[Mr.Chainsaw]

And what does it do for me?

The washing up would be nice

Peter

[Ramshackles]

Wiki usually explains better than me:
http://en.wikipedia.org/wiki/Dynamic_range_compression

It's an effect like EQ, reverb etc. Along with EQ it's probably the most commonly used - pretty much every song you listen to will have had all kinds of compression used.

In short, a compressor reduces the volume of sounds that exceed a certain threshold (which is normally selectable by the user). So, it kind of evens out the volume of a track, which is very useful when dealing with things that have a large dynamic range (vocals for instance), which can in parts be buried 'in the mix' or in other parts be much louder than anything else.
When you compress something, by reducing it's peaks you generally reduce the overall perceived volume so you usually apply something called 'make-up gain' after compression. It's just a volume knob on the compressor. So in effect you are increasing the volume of quiet parts relative to loud parts.

Compressors come in all shapes and forms and can have a variety of controls. Some common controls:

Threshold : Volume above which the compressor kicks in
Ratio: Basically the amount of compression - how strongly the compressor reduces the volume of parts exceeding the threshold. Low ratio = low compression, etc. If the ratio is 'infinite' (in general, that is anything above 20:1), the compressor becomes a 'limiter' (confusing...) which is basically a compressor which doesnt allow ANYTHING to exceed the threshold limit - all parts that would normally exceed it are totally squashed back down under it.

Attack: Usually in ms, it controls how quickly the compressor responds to peaks in volume

Release: Intuitively, this controls how quickly the compressor 'relaxes' after responding to the peak in volume

Make-up gain: The volume you add after compression to compensate


You can get all kinds of different compressors (both in the digital plugin world and in hardware) and when you get nerdy, you notice that they all react differently and have their own unique sound or signature. Which is why you might hear about engineers sending tracks through a compressor that isnt doing anything, or using lots of different compressors etc etc.

There are also many different ways to use a compressor: on single tracks, on groups of tracks, parallel compression, multi-buss compression, multi-band compression, on effects returns etc etc....

Most DAW's come with compressor plugin which you can play with (if you are using a DAW - if not, reaper is free ), and hardware compressors range from £100 - £10,000

[Mr.Chainsaw]

So, say on a metal track, a compressor is stopping the screaming vocals from distorting by reducing the volume? Then on the quieter verses it pushes the volume back up again, si?

Why wouldn't you want the attack anything less than instantaneous?

Cheers Ram, you da tech man

Peter

[Ramshackles]

So, say on a metal track, a compressor is stopping the screaming vocals from distorting by reducing the volume? Then on the quieter verses it pushes the volume back up again, si?

Why wouldn't you want the attack anything less than instantaneous?

Cheers Ram, you da tech man

Peter

Yes, it can stop something from distorting but it depends where the distortion originates from. I.E if the distortion is due to overloading the mic or preamp then it is already in the track before it gets to the DAW/Tape/Whatever. But if you put a compressor before the preamp and use it when recording, then yes, it can prevent distorting.

Its not only a tool to prevent distortion (in fact, you would more likely use a limiter for that), but to even out the dynamics in a track. If you have a very dynamic instrument playing over the top of a not-so-dynamic instrument (e.g. bass guitar, electric guitar, most things played as rhythm parts), then you might need to compress the dynamic instrument so that it can always be clearly heard over the top of everything. Theres also a trade-off between how much dynamics you want in a track. You dont necessarily want it to have as much dynamics as when you play it, otherwise listeners will be constantly adjusting their volume. But you dont want to reduce dynamics so much that it causes listening fatigue.
Theres this thing of 'perceived dynamics'. If you whisper, then scream and then compress the track so that they have the same volume, you will still perceive the whisper as the quieter part as it has less power. But you wont need to strain to hear it (or get your ears blown out by the following scream).

But of course, you dont want to overdo it as it can sound weird.

You would fiddle around with the attack for different reasons. Sometimes if the attack is too strong, the sudden 'jump' down caused by the compressor very quickly squashing the peaks can sound weird. Other times, it might be because the actual instrument doesnt react that quickly. E.g, the attack on a drum hit is faster than the attack on a bass guitar etc etc, so you would set the attack of the compressor accordingly. Some of the most famous compressors dont even have the option to change the attack.

Generally, the part that I compress the most is bass. It's natural to expect the bass to be playing at a constant volume all the way through the song. This isn't always the case when you actually play the bass - it's hard to make the strength of your plucking/picking consistent. A compressor helps out with that. Compressors can drastically alter the sound of a track - make it more 'punchy', or go the other way and make it 'smooth' or 'thick' sounding

I bet some of the other techy guys here have something to say about compression; I'm far from understanding how to utilize it well and being able to hear all the little differences it makes...

[Mr.Chainsaw]

I play a picked acoustic, vocal and harmonica. Is there anything I can really gain by playing with compression? Other than increasing my know how, obviously

R.e. hardware compressors; if you record through it, and the take sounds (as you put it) abit weird, fo you have do a retake? Where as with reaper you can just unapply what you've done?

Ta

Peter

[Ramshackles]

I play a picked acoustic, vocal and harmonica. Is there anything I can really gain by playing with compression? Other than increasing my know how, obviously

R.e. hardware compressors; if you record through it, and the take sounds (as you put it) abit weird, fo you have do a retake? Where as with reaper you can just unapply what you've done?

Ta

Peter

Well you can play around with compression on all of those things and see how you like the results. What you do might change from song to song depending on the sound you are after.

For the second point, essentially, yes.