This month in “Fix It in the Mix,” I’ll focus on recording music, like modern pop and electronica, that’s primarily non-acoustic but may have one or a few acoustic instruments thrown in (including vocals, the original acoustic instrument).
Once upon a time, this type of recording was done with something like a TASCAM Portastudio cassette-based 4-track. Now, for not much more money (or possibly, pretty much the same money adjusted for 30+ years of inflation), you can actually record like the pros do—straight into a computer-based digital audio workstation (DAW). I’m not knocking the cassette-based 4-track, as many fine albums were created using these, including Bruce Springsteen’s Nebraska.
Warning: Before starting any recording, make sure your recording interface (e.g., the Focusrite Scarlett 18i20 I recommended in an earlier article), computer, and DAW are all able to communicate with and pass signals between each other the way they’re supposed to. Do this well ahead of time if possible—like the day before or even earlier if you haven’t used the setup you’ll be working with recently, or at all. Be sure to make a test recording of any ol’ thing before you even start thinking about recording anything important. This can be as simple as saying, “testing, testing,” into a mike connected to each mike input of your interface and checking that it actually records and plays back without distortion. When you’re in a creative mood, you don’t want to be stuck trying to figure out how to connect things, why something’s not recording, or why it sounds like crap. Or even worse, imagining you’ve just recorded the best take possible, only to realize that what you got was actually bupkis.
I won’t be going into detail about how to mike every little thing, in part because there’s no one right way to do it. Also, thanks to the internet, there’s already more information available on how to mike any given sound source than you could shake a whole tree’s worth of sticks at. I don’t want to give you the same thing as everyone else. I’m also not going to be DAW-specific because there are a lot of them out there. Each has its particular strengths and weaknesses, and those depend on how you use your DAW. Steinberg Cubase is quite popular with composers. Others like Ableton Live, while Apple’s Logic Pro is popular with electronic dance music (EDM) creators, and commercial recording studios overwhelmingly use Avid’s Pro Tools. Instead of focusing on software, I’m aiming to help you build a solid foundation to start—or continue—on your artistic journey. What I will say is this: you should be recording at 24 bits and using a sample rate of at least 88.2 or 96kHz. Computer speed and storage are cheap, so take advantage of them. With 24-bit recording, your peak levels should be around -6dBFS to leave some headroom for the mixdown process. Six dBFS is equivalent to about 23 bits of dynamic range, which is more than any playback gear (or human) can resolve. Turn off all unnecessary plug-ins while recording. Plug-ins cause latency (delay) due to processing time, which affects musicians’ timing—and hence, their performance.
As I noted in the equipment shopping article, you won’t need a lot of mikes or an interface with lots of inputs when it’s just you and maybe one other person. I would go with an interface like the Focusrite Clarett 2Pre USB ($599.99, all prices USD) that has two mike preamps and the ability to pass through MIDI data. Focusrite’s Clarett line of products uses better mike preamps, line inputs, and converters than the lower-cost Scarlett line of recording interfaces (one of which I recommended in the shopping article). There are many brands out there, so get the interface that works best for your budget and sound quality targets. You’ll also be able to work with far fewer mikes than the ten I recommended in the shopping article and fewer stands, cables, and accessories. You’ll need at least one mike if you plan to record vocals and a pair of the same make/model of mikes for the stereo techniques outlined in the previous article in this series. The stereo pair can be the same mike model that you use for vocals or other instruments. Mikes like the AKG 414 TLII can do double duty, but they’re kind of pricey ($3148 for a matched pair; $1447 each for unmatched individual mikes). You can also use a single mike for everything and not worry about stereo miking. Most of the pop music recorded over the last 50+ years was recorded “multi-mono,” meaning that every sound source was miked with one microphone. If vocals, and maybe an electric guitar/bass-type instrument, are the only acoustic things you’ll be recording, then one good mike is really all you need.
I’m going to focus primarily on recording a good vocal performance since that’s what pop, electronica, and pretty much everything outside of classical and jazz have in common (and yes, I know choirs, opera, and Ella Fitzgerald exist). At this point, even though you may have everything adjusted for the sound you want (for details, see my last article), you’re still not quite ready to record. You need to get the monitor mix right. This is the mix of instruments/vocals that musicians listen to as they record, usually through headphones. The monitor mix for vocalists is likely the most difficult and the most important thing to get right. Generally, a vocalist needs to hear three things loudly enough to be able to hear how their voice and its intonation will sound: tempo (meaning a click track or percussion), the instrument(s) they need to sing in tune with—including pre-existing vocals for harmonies—and their own vocals. If you’ve heard your voice recorded and been surprised by the way you sound, then you know what I’m talking about. This requirement to adjust for tempo and pitch is one of the reasons vocals are recorded later or last.
