Why do people put delays and reverbs on a separate track instead of putting it with the track that has the EQs and compressors? Do you have to do this?
SOS Forum post
SOS contributor Mike Senior replies: There are several reasons why it often makes sense to keep delay and reverb effects on separate tracks. The main thing is that it allows you to share the same plug-in effect instance between several recorded sounds, by sending from their different tracks to a separate effects track via the DAW software’s auxiliary sends system — the so-called ‘send-return’ configuration. This not only makes more efficient use of your computer’s processing power, especially where reverb is concerned (decent reverb effects can be extraordinarily CPU-hungry), but also allows you to implement global changes to an effect across the board with ease. Want a drier sound overall? Just turn down the reverb channel, leaving all your instrument balances otherwise as they were — assuming that you’re using the effect in its ‘wet only’ mode, as is the norm when using separate effects channels in this way.
In addition, though, having your delay or reverb effect on a separate track gives you much more control to customise the effect without changing the dry sound. Most reverb plug-ins do now have some EQ at least on-board, but if you want more surgical spectral control — or indeed some distortion, chorusing, or triggered ducking applied just to the effect tail — then it’s much easier to do that when the effect is on a separate track. In more creative electronic styles, this is a big plus point of the send-return approach.
Having said that, there are some occasions where I still insert delay or reverb effects directly into a track, rather than using them in a send-return configuration. Usually this is where I’m using super-short delays or reverbs to adjust the tone of a sound via comb-filtering, rather than trying to set up an effect tail. In that kind of situation, the effect usually has to be heavily tailored to the specific recorded source I’m processing, so it wouldn’t be much use for anything else in the arrangement, and I would want any further processing (or indeed sends to more traditional send-return effects) to be fed with that altered timbre, not the initial dry sound. So in that scenario the send-return configuration makes less sense.