Compression: Dominate your Dynamics
Whether applied individually to a specific element of a mix or applied to the entire mix, there are a ton of use cases for compression.
Compression is one of the most important tools in a mix engineer’s arsenal. Every song you’ve ever heard was subject to the clamp of a compressor at one point or another. Whether applied to a specific element of a mix or applied to the entire mix, there are a ton of use cases for compression.
What is compression though? Put simply, compression is the reduction in volume of the loudest parts of an audio signal in order to manage the dynamic range of the signal. Dynamic range can be thought of as the difference in volume between the loudest part of an audio signal and the quietest part of an audio signal. By reducing the volume of the loudest parts of a signal in relation to its quietest parts we can achieve a more even perceived volume. This is useful when placing individual elements into the context of a mix and can also be applied to the entire mix to serve as a sort of glue holding the whole mix together.
Now that we’ve defined compression, let’s discuss the functionality of a compressor itself. Pretty much any compressor will have a combination of some or all of the following controls.
1. Input, this refers to how much of the dry signal is being sent into the compressor. Many compressors employ electronics that will color the signal based on how much input the compressor receives.
2. Threshold, this refers to a set volume which if exceeded by the signal within the compressor will engage attenuation (reduction of volume). For example, if a compressor’s threshold is set at -5dB, any part of the signal sent to the compressor louder than -5dB will be attenuated.
3. Ratio sets the amount of attenuation applied to the signal by the compressor based on how far the signal exceeds the compressor’s threshold setting. For example a 4:1 ratio on a compressor means that for every 1dB the signal goes beyond the threshold, the compressor will attenuate the signal by 75%. Or down to 1/4th of the input volume.
4. Attack is the time it will take the compressor to reach full attenuation after the signal passes the threshold.
5. Release is the time it will take the compressor to stop attenuating the signal after it has reached full attenuation.
6. Output/Make-up Gain, this controls the volume of the signal coming out of the compressor. When compression is applied to a signal the overall volume will be reduced. So, most compressors have a make-up gain control to bring the overall volume of the compressed signal back up to its original volume.
There are so many compressors out there it can be difficult to know which one to use for what. Let’s discuss the different types of compressor construction and the common uses for each.
VCA Compressors
VCA is an acronym standing for Voltage Controlled Amplifier. Basically it's just a fancy way of describing a volume fader. Essentially a VCA compressor is like a robot that rides a volume fader. It splits the incoming signal and sends one copy to the detector path which informs the compressor’s settings, and the output path which is the signal that actually gets affected. This is called a feed forward compressor. I won’t get too into the intricacies of the electronics and how they work but I will speak on how they are commonly used. VCA compressors are most commonly seen on channel strips and as bus compressors. These compressors are very precise and typically offer the most control over the dynamics of the signal. They are typically the cleanest, most transparent sounding compressors creating little to no distortion or noise. My personal favorite VCA emulation is the Slate Digital FG - Red based on the Focusrite RED 3 Compressor.
FET Compressors
The FET or Field Effect Transistor is the primary component of these compressors. FET’s are semiconductor devices, in which the output current is controlled by an electric field generated by the input voltage. The most commonly known FET compressor is the Urei 1176. By and large, these compressors only have 5 adjustable settings: input, output, attack, release, and ratio. This is because FET compressors employ feedback compression. On the 1176 for example, the threshold is fixed. To engage it you increase the input signal up to and beyond the threshold. As the compressor applies compression it feeds the affected signal back into its input. That means it compresses and recompresses the signal the more you drive the input. This makes for very fast and very aggressive compression with loads of coloration. It is often used to get really breathy vocals or to get really roomy pumping drums. My personal favorite FET emulation is the Purple Audio MC77.
