1. The best mixes begin with quality recordings of great performances.
You can hit up Google or YouTube for plenty of mic techniques for different instruments, but here’s a short guide to using your brain and your ears to mike an instrument up. Ask yourself: what are the major qualities of the instrument I’m trying to capture for this track? What do I want it to sound like? Have someone play and move your head around – where does it sound best? Just use your ears. Put the mic where it sounds best.
Ask yourself: is this a primary part of the track, or an accent? How many mics do I have at my disposal? Some instruments are easier to mike up with multiple mics – although rest assured, you can get a great sound with just a single mic. No excuses. How does your room sound? If it sounds bad, and you can’t find a place where it sounds good, get closer to the instrument. Or better, find a better-sounding room.
Practice your parts enough that you can play them without thinking about what notes you’re going to play. Know what’s going to happen when you sit down to record. A great performance comes when you can devote your attention to how you’re playing, not what you’re playing. And don’t skimp on thinking about arrangement! Ask yourself: what is each part contributing to the song? What register is each instrument occupying at a given moment? Is anything masking another part, like the guitar treading on the bass’ territory, for example? Can I change the voicing somehow so that isn’t happening?
2. Organize your session.
Comping parts, making sure everything is named and organized (and, if you’re into it, color-coded. I’m not into it, personally) appropriately, that all happens before you begin mixing. You will be happier for it.
3. Production and creative effects come before the mix.
(This is sort of a corollary to the previous point, and it’s also not a hard and fast prescription; work however is most comfortable for you.)
4. The essential tools in a mix engineer’s toolbox are, in order of importance: volume, panning, and EQ.
You ever hear a mix and wonder how they got everything so distinct and balanced? It happens here, with just those three tools. Here’s what I like to do:
Select everything and drag the faders all the way down. I like to start with the drums. Bring them up, get them at a comfortable level. If you have individual tracks for the various components of the kit, get them balanced. Feel free to consult a reference track if you’re unsure. If you’re ever unsure, consult a reference track. Once that’s feeling good, bring up your bass tracks. Try to dial in the volume so you feel like the bass and the kick are fairly level and no one’s talking over one another.
Now bring up the vocals. Get them sitting comfortably. Next, the guitars, or the keyboard, or the synths, or whatever’s filling in the midrange in your track. Here, again, you want to pay attention to how it’s balanced with the vocal (or whatever the melodic centerpiece of the track is). Spend some time tweaking things here, playing just with the faders.
Do you notice that? How when you bring a fader down, it loses some of that low mid presence? Here’s a great lesson from Gregory Scott of Kush Audio: your faders are EQs.
Once you’ve got the balance where you like it, it’s time to begin panning. Some general tips for panning: remember that we don’t perceive much directionality to low frequency information, so that bass is best served straight up the middle. This is nice, because the vocal wants to sit there too, and they occupy very different ranges. But what about those pesky midrange tracks, like the guitars and the synths and the keys, all stepping on one another’s toes? Well, this is where your panning helps. Put them off to separate sides. Hear how much space that cleared up?
Here’s a great lesson from Dan Worrall on his Fabfilter videos: hard-panned items will be dramatically quieter in mono. I’m not going to tell you not to do it, but know that if you do do it with an essential part of the mix, the balance of your parts is going to be very different when someone hears it off a phone speaker.
But we want that width! Wide mixes are sexy! Okay, I agree. But remember, width is something we perceive and we just need to create the perception of width. For instance: double-tracked guitars panned opposite one another will not sound wider. They will sound bigger (in stereo) and interesting and with character, but the similarity of the parts reflected across the stereo image will narrow our perception of it. If you want a wider image, pan very different parts across from one another. Their difference will exaggerate the wideness of the mix. And instead of hard-panning your core instruments, try panning them to 50 or 75 percent instead and bussing a reverb out to the other side – you’ll find it will give you not only more width, but more depth as well.
Now flip everything into mono. How’s the balance? Everything still sound good and clear? Good. If it doesn’t, tweak your panning until it does. Then leave it in mono and break out the EQ.
Here’s a great lesson from my own experience ruining mix after mix: you need to do less with the EQ than you think. This comes from not trusting your ears. It’s okay. You can trust your ears! You have to trust your ears. Use a reference track if you’re uncertain! But don’t suck the life out of your well-recorded, well–thought out, well-performed tracks with too much EQ.
I like to work in two passes: in the first, I’ll make a few surgical cuts in case there are any harsh or unnatural frequencies – but limit yourself to one or two narrow, surgical cuts per instrument. I promise all the weird, harsh and ringing frequencies you’re hearing by the end of your first sweep will go away after you rest your ears for twenty minutes. And don’t cut more than 6dB – that’s a dramatic cut, and it should be plenty if you’re working with good source.