If you’re an electronic musician, you’ll probably be using a MIDI (Musical Instrument Digital Interface) controller to record your performance as MIDI data. Keyboard controllers, those that mimic piano keyboards, are the ones you’ll see most often, but controllers that mimic most common instruments exist. Most DAWs, including Cakewalk by Bandlab, allow you to record MIDI data as well as audio. If you have a programmed drum track, start with that; this will also serve as your click track because drum machines are really good for that (and being on time, pun intended). You can use analog or MIDI output (or both); it’s your choice. Set your song’s tempo in beats per minute (BPM) using the drums, or just go with the tempo they’re programmed at. Once that’s done, you can add an additional MIDI or audio track for each instrument in your arrangement. There’s no need to worry about acoustics or mikes at this point; it’s all just signals in wires. If you play an acoustic instrument, such as guitar, violin, or saxophone, it can be recorded at any point after the percussion/click track. The main advantage of recording your performance as MIDI data is that it gives you a lot of flexibility in relation to sounds and arrangements during the mixing stage. Also, pretty much any acoustic performance can be converted into MIDI data using a plug-in like Melodyne 5 Studio ($849) if you change your mind later. Celemony Melodyne 5 Studio, like Antares Auto-Tune Pro ($399), is first and foremost a pitch-based plug-in.
One interesting thing that you can do by converting an acoustic performance into MIDI is improve an instrument’s sound. Say you’ve got a Fender Squire bass guitar knockoff (so, like, a cheap imitation of a Chrysler K-car). Record it plugged directly into your recording interface, convert the audio track into a separate MIDI data track, then later use the sound of a mid-’60s Fender Precision bass, Moog bass, or whatever you want in its place. You generally don’t want to start recording vocals until you’ve recorded at least a click track and the chord changes for the song so the vocalist has both tempo and key/pitch references for timing and intonation. A singer-songwriter singing and playing guitar would be an exception if you’re going for a live feel, but There Will Be Bleed. The vocals will be pretty loud in the guitar mikes and vice versa because there’s no good way to separate them, so make sure that both the vocals and the guitar are “keeper” takes. If you’re not sure, you can always record more versions on open tracks.
When you’re recording a part, try to ignore mistakes and keep going through to the end. Obviously, if making the mistake takes you or the artist out of the groove and you can’t get back quickly, stop and do it again. If there are mistakes, you have a couple of options. Most DAWs have a way of recording the same part over and over on the same track without erasing what came before, like separate files in a folder on your computer. Before DAWs, we used to record to separate tracks then mix and edit the individual tracks together onto another unused track to free tracks up again (losing a generation of quality in the process). Combining the takes is called “comping.” It’s industry slang for combining multiple performances into a single, hopefully perfect, composite. This can also be done on separate tracks, if you like. The other option is to “punch-in” and re-record the portion with the mistake. Punching in is a destructive method where, with the artist playing along to the song before the mistake, the recording engineer puts the track into record mode just before the clam—the misplaced note—and takes it out of record mode right after. Back in the day, we did this by holding down the Play button on the recorder’s remote control with one finger and hitting the Record button with another (usually middle and index fingers, respectively) to put the tape deck into record mode and “punch in.” To stop recording, I would press the Play button again to “punch out.” Now I can just specify the punch in and punch out points in the DAW and set the start point a few measures before the mistake, and it takes care of everything else. If the performer can match the feel of their initial performance before the “punch in,” and the engineer has good timing or sets good in/out points, the change will go unnoticed in the edited version. Think of it as editing on the fly. Punching in is usually quicker, assuming the performer can nail the performance quickly, but comping gives you more options when it comes time to mix. Double tracking is a variation on this, but you have two complete performances of the same part. It’s fairly common to use double tracking with vocals to make them thicker and stronger sounding, the same way a choir sounds denser and fuller than a soloist.
When recording vocals, it is very important that the intonation is as good as the vocalist can deliver. As a corollary, you should generally always tune your instrument(s) before recording. Yes, pitch can be corrected after the fact (and usually is on vocals), but the better your vocalist’s intonation, the less work you’ll need to invest fixing it in the mix. Also keep in mind that the further the pitch is moved towards correct intonation, the more “Auto-Tuned” it will sound. Most vocals are recorded with the singer wearing headphones, but this is not a requirement. You can record vocals using speakers instead of headphones, treating the session like a live club performance. This can help a singer achieve a more exciting performance, but just as you would in a club, you’ll need to be aware of the mike feeding back through the speakers. With instruments like acoustic guitar or violin, it’s better to use headphones if you’re using mikes and not a pick-up like an electric guitar uses. Otherwise, you’re likely to have a lot of the monitor mix recorded along with the acoustic instrument; this will cause you problems later on—unless you’re looking for that kind of sound. Remember, there are not a lot of hard and fast rules in recording.
Having everything prepared in advance allows you (the engineer) and the performer (possibly also you) to focus on the performance. All the fancy recording gear and golden ears in the world won’t matter if the performance is a steaming pile. If there’s one takeaway from this series, and this is the reason I stress preparation, it’s that your goal should always be to capture the best possible performances. Not having to worry about whether your playing is being recorded well means you can focus solely on what’s important, the actual performance. Next time I’ll talk more about recording larger groups, like rock bands, and the overdubbing process.
. . . Mark Phillips