Ableton Forum

Hi.
This is a bit long, but its something thats been in the back of my mind a long time.


When mixing down your normal projects, im not talking about exerimental and sound design mixing with sends like Robert Henke does, but normal "bread and butter" song mixing with standard aux effects like reverb and delay.

Some of my friends use Sends Only at the output on their tracks, when they mix down their song. Using one send to bring the level back up, so the volume fader can be used as normal.


Exapmle.
Send A: reverb
Send B: delay
Send C: empty

Tracks Audio To: Sends Only.
Tracks Send C to max, brings back the volume.
Send A and B will be wet signal only, adding volume to the mix, while the dry signal can be brought down by Send C.

What is really the benefit of this in a normal mixing situation?


For decades, the usual way has been to simply add parallel effects with the sends. This is in fact rather limited. Say that you add delay with send A and reverb on send B on a snare track: you will get one hit with only reverb, and another hit with only delays. This creates two individual rooms; a parallel effect. It would be more natural to have the delay inside the reverb room.


Say if you want to have the sends like this:
Send A: overdrive
Send B: delay
Send C: Reverb

And say you want some tracks to use them serial: overdrive to delay to reverb, and others just to delay, or overdrive do reverb. What can be done as a good work template that is a bit more creative then the standard paralell mixing of sends?

Sometimes I end up using a lot of duplicate effects on the tracks to archive different serial signals. And thats ok. But it would be cool to find a good simple solution with the sends, that could cut trim down the template. It would remove about 75% of my effects load.


For now, I just use the Sends in normal parallel mode. Delays and reverbs.

My tracks could look like this:
Snare: distortion-delay-chorus-eq, and might Send it to: reverb
Tabla: distortion-delay-autofilter-eq, and Send: reverb

When using cpu and ram intencive vsts, and say 8 or 10 tracks like this, it feels a bit of a waste. Especially if I use, say 6 instances of Guitar Rig 4, all with the same preset.
It would be nice to have at least GR4 on Send A, and be able to send any track to it, while adding serial amounts of delay or reverb individually for each track.

If sends could have their own sends and so forth, it could be archived. Like a second and thrid layer of sends. Like a cluster effect:
A to: B and/or C, and B to: D and/or E, or C to: D and/or E.

maybe its possible to build it with a combination of tracks and sends.
Hmm.

If you right click on the sends in the return tracks you can enable them and do exactly what you want.

Just be careful with the feedback!

The Leveller wrote:If you right click on the sends in the return tracks you can enable them and do exactly what you want.

I know, but I still dont get it to work as I described. Because it will then affect all tracks equally.

Say if you want to send tracks in diferent serial chains, like:
1. 1st through overdrive, then 2nd delay, and 3rd reverb.
2. 1st overdrive, 2nd delay.
3. Only reverb.
4. Only delay.


On track A youd perhaps use effect chain 1, with settings like:
overdrive 75%, delay 30%, reverb 20%

While on track B youd like the same effect chain, but with different dry/wet settings:
overdrive 30%, delay 50%, reverb 20%

And id like to controll this send chain mixing from each tracks sends. Maybe its not possible..

edit: similare to NIs The Finger, the VST effect for Reaktor.

I think I get you.

But isn't it a matter of mixing up the treatment of traditional insert effects such as distortion with traditional send effects such as reverb?

I don't get the whole sends only mixing you describe. I'd rather just set up groups in ordinary tracks and then control the send % to return tracks the traditional way.

1. Overdribe insert at 75% on the track and a delay at the right wet/dry and % setting, then send all that to send A with a reverb on it adjust the send knob to taste.

2. Insert the OD and delay again at the right % and send 20% top send A for the reverb.

It seems overly complex to send to distortion tracks, I would rarely use the same distortion on more than one track Unless it's for a group of instruments, in which case, group them and slam it on there.

i don't see the advantage of what your friends are doing...or i still don't understand.