In the second pass, I’m paying attention to the key frequencies each instrument occupies and only cutting where they are fighting for attention with another instrument. Low-passing the bass, putting mild high-passes on the guitars and the vocals if they need them, et cetera. Be more gentle with these than you think you need to. A little goes a long way and we want to retain all that life! If the vocal is fighting with the guitars, I’ll experiment with boosting around 1k and push- or pulling the fader. If it need more warmth, the same, but with 100 Hz instead. And remember – it’s far more natural to give a slight boost at 1k on the vocal and make a slight cut at 1k in the guitars than to do a dramatic version of either to just one track.
This is 90% of your mix right here. Everything can be accomplished with these three tools alone. Compression, de-essing? That’s just volume automation. A little reverb and delay are a great way to breathe life and depth into a mix, but you need way less than you think (unless you’re using it as a creative effect) – and you should reach for your delay more often than the reverb, even though I know you want to reach for the reverb. I will say, though, I love saturation. It’s like sugar, in that it’s delicious and a natural painkiller and really easy to overdo.
5. You don’t need more plug-ins, you need more time with one plug-in.
If I could do it all again, I would limit myself to one EQ, one compressor, one reverb, one delay, one saturator, and one multiband compressor. This is all you need to make a professionally competitive mix, but you need to grok those plug-ins. Know them inside out. They become intuitive extensions of yourself, the way writerly folks talk about warriors and their swords and warriors talk about writers and their pens. Narrow the distance between what you want to do and doing it. More importantly than even that, this will teach you to understand each of those tools and how they can be used fully.
Now, you can make a record entirely using the stock plug-ins in your DAW. I’ve done it with Reaper (and frankly the stock plug-ins in Ableton are really nice). So if you’re a little less financially flexible right now, don’t worry. You’ve got everything you need. But if you’re living a life of luxury, these are my favorites:
- EQ: Fabfilter Pro-Q (creatively: Goodhertz Tone Control)
- Compression: Fabfilter Pro-C (creatively: Goodhertz Vulf Compressor)
- Reverb: Sound Toys Little Plate (creatively: Goodhertz Megaverb)
- Delay: Sound Toys EchoBoy
- Saturation: Fabfilter Saturn (creatively: Soundtoys Decapitator)
- Multiband Compression: Fabfilter Pro-MB
Those are expensive plug-ins. If you’re in Reaper, I think you’re perfectly well off with the stock plug-ins. If anything, I might recommend grabbing a third-party compressor and reverb with more intuitive interfaces. There are plenty of great recommendations out there. I’ve really enjoyed Softube’s TSAR-1 Reverb and Saturation Knob, both of which I got for free. You don’t need a thousand free VSTs, I promise.
(Unless they’re being used creatively or in production, in which case who am I to say whether you need 400 free software synths.)
6. Do less than you think you should.
Have you ever wondered if a track needed compression, not been able to tell, and then decided to slap a compressor on it with unspecific settings just because you felt like it should have compression on it, probably?
Have you ever thrown a saturator on a track, pumped it until it sound absolutely righteous, and then come back later to find that everything in the mix was harsh and overdone?
You are an archaeologist of excellent musical performance. Restore it with the most non-invasive movement you can. Never mind the details of phase shifts and processing artifacts, just do less than your impulse. Dial it in until it sound right, then back it off a bit. Don’t make a move unless you know why you’re making that move.
And I’ll tell you what, you don’t need that fourth layered guitar part. Three is plenty.
7. Use reference tracks. Use reference tracks!!!
Using reference tracks is about the absolute worst feeling because it really highlights your inadequacies as a mix engineer. That’s okay. We’re learning. Hell, that’s what a career is – growing up in public. So ask yourself: do you want to make the best mix you can? Or do you want to feel the best you can about a bad mix?
Bring in a few well-engineered tracks to reference throughout the process. A good reference track is one that you think sounds good. If you’re not sure where to start, ask someone here what they like! Some stuff I really like: Phoebe Bridgers, Punisher; Perfume Genius, No Shape; Alabama Shakes, Sound & Color. You’ll notice that all of those records have really different tonal profiles – and also that they all sound amazing. There’s a lot of room to move around here. A great mix can be a lot of things.
Use your ears to match the volume of the reference to your track. Mastered tracks are way louder than where we’re working. Don’t sweat the volume difference right now, just get them even to your ears.
Listen to the balance of the instruments. How does it compare to yours? Listen to the tonal qualities of the mix and its instruments. How does that compare to yours? I’ll often take an EQ and separate the spectrum into sub, low mids, high mids, and high frequency bands. It’s borderline heresy, but I’ll solo each band and compare it to my mix to really highlight the differences there. I won’t make decisions based on this, though. I’ll just use it to inform my decisions.
8. The most important qualities of a good mix are quality songwriting, arranging, and performance.
In truth, all we need to do is articulate an idea. Neither Pinegrove nor Elliott Smith make the world’s cleanest sounding records, but the music is great and you probably never noticed.