Recording

RECORDING

Here are a few principles to follow in order to record under good conditions.

Preparing the session
Drums
Guitars and bass
Vocals
Conclusion

PREPARING THE SESSION

First things first, we'll prepare the recording. Here, we're not looking for a tune, improvising, or rehearsing... we're going to record a song that we've already written. The goal is to record the "clean" version, once the rough draft is done. To avoid having to redo the same setup every time I want to record, I've prepared a blank template that already contains all the tracks and buses I know I'll need. This doesn't prevent me from adding or removing elements if the template turns out to be incomplete, or on the contrary too complete for a new project.

What does my template look like?

On the tracks side:
- Rhythm guitar tracks (between 2 and 8 depending on my projects)
- Lead guitar tracks (usually 2 tracks played identically to thicken the sound)
- Two bass tracks (one with the raw direct sound, one with an amp simulator)
- Lead vocal tracks (usually one or two, depending on whether I double the take or not)
- Background vocal tracks (when needed for the project)
- Synth tracks (same idea — the number of synths depends on the project, anywhere from zero to five or six)

Then come the drum tracks. There's one track per drum element, automatically set up by the DAW when I insert my virtual drum plugin:
- Kick
- Snare
- Low tom
- Mid tom
- High tom
- Hi-hat
- Crash cymbal
- Ride cymbal
- Splash cymbal
- Overhead mic
- Room mic
- Piezo mic
- A MIDI track containing the drum part.

Among my buses, I have a group for the guitars, one for the bass, one for the vocals, one for the drums, each routed to a Master bus that goes out to my audio interface and monitors. Here's a quick general diagram:

What is a track, what is a bus?

Tracks are what you record audio or MIDI instruments onto. For instance, track 1 might hold a guitar, track 2 the recording of a second guitar, track 3 the bass, track 4 the vocals, and so on... Everyone organizes their tracks as they see fit.

Tracks in the "Console" view of the Sonar DAW.

Bus

Bus: well... yes, but no...

Buses, on the other hand, don't contain any recording. They're more like groups of tracks. For example, you can create a "Guitars" bus, to which all your guitar tracks are routed. That way, you can apply common effects to all of them without adding those effects on each individual track. You add the desired effect on the bus, and each of your tracks then produces the intended effect. It also lets you, for instance, mute all the tracks routed to a bus, or conversely solo a single bus to hear only those tracks, by acting on that one bus rather than on every track individually. Here too, the chosen organization depends on each person's needs, tastes, or habits, but using buses makes certain operations very convenient.

Buses in the "Console" view of the Sonar DAW.

Both tracks and buses can be mono or stereo, your choice. On modern DAWs, the number of tracks and buses you can create is unlimited. The only real limit is your own needs — and your computer's ability to handle the tracks, buses, and effects you pile on.

Sonar's "tracks" view.
The sizes of the "tracks" and "buses" panels can of course be adjusted.

Finally, before starting — and if your software and hardware allow it — choose to record at 24 bits; the sound gains in precision. Once the session is set up to your preferences, recording can begin. Begin with what, though? I always start with the drums, and I'll explain why below...

DRUMS

Recording a real acoustic drum kit is a real headache when you're not used to it. Even for pros, it's a time-consuming process: it's not uncommon to spend several hours positioning and adjusting microphones around the kit before you can even start recording. But we'll settle for canned drums since we're in a home studio. So it'll be MIDI-file based... For the lucky home-studio musicians who own an acoustic kit (electronic kits can be hooked up via MIDI, and you can grab the part directly to tweak it as you wish) and who, unlike me, actually know how to play drums, you'll be recording yourself with microphones. I can't help you on that just yet, I don't know enough on the subject, but I'm sure other sites will give you a hand. As for me, I'll just talk about the way I do it.

First thing: why start with the drums? Simple — they'll serve as our metronome. The sound of the drums will guide us in keeping the tempo we've chosen when recording the other instruments and the vocals. This will let the song be played at a steady tempo, and avoid having it constantly speeding up and slowing down. That said, fluctuations can be interesting and give the track a less mechanical feel than a perfect tempo. But still, let's assume that for a song recorded in a studio — even if it's just "home" — aiming for steadiness isn't a bad thing.

In my case, I usually have no idea what my drum track will sound like in the end. In fact, I only fine-tune it once the song is already recorded. But I still need its metronome function to record the other instruments. So I create a drum track that loops, trying nonetheless to use a rhythm that fits what I'm going to play. For example, I'll use one of these loops:

(mp3 file)

(mp3 file)

(mp3 file)

(mp3 file)

Each excerpt contains one or several "regular" loops, then ends with a "fill" loop that I'll generally place wherever a change occurs in the song — like the transition from intro to verse, or from verse to chorus. That makes it easier to find my bearings while recording. I then fill my drum track with the number of bars planned for the song, making sure to add a few bars at the very beginning (2 bars for example, as shown below) that will serve as a reference to lock in and start recording in tempo. These extra bars will be deleted at the end.

