Published Work

This article was written originally for the blog at Carvin Amps and Audio. I repost it here, and encourage you to check out Carvin’s amazing line of products!

By now it’s an age-old question: should we track the band together as if we were playing live or should we try to get the cleanest signals and performances possible by tracking separately?

The truth is there’s no right or wrong answer. How you go about recording depends not only on you –  your chemistry, skill, and tightness – it also depends on the studio. Is the space tight enough that bleed will be a major problem? Will you be able to properly hear each other? Is there a drum booth? Do you need eye contact can you get it?

Your genre is a big determiner here too. Jazz ensembles and jam bands who improv a lot and feed off each other may need to track together regardless of the tradeoffs. Pop groups who need to sound as clean as possible may want to opt for fully separated tracking.

The good news is, it’s really an either-or proposition. In fact, in reality most groups use some sort of hybrid, and most projects exist somewhere on a spectrum between fully together and fully separated tracking. Here we’ll go over just a few common ways to answer this question.

Authentic Live Capture

The most extreme end of the tracking together spectrum. An authentic live capture seeks to track the band exactly as it is on stage with no overdubs. The advantage here are authenticity – you’re really capturing the group as it is on stage – and efficiency. A well-rehearsed group may even be able to knock out 10 songs in the time it takes to play 10 songs. Potential disadvantages include instrument bleed, performance errors, and edit issues.

Some ways to mitigate these issues include separating drums and amps into booths, using headphones to monitor, and playing to a click or reference track. Playing to a consistent tempo reference can be crucial for editing takes together or making fixes later. Isolating amps and other loud sources to minimize or eliminate bleed gives you the opportunity to overdub later – even if your goal is not to. Finally, rehearsal is crucial if you track this way. Tiny mistakes and inconsistencies come through loud and clear on a record, so if you’re recording this way, you need to be tight.

The Basic Band With Overdubs Method

With this method, the intention in initial tracking sessions is to play together as a band, but the focus in the control room is capturing the main foundation of the band. This usually means drums, bass, guitar, and piano if it’s in the band. In this tracking session, the vocalist is singing, everyone is doing their part, but the main focus is to get the foundation right. You don’t go back on a bad lyric or a flubbed note from the singer, because the singer will come back later to overdub the “real” vocal take. Backup singers may or may not participate in initial tracking, and extra fillers like bells or extra percussion wait until later to overdub. This method may be the most common method for bands, as it captures the chemistry of the band while still getting clean vocals.  Extra leads and solos may be overdubbed too.

Guidelines for mitigating bleed still apply here, as does the advice about playing to a click or other fixed tempo source.

The Rhythm Section First Method

In this method, start with the drums, bass, rhythm guitar, and piano if you have it. The focus is entirely on putting together a solid rhythm section, without the “distraction” of vocals and leads and what not. This method may be good if the vocalist also plays guitar, piano, or bass and doesn’t always feel comfortable doing both at once. It’s even more useful if the vocalist is the drummer, because capturing a clean vocal take in the midst of a drum kit can be pretty hard.

Play to a click and isolate as well as possible without messing up chemistry.

The Hip-Hop, Pop and Everything Else Method

If the group is not a band, the whole question is nearly moot. Beat makers make beats which are always locked to tempo, performers perform over the top of those beats, and there’s no question about separation, unless there are multiple vocalists.

If there are multiple vocalists, you’ll have to decide whether they should sing or rap together and if so whether they should be in the same room, share a mic, be in booths, etc. It would be easy to decide to give everyone their own mic, but there are plenty of times you might choose to let vocalists share. For example, a really skilled 4-part harmony might be better captured with a great omnidirectional condenser than with isolated microphones. “Gang vox” where a room full of people chant together might be hard to deal with if there’s 40 separate signals come mix time. Instead, that group might be better captured in a room with a stereo pair.

Two rappers feeding off of each other, however, might be better tracked in isolation, but perhaps together in the same room.

The completely separated method is the best and sometimes only way to go about things if you’re capturing sources in a variety of home studios or collaborating remotely.

