Originally Posted by
Gojirosan
Well...
There are no hard and fast rules (as ever!) and compression is a weird thing. You have to use your ears and often if it sounds like it's not doing anything, then it's working just fine!
Compression is one of those things that goes in and out of favour and varies from musical style to musical style. All you can do is imagine what you want your end result to sound like and progrress from there. Glossy and radio friendly with no surprise dynamics? Then plenty of compression and limiting. High fidelity, realistic and with a whopping dynamic range? Then little or no compression.
Compressors really have to be learnt, and this involves a combination of study and farting about! I have gotten to the stage where I have an idea in my head about the compression I want and can dial it in and require little fiddling to achieve what I want. But it has taken years of using hardware and software compression and falling in and out of love with it! Using and analysing presets in compression plug-ins is a great help in tuning your ear, as it removes "blank page blindness" of starting from scratch.
The main controls on compressors can be (and almost always are!) extremely confusing. Threshold is the easist to understand and the hardest to use! But it sets a level below which the signal is unaffected and above which the compression kicks in. Finding the right threshold spot can make or break transparency in my experience and modern computer recording has helped a lot with its level meters and visual waveforms. But sometimes it pays to turn off the monitor and just listen really, really intently to the signal first. Threshold obviously affects how much compression gets onto your sounds and effects the sound itself, so experimentation is the way!
Ratio being the amount of compression actually applied when the compression comes on is self-explanatory, but can be tricky to apply. If you want an audible or pumping compression for an effect then you will over apply this control! But mostly you just have to aply the least you can get away with. A loose (very loose!) guideline is:
Drums - 4:1
Bass - 6:1
Guitars - 6:1
Vocals - 2:1
Limiting - infinity:1
Now, these are but starting points, and you will often find that you end up using ratios a fair bit above or below these guides! I will usually use a higher ratio on vocals and bass as I like these instruments pretty damned compressed! However, if I use any individual compression on electric guitar at all, chances are that I will go for a lower ratio! My personal tastes and always thought out with a mind towards the final mix and how these things are going to fit together. Also drums can be treated as a kit and given a general compression, but I'll usually compress each drum and cymbal individually (or not at all) first. I dislike most natural kick drum sounds, so will almost always compress the Hell out of a kick until it sounds like an 808!
Limiting is strictly an infinity:1 process, but you may well get the control you want over a vocal or bass track with just a high compression rating (say 8:1 through to 14:1). Controlling peaks is difficult, but you don't always have to go at it with a sledgehammer!
Attack and release seem comparatively subtle to the overall sound compared to threshold and ratio, but need care anyway. When do you want the compressor to come in? How rapidly does it need to recover - how close are the sounds, do you want it to audibly pump or do you want natural recovery? Think about the nature of the signal and the type of instrument. Then try to reflect the envelope of the sound with complimentary attack and release settings. Again, studying presets really helps here.
Samples and loops come in all kinds of forms some pretreated and effected others naked and dry. It usually says in the notes of commercial material, but you can usually tell anyway. The best way may be to leave them until you have a bit of a mix together. It's far easier to hear if something needs compression when it is fighting for its place than in isolation having been recorded by skilled people with top-end transparent kit!
Guitars are funny. The guitarist may well have a compressor pedal in his chain to begin with, any overdrive, distortion or fuzz adds compression and the low tech valve circuits used to amplify most guitars are also introducing a dollop of compression. You may well need to apply compression to acoustic guitars before they hit "tape". Think of acoustic guitars more like vocals than electric guitars in this case! They cover a relatively huge frequency range, through out peaks, and drift up and down in level. Electric guitars are less like this, and you can usually recod them and forget compression until you can hear the mix. Then come the nightmare of geting mid-hogging electric guitars to fit! :lol: It is personal again. I tend not to apply much compression to miked-up-amped or amp-modelled electrics, unless it really cries out for it or it is a sparkling clean Nile Rodgers-esque type sound that needs to be level and smooth. Then they get smothered! Remember, however, that I am not a fan of very prominent electric guitars in mixes. I prefer guitars to sit in the background a bit. For instrumental guitar soloing type stuff, you may well find you need much more compression than I'd ever use.
The best way to work out compression amounts is with the mix, as I've said, but this isn't always possible. You may need to apply compression earlier on in the recording process. Which leads us to my fave compression maxim: It is better to apply several stages of mild compression than one application of heavy compression. This leaves many more options open. Consider the process of recording and where compression may be applied, with, say a vocal line:
Singer sings into the microphone.
The signal will usually be uneven and may have dangerous peaks, so you add compression to tame it before "tape".
Recording continues and the vocal is getting swamped and muddy in the mix: another bit of compression.
Once the backing vocals and vocals effects are added the lead vocal(s) are getting lost and confused again. More compression.
The final mix is compressed, thus more compression.
The mastering runs the entire track through a limiter - more compression (of a sort!).
That is just an illustration from the top of my head: five stages of compression on the lead vocals alone! You may find yourself using fewer stages, or even more! It you had applied too much compression inthe first couple of stages you could ruin the entire vocal line. So you have to constantly think and listen and, I'm afraid, think ahead! Multi-band compressors can be a Godsend here, as application of compression to the specific frequency bands that need it can stop too much of it mounting up and knackering the sound.
Right. I am in danger of rambling now, so I shall leave it there. I hope it's been of some value!
ops:
Posting Permissions
Reply With Quote