Photo: Chuck Dahmer
“Flat,” “cluttered,” and “murky” are the most common complaints I hear from recordists and bands when they describe the self-engineered projects that they bring into my mastering lab. Typically, each instrument or vocal part may sound great on its own during the recording of an album, but when the tracks get mixed down, the whole can easily become less than the sum of its carefully crafted parts.
Sometimes making a few mastering tweaks — a high-end EQ boost or a mid-bass cut — is all that's required to revitalize a lifeless mix. But more often than not, homegrown mixes suffer from sonic clutter on many levels, not unlike an overstuffed closet that needs to be emptied and reorganized. So when time and budget permit, I recommend that clients with a serious “space problem” take a step back to their final mixes to do some much-needed cleanup.
To that end, one technique that works well for me is to expand the conventional concept of a “flat” space between the speakers into a living, three-dimensional space. Often I'll begin with clients by working together to identify and address the multiple factors and practices that suck the life out of their well-engineered tracks, bog down their mixes, and hasten listening fatigue.
Next, these problem mixes can be stripped down and rearranged from the perspective of creating clarity, or space, in the mix to draw the listener in. The end result is a revitalized sound world in which tracks no longer jostle each other for space or intelligibility, and the whole has a feeling of ease, openness, and balance in all dimensions.
I constantly encounter a number of common space-sucking issues while working this kind of sonic alchemy. Those recurring problem issues — and solutions drawn from my experience — are what this two-part article is all about. This first installment lays the groundwork for thinking, listening, and mixing in three dimensions, and addresses the issues of reverb and room sound. The second part will address compression use, arrangement, EQ, advanced panning concepts, and how these elements can affect space in the mix.
Bear in mind that these tips, tricks, and suggestions are offered as a way to think conceptually about mixing; they are not meant as hard-and-fast rules. And don't forget that rules are made to be broken — especially in recording. But it's more fun to break the rules once you know them well.
Creating and confronting the issue of space in a music mix begins with thinking and listening three-dimensionally. We are all familiar with the simple side-to-side stereo dimension (or imaging) that exists between the left and right speakers, which is controlled by panning. Consider this to be the horizontal axis or width of your mix.
FIG. 1: The three dimensions of mixing: horizontal space is represented by panning, depth by volume, and vertical space by EQ.
Photo: Chuck Dahmer
On its most basic level, the dimension of depth is derived from volume differences between the mix elements: the louder a track is turned up, the closer to the front of the mix it seems to be (ambience also plays a role in the perception of depth). The dimension of height can be understood much as it is heard (and felt) on a large P.A. speaker stack, with treble at the “top” of the sound, bass at the “bottom,” and midrange filling in the middle (see Fig. 1).
Being aware of balance in all of these three dimensions is a useful way to build an artful, advanced mix and start thinking and listening in 3-D. I think that most engineers would agree with me on the following points regarding how to address and structure each dimension for maximum spaciousness in the final mix.
Panning should be roughly symmetrical in terms of placement and frequency content. The more instruments in your mix, the more you can fill out the horizontal spectrum. For a solo performance, keep things centered; for duos and trios, avoid the extremes of hard-panning unless you're going for a dramatic effect. In denser mixes, use the entire left-to-right continuum, without bunching up a lot of center-panned mono information or panning so many tracks that a center-image hole is created.
FIG. 2: This shows the panning and perceived depth of the instruments and vocals in a hypothetical rock mix.
Photo: Chuck Dahmer
In full-sounding arrangements, the elements in the center of the mix tend to dominate, glue the mix together, and blend with other instruments (see Fig. 2). Conversely, tracks that are panned toward the outside of the mix are heard more discretely and do not blend well.
A special note about bass and kick drum relationships: in the days of vinyl LPs and singles, there were compelling reasons (related to vinyl mastering) to keep the bass and kick panned in the center. In the digital age, panning the kick and bass together still gives a satisfying low-end punch for many styles of music. But in 3-D thinking, panning the kick and bass apart slightly in a mix will make it easier to hear each of these instruments distinctly, and therefore increase clarity in the low end.
