Multiband Compression FAQ

Feb 1, 2009
179
0
16
Think a lot of people (inc. myself) need help with the idea of why certain areas of frequencies need taming and when it needs to be used. Mainly how do you know when multiband compression is needed and how it should be approached :)
 
I always use multiband compressor after EQ. First you cut the problem area (say 150hz in this example) with the eq as much as possible without affecting the sound negatively, then compress the same area with the multiband compressor so that the problem is tamed. Always check the exact problem frequency with a frequency analyzer.
 
The concept might seem a lot more complicated than it actually is. The Andy Sneap C4 preset is a good example on when multiband compressing is useful. Basically you take one frequency range (in this case around 65Hz to 280Hz) and only compress when sounds in that range go over the threshold, getting rid of the nasty boom-like sound when palm muting. If you'd try and get rid of it with an EQ, the tone would be affected around those frequencies all the time, not just during the palm mutes.

Another good example is a de-esser, which basically is just a multiband compressor with one band active. You check where on the frequency range the nasty sibilance is (let's say it's around 8kHz in this example), and apply the compression there. If you'd cut 8kHz with an equalizer, it would cut the 8kHz range from the vocals during the whole performance, not just when the S's jump out.
 
Multi-band compression rocks on the low frequencies. It allows you to retain a solid low end punch that is consistent and tight. If your bass guitar only booms out on certain notes, you can get the comp to attenuate only those associated frequencies, rather than cutting them 100% of the time with an EQ. Great stuff. After a while of using them you start to get a knack on when they're useful to bring out and use.
 
Multi-band compression rocks on the low frequencies. It allows you to retain a solid low end punch that is consistent and tight. If your bass guitar only booms out on certain notes, you can get the comp to attenuate only those associated frequencies, rather than cutting them 100% of the time with an EQ. Great stuff. After a while of using them you start to get a knack on when they're useful to bring out and use.

Hear, hear. It's retarded that ProTools M-Powered/LE doesn't come with a multiband comp. Gladly the sidechain on the single band is usable stuff.
 
I think you've got it explained to you in this thread. There really isn't much more to multiband compression than what has been said... atleast not that I know of :) Simply said, it's used to get rid of the occasional irritating frequency in a particular sound.

A simple example would be like, what do you do when you have a constant hum on let's say 60 hz on a sweet violin track? You can simply go ahead and cut it out, the violin won't really be needing that low frequencies.

And the other side of the example: What do you do when a bass guitar has some occasional nasty spikes in the muddy 300 hz area, for some reason? You avoid cutting it out since it's just popping out occasionally. You'd rather want to tell your Mband plugin to "Hey, look dude, whenever those spikes pop out, tame them by lowering the volume on that frequency, but when it's over, make sure to bring it back up to normal, ok?".

Don't know if that was a great example but it's something :)