One of the great things about being a hired gun is that you get to experiment with a plethora of disparate genres, honing the best tools for the job as you go. Eventually, certain truths become self evident—for instance, what’s good for hip-hop is poison for classical music (though that may have been obvious from the outset). As you go along, you begin to learn specific methodologies for scrawling out the appropriate signatures of each given genre.
In that spirit, I’d like to share a few things I keep in mind when working on a good ol’ fashioned rock and roll mix—the sort of thing of which The Kinks would be proud.
I believe in a war as old as engineering itself—a battle between the kick drum and the bass for “Who Rules the Low End.” I sometimes refer to it as “the skirmish of sixty and a hundred,” because those are (roughly) the two frequency ranges in question. This is a generalization: there's deviation between song and song, mix and mix, frequency and frequency, but, as a general guideline, I find one instrument's fundamental must be relegated to between thirty and sixty hertz, while the other should sit in the one-hundred slot. Usually a little curtailing is in order as well: if one instrument sits in the 100 Hz area, some attenuation might be needed in its 60 Hz region, and vice versa.
Again, these are generalizations, to be sure. While we’re at it, here's another one:
In the war between sixty and a hundred, let the bass win. For rock music, it always sounds more appropriate to let the bass hold the lowest of the lows, while the kick drum fuels the momentum by hitting (and clicking) higher up the spectrum. To put the kick in the lowest register is something far more suited for an electronic genre, one that uses 808s or some other similar, synthetic element.
This isn't to say that you should high-pass filter all overlapping frequencies out of existence. Nothing so drastic is required. Just keep in mind a general pecking order, and use subtle tricks to handle the overlap, such as a dynamic EQ sidechained to the kick to reduce masking frequencies in the bass. You can definitely have kick, bass, and guitar all occupying the sub-100 region (especially in drop-D); the key lies in carving out space like a Japanese printmaker.
Many engineers prefer to start with the kick/snare/bass balance, progressing from these three elements outwards. This approach works well for them, but in my own practice it leads to a sound more suitable for 70s-era R&B, or "tea-towel" style rock (when they used to use tea towels to mute the drum heads). If the sound of Nilsson Schmilsson is what you're going for, then by all means, start with the spot mics. Likewise, if your overhead picture is wonky due to inferior mic technique, perhaps you might need to gear the mix in that direction anyway.
However, I find the more natural modus operandi for rock is to start developing your drum sound with the overheads, followed by the room mics. Once those sound fantastic, move on to the spot mics. In recent years, I work my snare before the kick, because I’ve come to the conclusion that the kick and the bass are so dependent on each other, they must become their own dedicated time-drain. (Side note: I tend not to work on drums in solo; I prefer to bring the music buss and vocal down in volume so I never forget the context of my drum sound. That’s just me though.)
Working in this order helps you get a sonic snapshot of the overall kit right from the beginning; this process carries the benefit of creating a characterful sound. We all know rock and roll is about character. The sound of emphasized close-mics carries its own character too, but it's not nearly as individuated: A tom in your face tends to sound like a tom in your face. Working on the whole kit from the “big picture mics” first gives you the ability to forge a more individuated sound.
Clean, crispy vocals are not necessarily the stuff of rock music. Heft, meat, and midrange is where the money usually lies. Therefore, don't be afraid to let other aspects of the arrangement take front and center in the treble-range; cymbals can easily occupy the sparklies, while the voice can simmer well below the 10 kHz mark.
When just starting out in the mixing game, it’s a common pitfall to make every instrument as bright as it can be. Soon, though, we learn how ineffectual and brash that timbre-makeup becomes, and so we begin to pick and choose our trebles.
For R&B, neo-soul, pop, and even jazz, the vocals tend to win the war of the tops. But these are genres that often sound expensive by definition, whereas rock craves grit, grime, and other adjectives connoting distortion; distortion, in my mind, operates best not where it'll cut your head off, but below that threshold, in the mid and high-mid range. As the vocals carry the tune and its message, let them signify the genre not trolling the high end, but by carrying the meat.
Again, rock is rough around the edges. Croaks of the vocal give it flavor. Timing inconsistencies of the drums lend groove. While pop and R&B, especially in recent years, have started to re-embrace the imperfections of human performance, the inconsistencies of rock beget the intrinsic humanity that is its hallmark. These imperfects also contribute to the credibility of one band’s bona fides over another (I can think of no better example than Mumford and Sons versus the Avett Brothers—except for one: late Black Keys versus early Black Keys).
