Recording - October 2012 - 48

New tricks for old dogs The most familiar, and probably the most-used processors in mixing are those old stand-bys, eq and compression. Whether you’re soloing tracks to tweak the sound, or making the adjustments in context, you’ll be spending a lot of time with these two. I won’t even try to hit any specifics here (what bands should I boost/cut, should I compress this or that instrument, etc., etc.)—whole articles and books have been dedicated to these questions. But I will mention a few considerations about approaching the use of these familiar tools. I’ll start with just a brief nod to another greatest-hits-of-mixing question, “Which comes first in the channel strip, the eq or the compressor?” As with pretty much everything else in this article, the answer is “It’s up to you”, but there are some considerations. Conventional wisdom has always suggested that, for generic purposes, signals should pass through dynamics processors first, then eq—so, typically, gate (if applicable), then compressor, then eq. This conventional ordering is based on two things: the compressor’s effect is threshold-dependent (dependent on the incoming level from the previous effect), and making adjustments with an eq, even if they are made to specific frequency ranges, still changes the overall level of the signal. With the compressor first, you can eq a signal freely, without having to continually re-adjust the compressor’s threshold setting to preserve the compression effect you’ve previously dialed up.

However, sometimes you want the compressor’s action to react specifically to something in a particular frequency range (like taming a resonant peak that hits one or two notes more than others)—obviously this would be a situation where the compressor would need to follow the eq. Of course you could do both: a compressor for overall dynamic smoothing, followed by an eq, followed by a second compressor (or limiter) to control the effect of the eq. In fact, both compression and eq have more of an effect than just the intended settings would imply, as do many other types of effect processors as well. As mentioned above, using eq will not only make the desired tonal adjustments, but will also affect overall signal level—subtle, almost unnoticeable eq tweaks can sometimes be used to bring a track out a tiny bit in the mix, as an alternative to a small broadband level adjustment of that track’s channel fader. Likewise, compression can affect more than just dynamics—the tonal balance of a compressed signal will often change subtly but noticeably. For example, heavy compression can often “warm up” a track (thanks in part to its effect on transient energy), and compressors, even generic ones, as opposed to “character” models or multibands, can be used to fatten up an instrument or vocal, to good effect. If this effect turns out to be a negative, an eq (possibly a second eq, as suggested above?) placed right after the compressor

can compensate, restoring any sparkle that may have been lost. Mixers who have been at it for a while are used to this sort of “thinking outside the box”, when it comes to processing—many effect units can offer something to the mix beyond their intended effect, and when a traditional tweak (say, eq or overall level) doesn’t quite accomplish the intended change, making subtle use of another type of processor can often provide a slightly different, more effective take on that adjustment. A longer article (or book!) might have space to cover more such tricks & tips, but here I’ll skip ahead to further discussion about proceeding with the mix. Who’s next Assuming, again, a typical approach where the drums were addressed first, what would the next logical step be? Work in the rest of the rhythm section (bass, followed by guitars, keys, etc.), one by one, or something else entirely? While a common approach would be to build up the mix by gradually building up the instrumental parts, working up to lead vocals and solo parts, I often take another tack at this point. After getting a working version Figure 4. The lead vocal part (originally one take) was split between two tracks to accomodate the need for different EQ & Compression settings in the verses & choruses, without using automation of the drums or other basic rhythm part, I’ll jump to the lead vocal and work that up a bit before fine-tuning the rest of the instrumental arrangement. At this point, of course, I couldn’t shoot for a highly finished vocal sound (final decisions will have to wait until the arrangement is almost fully fleshed out), but incorporating a working version of the main melody at this time will

Figure 5. A typical finished mix (I 86'd one dubious track, and nobody noticed)
48
RECORDING October 2012
Recording - October 2012

Table of Contents for the Digital Edition of Recording - October 2012