Now that we have a metronome drum track in place, we can move on to recording the other instruments. In this case, guitars and bass...

GUITARS AND BASS

Let's remember that we're in home-studio conditions here, in an apartment, and that cranking up a big 100-watt tube amp is absolutely out of the question without making the neighbors scream. So we'll assume we're recording the guitars and bass directly through the audio interface, without going through a real amp and a microphone placed in front of it. The latter is probably the better solution, but one of the big advantages of recording direct paired with an amp simulator is that you can change the tone afterwards, without having to re-record. Just tweak the simulator's settings or switch simulators altogether.

So, bass first or guitars first? There's no absolute answer. You could argue that the bass-and-drums duo is the foundation of a track, the rhythmic base on which everything else rests. Fair enough. But it also depends on other factors: for instance, the person recording might be more comfortable with a guitar than a bass, and will therefore favor recording the guitar first. Or maybe the song features a really important bass riff that needs to be in place before tackling the guitar parts. Bottom line: it's up to you. When in doubt, the bass-and-drums duo is a safe bet. If that part is solid, everything else can easily build on top of it.

Either way, the recording will involve the same constraints. You plug your guitar into the preamp, with the preamp connected to the audio interface (or the guitar straight into the audio interface if you're using its built-in preamps), and you set the recording level. This is very important! Before recording, you absolutely must check that you're not at risk of going over the maximum level (0 dB, zero decibels). In a home studio, you don't have a sound engineer making the adjustments for you. So it's up to you to take the necessary precautions. How? Simple: do some test takes and adjust — it's quick to do, and it'll save you from recording a perfect take, error-free, only to realize afterwards that your levels were too low or too hot, forcing you to start all over.

You can do a throwaway take: for the rhythm guitar for example, play the parts you know will be the loudest, the most intense, and adjust the preamp and audio-interface levels so that when you play at your loudest, the recording level doesn't exceed -6 dB — and keep in mind that 0 dB is the absolute upper limit you must NEVER cross. If during the test you played the same way you'll play during the actual recording, then you can be confident your recording level will be correct. Try to keep your peaks somewhere between -12 dB and -6 dB, which should give you a more than sufficient average level while leaving you a safety margin before clipping.

Clipping is the term used to describe going into the red — that is, hitting or exceeding 0 dB. Clipping is evil :-)

No clipping With clipping
You can clearly see here the difference between a recording at proper volume and the same recording at too high a level. On the right, the sound is squashed, with almost no gap left between peaks and troughs. But let's take a closer look:

No clipping, zoom With clipping, zoom
Zoomed in, you can see on the right that the audio waveform is clipped — meaning everything above 0 dB gets unceremoniously flattened. And why is that bad? Because the sound becomes distorted, deformed — in a word: saturated. And this kind of saturation is to be avoided, unless you're deliberately going for a sonic mush that'll give your listeners a headache.

Example of clipping on vocals:

Without clipping (mp3 file)

With clipping (mp3 file)

And on a guitar:

Without clipping (mp3 file)

With clipping (mp3 file)

Alright, all set? Levels looking good? Then go ahead — it's your turn to record your bass and guitars.

Set up your guitar however you like by adjusting the bass and treble knobs, and the pickup selector on the guitar. Try not to end up with a recorded sound that's too dark (muddy) or too bright (piercing). The sound recorded without effects should already be pleasant to the ear. Obviously, by tweaking your instrument's controls, you can get different tones, and you should therefore be able to dial in your guitar to get a pleasing sound, in a tonal range that fits what you're after. Remember that during the mix, you won't be able to magically conjure up something that wasn't recorded in the first place. A bright sound won't turn into a deep, warm one, no matter how much tweaking you do. So plan ahead for what you want the end result to be, and choose your tones logically and consistently.

Now, stay in tempo, and if you mess up, start over. Once the song is finished and out in the world, it'll be too late to regret a bad take — so now's the time to get it right. Not everyone is named Jimmy Page (Led Zeppelin), Eric Clapton, Slash, or Matthew Bellamy (Muse), and everyone is allowed to make mistakes. There's no rush, and if you need to redo the same take 30 times, do it. Do it and redo it until you're fully satisfied. Record and listen back to the take, especially if you think it's the keeper. Above all, save your project regularly — you never know, a computer crash at the wrong moment can cost you hours of work.

If it helps, do the recording using your chosen guitar tone (see the section on amp simulators), but for now, what matters is the quality of the recording, not that of the amp tone. That's the beauty of amp simulators: you can change the sound after recording. We'll get to that later, during the mix.

There you go — you've recorded all the guitar tracks you needed. Maybe your song only requires one track, or maybe you need 8 — your call. It can indeed be interesting to double-track your takes, to add more body and consistency to the song. But also keep in mind that the more instruments you have, the trickier the mix becomes, because it's hard to make everyone sit well together. Going for a "big sound" doesn't necessarily mean stacking up tracks — it can also come from making good use of effects.