With this in mind, do not edit everything onto the grid. You will kill that very humanity. By all means, edit the obvious telltales of a bad performance—the train-like speeding up of a drummer after a fill; the bass note which comes in too early—but don't do this at the expense of the groove. You'll know the groove is dead when you stack the tune against a reference of similar tempo/feel; if you tap your feet at the reference, but strain to dance at your mix, you have a problem.
Conversely, sometimes you'll need to edit for groove when it isn't there. This is often best accomplished not by pushing the drums around (unless they truly are off-time), but in moving the bass and guitar slightly behind the beat, or more often, slightly ahead of the beat; this creates a tension that engenders groove. Usually—but by no means always—you'll want to move these elements earlier (ahead of the beat). This creates a feeling of intensity, rather than a relaxed swing, which is what you'd get by moving these elements later (i.e., behind the beat). Be careful though: you will not have to do this often, or throughout the whole tune. It's merely another tool at your disposal, and overuse creates more problems than solutions.
Moving on to the vocal: Do not put the whole thing through Melodyne, Autotune, or any other intonation software! These processes are never transparent; their FFT algorithms rounds the harmonics of the notes out in a manner that tends to belie the fake; listeners might not hear it, but they will feel it—something will seem off, seem more different, and feel less authentic to the genre of rock.
Instead, try to tune in a less aggressive manner, which means grabbing only the worst offenders and processing those on a separate track, then editing them back into the vocal. You can use the usual plugs, but I don't always reach for them; these days, I prefer to pitch the offender up or down with iZotope RX whose pitch contour module lets you move the note and its harmonics in a most natural way. It seems to mask the fakery.
I find the process so transparent that I've even used it when mixing a classical singer who was flat for an extremely short period of time. Now, this woman is a perfectionist when it comes to her voice, accepting nothing but the best takes. She would give me no end of trouble if she knew there was tuning software on her voice; in fact, I'm risking a lot by telling you this, as the vocalist is my wife. But in this instance, I think it's okay, as she was only flat because of the composer, who wrote the wrong note into score, and who happens to be her husband, who is me.
The takeaway: Don't put every instrument on every grid, and don't tune every note.
In many rock songs—though by no means all of them—the arrangement of the second verse is a carbon copy of the first. This is especially true in stripped down ensembles like power trios. Therefore, differentiating the second verse from the first in some sonic manner is helpful, or you could run the risk of mixing what sounds like the world's most expensive demo.
If you're working with excellent players who took the whole song live—or the tracks of an excellent producer who accomplished the same effect with rigorous comping—then this directive might not apply. But these days, the second verse can often feel cut-and-pasted together, and this requires your utmost attention.
Parallel processing can be your friend here: the introduction of a slight fuzziness to the bass on an auxiliary track might be enough differentiation, as might a squashed drum buss edged into the overall drum picture. Different ambiances and reverbs could be your friends too, as could modulation effects that, subtly employed, change the width or perception of the second verse. If you were presented with DI's of the guitar and bass, you can get very creative indeed. You could try "canning" the first half of the verse and opening out to the full heft of the arrangement on the downbeat of the second half.
The options are endless. The takeaway is this: if the band isn't giving you anything emotionally different for the second verse, it's on you to provide it. The band might have problems with any overt changes, so don't necessarily go bold; something eked into the picture might do the trick.
This rule applies to more than rock music, but it seems to work quite well within the genre. Try some sort of automation move—be it in frequency, width, volume, or a combination of the three—to deliver the emotional impact of a chorus. If you use a multi-buss technique, wherein you route drums to one buss, instruments to another, and vocals to another still (all to be processed differently and fed to a master buss), then you can easily perform these automation moves on a global scale.
A classic trick would be to juice the chorus section in level by 0.5 to 1 db. Other creative choices abound: Try putting an imager on the music buss and widening the guitars ever so slightly for the chorus. You could also try rounding off the low end on the drum buss around 80 Hz or so in verse two, returning to full, sub-harmonic force for the second chorus. These sorts of moves create the kind of energy rock music desperately craves.
There are many roads up the mountain of rock, and what I’m providing here are surely broad strokes to cut your path. More granular tips can certainly be related in future articles, especially if requested on Facebook, Twitter, and the like. For now, it is my hope that these tips/procedures will be of use to you on your journey. However you use them, rock on!