Drum Mixing: Part III - Compression

Over the last couple of days, we here at PAL have been discussing certain  tips and techniques as they relate to mixing drums. As anyone that has worked with mixing and EQ can tell you, unless we can directly look at all of the specific details of your project, we can’t just give you a meticulous step by step guide complete with specific settings to dial in. The reason for this is that every mix, every instrument and even every take of the same instrument for that matter is its own beast that must be attacked in its own specific way, but don’t worry, although creating music is an art that can be approached in a seemingly endless amount of ways, there are still some basic principles that can be applied with every project as well as common problems that you might encounter and ways to maneuver around them. Even though it’s an art, there is still plenty of science in there and by exploiting these constants, you can begin understand the mixing process in its entirety. An example of one of these scientific constants might be something like any frequency that registers below 100 Hz starts coming off as a rumble that you feel instead of hear or how the middle frequencies of a drum track is where most of its unwanted ‘boxy’ sound resides. This is where our guides come in; yes, we can’t tell you what settings to specifically dial in or what how to graph your EQ, but we can point you in the right direction by giving you an idea of what things you should be looking for, universal techniques that will most likely apply with your project and even some suggestions on what to do for certain situations – basically giving you general guidelines that can be reapplied with any future project.

In case you haven’t read our last two drum mixing guides, you can check them out after the break:

Drum Mixing: Part I – Leveling and Balance

Drum Mixing: Part II – EQ


Understanding Compression

Ah, good ol’ compression. Much like EQ – or like most anything else actually – when used correctly, it can be a godsend. When used incorrectly on the other hand, it can create a complete disaster. Unlike EQ though, whose figures have to be either plotted on or inputted on a graph manually, compression is more of an automatic volume leveler of sorts (instead of a manual frequency shaper) that just needs a bit of direction before doing most of the actual work. You still have to come up with the numbers to plug-in yourself, but it’s much less of a hassle than EQ and a bit easier to comprehend for newer home studio producers. In very simple terms, dynamic range compression – as it is formally known as – reduces the volume of loud sounds and amplifies quiet sounds by “compressing” a signal’s dynamic range.

The way a compressor knows exactly how to compress a given track is delegated by a few settings that tell it exactly how much to compress, when to come in, and how long to fade out. First of all you have your threshold setting which is like a volume cap for your track. The units measured are usually seen as decibels (db). If you set the threshold at -15 db for example, any signals that go beyond this volume cap will be compressed. By how much? Well, that brings us to the ratio setting. Any of the signals that go past the threshold are then compressed according to this ratio setting which is seen amazingly enough, as a ratio (such as 2:1, 10:1, or even something like 4.5:1).

 Keeping in line with our example, let’s say a certain signal went past our -15 db threshold all the way up to -5 db, meaning it’s essentially 10 db over the cap. According to our 2:1 ratio setting, it will compress the signal 2 to 1 (or by half), meaning that for every 2 db it goes past the cap, it will compress it to 1 db. In our example, this means that our 10 db overage will be compressed to 5 db, which will result in our signal leveling off at -10db (because -15db + 5db = -10db). If our ratio were set on 5:1 on the other hand, our initial -5db signal will be compressed to -13 db since the 10db signal overage would be compressed 5 to 1, leaving us with 2db, and since 2 db added to our threshold of -15db is -13db, this is the volume we are left with. The same is done with upward compression to make quiet noises louder. The next two settings are known as attack and release/decay. Attack tells the compressor how long to wait before it starts compressing and the decay or release tells it how long to wait before it stops compressing. These are described using seconds, usually milliseconds. We’ll touch more on those two a bit later.When used right, a compressor will reduce the amount of peaks and dips in the track’s volume range and should give it a more consistent overall volume


Using Compression on a Drum Mix

Simple software compressors can still be used with great results*

Besides taking care of the volume levels of a track, compression can also be used to change or enhance the tone of an instrument. This change in tone can be different from instrument to instrument with some being changed very slightly while others are changed more dramatically. But what will compression do for our drum track? Since compression pretty much makes loud stuff quieter and quiet stuff louder, when used on drums, you will hear more of the body of the drums and less of the transient attack sounds – such the sound of a stick hitting the drum head. That’s might be putting it in very simple terms but the idea that we are aiming for is there. By lowering the volume peaks of the transient attack sound and rising up the level of the body, you will have a much more balanced drum track that will allow you to raise its overall volume in order to better cut through a full instrument mix. A common example of this is the floor toms. On their own they can resonate pretty loudly but once they are combined with the entire mix, you might find they get lost in the rest of the kit. Adding a compressor to the floor tom will give it more body and thus help bring it out in the mix.

Remember, compressing drums is very much like a balancing act; essentially, you want to compress each part of the drum kit enough so that they will stand out and be heard more clearly though the mix but if you over compress, they won’t have much range, meaning you’ll end up with a flat, dull, lifeless drum track – but probably very loud. There are some home studio producers out there who tend to love adding tons of compression to the overheads. While this will result in a pretty cool “full room” effect – giving the drums a very organic and natural sound – the more you compress the overheads the less likely the other drum tracks will cut through, so keep that in mind.


Don't Forget about Attack and Decay

Let’s say for example that you have added compression to the snare drum and you are very satisfied with its overall tone and has a good balance between the body of the drum itself and the rattle of the snare. Alright, so far so good, but let’s say once you start trying to mix it in with the rest of the kit, it doesn’t cut through. A common mistake made by newer home studio producers that haven’t had much experience with compression is that they tend to ignore the attack and decay/release settings. For example, if you happen to set the attack time to low, the transient sounds of the kit will get turned down. The compressor will kick in before the full sound of the transient (such as the stick hitting the drums) comes in, making the snare sound muffled and less punchy. Too much attack time on the other hand and chances are the entire drum kit might pass through the compressor before if even had a chance to kick in, meaning you’re pretty compressing nothing. The decay and release times aren’t as big of a problem as long as you understand that a very long decay/release time might not give your compressor enough time to recover (stop compressing) before the next drum hits, so try to keep it more or less on the short side. Anyways, the reason the snare in the example probably wasn’t cutting through was because the attack time was too short; by making it a bit longer, you’re allowing the initial crack of the snare to pass through the compressor, while the rest of the sound gets compressed. Your goal with the attack time setting is find a good balance between the transient snaps of the drum and the fullness of the body.


One Last Thing

There are certain home studio producers out there who tend to like to compress the entire kit together. Whether that’s a good idea or not is pretty much up to the producer’s ears but if you find yourself curious and wanting to take a similar path, all you need to do is simply route all of your drum tracks through a single audio track, essentially creating a full drum kit ‘submix’ of sorts. After you have this single track submix ready, apply compression. Doing some EQ is not a bad idea either. One of the few drawbacks to applying compression to the entire kit is that if you apply the wrong settings, you can screw up the entire sound pretty badly. I would go for longer attack and release times than you would normally use on single drum tracks so that the compression comes out sounding very smooth and doesn’t interfere with the transient attack of the kit.



*Compressor featured on freeware audio editor Audacity

Leave a Reply