Dynamic range compression is one of the most essential effects used in modern audio production.
And if used correctly, it is one of the best tools you can ever have in your studio arsenal. However, if not used correctly, it can destroy a mix, leaving it a lifeless corpse of an audio recording project. Here’s how I look at dynamic range compression when used incorrectly. Remember the old video game Frogger? Yeah, of course you do. If you don’t, just pretend that you do. Anyway… that little rascal, full of life, has to merrily hop himself to the other side of the highway. But… one wrong move and he is finished. He gets squashed to death by oncoming traffic. Yep, that’s exactly what bad compression does. It squashes the life right out of an otherwise good recording. If the recording already sucks, then it doesn’t matter. Kill it!
When you apply dynamic range compression, you simply bring the top down and the bottom up, thus decreasing the range (area between the peak and valley). By increasing the bottom level of the dynamic range, everything with the signal, including noise, is made louder. This is good if you would like to increase the level of room ambience. But you might also increase the level of unwanted noise, such as a nearby air conditioning unit. Keep that in mind while squashing. Now, moving on to the main features of dynamic range compression. If you do not already know the three major features, then let me inform you. The first is threshold, which is the dB level at which a signal must exceed before compression begins to decrease the level. The second is ratio, which is the amount that the signal is reduced. For example, a ratio of 2:1 means that for an input level of 2db above the threshold, the output level will only be 1db above the threshold. A ratio of 4:1 means that for every 4db of input level over the threshold, the output level will be just 1db above the threshold. So remember this: the higher the ratio, the more the squash. With higher ratios, the compressor becomes a limiter, which is a more extreme way to keep the level at the threshold. And the third main feature of dynamic range compression is attack, which is the amount of time it takes for the compressor to do its job once the level exceeds the threshold. The attack setting needs to be approached with much precision. If set too slow, then too much of the signal will get through. If it is set too fast, then not enough of the transient will get through, effectively killing the dynamics of the signal. So be careful. In the music industry, an incessantly increasing volume level and too much dynamic squashing has led to what is called The Loudness War.
There is a very easy way to protect yourself from destructive compression. This is by using what is called parallel compression. It is simple. Just copy and paste the track in which you would like to apply dynamic range compression. Make sure the new track stays panned the same way as the original track. Otherwise, funky things might happen to the phase. After creating the new track, you now have two tracks of the same signal. You can compress the crap out of one of the tracks and still have one completely dry track. The two tracks can be blended together for the ideal level of compression. I keep the dry track higher in the mix than the compressed track. This way I do not lose any essential transients. I started off doing this for just vocals, but over time I have expanded it to other parts of the mix. It works great for guitar and on the overhead drum mics. Placing the compressor VST plugin into the effects bin also allows you to have a dry track and a compressed signal, which can then be blended together. However, I tend to only use this method if I have a bunch of tracks that need to be compressed in exactly the same way. This rarely happens. I need more control. However, with this increased control also comes increased PCU usage. An adequate computer setup with a crap-load of RAM is required.