|
If you look at dynamics processors, like compressors and gates, they have something called the sidechain. Unlike some people think, every dynamic processor has one... But on better models (either software or hardware), this sidechain is accessible by the user (the sidechain input, sometimes called the key input).
What is the sidechain? It's the circuit that "triggers" the processor. In hardware it can be based on a VCA, an optocoupler, etc...
Let me give you an example, with a hardware compressor, and I think it'll become quite clear.
In normal use, let's say you want to compress a kick. You present the kick to the input of the compressor (usually via an insert). What happens is : the kick signal runs through the sidechain, which detects the level. According to that detected level and your settings (threshold, ratio, ...), an amplifier will be controlled that applies gain reduction on your signal. In this example (the normal use) the kick signal triggers the compressor, to compress the kick signal.
Now, when you have access to the sidechain (there is a sidechain in/key in, on your compressor) you can use this to compress a signal, but let the compressor be triggered by another signal than the one you are actually compressing.
One of the most common uses for that is bassducking (sometimes called kick over bass). In this example, you want to compress the bassline. So you present the bass signal to the compressor input. Yet, you want the kick to trigger the compressor. So you present the kick signal to the sidechain. What happens is, the detection circuit will react according to the kick signal. It then drives the amplifier that applies the gain reduction, but to your bass. In practice this will mean that your bass will go down each time the kick plays. Very good to make get the bass louder without losing the kick punch, or to get a pumping effect. That's basically one of the uses of sidechaining.
Other uses :
- a de-esser : let's say you want to remove sibilance from a vocal. That can be done with a compressor where you have access to the sidechain. Just split the vocal. The "untreated" vocal goes into the compressor. The second instance of the vocal will be heavily EQ'ed. You will boost the "S' es" a lot, and remove most of the bass. This EQ'ed vocal will be presented to the sidechain. What happens is, the detection circuit will be mostly triggered by S'es (as it merely looks for levels). In practice it means the compressor will apply gain reduction on the "untreated" vocal on every S that goes over the threshold. In specialised de-essers, all this is done for you, but you can perfectly make one yourself with a decent compressor. It's better than EQ'ing out the S'es, as that will result in a more muffled sound. The de-esser only works when a harsh S is present, and doesn't affect the sound any further.
- Voice over / talkover : mostly done with a gate in ducking mode. The vocal trigs the sidechain, to pull the music down when the narrator/dj is speaking.
- Rhytmic patterns : you can do this with a gate in gating mode, or in ducking mode. With a heavy compressor another kind of effect will be created. Say you have a synth playing a steady tone. And you have a rythmic loop somewhere. Put the loop into the sidechain, and adjust your gate settings so it opens on the rythmic hits. The synth goes into the gate input. In gate mode, what happens : each time the loop hits, the gate opens, so you'll hear the synth. In ducking mode, the opposite will happen. The synth will play, except when the loop hits. The effect is you're steady tone will be choppped up rythmically.
People who know Fruity will immediately think of the peak controller. Similar stuff, exactly.
Hope you understand now.
|
!!!
