routing and subgrouping

All the way back when I was just starting to mix, I used to do this really simple mistake.

It still makes me cringe to this day.

I sent all my tracks to an aux channel, added some EQ and compression and then wondered why my filters weren’t clearing up the low-end and my compressors weren’t taming my dynamics?

Because sending isn’t the same as routing. -Click to Tweet!

Many engineers don’t realize that by sending tracks you are still left with the original track.

Now you have the original on the same track, but you also have a copy playing in your aux track. What you do to the copy doesn’t affect the original at all!

You just end up with a louder signal.

If you want to send the original somewhere else, to an aux bus like I mentioned before for example, you have to route the outputs of the original to an auxiliary bus.

Don’t send a signal to a bus if you want to change it completely.

You will only change the copy. 

You will only be left with more of the same, and any processing you do to the sent track will not affect the original.

Learn from my mistakes. Route your tracks if you want to change a group of instruments, but use sends if you just want to add parallel processing alongside the original.

But there’s more to it than that.

Gain Staging With Sends

I try to mix pretty quietly. We’ve discussed the importance of headroom before and I practice what I preach. I let the volume knob take care of the volume, not the faders themselves.

And by leaving the faders pretty low it gives the master fader some room to breathe.

But it’s different with subgroups.

When your route your tracks to a subgroup make sure that the tracks are loud enough. Because if all your tracks are coming into the sub-bus super quiet, you’re forced to push the aux bus really hard to get some volume.

You’ll often increase the aux bus above unity gain, which is completely unnecessary if you just follow simple gain staging.

Don’t Do This:

subgroups and routing

Keep the actual channels loud, while the group bus should be quieter. It will give you more control. Of course you should mix the volume of the individual tracks themselves so that they blend together. But make the loudest one the one at 0.0 dB instead of mixing them together at such a level that makes you max out the aux bus.

Instead, Do This:

subgroups and routing

Now you have the same mix of backing vocals but you’ll have an easier time using the group fader. Grouping and routing is all about creating an easier mixing experience.

Don’t be counter-productive and make it harder on you or your DAW.

Using Reverb and other Time-Based Effects as Inserts

There’s a similar mistake made with sends and inserts. Especially when it comes to using reverb and other time-based effects.

Reverb can be pretty confusing sometimes. Many beginners don’t know where to put the reverb when they’re mixing, and they usually end up inserting it after their compressor or EQ.

This is, almost always, regarded as a big no-no. 

But why?

To understand why we need to know the difference between a send and an insert.

What’s a send effect?

Send effects are used when you have an original signal that you want to modulate, modify or process in any way without affecting the original signal.

Your original signal sits there by itself unchanged but it has a separate copy being effected.  This is why we use it on reverb; the original audio is there as well as the delayed signal working together to create a nice effect.

What’s an Insert?

An insert is used when you want to process a signal directly. You insert a processor, such as a compressor or an equalizer directly onto the original audio. This is when you do not want to retain the original audio file. You want to change it completely(or subtly) and only hear the processed part.

So when is it smarter to use sends or inserts?

Well, you want to use sends when you are using time based effects. Time based effects, such as delays need an original sound source to work.

How would you know a signal was delayed if it didn’t have an original un-delayed track. You would only hear the delay, which is weird in the first place because it would just sound like the musician was consistently playing behind the beat.

Inserts are used for processing directly. You use inserts when you need to compress a signal, gate it to eliminate bleed or process it in any way that doesn’t affect the time of the signal.

Compressors are not time based, they are dynamic processors that work directly on a signal. That’s why you should usually use them as inserts. If you were to use a compressor as a send you wouldn’t be compressing the original signal, you would have a copy that’s compressed but it wouldn’t matter because your original is still unchanged.

Reverb is a time based processor, meaning that it affects the source by delaying it in time and creating a space around it. As you might know, reverb is basically just a combination of echoes.

And echoes are just direct delays of the original signal.

Since reverb is time based, we use auxiliary sends to send our original track to a separate reverb where we process it individually.

That way we have both the original and reverb’d signal to work with. So next time you want to add reverb to a signal, use sends NOT inserts.

Breaking the “Rules”

Of course, there are instances where you can break the rules. Using compression as a send is a well known trick called parallel compression. Sending your drums to a send where you compress them separately will give you a punchier sound when you blend the compressed drums underneath the natural sound from the original tracks.

Alternatively, time based effects have been used since the dawn of the guitar stomp-box.

Guitar players that use reverbs and delays as a part of their stomp-box chain are using time-based effects as an insert. But all of these stomp-boxes have a “mix” parameter to control the amount of effected signal that’s blended into their guitar tone.

You can do the same thing with your DAW. For instance, if you have a guitar track you can usually insert modulation effects as an insert. Just be sure that you’re blending the two signals. You don’t want an insert to be 100% wet(or effected) since you won’t have any of the original signal.

In conclusion

Inserts are great for completely modifying a track, but if you want to add a touch of reverb or delay on top of an already great sounding instruments, sends are the way to go.

Go check out this week’s video content for more information on sending/routing and the mix review with Mike Rizzo about his live recording and how he used just a few EQ and delay tricks to declutter his mix.