Effective Techniques to Learn Compression

Learn compression

Compression has become an integral ingredient of most music these days. At the same time it remains somewhat a secret science – a commonly misunderstood one. It’s a great pleasure to have the mastering engineer Barry Gardner from SafeandSound online mastering services to talk to us about how to learn compression. During his long career in music Barry has worked with people such as Terry Callier, Ronnie Wood, Craig David, Kano, Chick Corea and many others. Needless to say he knows his stuff. Time to pay attention.

Compression can be used in both technical and creative ways and is a much misunderstood process in audio production. Hopefully this article will help you learn what compression is capable of and introduce ways of understanding it based on a goal, rather than trial and error.

The best way to understand compression is to deconstruct it and then reconstruct it with a goal, be it technical or creative.

 

The Common Controls of a Compressor

Let the deconstruction begin. It is good to have in mind that job of a compressor is an automated level control which has and initial downward action – much like pulling down a fader. However there are certain controls which fix the response of the compressor. In this case what goes down must come up! We shall start with each control found on a basic compressor. We could go into great depth about the variations you can find, but for this article we will only be concerned about the controls that are relevant to all audio compressors.

  • Threshold: level at which compression starts to take place.
  • Ratio: amount of compression applied when the threshold has been exceeded.
  • Attack: Speed at which level is reduced (a time constant) practically expressed in milliseconds.
  • Release: Speed at which unity gain is restored (second time constant) also practically expressed in milliseconds.
  • Make-up gain: A post compression gain control allowing the gain to be restored to that similar as the pre compression level (remember compression reduces the level of sound peaks).

Tokyo Dawn Labs Feeback Compressor

 

 

Uses for Compressors and the Importance of a Sonic Goal

For a compression novice there will be significant online producer peer pressure to use compression whether you understand it or not. Make no mistake, compressors can make your music sound worse quite easily if you do not understand them. Do not rush here if you are starting out. A much better way to learn compression is to understand some basics and then apply.

In technical context a compressor can even out the levels between the naturally occurring peaks and troughs in a piece of audio. For example in a voice recording, the voice will naturally be softer and louder at times and it can be good to even this level out for the purposes of intelligibility – for example against a music back ground. This is a good technical use of a compressor.

A creative use of a compressor could be to adjust the tone of an instrument. This can be achieved by enhancing the micro dynamics of for example a snare drum. It is possible to bring out some extra snap in the drum attack when the stick hits the skin and maybe some extra body from the decay of the sound.

Initially it will be difficult to practice a goal driven need for compression. When you first start using a compressor at degrees where it would commonly be used it requires significant listening skill to determine the subtle changes. So some fairly extreme settings should be used at first so the audible changes are more dramatic. This can be backed off when you improve your ear to hand coordination skill set.

 

Initial Practice

For the following practice I recommend a good pair of headphones so you can focus on the sound changes with less disturbance from room acoustics and/or other room noises.

The human voice is a good sound source to practice on. The human ear is very sensitive to any unnatural artifacts introduced into this important human sound.

I suggest setting up a track with some speech or singing on it and applying a compressor with these settings (compressors are employed in the insert point of a track most commonly). Ensure the loud peaks in the voice are peaking at -8dBFS and set the compressor as follows:

  • Threshold -16dBFS
  • Ratio to 10:1
  • Attack to 5ms
  • Release 200ms

When you play the voice track you should see activity on the gain reduction meter (amount of compression being applied). You should also be able to hear the voice level drop in synchronization with the meters activity. The main practical use of this meter serves to provide an indication of how much make-up gain to add in order to match the original signal level.



 

Make-up Gain Is Very Important When Learning Compression

Make-up gainI cannot stress this enough, the make-up gain is the control which allows you hear compression action as opposed to level difference. We know the level will drop when compression is applied. But we want to hear the improvement or detrimental effects of the action of compression and not the volume change. If the predominant change is level drop, it will be difficult to judge if the action is positive or negative. Once the input and output levels are matched, you can then use the bypass button to listen to pre and post compression sound.

The next step is to start gently adjusting the attack time and listening to what audible effect this has on the action of the compressor. As the attack time becomes longer you should hear a little more punch coming from the somewhat muted (at 5ms) snare attack. Then try adjusting the release time and considering what happens audibly. When adjusting the attack and release, always consider the make up gain control and A/B the input signal versus the volume matched compressed version.

Ok – so you might feel that was a lot to take in! Well it is, and learning how to effectively use compressors usually takes most producers and engineers a few years to become competent. This is made more complex by the fact that most compressors have different sonic signatures. Some are smoother and some are more aggressive and so on.

Learning how to use compressors is an iterative process as opposed to a smooth learning curve. The reason being is you will try a compressor on a new sound source and it will probably not respond the same as on any other source. The frequency content and dynamics of that particular instrument are different. Try another compressor and again it sounds quite different at the same settings. So be prepared that there is a lot of practice to be had before you will master it.

 

Suggested Second Practice

I recommend getting a drum loop playing and peak it at -8dBFS. It is best this is a sparse drum pattern. A kick, snare and hihat is ideal. Set the compressor controls as before. Listen to what happens to the attack of your drums when compression is applied and make-up gain is matched. A/B the input signal and listen carefully to what you hear. Then you can start adjusting the attack and release times and considering the sonic results.

Listen to the following aspects:

  • Punch/snap (or lack thereof) of kick and snare drum.
  • Room ambience/acoustics in between sounds.
  • The pressure of the sound as a character of sound.
  • The tone of the sound and lose or gain of highs.

