely no difference whatsoever how much you’ve tweaked the other settings on the compressor, if the threshold is set at 0db (or the top) then no compression is actually happening.
Threshold is the volume at which compression kicks in. It’s like a “master control” on the whole darn plugin.
Turn it down a tiny bit, and the compressor will only affect the loudest peaks of the track. Turn it down a lot more, and you’ll get even more squash to your signal.
A great rule of thumb for natural sounding compression is to turn it down and watch your “gain reduction” meter until you’re seeing around -3db to -4db of gain reduction.
Ratio
Another knob that works in tandem with threshold is the ratio. The ratio knob controls how hard the compressor hits when signal crosses the threshold.
A 1:1 ratio means the signal isn’t turned down at all. It’s a flatline. A 2:1 ratio tells the compressor to turn down any signal that crosses the threshold in half. (For example if signal crosses the threshold by 6db of volume, a 2:1 ratio would turn that signal down to only 3db above the threshold.)
The higher the ratio, the more squashing you get. Basically you can play with the ratio and threshold knobs together to get as much or as little compression as you like.
Attack And Release
If threshold and ratio control how MUCH compression (or volume being turned down) is happening, then the attack and release knobs control how fast the compression kicks on and off, thereby affecting the tone.
The faster the attack setting, the quicker the compressor pounces on your audio when it crosses your set threshold. And the faster the release, the quicker the compressor resets or “lets go” of the signal.
Attack and release are game changers because they basically change how your compressor sounds and what it does to the audio.
The best suggestion I can make is to keep attack and release in the middle as a starting point, and then experiment with much faster (and slower) settings, noting what affect they have on your track in question.
You’ll soon discover what sounds cool on what type of audio (and what doesn’t).
Makeup Gain
The last compressor setting I want to address today is makeup gain (or sometimes simply “gain”).
With all this turning down of the signal (because remember, that’s all a compressor is meant to do), you’ll have a much quieter track than when you started. The solution? Simply turn it back up a bit.
Thankfully for us, engineers were smart and put a simply output gain stage at the end of compressors to “makeup” any lost volume or gain.
If your compressor is on average turning down your snare drum (let’s say) by 4db with every hit, then it’s safe to say you can turn the makeup gain knob up around 3db to 4db. The goal is to level match it to the original signal going into the compressor.
The difference, though, is that now you have a much more consistent signal coming out of the compressor, than you had going in. – Which is the whole point of compression.
Forget The Rest For Now
While many compressors have other settings like knee, key input, etc, ignore them for now. All that matters for 90% of the time are the four main settings I’ve shown you here.
In my next post I’ll give you a little demonstration of how these four elements can give you the power to totally control your audio.
In the mean time, I’d love to hear from you what has been the most frustrating or confusing thing about using compression during your mixing adventures. Share your thoughts below!
Do Not Create Another Song Until You Read This Guide
(Price: $17 Free Today)
Discover the 6 Steps for Creating a Radio-Worthy Song From Scratch
My name is Graham Cochrane and I'm the founder of The Recording Revolution, one of the world's most loved audio recording and mixing blogs with over 200,000 readers each month. 
Great beginner’s guide to compression, G!
When I was starting out as a mixer I was so confused with all that stuff I don’t even know what my biggest frustration was… I just didn’t know what any of the controls where meant for and usually ended up slapping a preset on the track that claimed to be good for kick drums or whatever it was I was working on. However I didn’t know I had to adjust the threshold for the presets to actually work well and so I usually ended up with the compressor just turning things up a whole lot or squashing them to death… It sounded horrible. Your Jump Start Series really helped me wrap my head around this and now I’m quite confident with compression and all the fun stuff you can do with it 🙂
Just great explanation! This should be ticked as “sticky” on the entire internet 😉 thanks Graham, as usual, simple and direct to the point!
The first compressor that I came across had no make-up gain knob and it just did “auto-make-up” the volume somehow. The result was always louder, which tricked me to think that the sound was actually better. Not to mention that it changed the balance of the mix.
