Dubstepforum.com

Compression.

I have been trying to get my head around it. When I use compression on my drum parts ect I always get scared of 'over compressing' them and sucking the life out of the sound so I tend to find my self using the same compression settings for everything which I know isn't right. I play around then always get drawn back to that subtle compression of very low ratio, slow attack and fast release with threshold of about 12, just because it sounds like its hitting hard there. I know with these settings very little compression is going on at all. But when I 'compress the shit' out of the parts like everyone loves to talk about on here The sound looses that crispness and bite, to me it just sounds like a knocking piece of shit lol. I'm obviously doing it wrong?

I know there are probably loads of threads about this but I just wanted to make this specific to what I'm looking to learn. So any help will be much appreciated.

I know it depends on what sound your trying to get but what general compression settings would you use on these individual parts?

Kick:
Snare & Percs:
Sub & Bass sounds:
Synths:
ALSO WOULD YOU BOTHER TO COMPRESS THE FOLLOWING?
Hats & Cyms:
Pads & Atmos:

And if you find a really meaty sounding kick and your happy with how it punches you would you bother trying to compress it any further?

I'm new to 'getting' compression, finally. And the improvements it makes on sound are always incremental. They're the icing on a cake.

If it's still too hard, just try tansient shaping, clever enveloping, or if you wanna add snap, layer the attack of a snappy sound (a bang, gunshot, etc) into the sound.

The settings you would use all depend on the effect. Want a snappy or full kick?

And before you do ANY sort of compression, first just ask yourself.. does the sound need it? In most cases it doesn't. I mostly put it on drum buses for slight movement in the overal sound or when I'm layering sounds. Only very rarely for taming peaks. If you want a snappy snare for example, pick a snappy sample. If that doesn't work, only then try to compress another snare.

Anyone who knows shit all about compression will tell you that overcompressing is worse than undercompressing. Don't sweat it if you don't need it. There's no guideline whatsoever, it's all down to your source sounds and what they need. If you're using drums from a commercial sample pack or something, they're already compressed to high hell. You may find some buss compression is still useful to glue things together.

I probably wouldn't compress a pad that much unless it had some phasey chorus type shit that's causing too much dynamic swells. I'll definitely compress cymbals though. I use live sounding drums though so I'm compressing overheads and room mics. Again, if you're using sample packs, the cymbals might not need it. You can still simulate a room with a reverb send and then compress that though.

Just keep experimenting and try to get the dynamics right at the source. It's foolish to try and compress a synth if you can just adjust an amp envelope on it or something. Use velocity in all your midi programming, especially drums.

like you, 99% of the time, I am compressing lightly with a ratio of 1.5 - 2.0 and threshold at -10 - -15. The vast majority of my compression is on groups of sounds, not a single sound. but sometimes I will put compression on a single sound, like a bassline or whatever, if there are high frequencies in it that are way too loud or something like that.

Read about compression in terms of recording musicians playing instruments. It'll make more sense.

Welcome to Dusbtep forum

This thread on compression may be useful for you, particularly page 2 (look half way down)? http://www.dubstepforum.com/viewtopic.php?f=8&t=151890

Also check out the huge guide to producing dubstep in case you missed it? ---> http://www.dubstepforum.com/dubstep-pro ... 59713.html

Never compress pads and atmos because compression fucks up their texture.

DOLGAP wrote:Never compress pads and atmos because compression fucks up their texture.

Not that bright of a thing to say.

If you can't hear what it does don't do it.

Deleted.

u need to do it 6 times

AxeD wrote:If you can't hear what it does don't do it.

Yes! And not just "it makes it better cuz its moar louder" either. Don't rely on compression just to make an instrument be heard in a mix. Start with eq and gain staging..

You know how some waveforms are nice and relatively smooth, and other waveforms are really hairy and look like funky strings of melted beads with bits of lint stuck to them? Typically it's the latter that needs compression to 'play nice' in a mix and make sure those hairy linty spiky bits (or the dropouts between the metaphorical beads) don't muddy up or otherwise mess with the same frequencies on other neighboring tracks.

However something else many people aren't aware of is the natural compression that comes from various gear, software, and playback media. Like, tape compression is a desirable thing for many people as an inevitable byproduct of the process of recording to tape. Mp3's compress by merit of their encoding algorithm. Our soundcards often have an innate degree of compression happening as a byproduct of the sheer physics of trying to cram 5.1 sound through some tiny ass wires where the circuitry itself acts as a compressor/limiter/bandpass...

So yeah: if you can learn to mix nice sounding tunes without ever having to touch a compressor, people will actually be jealous of you - because there's a whole school of pro that made classic, amazing sounding music without ever touching a compressor.

Takes a dinkload of automation and attention to dynamics. But the results from doing it by hand usually beat the piss out of the equivalent result from a compressor.

Ideally you'd have a nice balance of both. Electronic music typically likes to be a bit more compressed but the synths and samples in use usually don't need much compression compared to recorded instruments. It's nice to use compression to shape percussion and to shave little off here and there to shape the overall character of the track, but I don't like it as a crutch for poor mixing. You can't go back and edit a performance's velocity when it's a recorded instrument. You can with midi instruments and you also don't have to worry about clipping your preamp on the way in because there is no "way in". Take advantage of producing in the box, put in the work to shape your dynamics the best you can before they even hit the mixer channels, and keep compression as a creative tool.

alphacat wrote:

AxeD wrote:If you can't hear what it does don't do it.

Yes! And not just "it makes it better cuz its moar louder" either. Don't rely on compression just to make an instrument be heard in a mix. Start with eq and gain staging..

You know how some waveforms are nice and relatively smooth, and other waveforms are really hairy and look like funky strings of melted beads with bits of lint stuck to them? Typically it's the latter that needs compression to 'play nice' in a mix and make sure those hairy linty spiky bits (or the dropouts between the metaphorical beads) don't muddy up or otherwise mess with the same frequencies on other neighboring tracks.

However something else many people aren't aware of is the natural compression that comes from various gear, software, and playback media. Like, tape compression is a desirable thing for many people as an inevitable byproduct of the process of recording to tape. Mp3's compress by merit of their encoding algorithm. Our soundcards often have an innate degree of compression happening as a byproduct of the sheer physics of trying to cram 5.1 sound through some tiny ass wires where the circuitry itself acts as a compressor/limiter/bandpass...

So yeah: if you can learn to mix nice sounding tunes without ever having to touch a compressor, people will actually be jealous of you - because there's a whole school of pro that made classic, amazing sounding music without ever touching a compressor.

very informative way of looking at it. thanks.