How To Use A Compressor | Threshold, Ratio, Attack, Release & More

[00:00] Dynamic range refers to the difference between the quietest and loudest portion of a signal. The most basic use of a compressor is to reduce the dynamic range by turning down the loud parts in relation to the quiet parts. This has the effect of smoothing out the signal level throughout

[00:14] the performance. The action of compressing a signal alone will reduce the overall signal level, but when you boost the signal after compressing it, you'll be left with a louder signal than you started with. That's because you're turning down the loud parts in relation to the quiet parts

[00:28] and then bringing everything up together. Let's take a look at the controls that you'll commonly find on a compressor and how to use them. I'll be using one of my favorite compressors for this demonstration, the FAB filter Pro C2. You'll find a link for that compressor in the description below

[00:42] the video if you want to try it out for yourself. The reason I love this compressor so much is because it sounds great, it's very flexible, and the metering is unmatched. For the sake of visualizing and understanding what a compressor actually does, let's start with a snare drum track that I put together.

[00:58] You can see and hear that these snare drum hits alternate between quiet and loud. Right now, the compressor isn't doing anything because the threshold is set to zero, and the ratio is set to one to one. The threshold is the point at which the compressor will start compressing. With the

[01:14] current setting, even the loud snare hit doesn't exceed the threshold, hence the compressor isn't doing anything. Let's turn the threshold down so that the peaks of the signal reach it. Still, nothing is happening, and that's because we need to set the ratio. The ratio determines how much compression

[01:30] will occur when the signal level exceeds the threshold. Right now, it's set to one to one, which means that for every one dB past the threshold, one dB will pass through. Again, this amounts to the compressor doing nothing. But if we set the ratio to four to one, we'll start to see some action. Now, for every four

[01:48] dB past the threshold, only one decibel will pass through. Over here, we see the input signal level on the left, the gain reduction level in the center and red, and the output signal level on the right.

[02:00] The gain reduction meter tells us how much the compressor is actually reducing the signal level. These gray lines show the input and output signal level over time, while the red line shows the gain reduction over time. Maybe you're starting to see why I like this compressor so much, especially as a

[02:16] teaching tool for teaching what a compressor does. Let's play around with the threshold and ratio settings before moving any further. We'll see more gain reduction anytime we lower the threshold

[02:30] or anytime we increase the ratio. Let's say I want to make the loud and quiet snare hits the same level. I'll set the threshold to the peak of the quiet snare hit so that the quiet snare hit

[02:44] isn't reduced much at all. Then I'll set the ratio as high as possible, effectively creating a brick wall limiter that doesn't let any sound pass the threshold. By the way, this is considered a limiter rather than a compressor because the ratio is greater than 10 to 1. I want to take a moment

[03:02] to mention that some compressors don't have a threshold adjustment, but instead have a fixed threshold. This is the case with the classic UA1176 and any plugins that model it. Instead of bringing the

[03:14] threshold down to the signal level in these cases, with these compressors, you'll boost the signal level up to the threshold using the input gain knob. Okay, back to the fab filter proceed to. So far, we've just turned down the louder parts, making the track overall quieter, but you might be

[03:31] thinking I thought compressors were used to make things louder. Well, this is where makeup gain control comes into play. The makeup gain control is used to turn the overall signal level up to make up for the

[03:43] reduction we've applied with the compressor. I find it helpful to think about it this way. If we wanted to make the quiet snare hits louder, we could just turn up the overall level, but then the louder snare hits would exceed 0 dB, which would cause clipping in a fixed point digital system,

[04:00] or some distortion in an analog system. This is why we use a compressor to turn the loud hits down and use the makeup gain to turn everything up. If all this makes sense so far, leave me a comment that

[04:12] says I understand ratio and threshold. Next, let's look at the attack and release controls. The attack knob determines how quickly the compressor will begin compressing once the signal level has

[04:24] exceeded the threshold, and this compressor's attack ranges from 0.005 milliseconds to 250 milliseconds. When I was trying to crush that loud snare drum hit down to the same level as the quiet snare drum hit,

[04:38] I used a very fast attack time, but you may sometimes want to let the transient through and only compress the tonal part of the signal. The transient is the high amplitude short duration sound at the

[04:51] beginning of the waveform. The tonal part of the sound comes after the transient. Using the attack on a compressor, you can decide if the compressor will quickly clamp down on the transient, or if it will let the transient through and only kick in during the tonal part of the sound.

[05:06] The release knob determines the time the compressor will take to recover after compressing. The release time on the FAB filter Pro C2 ranges from 10 milliseconds to 2500 milliseconds. A fast

[05:18] release time will recover more quickly, while a slow release time will take a little bit longer to recover. The whole adjustment controls how long the peaks and gain reduction will be sustained.

[05:31] While the release will usually begin right after the peak and gain reduction, the whole time can push that back in time so that the gain reduction is prolonged before the release begins. To visualize the knee setting, I think it's easier to look at a more traditional compressor

[05:48] plug-in layout. On this graph, the input signal level is shown on the x-axis and the output signal is shown on the y-axis. The threshold is the point where the line bends while the ratio determines

[06:00] how much the line bends. That brings us to the knee setting which can be used to soften or harden the compression around the threshold. A softer knee will generally create a more gradual and transparent

[06:12] compression. Many plug-ins have a wet dry knob for parallel compression. The wet signal is the signal at the output of the compressor that's been compressed, and the dry signal is the signal at the input of the compressor before it's been compressed. You can sometimes achieve a more realistic

[06:29] product by blending the two together, preserving the transient of the dry signal while adding in the thickness of the compressed wet signal. This is called parallel compression and I'll go into more detail in the future video. You can find links to all of these plug-ins in the description below

[06:44] this video. I hope you enjoyed the video. If you did, hit the like button and watch the video that's on your screen now. I'll see you over there.