Then I learned from you or Joe (or both) that I should level match the original and the processed sound – that was a revelation to me! 🙂 Since then, I always check for a sort of output level control when I see a new plugin. They are so under-estimated.
most frustrating thing for me in using compression has always been using the compression as a tone enhancing effect (punch in a drum or sustain on an instrument). It is so subtle that changes tend to destroy my audio by the time I am hearing the effect. Enter Graham’s half rule!! I do fine with using a compressor to tame a track with a lot of dynamic a bit, but I do find it harder to use good judgment when I am dealing with a high track count and using compression on multiple tracks. Buss compression is a little tough for me to use properly for the same reason. The main thing I have learned is to do my best to capture REALLY REALLY good performances from the start and let the dynamics speak for themselves using compression sparingly and with purpose never just because I “should” use a compressor. Then I use half what I think I need.
This is an excellent article on compression! My biggest struggle with compression has been learning how the attack and release correspond with each other. Perhaps you could give us a few quick and general examples? Thanks Graham!
Graham, it would be nice if you could show us a demonstration of using a compressor and compressing different signals. Up until now, I only use the stock BF-76 compressor. Two reasons… One, its very basic (input, output, attack, release). Two, i dont use the others because I don’t understand compression enough to be able to implement them without ruining something. Also, if you can show when or when not to compress as well. Thanks a ton Graham!
Stay tuned for later this week!
This is the same explanation you can get anywhere, and it really doesn’t tell you anything about HOW to actually USE compression. You say your next post will get into that, and that is good, because that is what beginners really grapple with. The concept of compression isn’t too hard to grasp, but learning how to use it can be daunting. Especially when the first compressors people use are usually the ones that come with the DAW — those usually have too many bells and whistles.
I know that when I started, I was so worried about what ratio and attack/release settings I SHOULD be using (based on stuff I read) that I couldn’t see the obvious: you should be using your ears to dial in the compression rather than your eyes. The other thing that helped a lot was using plug-ins of the classic analog compressors — the LA2A, the 1176. These compressors are FAR simpler to use than the digital beasts that come with the average DAW. The LA2A, for example, just has two knobs! One to compress, the other is make-up gain. Simplicity itself. My advice to anyone wanting to learn how to use compression is to start with an LA2A, because it’s SO easy to use it kind of forces you to use your ears rather than your eyes. In fact, you can do it with your eyes closed. You literally just turn the compress knob till it sounds good to you then readjust the gain. Once you get used to that, move on to the 1176 which is only slightly more complicated.
I agree that there are a ton of sources that tell you what ratios you should be using, but to me those are just general starting points. Being that there are no absolutes in mixing because sound is subjective, the best tool we have is our ears. To me the information that Graham has provided here is probably the most important. With compression it’s all about how squashed we want the audio. So having a basic understanding of what ratio and threshold are is key. I also agree that analog based compressors like LA2A are way easier lol. But some DAWs don’t have a analog based compressor available, so it’s good to know how to tweak a stock plugin as well.
Thanks a ton, Graham! This definitely refreshed me on my knowledge of how the Compressor works. Always learning from you! Thank you so much! 🙂
Hi Graham
Great explanation
when i started my first mix, i was like “ok, i read everywhere that i have to use compressor, but… how?”
im pretty sure that this information will help a lot of people, its very easy to understand when the person (you in this case) who is trying to teach you or help you really want to do it…
thanks!
Great tutorial!
Easy and nice 🙂
Just a quick check… in the text you say “The higher the threshold, the more squashing you get. ” Shouldn’t it be the other way round? Lower the threshold and your signal gets squashed more?
Good eye! Should read “ratio”. Edited now.
Perfect, thanks Graham!
I discover compression by myself on audacity because i had some loud parts on my track.
Using a compressor in audacity i just applyed on the necessary parts. That cause me a lot of problems because tracks sound diferent.
Later i discover how to combine compressors and how to avoid ultra-loud parts on recording by moving the mouth or mic.
My first compressors were a pair of DBX 163, single knob, soft knee compressors I used on live recordings. When I got into the DAW’s I had a lot of trouble adjusting the four knobs, lack of punch and volume loss. The biggest problem I had was “breathing” from attack and release being mis-adjusted.
I really liked the single slider DBX 163’s but am getting an ear for the multiple knob plug-ins.
Hi, nice article.
I think the biggest problem I had when learning about compression was when to use it. A majority of compression tutorial on the internet give the impression that everything needs compression.
I finally had a break through after reading a section in Mixing Secrets For The Small Studio. Basically it boiled down to this.
If you set your initial mix balance and you have a track that you can’t stop adjusting the fader to make it fit in with the mix, that track is a good candidate for compression.