Exploring the Depths
The first rule of mixing is to make sure that every instrument in the mix is audible and presented at a relative volume level that is appropriate for the particular musical style. Certain elements have to be in the foreground and louder than other instruments, such as the vocal on a radio-oriented pop or jazz song, the guitar solo on a rock or blues record, or even the echo return on a classic reggae-dub piece.
As a mixer, I never hesitate to let the primary foreground element(s) be the loudest part of the mix so they make a strong and musically focused statement that will guide the listener's ear. One tried-and-true way to do this is to bring up the most important track(s) first and get that lead instrument or vocal sounding exactly the way you want. Then you can arrange the backing tracks to support the foreground element and enhance your sense of spaciousness.
Don't make all the mix elements equally loud all the time unless the musical style (loud rock, classical ensembles) dictates doing so. And try treating background parts (especially hand percussion, keyboard pads, and simple chordal instruments) as a pinch of spice in a mix rather than as an identifiable flavor, and position them accordingly. Depth is a wonderful quality to preserve in a mix. And depth can be accomplished only by establishing a decisive hierarchy of foreground, midground, and background relationships — determined primarily by relative level.
Another way to call attention to a certain instrument is to undermix it or intentionally place it in the background. As the old saying goes, if you want to get someone's attention, whisper instead of shout. For example, literally reducing a vocal performance to a whisper — either in performance or with radical EQ or distortion — is an effective technique to draw the listener into the interior space of a mix.
Although it is rarely considered until mastering day comes around, the vertical frequency aspect of your mix is an equally important space consideration. To generalize, frequencies in the bass range (20 to 320 Hz, four octaves) provide power and physicality. The midrange (320 Hz to 2.5 kHz, three octaves) yields tone and harmonic and chordal information. The treble range (2.5 to 20 kHz, three octaves) provides intelligibility and clarity to the power and tone ranges. Treble information also enables your ears to localize sounds in the stereo (horizontal) spectrum.
One of the biggest challenges of mixing is to provide a consistent and pleasing balance of frequencies that will translate well to the variety of listening systems out there. While high frequencies are very exciting and lots of low end is physically stimulating, listening to a speaker with a blown woofer or tweeter quickly reminds us that there has to be a balance of frequencies for any music to work.
Many home-recorded projects have inequalities in the overall frequency spectrum. This is most often attributable to inexperience, substandard mixing environments, or both. Mixes with too much low end or not enough highs usually don't convey a sense of air or space, and therefore fail to keep a listener's interest. Overly trebly or bass-lean recordings are initially stimulating to the ear, but too much treble can quickly induce listening fatigue and mask the subtleties of tone, depth, and space.
Fortunately, lo-fi recordings with too much midrange seem to have died out with the 4-track cassette and are now relatively uncommon. Much more common is the opposite case — the use of the “smile curve,” named after the segmented visage of a graphic equalizer in which the bass and treble extremes have been boosted and the midrange turned down.
Due to the ear's nonlinear response at quiet volumes (which is demonstrated in the Fletcher-Munson curve), the smile curve approach enables mixes to sound full on small speakers and at low listening levels. However, on a good pair of monitors, a mix that relies too heavily on smile curve aesthetics risks both the low- and high-frequency-based flaws just mentioned.
Filling the Spaces
By now it should be clear that the balance of instruments, imaging, and frequencies lays the foundation for a spacious mix in three conceptual dimensions. It's also important to realize how a change within one of these dimensions can affect and interact with the other dimensional axes in the mix.
For example, shifting the frequency range of a track with EQ not only moves it vertically but also moves it between background and midground positions. Changing the panning can have a similar effect. The simple act of turning a track up or down to change its depth position transforms the frequency balance and affects imaging as well.
