Don’t Buy A Hardware Compressor…Unless

Brandon Drury —  October 28, 2007

It’s Day #1 of your recording. Actually, it’s kind of nice looking back to my Day #1. It kind of reminds me of when I was nervous around chicks when I was in junior high school. Now I’m older and my more of a cynic. Oh well.

Anyway, it’s day #1 and you are pulling all of the gear out of the box. You don’t know what half of it is and you aren’t really for sure if you bought the right gear. You spend all day tearing your hair out because this whole recording thing is a little more complex than you had originally bargained for. (Don’t worry, we are here to help. Post on the recording forum if you are having trouble. )

After you go through the depths of hell, you finally get everything working. Then you start trying to figure out how to use the recording software and actually get something that sounds good. You were told by some idiot that you MUST have a hardware compressor, so you purchased a cheap one. (Relatively cheap, that is).

You may be asking “How in the HELL do I work this thing?”.

Unfortunately, using a compressor is a lot like riding a bike or maybe a better analogy is the first flight of a baby bird. The only way to learn how to use a compressor is to crash and burn repeated times. In fact, this is how you set a compressor EVERY time. Let me explain.

The whole idea behind a compressor is limit the dynamic range. In English, this means we want to knock the peaks of the loud stuff down. Of course, this is tremendously oversimplified, but it’s the general concept. No one can tell exactly what settings on a compressor are going to work because it’s highly dependent on a million factors. You have to sit there and tweak the compressor to get it right. So what’s the big deal?

The Problem With Hardware Compressors In The Digital Age
#1 Can You Even Use A Hardware Compressor With Your Audio Interface?
In a typical home recording setup, you will run a budget microphone to an audio interface which will have a preamp. From there the signal is wired directly into the computer So, there is no way to get to the signal directly after the preamp (gain knob thingy). Your options? If you audio interface doesn’t have some sort of “insert” on it (most do not) you’ll have to buy an external preamp to even use your compressor. This preamp probably will be of similar quality to the preamp in your audio interface unless you spend some big bucks, but if you are beginner you won’t even hear a difference with the big bucks preamps either.

#2 How Can You Actually Hear The Compressor?
We’ll assume that you have an external preamp or your audio interface has inserts so that you can actually use the compressor. How will you hear it? In other words how are you going to listen to make the changes to the settings on the compressor (especially if you are brand new to recording and do not know what to listen for)? The singer must continue to sing while you play with the compressor (if you are recording vocals). If they are in the same room as you are, you may have trouble using your studio monitors for this. You may also be effecting their vocal performance, so while you take the time to figure out what a compressor actually does, you are going to be screwing with the singers mix and they will have problems and maybe even stop because they can’t hear themselves.

While an experienced recording engineer could probably set this up without too much trouble, this is NOT the place to learn what a compressor actually does.

#3 What Will You Do When You Overcompress A Good Take?
So your rig is hooked up to properly use a compressor and you’ve figured out a way to hear what is going on. You are still learning how to use the thing. What will you do when you overcompress the signal on a good take? Once again, do you really think it’s a good idea to practice while tracking? I don’t.

#4 What If Your Singer Needs Aggressive Compression Only In The Headphones?
What if your singer is VERY dynamic and needs to be crushed pretty hard in the headphones? This is fairly common. What if the headphone mix requires an amount of compression that you really aren’t comfortable using on the real track? If you have a mixer of some type, you can pull off, but if you require different compression in the headphones and on the recording, you will need two compressors to pull this off.

Why I Recommend Compressor Plugins For All Beginners

#1 The time to practice using compressors is 2am when no one else is around. You can REALLY experiment with a compressor during mixing. You can really learn what the attack and release times do. You can get a feel for how much gain reduction is needed.

#2 In my opinion, compressor plugins are really good! (See Compressors: Shooting Down The Anti-Plugin Mentality I’ve heard many very expensive compressors. While I’ve never heard a plugin that could compete with the outstanding, bold character of a $4,000 EMI compressor at Michael Wagener’s studio, I think compressor plugins do the job fine. Until you get to a point where the biggest problem in your mixes is compression character, I wouldn’t worry about it. You’ve got a while!


Brandon Drury

Posts Twitter Facebook Google+

Brandon Drury quit counting at 1,200 recorded songs in his busy home recording studio. He is the creator of and is the author of the Killer Home Recording series.

13 responses to Don’t Buy A Hardware Compressor…Unless

  1. …but, isn’t one of the functions of the hardware compressor to ensure that the input isn’t clipped before the digital conversion process? I’ve been told that, once a clipped signal gets into your computer, it can’t be “un-clipped” by a software compressor.

  2. I wouldn’t go so far as to say that the purpose of a hardware compressor is to eliminate clipping, necessarily, although it can do this.

    I’ve recorded well over 500 songs without a hardware compressor. There is no need to jack up levels so high that they would clip without compression or limiting.

    A compressor is used as a tool to improve tone. To use it to make up for poor gain structure is not really the point. Does that make sense?

    If I’m recording a very dynamic vocalist, I’d set the gain so that they are just under clipping at their loudest parts (maybe a scream or something). Then, I’ll use a software plugin to crush the vocal pretty hard when it gets up to that point so that the mix in the headphones is about right. Of course, this compressor plugin can be removed at any time.

    Brandon Drury

  3. Of course, another good way to do it is to have the singer sing the quiet parts only on one track, then adjust your levels for the loud stuff, then record the louder/screamo stuff a separate take on another track. Then, not only are you getting more even levels, but you are also maximizing your bit depth (assuming of course it’s a digital setup). Third, you get more pronunciation from the singer, because he has time to finish words and also start new sentences because of the time break.