Of course, since the bass is also a guitar (a bass guitar, that is), the recording constraints are identical to those for guitar. You check the input level before recording to avoid clipping, and above all — even more so than with the guitar — the bass needs to be locked into the tempo, otherwise you risk making the song feel wobbly.

VOCALS

I prefer to record them last but there are no rules. If you prefer to record them first, then do so.

To record vocals, make sure the place is quiet, shut the door, tell the people who live with you to be quiet, and do not record while your neighbor is drilling holes through his kitchen walls! Also, turn off your monitors and use a headset instead to avoid recording the playback with your microphone.

Condenser or dynamic microphones?

Dynamic microphones are solid, they don' need a power source, they can take heavy acoustic pressure (like a kick drum or a saxophone) and they are not too expensive. They are also less sensitive to surrounding noises than condenser microphones. The cons are they lack clarity in the high range, which renders takes less clear and defined than with condenser microphones. They can be used with Jack or XLR plugs.

Condenser microphones are much more responsive and accurate. Their high sensitivity is double-edged, because they will capture any noise when recording. The fans of your PC are noisy? Chances are this noise will be recorded. Sound comes out of your headset? It will be recorded by your condenser microphone. Children are loudly playing outside? You might get that too. However, some condenser microphones are called "cardioid", or "hyper cardioid", and they only record what comes from a specific direction, ignoring (more or less) other sound sources from other directions. On the contrary, omnidirectional microphones record what comes from anywhere. Not ideal for a home studio. Condenser microphones are also more fragile (don't knock them) and must be powered through a "phantom power", whose standard is 48 volts. This kind of power is either present on your audio interface and can be turned on and off with a button, or it will require the use of an external phantom power source that you will then connect to your audio interface. You have to use 3-pin XLR plugs that carry the phantom power current. Finally, condenser microphones are usually rather expensive, some of them cost several thousand euros (or dollars, or pounds), but only professional studios or rich amateurs can afford those. On the plus side, the sound you get with a condenser microphone will have the best quality.

Be cautious though, a good dynamic microphone is worth better than a bad condenser microphone. No big secret here, for microphones like for anything else, very low prices are rarely synonymous with good quality.

A few known and renowned microphone brands: AKG, Milab, Neumann, Rode, Sennheiser, Shure...

Jack plug (left) and XLR (right)

Some pieces of advice: buy a microphone stand and a pop filter (you can also make one yourself with wire and a piece of tights from your wife / girlfriend / mother / daughter / neighbor). The stand will prevent you from manually holding your microphone and thus produce handling noises. As for the pop filter, it prevents the air to hit the microphone and produce unwanted blowing sounds when you pronounce some letters such as "p" or "b".

Microphone stand and pop filter

Furthermore, try to stay in front of the microphone when you sing, don't move from right to left or back and forth in order to avoid big volume variations. Dynamic microphones will not record you correctly if you stand too far from the microphone (8 inches, 20 centimeters would be a lot, meaning you can easily go too far). Don't let this piece of advice prevent you from "feeling" your song. If thinking about your position prevents you to have emotions, just forget about it but try to remember that the less you move away from the microphone, the better the quality. Nevertheless, if you have to really scream all of a sudden (I think about Frank Black from the Pixies, who goes from whispering to howling in an instant), you should then step back from the microphone in order to avoid clipping.

Just like the guitar recording, you have to set the recording level of your vocals before the take. Have some tries and check once again that the level doesn't go beyond 0 dB. Now, if you plan on having very different volume levels, using the howling moments as a reference will render the quiet moment barely audible. So, either you live with that (but you could bring some noise when compressing the quiet sounds during mixing), or you make several takes, with different level settings for the quiet and loud moments. You can also get the help of a friend who knows the song and will manually adjust the levels on the fly.

I also advise you stand up when recording vocals. Singing while sitting is a bad habit, as it prevents you from breathing deeply and could compromise your vocal amplitude. When standing, singers can better "live" their song. Look at studio recording videos: singers stand. Not only because it looks nice on the video, but for breathing reasons. Also think about carrying your voice. I am not saying you should scream (unless it is required in the song), but you should carry the voice. If you're mumbling, people will hear it. Think of Mick Jagger, David Bowie, Elvis Presley or Freddy Mercury, they carry their voice. You can clearly hear when they refrain, then just let go. So let go, let your future listeners feel that you gave it all, that you lived your song. But once again, unless it is voluntary, do not force it. You are not taking part in a shouting contest!!!

Unless you have a natural gift (and even then...), be aware that singing requires work to be mastered. Nobody becomes a great singer without practice. Do not overestimate your capabilities. Each of us has a singing range (tessitura) that can only be broadened through training. This is the range of notes one can sing, from the lowest to the highest. Needless to sing a note that is too high-pitched for you. If you cannot reach it (not yet), forcing won't help you much. At best, it will sound awful, at worst, you'll damage your vocal cords.

No need to go on and on forever, recording is rather easy. As long as you pay attention to your recording levels and take care over your takes, you should get a satisfying result, good enough to finalize the song