Optical Compressors
Optical compressors are easily my favorite to work with. Not only do I find their design fascinating but I quite enjoy their musicality. Optical compressors employ a light element such as an led that shines brighter the louder the input signal is. This light element is paired with a photosensitive resistor to actually perform the attenuation. The relationship between the light element and resistor makes for a slower more meandering style of compression. This is because of the time delay between the light being emitted and the resistor reacting to it. It also means that the attack and release times are not constant. Typically optical compressors don't even have attack or release controls! Example, an optical compressor is doing 10 dB of compression on a signal. The first 5 dB might get squeezed really quickly while the last 5 dB will attenuate slowly. Same goes for release time. It’ll let go of the signal back up 5dB quickly and begin slowing down the rest of the way until it disengages completely. These compressors are not for precise transient shaping. Where these shine is when a sound needs a gentle squeeze to round out the louder parts. I often go to optical compression when I’m trying to set a vocal into the context of a mix. It helps balance the dynamic range without squeezing the life out of it. Some of the more famous optical compressors like the Teletronix LA2A are known for being hard to get them to sound bad. My go to optical compressor emulation right now is the Brainworx Opto compressor but it’s hard to go wrong with optical.
Delta MU/Vari MU Compressors
While many compressors use tubes in their circuitry, they are typically just for coloration. In a Delta MU compressor the tube is the engine driving attenuation. As the input signal is increased so too is the ratio of compression. This is referred to as program dependent compression. Because of this design, most Delta MU compressors don’t sport ratio controls. This is also where the name comes from. Delta (greek for change) and MU (greek for gain). The level of attenuation is completely dependent on the input signal level. But what do they sound like? In my experience Delta MU compression behaves similarly to optical compression. It’s not heavy handed on transients and moves fairly musically. However, unlike their transparent optical counterparts, Delta MU compressors offer lush coloration through tube distortion built into the attenuation circuitry. My go to emulation is the FG - MU by Slate Digital. It’s an emulation of the famous Fairchild 670.
Ear Training: Listening Critically to Compression
Understanding the different types of compressors and how they function is all fine and good. However, without developing an ear for what compression actually sounds like all that information is just trivia knowledge.
When listening for compression I focus on the two areas of the sound that are affected by compression. First, the transient or very beginning of the sound. These are especially prominent in drum sounds. Then, the sustain or tail end of the sound.
Let’s first focus on the transient. In the audio clip below I have a drum mono overhead going into the Kush Audio Novatron doing about 5dB of attenuation. When the audio starts, the compressor is set to a gentle ratio with the slowest attack setting and the fastest release setting. This means the compressor is engaging very slowly, allowing the transients to come through and disengaging quickly allowing the sustain to come through. Basically there’s not much audible compression going on. Listen as the attack is slowly sped up. The punch of the transient will slowly begin to disappear. Focus on the volume relationship between the hi hat and the kick and snare drums. The hi hat will stay pretty much the same volume while the kick and snare volume will be reduced.
Now let’s examine how release speed can control the sustain of a sound. In the audio clip below I have the same drum mono overhead going into the Novatron doing about 5dB of attenuation. This time I have the Attack speed set at about 10 milliseconds to get the punch I want. At the beginning of the clip the release is set to its slowest setting. Meaning its staying clamped down on the signal and holding the sustain of the sound down. Listen as the release time is slowly sped up. Focus on the relationship between the transient of the snare drum and the ring of the snare drum. Ast the release is sped up the ring or sustain of the drum will come into focus.
By using the attack and release functions in conjunction with each other you can dial in the desired punch from the transients and the desired ring from the sustain of the sound. Say you’re working on a sound that, to your ear, has a lot of transient energy but lacks the sustain you’re looking for. You can reach for fast attack and fast release settings to push back the transients and bring out the sustain. If you’re dealing with a sound that has no punch and tons of sustain you can go for a slow attack and slow release. This will hold down the sustain of the sound while allowing the transients through.
Now that you’re more familiar with the fundamentals of compression it’s time to start practicing. Follow the link below for a free download of 5 sets of drum loop multitracks from the DBA Drums Vol 1 drum pack. I’ve included fully mixed versions for reference while practicing.