Thinking and listening in 3-D like this makes the space between the speakers more of a complex challenge to structure and fill. Not only are there more interactive elements to consider, but with greater listening experience, the gradations of imaging, foreground-to-background relationships, and frequency become finer (see Web Clips 1a and 1b). Ideally, however, these concepts also increase the fun factor.
So far, I've discussed mixing only raw tracks and haven't brought in any of the other tools that define the art of mixing. To conclude the first part of this article, I'll look at reverb and room sound and how these elements can be used to either create space or increase clutter in a mix.
Perhaps you remember this science experiment from your school days: a container is filled with marbles, and when the teacher asks if the vessel if full, chances are that some members of the class will answer yes. Then sand is poured in, and the sand fills up all the spaces between the marbles. Is the container full now? Don't raise your hand just yet. Water is then added to fill the remaining spaces between the grains of sand.
Like the water in this demonstration, digital reverb is a powerful tool that, if overused, can fill up all the available space in a mix. In 3-D thinking, it's easy to visualize that stereo reverb inhabits not only the horizontal and depth dimensions, but also the vertical frequency spectrum (most audibly in the treble and midrange). By sustaining any or all sounds that already exist in a mix, and doing so in all three dimensions, reverb has the potential to be what I call a primary space sucker.
Of course when used tastefully, reverb can glue a mix together, add a lush professional sheen, and contribute a unique feeling of palpable space. To achieve this ideal of artful reverb use, two things have to be learned: basic reverb parameters and restraint. For those who need a refresher course on reverb settings and usage, see the online bonus material “Reverb 101” at emusician.com.
The Long and Short of It
While reconstructing a mix with clients, I commonly find long hall or plate reverb settings used across the board, and on many more tracks than is necessary. On snare, tambourine, rock guitar, bass, and keyboards, among others, these long-decay reverbs create a murky, sibilant wash that eats up definition and overwhelms background subtleties. When questioned on this practice, the engineer often has no real explanation for why a long reverb was chosen; it just “sounded good” or was a default plug-in setting. For instance, Digidesign's D-Verb plug-in defaults to a large hall setting when opened.
FIG. 3: Using a reverb with a decay time of less than 1 second—such as this setting on Lexicon''s Pantheon plugin—helps keep reverb wash from cluttering up a mix.
Employing short reverbs keeps the background clear, which also improves imaging and foreground definition. My personal default setting for producing spacious, uncluttered mixes is almost always an ambient or small-room reverb, with a decay of a second or less for most instrumental tracks (see Fig. 3).
Up-tempo vocals also get the ambient or room treatment a lot, often in combination with lively sounding gated reverb or chamber algorithms. In these cases, the decay time can be longer than 1 second, depending on stylistic considerations and mix density.
I tend to use longer concert hall or plate settings only for certain types of ballad or operatic vocals, bowed violin or cello, sustained organ sounds, or any dramatic foreground element that would benefit from a lush, lonely, or haunting reverb tail. Of course, the presence of one richly reverberant track makes it tempting to start sneaking up the reverb level on everything. The end result of this is often a soupy mix that mimics the uncontrollable wash of a school gymnasium.
To keep this from happening, I often try to balance out a lushly reverberant foreground or midground track with other very dry mix elements while keeping an ear open for the symmetry of wet and dry tracks in imaging. This helps remind me to retain some space in a mix. And the distinctive contrast often heightens the dramatic impact of the wet elements.
When employing long reverbs, a judicious use of predelay helps to preserve the detail and transients of a track. Think of predelay or delay-time settings as a “dry window” that can let the crucial first 10 to 50 ms attack of a word, note, or drum hit slip through without reverb. A predelay of longer than 70 ms is perceived by the ear as a slapback echo, and, though useful in some cases, can also be heard as a distinct, often fluttery, effect.
