After hours of hard work, yes, we’ve all been there: that which we have mixed—which we have spent hours upon hours fixing up—sounds more like a demo than a record.
First of all, don’t be too hard on yourself. Demos are very good these days! In fact, some maintain there is no such thing as a demo anymore (what you send out is what gets heard, so you better make sure that it sounds like a record!). Home studios have also come a long away from four-tracks and flash-drive recorders; today’s average interface beats the heck out of those all-flattening, stereophonically sterile 002s and 003s of erstwhile epochs. Plus, tutorials are everywhere, so everyone is learning how to mix at a rapid clip. It stands to reason that today’s demos and rough mixes will force us to work that much harder.
I propose, then, that we stop for a moment and ask ourselves an essential question: what separates a demo from a polished mix?
Let’s take it further: if we can identify some key things, what can we do to transmute our mixes from demo-quality blandness into exciting, lovely vehicles for the music at hand?
Read on to find out!
But be careful: we can’t edit everything so carefully as to kill the groove, and the grid is not our best friend; uniformity of downbeat is boring and sterile. Indeed, it can sound even more like a demo, since so many producers and artists tend, when working with MIDI, to quantize themselves to the hilt.
The key concept here is intentionality—your edits should have intent behind them, and that intent is always to enhance the music.
Game plan: if you’re given a rough track/demo as a reference, identify first what you love about the song. Next, put up a static mix as quickly as possible (I’ll admit, sometimes basic editing tasks will take precedence, but try as best as you can). Now identify what is detracting from the elements you love—which notes rush too quickly, which sounds need de-noising, and the link.
This is what you’ll need to edit.
Once it’s been identified, you can get to work culling the tracks into shape with every tool you have, but do this with as light a touch as possible. That heavy-footed drum part might actually groove once the hi hats are sidechained to the snare, so don’t get too zealous.
If this is new to you, use a mastered reference track in the same ballpark as a guide. And if all this takes a while, that’s fine! It should. Many top mixers have assistants who do this for them. The rest of us assign hours to the job.
Edits to Hide the Edits
Yes, you need to make edits to hide your handiwork. Nothing calls attention to itself like a badly-edited breath, or a digital pop between contiguous regions. This is basic maintenance for achieving a polished sound, but still, it bears paying mentioning here.
Game plan: at the end of your editing session, before you really dive into the mix, apply fades both into and out of every region. Be judicious here. For two regions abutting each other, apply crossfades to smooth the transitions.
For edits between drum takes, make doubly sure these takes match in timbre. Listen for a lingering cymbal that’s suddenly ringing out of context, or a kick sliced right on the transient. The same applies to the vocal: if you’ve got a region obviously recorded with a different preamp chain or mic position, do some basic EQ matching in a software suite like RX.
Look for any regions that aren’t muted and make sure they’re silent. Don’t forget to look for hidden tracks or unchecked groups. Bonus tip: a pen and paper helps with this (as you can tell, I rely heavily on checklists).
Bass that Fits the Context
Turning our attention to the mix itself, let’s examine an element sure to attract attention: the low end. If your low-end isn’t sitting right, you’ve got something more like a demo and less than a record. Mastering Engineer Dave Kutch, in his infinite wisdom, calls the phenomenon of badly treated low-end “sloppy bottom.”
I shouldn’t have to say this, but you definitely don’t want a sloppy bottom. First of all, that sounds awful. Secondly, the low-end of a mix is something that, in sociology terms, we might call a “master identifier”—a governing characteristic that will be judged immediately. It’s no accident that many demos don’t stack up in the low-end department. You’ll notice they might consume too much bandwidth, sound too thin by comparison, or boast certain notes resonating in weird places (a phenomenon abetted by badly treated rooms).
It doesn’t help that low end is so hard to hear well; our rooms, if not properly treated, do strange things to our perception of the bass. Our monitors, if not placed correctly, fail to give us an adequate sense of power.
Now, the easiest thing in the world would be for me to say, “get proper treatment and good monitoring!” But I’m not going to do that, because ideal monitoring situations are just that: ideal. Most likely you mix out of a shared space, like a bedroom or a living room, and in these situations, compromises must be made. You won’t be able to drop expensive mains in a room easily overwhelmed by their output. Similarly, you’re not going to block out your window for sonic purposes, as this would crush your very soul.
But good mixes can be achieved on compromised monitoring systems—it happens all the time. What you need, in these cases, is a comprehensive strategy for dealing with room inefficiencies. I recommended ample use of reference tracks, an unwavering loudness-matching practice, consistent monitoring levels, multiple pairs of cans, and multiple speakers.
Game plan: pick out two references. The first should represent your ideal bass sound, regardless of genre. The second should be more genre-appropriate—either recommended by the artist or your own choice. When you’ve got a mix going, level-match the reference to your mix (this is easily done in Ozone 8 with Tonal Balance Control). Now, A/B your work. Use different monitors, cans, and listening positions.
What do you notice in the low end by way of comparison? If there’s a problem, can you identify what instrument is causing the hiccup? Mute that instrument and listen. Did the problem go away? If so, unmute the offender, and work on that instrument till it fits.
A Tightly Sculpted Mid-Section
A poorly-shaped mid-section also carries sonic problems that scream “demo.” Indeed, you have to be careful with the midrange because it communicates so much of what’s necessary—it’s the main aspect of any mix that translates from system to system.
If there’s a game plan, it’s the same as the previous one, with a few caveats: keep in mind that you’ll have more to contend with than frequencies in the midrange, because panning comes into play in a heavier way. So, when you’re comparing between references, play with panning positions to see if that clears up your midrange—it might just do the trick, even in mono (as mono fold-downs roughly translate into level changes).
Much of what we could say about vocals falls into the purview of editing, so let’s not rehash that. Let’s also assume you know the basics of mixing vocals and skip to what might be the most important—and yet most basic—aspect of successful vocals:
To my ears, the biggest difference between the voice on a demo and polished vocals is the level. Not the expensiveness of the sound (Steven Wilson, a famously “expensive” sounding mixer, often uses canned vocal parts), not the microphone chain, but the simple, nuts-and-bolts level. Simply put, the relationship of the vocal to the rest of the mix has to be perfect at all times.
This doesn’t mean the vocal has to be loud; quite the opposite. Some of my favorite Tool songs bury their frontman under the guitars. Conversely, go too soft on a ballad, and you’ll often engender unintentionally unintelligible results.
It’s objectively tough, because you don’t want to give the illusion of creating a karaoke track, nor do you want people to strain to hear the lyrics. Also keep in mind that vocals should be dynamic. You can’t just pick one level position and stay there; you have to make sure the level of the vocal matches the intensity of the phrase, the word, the music, and the moment. Sometimes the vocalist has taken care of this for you; you’ll find that if you haven’t over-compressed the track, you don’t need to move the fader all that much. In other situations, get ready to make friends with automation.
Game plan: Mix with the vocals in from the start. Yes, it’s annoying to focus on your drums with a vocal going on—and you can certainly dim the vocal when you’re working on other elements—but keep the vocals somewhat audible, and you’ll always have a feeling for the general timbre of the voice, and you’ll be able to mix them in a complementary fashion.
Also, I wouldn’t rely on headphones to give you an accurate sense of vocal placement, as they tend to bring elements closer to you. I prefer judging my vocals on my monitors, in car speakers, on consumer stereos, and over laptops. To check vocal integrity, I’ll often beam the mix to my clock radio in my kitchen, boil water, fry an egg, and do other things to create a noise bed. I want to make sure I hear more than the vocals while this is happening—I want to hear the song.