Having been a music director for a couple of open source game projects, I’ve reviewed many tracks made by people who are new to virtual orchestration. Two of the most common issues I’ve heard are: 1) people having little or no idea how to position orchestral sections within a mix, and 2) people underestimating the importance of reverb.
For that reason, I figured I should post some guidelines on how to approach this aspect of working with a virtual orchestra. It’s not a definitive how-to and it doesn’t have step-by-step instructions. It should be considered a bunch of tips for getting you on the right track. Nothing more, nothing less.
I will be taking some things for granted here. For example, I will just assume that the reader is familiar with basic concepts like send effects and EQ and what different reverb parameters do. If you’re not clear on what something means, Google it. Comments like “what is a reverb?” or “what DAW should I use?” will be politely ignored. My main concern is to get into the juicy details and provide some tips for people who need them right now. In other words, this should be considered intermediate stuff.
When reading this guide, it’s good to keep in mind that I am not an audio engineer; I’m just a musician with an interest in recording technologies. So don’t take anything stated in here as a universal truth, because I’m not an authority on the subject. Most of the stuff I talk about here is the result of my own experimentation and knowledge gleaned from articles by other people.
Also, always remember that if it sounds right, it is right. There’s more than one way to skin a cat, as the old saying goes. Be open to new ideas and learn as much as you can from others. But don’t let me or anyone else tell you how to mix your music. Have fun. Be creative. Think outside the box, and learn by doing.
Got that? Good. Let’s get started.
If you’re new to music recording, it’s deceptively easy to think of a mix as simply multiple sound sources playing back at the same time — because that is, basically, what it is. And make no mistake, a completely dry recording in mono can sound great if mixed by someone who knows what he or she is doing. But in most cases, all mixes rely on stereo positioning and various psychoacoustic effects to achieve a wider and deeper sound, partly for minimizing the risk of running into frequency cancellation and other issues, but mostly for the plain ol’ reason that it sounds better.
Our brains rely on direction and distance to pinpoint a sound source, and if a recording lacks this information, it doesn’t sound “natural” to us. That’s why you shouldn’t think of a mix as a bunch of layered tracks. Instead, imagine that it’s a three-dimensional space. Your job is to position all instruments along the X and Z axes of this space so that they are out of each other’s way while still sounding like a whole. Your two main tools for achieving this are panning (left-to-right placement) and reverb (front-to-back placement).
Compared to getting the reverb right, panning is quite straightforward. You move the pan knob (or slider) left and the sound moves to the left; you move it right and the sound moves right. If you leave it sitting in the middle, the sound will remain centered, equally loud in both speakers. Anyone should be able to grasp the basic principle behind panning right away. When, why, and by what amounts you should use panning is trickier, though.
Panning and orchestral seating
When it comes to panning sections within an orchestra, “less is more” is an excellent adage. Panning the 1st violins hard left and the basses hard right might sound impressive at first, but if you’re aiming for at least some amount of realism, it’s not the way to go — especially if you’re using a generous amount of reverb (which I think you should — more on this later).
The laws of physics dictate that the farther away you are, the less you hear the direct sound, the more you hear the sound’s reverberations, and the closer together the left and right sound sources appear. So violins hard left and basses hard right, both with a lot of reverb on them, doesn’t make any sense. This sound information tells our brain that we are standing between the two sections, and that they are spaced a huge distance apart. If you have ever seen an orchestra perform, you know that this is not normally how violins and basses — nor the listener, for that matter — are positioned.
Which brings us to seating. Long story short, the players in an orchestra are seated the way they are for a reason, and unless you know what you’re doing you should be careful about tampering with this. The typical orchestral seating plan has evolved over centuries, and while it’s definitely not set in stone, you can’t just position sections willy-nilly and expect it to make sense to the listener. To achieve a good balance in terms of volume, loud instruments like brass and percussion traditionally go in the back, with woodwinds and strings in front of them. To avoid clashing, sections with similar ranges/timbres are positioned opposite each other in the binaural field. For example, violins left, trumpets right; horns left, celli right.
Of course, positioning instrument families according to loudness is a completely moot point in a virtual environment, when by moving a couple of faders, we can make a pp woodwind passage clearly audible over an accompaniment of ff brass. But even if it’s fine to take liberties with realism, it’s important to remind ourselves that our goal here is to make orchestral music, and thus, we need to respect at least some basic rules or it won’t sound much like orchestral music anymore.
Positioning sections “correctly” will simply make your music sound more convincing, and that’s a point you really can’t argue with.