Looking back to when I was learning how to use a compressor, I wish I had your explanation then haha. I have a pretty good handle on compression now, but in the past I remember seeing all of these mixing tutorials telling me I need to add compression to give my mix more body. Before I understood what compression actually did to the audio compression was my worst enemy. I was doing what I thought everyone else was doing but in return my mixes just sounded muddy and squashed by the time I was done. Talk about being frustrated, there were times I thought I just didn’t have what it takes to mix because of that. Then I also tried to blame my equipment for my lack of understanding. I thought well they have a $10k outboard compressor and I have stock plugins that’s why mine doesn’t sound good. Of course I realized later that I just didn’t really know what it did. It took me a while to learn that compression used sparingly here and there in the mix really does add up to fatter sounding mix, but we have to be careful because it all adds up exponentially in the end.
Thanks for the simple analysis. It helps. I record synthesized virtual orchestrations of classical music and the huge dynamic range is always an issue for me. Any thoughts on the best approach for compression when recording classical music? Is it best to compress each instrument or section or to apply compression to the master bus?
This is a very good article! 🙂
As a rule I don’t really like compressors very much, because for me they can crush the musicality out of performances, and I generally prefer using volume automation instead when possible. However, they are often a “necessary evil” and I do find them incredibly useful when I have a part (such as a lead vocal) that I want to “jump forwards” in a mix. 🙂
When I first started learning how to use Logic (my DAW of choice) I was told that you would normally employ EQ and compression on every track. I would have recordings of parts that to my ear would sound great and then start to sound WORSE when I applied compression to them, for a simple reason – the part didn’t need compressing! 🙂
So, some of my earliest recordings sound too full and too fat to my ears, because everything is being compressed and thus everything is being pushed to the front… 🙁
So, the biggest challenge I faced with compression was judging when to use it and when NOT to use it…which I find is partly dictated by the genre, and partly by personal taste. 🙂
Thanks for another thought-provoking article, Graham! 🙂
I’ve never heard such a simple explanation of Compression, very excellent man. Always great to see this type of stuff right in my email.
Hi Graham
Thanks again for some useful enlightenment!
I didn’t use compressors during recording sessions or mixes for a long time because no one could really explain to me how they worked exactly (although I’ve had -an analogue- one lying around in my studio)
Then one day I got the bright idea to send an email to the manufacturor asking them if they could help me out.
They were kind enough to send me a manual (of a later model) for free which really helped me out (the basics were there)
From that time on I started using the compressor during tracking sessions.
Nowadays – in the digital realm – I can’t imagine not using any compressors during mixing or master sessions !
I might use two or even three on one track – each one compressing just a little for different purposes.
And on a master track I really can’t go without a multiband compressor anymore!
Err… not really a story about frustration or confusion but a nice one to tell anyway
and it shows how compressors have grown on me..
Probably my biggest frustration with compression is simply being able to recognize what it’s doing once applied to a track, even with the proper threshold and ratio settings. Most other effects tend to be quite obvious to me, but compression is much less noticeable even though I know what it does.
yes, that’s the difficulty to me too!
My biggest self caused problem comes from having too much compression at once on one thing and having what seems like enough become too much by the time it hits the instrument bus compression and once again the mixbus compress or limit/leveler. Rookie mistake helped by mixing backwards and what I guess would be called “compression staging?” As well as better gain monitoring on the makeup gain control to catch “level creep”. Thanks again for the clarity!
Hey Graham, as always- great article!
I’ve been following your blog for about a year or so and already understood the basic concept and “knobs” of compressions from your other articles and videos.
I guess my biggest frustration is recognizing what is the exact amount of threshold and ratio I need to apply to the track I want to compress. Any tips? Maybe I’m looking at the wrong meters?
I use the Gain Reduction meter to judge how much compression I’m doing.
Shouldn’t I look at the “Out” meter to see what actually the output volume is? So I can compare the volume on the entire track to see that everything is about the same…
The most annoying thing about the compressor is knowing when to use a knee or not. I get the concept of the compressor with a hard knee but everything after that is unknown.
My most frustrating issue with understanding compression is definitely soft knee and ratio.
For Example:
when I need to fatten up my snare to a point where I wanna achieve a sound that has some weight, some sustain and lots of energy I always end up squashing the signal and the worst of all, can’t get it to sit right in the mix,
If I raise the fader then It’s too much and too ugly
if I lower the fader and use make up gain then it disappears in the mix.