    “Well, what if the singer doesn’t feel it as much, or loses their feel, and blah blah blah..”
    All I can say for this, is use reverse psychology on the guy (or girl). Boost their ego up way huge and then quickly shoot them down by saying, “What, do I need to come in there and sing it good for you?” Then they feel the need to prove themself and will say, “pfft, whatever dude, hit record..” and then their agression is a lot more even keeled.
    Actually, I’m totally kidding (sort of). All jokes aside, The vocalist should be able to pull it off, especially because the singing style is changing, therefore, the voice will naturally be different anyway. It shouldn’t be an issue.
    If I could recommend a really bitchin software compressor, it’s the PSP Audioware Mixpressor (comes in the mixpack for $150). I seriously, seriously recommend this compressor. Why? Because it’s very musical sounding, and it does a great job of replicating analog. Brandon is totally right, most people wouldn’t hear the hardware anyway.
    The one thing I could say that’s good about the hardware compressor is if you have multiple outputs, you could probably save yourself some CPU cycles by using it as an insert…at the expense of taking up another input track as the signal returns to the computer.

    So….”Never use a hardware compressor unless… buy a really great one, have multiple outputs, an extra input, have super sick D/A converters, and can feed the proper signal to the compressor and back in again.” Then you can use it during mixdown…but don’t even think about tracking with one.

    P.S. Brandon, I love the site dude. You are funny as hell.

  4. Never use a hardware compressor unless… buy a really great one, have multiple outputs, an extra input, have super sick D/A converters, and can feed the proper signal to the compressor and back in again.

    Dude, you hit the nail on the head. I’ll definitely check out the PSP Audioware Mixpressor. I’ve been eyeballing the Focusrite Liquidmix for some time now, but if I could get a super musical sounding EQ and compressor for less, I’d do it!


  5. I agree with the point of not using a hardware compressor during tracking unless you know how, but I believe that if one wants to be a proficient recordist, they need to understand compression and how to use it at every phaze of the project. I personally use compression at tracking(“printed”), on the independant track and on the stereo buss of all projects I do. Transparent compresson technique takes many years of practice and one can never really stop learning it, as each waveform affects the unit differently.

  6. I am pretty much with all the seasoned pros on this one – you can compress after the tracking in the mixing stages. It would be nice if your audio interface had a compressor/limiter on the input stages so that way you can feed a hotter signal but if you don’t know how to feed this it could be detrimental instead of an asset to your session.
    I personally when working in the studio set limiters on all channels to prevent digital overs (or more commonly known as “clipping”) and maybe some mild compression on the tracks that require it. On the other hand, at home, I record straight from the interface and watch the levels. Once in a while you’ll overload it and have to redo a part but more you record the better you get at it.

  7. Software compressions seems fine but what would be the a recommended approach to prevent occasional clipping, like when working with a group or if you have a singer a gets a little more excited than anticipated at a certain verse.

    Do you just say “Amazing performace! You really ‘felt’ that…. but unfortunately it’s clipped, can you just do it again?”

    Do you record at an overall lower volume and sacrifice the bit depth?

    Or maybe you can but limiter between the pre-amp and DAW – can you recommend one?

    Thanks for the help.

  8. Software compressions seems fine but what would be the a recommended approach to prevent occasional clipping

    The best solution to prevent clipping is setting proper gain structure. That may mean using less gain on the preamp. I think the myth that you are actually going to get audible benefits for loud music for tracking at high levels causes people to set their gain structure way too high.

    If I’m tracking a dynamic source, I always plan ahead with the idea that the loudest sources simply will not clip.

    Hardware compression is not, in my opinion, a utility as much as it is there for tonal creativity.


  9. Do you record at an overall lower volume and sacrifice the bit depth?

    “Sacrifice” is a VERY strong word. I’ve seen no correlation between fidelity and tracking at -6dB or -15dB. None. Maybe somebody can hear it, but I think this bit depth benefit is dramatically (and maybe even infinitely) over stated.


  10. Brandon – Thanks for the response.

    I was under the impression that you lose resolution on your DAW as your gain goes down. Therefore I always try to feed a very hot signal to the DAW. Are you saying it’s better to feed overall lower dB signals to you DAW (enough headroom so it can never clip) instead of using a limiter for the rare gain spike?

    How low does the input have to be to really notice loss of resolution anyways? -12 dB? -32 db? -64 db?

    p.s. Where’s the article on gain structure :-)?

  11. Our post crossed paths. Thanks for addresing the “sacrifice” concept. I’ll try tracking at lower levels and see how it sounds. I guess you pump up the juice at the mixing mastering phase more anyways (right?)

  12. I guess you pump up the juice at the mixing mastering phase more anyways (right?)

    Definitely! The actual level of a mix is more about maximizing RMS level and you get into compression and such. The level of the preamp during tracking is almost totally independent from what ends up in the final mix. That’s why faders are so big on a console or your recording software.

    This bit depth business is dramatically overstated like pretty much everything else that isn’t important to recordings.

    You may want to check out this in regard to bit depth:

    How low does the input have to be to really notice loss of resolution anyways? -12 dB? -32 db? -64 db?

    How much gain does a guitar need? How much low end in a kick? How much reverb on a vocal? Use your own ears and decide.

    Here’s the gain structure article:

  13. Cool. I’ll check those articles out.

    Next time I purchase something I’ll go through your ads ;-)

    Thanks Again!