Compression is perhaps the hardest thing to grasp in recording because there are multiple reasons for using it. Also, it takes a while to learn how much or little to use and what settings to use and why. I am still learning.
Here’s a few pointers:
It is wise to use a little compression via an outboard compressor when recording to digital as a precaution to make sure you do not overload the converters because digital distortion is the worst type of distortion there is. Granted, if you are really careful, it is still best to avoid any processing when recording a track but as long as you are prudent, this method is often accepted. And, all rules are meant to be broken if you have a deliberate reason. When recording thru a mic, I tend to use my Tube-Tech CL-1B compressor – certainly one of the best on the market. I set it to a ratio of 4:1, relatively fast attack, relatively moderate release, and a threshold that will only catch the hottest signal – in other words, I do not want to compress the whole performance – only the possible over loads or hot bits. Think of the compressor as a safety net to keep any distortion from happening to your signal going to disk.
Another reason to compress when recording is because you want to commit the compressed sound to disk. There are purists who disagree with this method but many pros agree. Here, you are intentionally compressing the sound. Reasons are a plenty but a good example is when you have a performance that contains a very wide dynamic range. You want to rein in the highs so that the dynamic range is not out of control. This is needed if you have a vocalist not used to recording and who does not possess good mic technique.
Parallel Compression is very common. It is great for tracks with wide dynamic ranges. Vocals, Bass, and Drums are frequent candidates. Your original track will contain the large transients that occur at the beginning of a sound – say the crack of the sticks on the drum head or the attack of the pick on the guitar strings. The parallel compression is used to raise the levels of the quieter parts of the sound. In essence, you hit the transients hard with the compressor by using lower thresholds, and faster attacks. This allows you to add gain to the compressed signal, thereby increasing the quieter passages. Here, you generally will be looking at a longer release time so that sustain of the quieter passages is maintained. The exception to the sustain bit would be on a drum kit where you want a really sharp and snappy drum sound. Here, you’d want short release to add the punch to the drums.
Another common use for compression is to smooth things out. Most vocals are compressed and often they are compressed more than once. If you really listen to mixes when listening to music, you will start to notice that most vocal tracks are compressed. While the idea is to make the compression as non noticeable as possible, most professional vocal tracks are very smooth – even between the loud and soft parts. This is probably one of the things that separates the men from the boys. The art of using compression in this manner is I think the hardest one to grasp and achieve. There is a happy medium here between to much and too little and finding that spot is an art. I am still chasing this one.
Sustain is another reason to compress things. Guitars – especially electrics often use compression to increase the sustain of notes so they hang on longer than they should. Here, is is more about the release. Mark Knopfler’s sound is a great example of this as it was an integral component of his sound. Oddly, before he came along, this was in my bag of tricks as a guitarist. When Sultans Of Swing hit the air waves, everyone was coming up to me to tell me of this great guitar player who sounded like me. Needless to say, when I met Mark back during the Brothers In Arms tour, I did not bring this up! 😉 In the beginning, Knopfler alsi used a lot of compression on the attack to squash the transients of the pick hitting the strings – which as I later learned, was really is fingers plucking the strings as he does not use a pick.
These are the most common uses of compression. There are others – say using a compressor as a ducker. This is where you use the signal of another instrument, say the snare, to invoke the compressor while compression acts on the signal of the instrument in the channel – not the snare that turns the conpression on.
Also, D-Essing of vocals. In essence, this is like ducking only you use a frequency to invoke the compressor instead of merely a snare or other instrument. Esses have their transients within a very small frequency range and almost every vocal you have ever heard since the creation of de-essing uses it. You run the vocal signal thru a side chain and set the frequency so that when sounds come in within the range of the targeted frequency, the compressor gets turned on and compresses only the part of the signal in the target range.
There are other tricks and uses but the ones I have mentioned are the most commonly used ones.
Classical music mostly steers way clear of compression. Jazz too. Everything else is pretty much compressed to some degree. Again, you will always have the purists who detest its use. The past 20 years or so has brought on a wave of things being too compressed in my opinion. It has a lot to do with the ability to make things sound louder on the radio. Things played on the radio have no dynamics so they sound loud – and probably so that quieter passages can be heard over the sounds of driving automobiles.