Sends vs Inserts For Effects

LinearLineman

Send effects are used when you want to add something to a signal, but the signal itself is not modified. To achieve this, every channel in a mixer has "Send" or "Aux" knobs that take a portion of the channel's signal and send it (hence the name) to the send effect (which then returns the "addition", for exmaple, reverberation -- on a special "Return" channel (this is often invisible in virtual mixers such as in apps)). Examples where this is useful would be reverb and delay (echo). The big advantage compared to insert effects for this is that you can use a single reverb or delay for as many instruments as you like, because the signals themselves are never modified, only something is added. So if you have 10 instruments that you would like to have reverb, you only need a single reverb plugin, thus save 90% CPU. By varying the amount of signal you "tap" off each channel into the effect (using the send knob), you can vary the amount of reverb (or echo, etc.) each channel gets, all through a single effect plugin.

In contrast, Insert effects are used when a signal is to be changed completely, i.e. none of the original signal is left after the effect. Examples would be distortion, bitcrushing, pitch-shifting and the like. This cannot easily be achieved using sends because even if you mute the original channel signal (by using a pre-fader send, for example), all processed signals will still end up on a single channel (the send effect's return channel) and thus the ability to separately mix and further process them will be severely limited.

May sound a bit daunting but this is really the easiest way to describe it while stating all important facts

[Deleted User]

The use of a bus can be an effective way to manage these issues.

[Deleted User]

The user and all related content has been deleted.

[Deleted User]

@Max23 said:

@InfoCheck said:
The use of a bus can be an effective way to manage these issues.

not really
if you have reverb on the aux you can apply different amounts of reverb to each instrument individually (!)
if you have some instruments on a buss everything gets the same glue (that's kind of the point of a bus)

This is a reasonable and appropriate explanation yet represents just one usage case for a bus. Another example could be sound sources you want to mix together and route to the same bus where you can apply the same effects to them and then route to the main output or even another bus. In the digital world, audio streams can be used to create copies which can then be transformed with various effects to create variations which can be combined in various ways outside of traditional mixing workflow considerations.

espiegel123

Sends are useful in other situations, too.

They can be convenient for diverting a copy (or portion) of a signal you are recording to another pathway so that you can, for instance, hear reverb or delay or distortion on the instrument you are recording without actually recording the effect. A lot of singers like to hear reverb on their vocals when they record. You can send some of the signal to a separate reverb channel that isn't record. That lets you record dry vocals while the singer gets to hear reverb when they record. Or you might want to hear a spacey delay on your guitar when you record, but you don't want to print it to 'tape' because you aren't sure that the delay settings will sit right once you have recorded everything else.

p.s. @LinearLineman : I would suggest a subject rewording: Sends vs Inserts for Effects. The current wording can confuse people that don't know about sends/returns/inserts. They might think that some effects are send effects and others insert effects, but the issue is signal routing and not the effects themselves.

LinearLineman

@espiegel123 , Done.