These are just a few of the most common answers to the question of whether to track together or to track everything separately. As you can imagine, there are infinite ways you can tweak to make it work for your group. The best way to come up with the plan of action is to first think about the abilities and idiosyncrasies of the players, the genre, the end goal, and of course the budget. After all, if you’ve only got the studio for an hour total – the question is answered. Similarly, if you’re working with 4 or 5 people all in different towns, you know what you have to do. In any case, there’s no one right answer to the question “should we track all together or separately?” – only your best solution.

If you have questions or want to talk, just hit me up on Facebook @AaronJTrumm – or email me aaron @ aarontrumm.com

This article was written originally for the blog at Carvin Amps and Audio. I repost it here, and encourage you to check out Carvin’s amazing line of products!

In signal flow part 1, we covered the basics starting from ground zero. In part 2, we’ll cover some more advanced topics that shouldn’t be too hard to understand now that the basic idea of the directionality of audio signal is down.

Many of these more advanced issues come up in the DAW, but we can still get a better handle on them by starting in the old-school analog studio. We’ll also look at a couple of things to be aware of in live settings.

Splitting and routing

In part 1, we discussed simple set ups where audio flows in one “straight” line and doesn’t deviate. Now let’s introduce the concept of splitting, which will set us up to understand nearly everything else that comes up.

Splitting is simple. It just means that an audio signal can essentially be “copied” so that it goes to two places at once. Think of it as if you were walking down the street with your twin. At a fork in the road, you go right and your twin goes left. In the analog domain, splitting can sometimes affect signal level or quality, but its rarely a problem, and only when it’s overdone. In the digital domain, a perfet copy can be made.

Splitting is what allows things like aux sends, busses, monitor mixes that are separate from house mixes, and inserts. It can be as simple as using a y-connector or as complicated as multiple routing options on a mixing board.

Routing, meanwhile, just means telling a signal to go to a certain place. On a mix console (and in a DAW), routing is done with buttons. This is a kind of splitting, but it’s easier to think of it as multiplying. You could push the button that sends your signal to the main output, and also push the button that routes it to a buss.

Busses

A “bus” or “buss” in audio is a channel that mixes together other channels. So, technically the main output fader on a mixer is a buss, but when we talk about busses, we’re mostly talking about any extra busses. A console may only have the one main output, or it could have several busses. A mixer referred to as an “8 buss console” would have 8 busses besides the main output. Most DAWs let you create busses to your heart’s content.

Busses are useful for a variety of purposes. Probably the most common is mixing various tracks into one buss for easy manipulation of the whole group. For example, you might route all your drums (kick, snare, hats, overheads, toms, etc) to one stereo drums buss. This would allow you to control the volume of the entire drum set and apply processing to the whole thing at once. You would then route the drum buss to the main output for your final mix.

Splitting comes in when you decide to route those drums to a couple of different busses at once, or to the main output as well. You’ll have to pay attention, because on most mixers and DAWs, it’s possible to route tracks individually to busses and the main output – something you may or may not want. In our drum example, we’d want to make sure they were not routed directly to the mains.

In a DAW you could use busses for that same purpose, or to do things like route click tracks to the headphones and not the mains (if your audio interface has multiple outputs).

Bussing is also useful in a live setting for the same reason. Busses could also be used for a variety of other purposes but some of them would be awkward. Let’s say you have a great house mix but the band wants it a little different. Instead of a buss, the best way to handle this is an aux send.

Auxiliary sends

In that last example, you could use one of your busses (if it has its own physical output) as the “monitor” mix and use the main output for the house mix. That’s ok, but routing to busses is either on or off, typically. So, you couldn’t create a whole different mix – only turn off certain parts.

To create a separate monitor mix, you would need to be able to send different levels to your monitor buss.

Enter the aux send. An aux send is essentially a splitter with a volume knob. Most large mix consoles have several aux sends. The way they work is on each channel strip is an extra knob labelled something like “Aux 1” – there could be several. For every one of those, there’s an extra physical output on the console. Again, this is splitting, because signal goes to the mains (and/or any busses) and to the aux send.

