|
Ok, let me try to explain a bit. The terminology is taken from hardware mixers, but still retained in the digital era, so you have common applications for both.
You should really see the signal path in two ways.
A serial path : everything is sequential, one behind another. If you insert (hint hint) something in that serial path, everything behind it will be affected. To make it easier to understand, imagine a mixer channel. You plug in a source (mic, synth, whatever) and it goes through a gain stage, EQ, maybe dynamics, pan, and finally the fader. Everything follows one and another, you enter above, you exit "on the bottom".
A parallel path (usually more than one) : unlike the serial path, you can take the signal, and route it to another part of the mixer, without actually interrupting the original "stream". A bit like splitting the signal and take it to different parts. So you have different parallel signal paths, and such a path is called a buss. Mixers have different busses (so paths you can route the signal to) : the master buss, the solo bus (prelistening), aux busses, maybe group busses. The higher grade the mixer, the more options you'll have.
Ok, those basic principles need to be understood to be able to explain sends/returns and inserts.
First the insert. I gave you a hint already. An insert is a "breakpoint" in the serial path. For example, when you use a compressor, you put it on an insert. What does it do? It takes the full signal at the insert send, that goes to the compressor (that's why it is called insert SEND), and the processed signal is sent back to the following stages of the serial path (insert RETURN). As a matter of fact, with most mixers, if you use the insert send, but forget to return the processed signal, you won't have any sound on that channel (proving it's indeed serial, you interrupted the signal path, and forgot to close it again). For example, using a compressor on the master insert send/return (and master goes to amp for example), would be the same as master output -> compressor input, compressor output -> amp. Except one difference, the insert point is usually before the meters, so you see the effect of the compressor on the meters when you use the insert.
So you must be careful, because send/return is used both with inserts and auxes (or whatever other parallel bussing).
With aux sends it's different, because they are parallel busses. You usually have a pot called aux send (or more than one, each for an apart aux buss). That pot controls how much signal you take and send to the corresponding aux buss (on better mixers, the aux busses also have a separate master control each, called aux master sends. You can compare it with the normal master fader of the master buss. Useful if you already made a good balance on the auxes, but need to reduce the overal level). The output of that aux buss can for example go to a reverb. You can have pre or post fader aux sends. Prefader means the level you set with the aux send pot is independent of the position of the corresponding channel fader. This is useful for example for monitor mixes for live PA. The musician won't be happy if he hears his instrument disappear in his monitor when you decreased the fader... Usually for music production (except creative tricks like always) you'll use post. For example, you want the reverb level to follow your fader level (if the source gets quieter, the reverb gets quieter too, for example).
Now for the aux returns it's a bit of a mixed bag. The aux return is nothing more than an extra input. Basically (to keep it logical) you get the process signal back from the device that received it's signal from the corresponding aux send (for example, you connected a reverb A on aux 1, the processed (called "wet") signal will usually be returned on aux return 1, to keep it logical.
BUT... like I said, usually the aux returns are only an input with a volume control. That's why most of the time, people will prefer to return the wet signal to a normal channel, because you get more control over it then (compare, a little volume pot vs a big fader, full EQ, etc...).
Understood? Ok, but when do you use inserts or auxes?
To keep it simple and standard (99% of the time it follows that rule)
- all frequency and dynamics related processor go through inserts : compressors, limiters, gates, expanders, eq's, aural exciters, filters, distortions,...). With those processors you want to fundamentally change the signal that is coming in, and those changes need to be kept through the rest of the chain. (on a side not, parallel compression can be used, but that's really for very advanced people. Keep it simple).
- all temporal effects go through auxes : reverb, delay, echo, flanger, chorus, etc... Basically you want to add something extra to the existing signal, not change the original signal.
Why really? I know that most reverb (etc.) units have a dry/wet setting, but they put that in for people that want to do the creative tricks, or that have such crap equipment they don't even have auxes. Normally you will use auxes, so the mix should be set to 100% wet (100% effect). Imagine, if you have 24 tracks, and you want to set them all in the same room, you'd need 24 reverb units (maybe doable on a computer, but hardware, you don't shit money do you). No, with aux sends, you maybe take one or two reverbs, with two different settings (a usual setup is one longer and one shorter reverb), connected to two auxes.
Let's for example say, the longer reverb is on aux 1, the shorter one on aux 2. For all the sources you want to place in the larger room, you just open their corresponding aux 1 pots, for all the sources you want to have the short reverb on, open up the aux 2 pots on their respective channels. You see you can make a mix of what you send to the effects. How you return the wet signal is up to you (back to a channel or via aux returns. With some softs, like Cubase SX1, a send usually had it's own return integrated, but you can still trick it to be returned to "normal" channels by using groups, if you wanna know more, just ask, in SX2 that changed with the arrival of the FX channels, that's basically the same as returning the wet signal to a normal channel).
This goes for all the other temporal effects to of course. You see you can use a couple of effects to drive a huge amount of sources.
Voila, that's basically it. If you have more questions, feel free to ask.
|