A rough panning example
When I set up an orchestral template, I usually position everything something like this:
1st violins – halfway left
2nd violins – less than halfway left (close to 1st, but you should be able to tell them apart)
Violas – center or slightly right
Celli – less than halfway right
Basses – halfway right
Trumpets – one third right
Horns – one third left
Trombones/tuba – halfway right
Flutes/clarinets – slightly left
Oboes/bassoons – slightly right
Percussion – timpani and bass drum centered or to the left (so as not to clash with celli/basses), anything else to taste
As you can see, this is very vague.
I can’t give you any exact numbers because it depends on what samples you’re using. Some samples have a wide stereo image and need to be panned more aggressively to sit well with the rest of the orchestra. Sometimes it might even be necessary to use your DAW’s stereo width control (or a stereo imaging plugin) to make the section narrower. I sometimes do that with violas, as they are normally centered in terms of seating, and if the stereo image is too wide, it’ll sound like you have viola players all over the place.
Likewise, I usually “mono-ify” my woodwind sections. That is, I turn down their individual stereo widths to zero because these sections are small, sit close to the center of the orchestra, and are not as loud as the brass nor as diffused as the strings. To be clear, I’m not talking about mono as in centering the woodwinds. I always keep flutes and clarinets slightly left off center and oboes and bassoons slightly right.
But I do not want the individual sections to be in stereo, because at the listener distance I’m aiming for, you would not be able to hear, for example, section flutes as separate instruments with discrete positions anyway. To each his own, of course.
Some composers like to have 2nd violins to the right and/or the basses centered, as low frequencies are less directional than mids/highs, and we’re used to hearing centered bass in popular music. Some sample libraries (or at least one: Miroslav) have the horns on the right together with the rest of the brass.
If I were to alter something in the traditional seating, I would probably put trombones and tuba to the left. That way you’d have high strings + mid/low brass to the left, and high brass + mid/low strings to the right. It theoretically makes sense, but like I said, there are usually good reasons for instruments sitting where they are, even if they are not obvious to a hack like me. This is, after all, a genre of music that’s been around for hundreds of years, and has been explored and refined by minds (and ears) much greater than mine. So playing it safe and going by the old “if it ain’t broke, don’t fix it” rule is probably a good idea for most of us.
If you really want to try a more unorthodox seating plan, by all means give it a shot. But make sure to put some thought into the whys of it all, and whether it would make sense in the real world.
Putting things in place
What’s most important here is that you let your ears guide you, not your eyes. A DAW will always give you very detailed visual feedback, and this can sometimes be misleading. If the panning readout says the 1st violins are at 50% left and the 2nd at 40%, yet they sound like they’re in the same spot — or even reversed — ignore the readout and adjust the panning until it sounds right. Your DAW only reports what it’s doing with a signal; it has no idea what one signal actually sounds like compared to another. That’s up to you.
But I hear you asking, “How do I know what ‘right’ sounds like?” This is where reference music comes into play. Dig out some movie soundtracks or classical recordings that you like. Use a good pair of headphones, close your eyes, and try to picture the concert hall and all the players in front of you. Pay special attention to how the sections are positioned relative to one other, and their size (e.g., a section of three flutes will sound smaller than a 16-piece violin section, even though their loudnesses are balanced).
You will quickly notice, if you didn’t already know, that orchestral recordings can sound as radically different as pop/rock albums. Some are close and intimate-sounding, to the point where you can pick out the rasp of the bows and key clicks from solo brass and winds. Others are huge-sounding, distant, and somewhat blurry.
Even if you prefer the latter kind and want your music to sound that way, I recommend digging into the “intimate” recordings. For example, I heartily recommend all the ’70s and early ’80s John Williams scores (Jaws, Close Encounters, the original Star Wars Trilogy, the first two Indiana Jones movies). These are great for learning positioning, as they are pre-“wall of sound” Hollywood soundtracks that you’d normally hear today.
Learn more about modern mixing production techniques (like EQ, compression, level, pan setting, digital signal processing, FX sends, and so much more) from some of today’s leading sound engineers, and get your record sounding crisp! Preview Soundfly’s newest and most in-depth mentorship-assisted online mixing courses, Modern Mix Techniques and Advanced Mix Techniques, for free today.
Mattias Westlund is a Swedish musician and composer specializing in soundtracks for computer games and other media. He lives in Gävle, a seaside town in eastern Sweden, with his daughter and three cats.