When I first started learning about compression, There was a confusion as to whether the ratio would decrease the whole level or only the level that EXCEEDS the threshold.. You’ve nailed that here for a lot of people who are learning it for the first time 🙂
I don’t like that word “squash” because to me it immediately implies a ‘dulling’ of the tone, which is not the case within normal parameters. For me release is the killer to understand…no matter what anybody says, I’m still not convinced at exactly which point it starts the countdown to release: is it below the threshold, or actually from the moment compression kicks in? As it is generally about shaping sustain, for now I rely strictly on my ear…and perhaps that is all i need to know ha ha. Love TRR!
Very very good, thank you!
My frustrating story with compressors basically revolves around me understanding which instruments/tracks need them and which don´t.
Like many of us here, I´m doing mixes on my own now, at home and there´s a lot of trial and error going on. I put some compressing on some drum machines and it sounded good, but not as cool on some synths, you know? But now with your basic and easy explanation, I have a bit of a deeper understanding and will definitely be tweaking with more confidence from now on! 🙂
Thank you!
Compression has always been slightly mysterious to me. I could “copy” (or at least try) what I heard on pro recordings, or pretend I knew about compression during a live performance, but your explanation helped the lights turn on! Thanks for the lesson!
Graham you’re right on! thank you for helping me understand the basics, I’m using a Tokyo dawn labs comp that let’s me hear what I’m doing when adjusting the threshold, some of my tracks are at -12db to -17db that’s where the threshold starts for me, Graham you’re the best!
Ah Compression – The secret sauce for me in live mixing that I cannot live without. Thanks for the great explanation and for touching on the “key” components. I agree: You covered 90% of what is needed. Knees, key in’s, and side chaining are there and do have their specific place. Thank you, as always, for humbling sharing your knowledge to the recording community so freely. God Bless!
Simple but effective way to explain compression! Nice Graham.
I’m having trouble with the use and effectiveness of compression within a certain frequency range, similar to the multiband compression. It’ll be really cool if you can cover that topic on compression!
When I went to school for audio, there were 3 things that seemed to be major points of confusion for all of us in class: The Patch Bay (ಠ_ಠ), Impedance, and Compression. Compressors are confusing at first.
My friends that are a little bit interested in audio always end up at some point telling me they don’t understand compression. It took me forever to fully understand it myself, until I realized, ignoring all the knobs they have, all they do is turn things down automatically. Boom. Simple. After that, it’s just learning about how it goes about doing that.
Need a flatter, more consistent volume? Up the ratio! Need a little more pronounced pluck on the string? Slow down that attack!
Once again, thinking in simple terms is the key here. Graham, you definitely helped me to do that! After spending so much time on this site, I kind of regret being thousand of dollars in the hole for an audio engineering degree 😉
I’m sure you learned some great stuff in school though 😉
Oh, absolutely, and from some fantastic teachers and mentors. In fact, I wouldn’t even be interested in engineering today if it hadn’t been for my time at school.
At the present time, I don’t have a proper DAW. I have the complete Nero suite, which includes SoundTrax and Wave Editor. These work quite well, but the “dynamic processor” feature is a PitA in one crucial way.
Instead of knobs to adjust threshold and rate, which is how I learned to do things, this tool has a fully graphic interface. When I say, “fully graphic,” I mean all you get are dots and lines on a grid. There are some numbers along the sides, but they don’t line up with the grid, so they’re nothing but baubles and bishbosh. The grid lines also don’t correspond to any values, either; this is demonstrated by their own presets! There are no values given for the dots, so it’s hard to adjust them to anything. I’m forced to use just my eyes and ears, which can be good and bad.
The one good thing about all this is that I can create complex curves using as many dots as I like, and it can also create a type of reverse compression not unlike expansion.
Attack and release times are virtual knobs, but I rarely hear any discernible effect.
Needless to say, I don’t feel like I’ve had enough practice with normal compressors. I learned all the stuff Graham explained above elsewhere, and I have used it to try to make stuff happen as best as I can, but I’ve always known in the back of my mind that it’s not the same.
So, for me, this is gonna be great!