Lots of subtle changes can be identified and this is why these tools are coveted by sound engineers.

 

How to Learn Compression – Summary

To learn how to use a compressor will take time and practice and progress will be somewhat dependent on grasping the fundamentals and your ability to hear the subtle changes. As one progresses you can use less extreme settings ( i.e. lower ratio, say 3:1 or longer attack times with shorter release) and continue to listen to the changes as you do.

Eventually you will slowly be able to discern all of the changes the compressor invokes. So put some time aside for some daily practice to learn compression and in the not too distant future you will be a competent controller of dynamic range in your tracks.

 

Barry GardnerBarry Gardner operates SafeandSound online mastering services in London, UK. He has worked with many world famous artists including: Ronnie Wood (Rolling Stones), Lemar, Craig David, The Rumblestrips, The Coral, The Hoosiers, Chick Corea, Kano, MC Skepta, Avishai Cohen, Incognito, Martha Tilsden, Still Remains, Terry Callier, Amp Fiddler, Billy Cobham, Marshall Allen and Sun Ra’s Arkestra.

Accept responsibility of your success. We Spin shows you the rest. Join me there:
  • SCH

    Nice article.
    As an ultra-beginner, I love to read about basics. Really appreciated !

    • Ilpo Kärkkäinen

      Great, glad you like it and thanks for letting me know about what you like to read!

  • Harry Ransom

    Thanks for including a couple exercises too!

    • Ilpo Kärkkäinen

      Good move from Barry. I need to start doing that too.

  • http://redmastering.co.uk Red Mastering

    it is good article explaining basics of compression. I would also add that there are different type of compressors (vca, vari-mu, optical, fet, etc) and those work in bit different way, sometimes there’s no ratio, no threshold, and attack/release could be semiautomatic. Therefore what you learn about 1 type of compressor won’t necessary translate on another type.

    • Ilpo Kärkkäinen

      Cheers for the comment. That’s very true. Now that the basis are covered I will start thinking about an article that goes a bit deeper. There’s a lot of ground to cover though!

  • barry gardner

    Red, your first sentence was right and then you simply confuse people. The goal here is learning the basics of compression globally as opposed to learning a specific compressor type, which most people are not going to own. Additionally, have the respect to sign in with your name as opposed to your mastering sideline, otherwise it looks like you are trying to ride off the back of someone elses work.

  • http://www.soundcloud.com/cosmology Ben Thomlinson

    I used to just use a compressor for sidechaining because I didn’t have the knowledge to use it properly so steered clear of it but this is a great read and the practical exercises have been a great help to get me started.

    • Ilpo Kärkkäinen

      That’s great Ben. Remember if you have any questions feel free to drop a comment here and we’ll do our best to help you out.

      • http://www.soundcloud.com/cosmology Ben Thomlinson

        Thanks for the reply :) I do have one question which I think is still on topic? I was recently asked in one of my tunes to make all the notes in the bassline the same volume. In the end I had to go in and edit the audio for each one manually looking at the spectrum to see if the levels were the same. My question is do you know if there is a way to do this using a compressor or limited or am I barking up the wrong tree??

        • Ilpo Kärkkäinen

          Yeah you could use a multiband compressor for that purpose. With that you could compress just the lowest end of the bass in order to tame notes that have a louder bottom end than others.

          • http://www.soundcloud.com/cosmology Ben Thomlinson

            Ok cool thanks I’ll give that a try.

  • http://www.soundcloud.com/cosmology Ben Thomlinson

    I’m having a bash at this again with a compressor and I’m adjusting it so I can see a reduction in gain in the level meter of the notes I’m trying to tame but in voxengo span the louder notes still look louder and when I bring the threshold right down it just seems to reduce the volume of the whole bassline not just the notes I want it to effect. Any thoughts would be greatly appreciated! :)

  • Barry Gardner

    Hi Ben, this is somewhat off topic as multi-band compression is somewhat beyond the scope of the article. However I can respond…..

    Firstly the person asking for this… do you know they have a very linear low frequency response in their listening room ? If not how do you know if this is good advice ?

    I believe multi-band compression is an inappropriate tool for this job. It is much better to manipulate midi note velocity or volume (time consuming though it is). Multi-band compression will always be a compromise across an entire instrument and only to be used as a corrective tool in the last resort.

    To add, it will be extremely difficult to get a single band to control 1 note without effecting the others as it is unlikely that the upper and lower band filters for the compressing band will be steep and/ or capable of honing in on specific notes.

    What does it sound like ? Thats much more important than what it looks like, though this assumes some degree of accuracy in your monitors and room acoustics.

    cheers

    • ben thomlinson

      Thank you for your reply Barry, I’m sorry for going a bit off topic but this is a senario specific to my needs that I thought my be able to be dealt with using compression. I’am not sure why the person is asking for this or if it is a good end result, I’ve never been asked to do this before, but I guess I want to be on this particular label so I’m happy to comply. I will go back to editing the audio then I think, it is a bit time consuming, but I can think of worse tasks to do! Tbh the first two notes of the bassline sounded fine at the same level, but the third low note sounded too loud and grinding so by ear it is better to have this at a lower level. I use M Audio Ie 40s for my headphones which I have found to be accurate having tried a few different types of headphones and the blue sky media desk which covers pretty much the full spectrum of sound so is good for hearing the low end. Once again thanks for your reply, its been a great help.

  • Barry Gardner

    No probs about going slightly off topic, it’s a free commentary box : )
    Do not out rule good old equalization either, though direct volume/velocity control at source is optimal. All the best with the release.