Is not using return tracks bad practice.

[amigo]

I find that I very rarely use return tracks and due to modern computer capabilities I feel I don't have to.
I might do it just for the sake of doing it if I find that I am using the same effect on multiple tracks but this does not happen very often. Every effect is tweaked specifically for each track/instrument/sound.

Is this bad practice and should I be using them for my all of my effects. Am I really missing out on anything?
Are they just an artefact from a time when using outboard mixers and effects units meant that you simply had to use them if you were to get the most out of that single reverb/delay unit that you spent a small fortune on?

I'm interested in hearing what you think.

[NRJay]

The most common practice is to use a return for reverb, everything you put through the same reverb sounds like it's in the same virtual space.

[oddstep]

Yes - you could see them as an anachronism , and yet.....
I use them a lot for the ability to send a signal back on itself, the exposure of a global dry-wet control on the mixer view and the ability to use them as a bus that brings multiple tracks together.... which can be handy if dynamics are involved. I think you're missing out - Live's routing makes them really flexible tools within that whole 'studio as instrument' paradigm ... particularly in a performance context.

[amigo]

That's interesting oddstep. Perhaps my view on return tracks is very one dimensional and needs further investigation.

[pgmjsd]

Yes - you could see them as an anachronism , and yet.....

Anachronism? Yeah I suppose you could say that. However, all the analog mixers I have used supported both insert effects and aux send/returns. The big difference I see is that now you can make copies of your effects, which we couldn't do easily in the old days. It *is* really nice to be able to give each snare it's own special reverb unit.

The most common practice is to use a return for reverb, everything you put through the same reverb sounds like it's in the same virtual space

+1 Glues the mix together, IMO.

I also usually have an echo on a return, and that's sent to the reverb to soften it up a bit. I give drum racks a return that passes through to the main 'verb so I can give a little space to individual sounds (although often kick & snare may have their own).

Recently I started using returns for sidechain compression. This works great, IMO. The SC return is not routed to the master, and it's a submix of kicks and a little snare maybe. Then make compressors on bass tracks or groups, pad groups.

[Vios]

I have made tracks where I only had one reverb return, but that's not typical. Most tracks I have three reverbs - typically ableton's built in reverb and omniverb in return tracks. They each sound better depending on the instrument, but together mesh really well. Anything I send to one I send to the other at at least -10dB from the other level. The third reverb I place in the return built into my drum rack, with a short delay time. Sometimes I need a fourth in the drum rack for a loud dubsteppy reverb snare. I'll keep my kick seperate from the rest of my drums and still send a little bit from my drums to my two return reverbs, just to keep everything glued together. Typically everything but the kick gets some reverb applied, even if it's -40 or -50dB on the send.

I typically place delays in returns as well, even if I'm only using them for one synth. I find adjusting the return send easier to get the exact amount I want versus using the wet/dry built in to the delay/effect. The problem with the wet/dry adjustment is it tends to be louder closer to the 50% mark, which makes it easy to end up with too much delay/effect.

[JuanSOLO]

I use returns for a few things.

Post Fader Delay (dub style)
Post Fader Reverb
different kinds of reverbs (i.e. realistic room verb, long tail atmosphere verbs)
I also use Sends to toggle between a "Clean" output vs. a MashUp effect output. Each track has the ability to toggle sending to a Mashup device or Clean, I really only want 1 instance of that Mashup device in my set, not one on every track.
I also put a Looper device on a Return, I can send ANY track to 1 Live looping device.
I am also using returns to consolidate my set and CPU consumption. I know you posted something about computer in the modern age, but even with the latest MBP, I have got to stay CPU conscious of my Live set or my system performs sloppily.

[Samaritan Sound]

I use returns for verb, delay lines, and my Phat bus.

[oddstep]

I use them to hold a saturator, resonator, reverb chain. I can then just turn a dial on a track and generate a massive drone that bears a slight resemblance to what was previously playing. I also use them to hold loopers (with the send dials being used as faders for the mix of sounds on the looper). You're messing around if you don't spend some time making infinite filtered dub delays with them.

[mholloway]

they are vital because they offer Parallel FX Processing whereas inserts provide Serial FX Processing. The two processes are very different and lead to very different results; using a delay as insert, for example, is very different than a delay as send -- in the latter case you preserve all the original material (if wanted) while adding delayed signal in the return channel. In the former (insert) you lose the original signal as you increase the amount of audible effected material. That's the key difference between the two FX routing approaches. Just try this on a drum channel, insert vs. send, and you'll see how huge the difference is. In the insert, your drums disappear into the effect as you increase the effect amount. With the send, you preserve the original drums but add delay of them into the mix.

-M

[oddstep]

Its true that sends are a good way to do parallel processing. You could make a rack that did this though.