This gave me some relief I now have some answers to questions I’ve been trying to figure out thanks Graham u da boss ^.^
I have to say actual compression has not been that hard of a concept to grasp, hell, parallel compression (or NYC compression) has been a fairly easy one to sort of get down. The only thing regarding compression that still boggles my mind to this day, is sidechain compression. It is usually only noticeably heard in electronic dance music (what I mostly produce, you can here my stuff at http://www.soundcloud.com/5skar ) but can also be a fantastic technique in preventing similar frequencies from conflicting with each other, producing a bad sounding, muddy mix. My main struggle with that type of compression has been with the really obvious ducking heard in electro/progressive house style of music, and probably more so leaning towards wrapping my head around the attack and release parameters. Basically the problem has been that when I went to make a preset to use for sidechaining later on in my DAW, by the time the volume came back up from being ducked down by the ghost signal (typically a 4×4 kick track assigned to No-Output) the whole effect sounded extremely off time. The idea I initially had I’m mind was a consistent 1 and 2 and 3 and 4 and sort of feel, just the bass or synth ducking in volume on the downbeat, then coming back up right between the previous downbeat and the one to follow, in other words the “and ” of the rhythm. In the past I have experimented a bit and noticed that the velocity and duration of the signal that triggered the sidechain had definitely had an effect on the timing I was hoping for every time. Hopefully that made any sense haha, to me I feel as if I came off a little long-winded, just trying to make sure I illustrate my point clearly. Let me know if any of you guys have tried this type of compressing out and have experienced similar problems and solutions, or just want to chat! 🙂
Well said. There’s really more to compressors than meets the eye!
I always explain compression to beginners as follows: it is a tool that helps you control a performance by evening out the highest peaks with the lowest ones in your audio, thus giving a more balanced gain.
Biggest frustration/something I tend to do is, over compress everything. I love the control of dynamics and getting a static mix that functions without many fader moves that I tend to just squash everything. Also at the mastering stage I can’t seem to boost levels to a good point between a squashed mix and something too quiet.
Dear Graham and others,
Though I read some comments saying this can be found anywhere on the internet which I guess is true but I want to point out Graham is the only one so far I know of to implement this in a big whole package to get your mix done from start till end leaving no subject you can encounter uncovered. To use a tool whatever it is, you first need to know what it does and how to approach it, and I think Graham nailed it with his effective methods.
The most difficult part I find about compressing are the slide knobs in stead of the ‘this’ or ‘that’ setting. It can quickly tend to an tweaking marathon. This was of course before I met Graham and his methods to limit the amount of time for a mixdown and mastering and just focus at the major disturbances getting solved.
Nowadays I can safely use slide knobbed compressors and A/B compressors without destroying a mix and have time left too!
thanks for the knowledge, i definitely learned alot.
Awesome help, thanks GC!
Thank you, thank you, thank you, for giving us these VERY helpful instructions about compression. As a newbie, i had no idea what they were for! So now, i can’t wait to get in to the studio to apply these rules, and see how much better i can get my recordings to sound! God bless you and yours, Graham!
Nice and simple Graham, compression to me at one time was what was in my 57 Chevy, thanks.
I don’t normally comment twice, but I have a question about normalization vs compression. While I’m at it, I’d like an opinion on the studio use of the following bit of freeware as well as normalization in general.
http://web.archive.org/web/20130729204551id_/http://www.conversationsnetwork.org/levelator/
I realize normalization and compression are two different things, or at least are supposed to be, but the differences can be a bit fuzzy sometimes. It seems to me like normalization as I’ve seen or heard it, is a cross between a hard limiter and a volume control, yet descriptions of it seem to indicate some other esoteric mumbo jumbo is involved. I’m not one for believing in alchemy, so I figured I’d bring it here.
Normalization is very different from compression. It simply raises the volume of the entire audio file until the loudest peak hits zero, then it stops.
So the dynamic range stays intact. No compression happens. Just volume increase.
You’re describing peak-style normalization, right? In any case, that explains why it never seems to work right–LOL! Thanks for that clarification.
Now I have to ask, is there some way to accomplish the same thing with a compressor and gain control? As I said, it does seem to behave similarly to a limiter sometimes, depending on who designs the algorithm (I’m not a coder–I just know what I see in the graphic or hear in the speakers with some of them).
Also, is RMS-style normalization better than peak-style? If I understand correctly, it should be the same thing, except it stops raising the volume when the RMS hits a certain level, right?
Thanks again!
Well, since I am a beginner in the field of mixing, I have problem to decide if the sound after the comperssion is better or not (I guess I am lacking some sort of the “feeling” for the working with audio :D).
nice graham reli helped me alot mostly by d mathematical demonstration bt pls can u gv us sum info abt compression wen mastering and if possible sumtin abt multi compression in d mastering stage to help d lyks of us with home recording studio.