Aux sends have a multitude of uses, including creating separate monitor mixes (if you have multiple sends you can create a different mix for each player, for example), making separate headphone mixes in the studio, and for effects like reverb. For that to work, we need something called a return.

Aux return

If you’ve only used the aux sends in a DAW, you may not know that in analog mixing, sends usually have returns. This is a special input on the console which is connected to a gain knob labelled something like “Aux 1 Return”. This is designed specifically so you can route audio to effects like reverb and return the processed signal to your mix. Aux returns are then routed to either a buss or the main output.

In a DAW, this is handled differently. Individual tracks can have as many aux sends as you want to create, but instead of being routed to a physical output, you route them to a buss. Then you put the effects plugin on that buss, turning the buss into a “reverb” buss, and routing it to the main output. This recreates the “aux return” in an analog mixer set up.

Since this has gotten a little complicated, let’s review with two simple diagrams for analog and DAW mixing scenarios.

In a DAW, it would be more like this:

Inserts

We just have one more monkey wrench to put in, and that’s the insert. You probably won’t see an actual insert if you’re working in a DAW, but the plugin section on a DAW’s channel strip is akin to an insert in the analog domain.

An insert interrupts the audio signal and sends it out of the mixer via a separate output and accepts it back via the same connector. This is accomplished using a special cable called an insert cable.

The insert is used in recording and live scenarios for two main purposes:

During mix down or a show, inserts can be used to process a track, similar to the way you would use a plugin in a DAW. For example, you might want to use a compressor on our vocal track. You would connect the output side of the insert cable to the compressor’s input and the input side to the compressor’s output. This is exactly the same routing as a plugin chain, and just as in a DAW, you could chain multiple processors together in the analog domain by connecting the first processor’s output to another unit’s input and so on – finally connecting the last piece to the input side of the insert cable.

During tracking with a mix console, the insert is often used to route audio to a recorder while still being able to hear the live audio in real time via the main outputs of the console. This way, a player or singer can hear themselves separately while still recording. In this case, the input side of the insert is simply not connected, which allows signal to pass through (turning the insert into a splitter). This can also be accomplished via the “direct out” of each channel – but many mixers have no direct output, instead letting the insert act as a direct output.

In a DAW you might simply turn on software or hardware monitoring and listen to the live signal that way. The problem with that is you may have to deal with some latency, which could throw timing off. So even in a DAW recording environment, when a studio is equipment with a console, the insert method is often preferred to monitoring from the computer.

Similarly, during a live show, inserts are often used to split the audio out to a recording device, to record a fully separated multi-track performance (when there’s no direct output).

A note on live splitting: Sometimes, signal is split before it gets to the main console, in order to allow a totally separate monitor mix. In this case, splitters are applied at the mic connection, and that signal is sent to two separate mixing consoles.

Final notes, feedback, and sequential power-up

As a final note, you will have noticed by now that given the right equipment, it’s possible to route and audio output back to its own input. This what’s known as a feedback loop and it should be avoided at all costs, for obvious reasons.

This is related to the feedback you might hear in a live setting. That feedback is because microphones are picking up their own output from the speakers and feeding that signal back into their own input.

It would also be good to mention sequential start up here. Since devices such as mixers create a little surge of audio signal when they’re powered up, it can be damaging to equipment down the line to power up in the wrong order. For example, if a powered speaker is on when you turn on the mixer feeding it, you’ll hear a pop. That pop could damage the cone.

So, audio professionals always power up gear in order of audio signal flow. So, it’s crucial to remember what direction audio flows. There are devices that do this for you, like Carvin’s AC120S power conditioner. Just remember to hook up the power in the right spots!

There you have it. We couldn’t possibly have covered every possible scenario, but you should have enough understanding now to understand and adjust to any situation you might encounter.

If you have questions or want to talk, just hit me up on Facebook @AaronJTrumm – or email me aaron @ aarontrumm.com