Once you have the reverb in your mix just the way you want it, print the mix for safety's sake (or save the DAW file and then save it with another file name), and then try a little reverb restraint to create more space and contrast between wet and dry elements. Try turning the reverb down on an instrument in the rhythm section or background, shorten its decay time, or even turn it off. That move might either suggest another similar reduction or inspire a change to the vocal or other foreground reverb. Once you get over the habit of thinking that more reverb is better, you might just like the clarity and spaciousness of your new, drier mix. If not, you can always go back to your first mix or try adding a little more reverb in mastering.
And remember, there's no rule that says you have to use reverb on every track, or even at all. I rarely use digital reverb on kick drum, bass, or any predominantly low-end material. In addition, I religiously avoid the use of long hall or plate reverbs in any percussive track unless it's for a particular effect.
Using a mild chorus on chordal instruments or a slapback delay on vocals makes it possible to mix without artificial reverb in some circumstances — particularly with dense mixes in which reverb tends to swallow up subtle details.
Roomy from the Start
You can greatly reduce your need for digital reverb by using room mics and creative miking to create a sense of space when tracking. Because this is a topic I have covered in previous EM articles (see “Underground Drum Sounds” in the July 2001 issue and “Tracking in the Unplugged World” in the November 2003 issue, both available at emusician.com), I'll just throw out a few ideas here.
A room mic placed 5 to 10 feet away from an overdubbed source (or 50 feet away, if you have a room that big) will yield a unique timbre and ambience that standard reverb plug-ins and outboard processors can't duplicate. Depending on your taste and how adventurous your spirit is, that room mic can be mixed in to be barely audible, equal in level to the close-miked source, or used instead of a close mic. In a DAW, any room mic track can always be nudged to be farther in time from the original track, creating the illusion of a larger room. Keep in mind that a nudge of 1 ms amounts to a virtual distance increase of roughly 1.125 feet between the room mic and the source (based on the speed of sound as 1,125 feet per second), giving the illusion of a larger room.
Adding reverb to the room mic or using it only as a reverb send yields an interesting variety of predelay and coloration. Unusual mics (such as figure-8 ribbons, cheap lo-fi models, and omnidirectional models) and unconventional placement (on the floor, inside a metal tin, facing a windowpane) can add additional colors to your sonic palette.
As exciting as these ambience-miking ideas can be, they can also be big space suckers in a mix due to overwhelming bass or midrange coloration. I use low-cut filtering liberally on room mics (and reverb returns) to avoid adding muddy frequencies to my mixes.
In my experience, most free reverb plug-ins and inexpensive outboard reverb processors have high coloration and can be overwhelming as soon as they are audible in a mix. If you feel the same way and find yourself with a mix that has no room mics and no decent digital reverb options, try reamping for the room. This entails running selected tracks (or the whole mix if you want) through a clean guitar amp or monitor speaker, placing a mic in front of the speaker, and recording the result for use in your mix. (For more on reamping, see “Better Tone Through Reamping” in the October 2008 issue.)
Next month in the second part of this article, I'll discuss, among other issues, how compression, EQ, and a song's arrangement can impact the perception of space in a mix.
Myles Boisen (mylesboisen.com) consumes significant quantities of space and time at Guerrilla Recording and the Headless Buddha Mastering Lab in Oakland, California.
Tips for a More Spacious Mix
- Conceptualize your mix in three dimensions: horizontal (left to right panning), depth (front-to-back or relative level), and vertical (EQ — with highs being up, lows being down, and midrange in the middle).
- Pan the elements in your mix in a roughly symmetrical fashion, both in terms of stereo placement and frequency content.
- Try moving the kick drum and bass a bit off center to remove clutter from the center.
- Keep primary elements loud enough to be in the foreground. You might want to bring them up first when starting your mix, and then work the other parts around them.
- Consider mixing background elements — such as hand percussion, pads, and some rhythm instruments — low enough so that they're more textural than up-front.
- Reduce reverb decay times to 1 second or less where possible to limit excess ambient wash. Try using delay or chorus instead of reverb. Use room miking to create natural reverb when tracking (or after the fact in a reamping scenario).
Pitch-a-Tent Records' Site
Mag Net Ic Site