I have also recently been wondering the purpose of sending the Audio To Sends Only. What is the advantages of doing this as opposed to just using automating the send dial as and when you want that track to go to the send. I am playing around with it at the moment. Audio out to sends only which cuts off the sound until the dial is taken up. However I don't see what the point is.

The Leveller wrote:If you right click on the sends in the return tracks you can enable them and do exactly what you want.

Just be careful with the feedback!

This will indeed risk causing phasing issues as the send on the return adds a sample of processing. I think it might not if you use send only on the return track, but I haven't tested this.

I don't see the point of this anyway as the processing could all just be in the same return track. All you gain is a send. Or…

ze2be wrote: If sends could have their own sends and so forth, it could be archived. Like a second and thrid layer of sends. Like a cluster effect:
A to: B and/or C, and B to: D and/or E, or C to: D and/or E.

maybe its possible to build it with a combination of tracks and sends.

If you set one or several return tracks to Sends only and activate only the sends you need you'll be turning each return into a buss that you can send in different amounts to other returns, that we can call Aux tracks as well. Only activate sends you intend to use. Sends only should hopefully prevent phasing issues due to delayed processing, which is something to look out for when doing "advanced" signal routings.

The buss could do its own processing before sending the signal off (traditionally in many analogue boards a "buss" reroutes only). The receiving return track is then sent to master or alternatively to another audio track buss with sends turned off (again to prevent phasing issues). I think this might work, but needs to be tested.

Something like this with all sends not intended to be used disabled also in instrument tracks:

If you set one or several return tracks to Sends only and activate only the sends you need you'll be turning each return into a buss that you can send in different amounts to other returns, that we can call Aux tracks as well. Only activate sends you intend to use. Sends only should hopefully prevent phasing issues due to delayed processing, which is something to look out for when doing "advanced" signal routings.

The buss could do its own processing before sending the signal off (traditionally in many analogue boards a "buss" reroutes only). The receiving return track is then sent to master or alternatively to another audio track buss with sends turned off (again to prevent phasing issues). I think this might work, but needs to be tested.

Really like this idea and it seems to work really well. I wanted to create something like the new Waves CLA Epic where you can feed one effect into another and control the send and return amounts and adjust them in real time.
I setup different send channels that can be fed audio from the same track to either blend or try out different ideas by adjusting their faders.
==========
Amendment - i think this is over complicating things too much, the original post was more on track. I will be posting my thoughts and tests shortly

==========

Some of my friends use Sends Only at the output on their tracks, when they mix down their song. Using one send to bring the level back up, so the volume fader can be used as normal.
Exapmle.
Send A: reverb
Send B: delay
Send C: empty

Tracks Audio To: Sends Only.
Tracks Send C to max, brings back the volume.
Send A and B will be wet signal only, adding volume to the mix, while the dry signal can be brought down by Send C.

What is really the benefit of this in a normal mixing situation?

This idea can be modified slightly to give you very creative control of your sends.
You have an audio track colour blue called kick and you want to send some of that signal to reverb on send A - you would normally do this by changing the send amount.
Create a new audio track ( colour blue name Buss+EFX)change its input to the track "kick" and send from here instead.
I have set my return track E to have no effects on it - so it only carries the original signals sent to it.
At first this might seem pointless but it opens a lot of creative possibilities.

• You can put EQs or other processing on the Buss+EFX before you send to Reverb on A
  You can create multiple busses (all slightly different) and switch between them to hear which sounds best
  You can put a delay on the bus and send it only to the reverb - by not returning the delay signal on Send E
  All the effects are easily controlled on their own faders - for precise control.
  Creating and Deleting one of theses effect busses does not affect the original track ( its left alone)
  Drag and drop in saved 'Buss+EFX' channel strips ready to go - just assign the input track
  Take control of your effects rather than accepting the way they sound when just using sends!

Oh, nice! Good stuff. I didn't see these last replies before now! My original post was from 2011 and I haven't been hanging here on the forum the last few years. But thanks guys for the nice ideas!