Part of listening in the real world involves locating sounds in space. Although 5.1 surround sound excels at providing spatial cues, most of the music world still runs on stereo, which has been with us for four decades. While stereo's spatial options are limited to left, right, or somewhere in between, it remains an important part of recording -- and we need to make the most of it. Consider what happens when you sit down and play the piano. There's a definite sense of lower notes emanating from the left, and higher notes coming from the right. Also, the undamped high strings, which are on the right, ring faintly even when you play a note on the extreme left. Try recording a piano in mono, then again in stereo. If properly miked, the stereo version will be far more realistic. Stereo placement (panning) isn't just about realism, though. It's also used to keep instruments from interfering with one another, as well as to add special effects. So, let's look at some tips designed to help further your skills in the art of stereo.
Audience Perspective or Performer Perspective? As you set up stereo placement for instruments, think about your listener's position. For example, for a right-handed drummer the hi-hat is on the left and the toms on the right. For the audience, it's the reverse. I generally go for the performer's perspective, unless the object is to emulate a concert experience. Frequency Response & Panning Low frequencies are fairly non-directional, whereas highs are very directional. As a result, low-frequency sounds (kick, bass) generally sound better in the center of a mix, whereas higher frequency instruments (shaker, tambourine) can go further out to the left and right. With synthesizers that have stereo outputs, keyboard split functions can be very useful for spatial placement. One option (especially if you're handling the bass line with your left hand) is to send the lowest split to the center, a middle split to left of center, and the top split to right of center. This is not necessarily the most realistic option with respect to imitating the real world, but hey, it's a synth. If there's also a bass part to contend with, then move the bass to center, and spread the keyboard from left to right (or right to left), going from lower keys to higher keys. This keeps your low frequencies spatially separated from the bass player. The easiest way to spread a keyboard is if you can modulate the pan position from the MIDI note number. This creates a spread when you feed the keyboard's stereo outs to two mixer channels. However, unless the keyboard is the major focus of the music, you may want to narrow the final range down a bit. For example, if guitar is another major melodic instrument in the piece, try spreading the guitar from left of center to center, and the keyboard from center to right of center. Panning & Delay When a delayed sound is placed in the same spatial location as the main sound, the end result may become somewhat indistinct if the echoes are on the same note that's currently playing. There are two main options: If your instrument is weighted to one side of the stereo spread, weight the delayed sound (set to delay only, no dry signal) to the other side of the spread. If you're using stereo delay on a lead instrument that's panned to center, you can get some lovely results by panning one channel of echo toward the left, and one toward the right. If the echoes are polyrhythmic, this can also give some lively ping-pong type effects. Of course, this may sound gimmicky (not always a bad thing, mind you!), but if the echoes are mixed relatively low and there's some stereo reverb going on, the sense of spaciousness can be huge. Plan Ahead One way to pan is just to arbitrarily move panpots around until things sound good. But I prefer to plan ahead by drawing a two-dimensional diagram of the intended "soundstage," much like the way theater people draw chalk marks where characters are supposed to stand. When it's time to mix, using this diagram as a "map" helps me stay on track. The further back I want to position the sound, the lower in level and, sometimes, the more reverb. Closer sounds are louder and drier. You can also experiment with adding predelay to the reverb for a sound that you want to seem close to the listener. This works best when each of the main sounds has its own reverb. Be careful with adding one reverb to the entire mix: When done carefully, this can give a sense of space, but it's easy to overdo it, drowning your mix in mush. When Mono Matters There are three reasons to pay attention to mono. One is that the final product may play over a mono system (well, you do want to hear your music on TV, right?). Another is that some methods that create stereo from mono signals, or ultra-wide signals from stereo, can create mono incompatibility problems. But the main reason for using mono is when you begin a mix: I suggest starting with all instruments panned to center. Don't do any stereo panning until you've used basic EQ with all instruments to make sure that each part stakes out its own part of the frequency spectrum. For example, if two keyboard parts clash in mono, initially use EQ to differentiate them, not stereo placement. If you're able to differentiate parts with EQ first, then stereo becomes the element that brings true spaciousness, not just separation, to a mix. Creating Wider-Than-Life Sounds Many signal sources are still essentially mono (voice, vintage synths, electric guitar, etc.), but there are ways to "stereoize" sounds. With a hard disk recording system, the easiest is to copy a track and "slip" it ahead or behind the original track to create a slight delay between the two (see Figure 1 above), then pan the two tracks to opposite sides. In some cases, it's most effective to slip the original track ahead of the beat a bit and the copy a little late, so the two end up "averaging out" and hit in the pocket. But you can also use slipping to alter the feel. To drag the part, keep the original on the beat and slip the copy a little later. For a more insistent, on-top-of-the-beat feel, slip the copy ahead. How much slip to add depends on the instrument's frequency range. If the delay is too short, the two signals may start to cancel each other and create comb-filtering effects. This results in a thin, peaky sound, much like a flanger stuck on a few milliseconds of delay. Lowering the copied signal's level can reduce these negative effects, but then the stereo image will be less dramatic. If the delay is too long, you'll hear an echo effect. This can also be useful in creating a wider stereo image, but then you have to deal with the rhythmic implications -- do you really want an audible delay? And if the delay is long enough, the sound will be more like two mono signals than a wide stereo signal. Thankfully, hard disk recorders make it easy to experiment. Just be sure to check the final result in mono. If the sound ends up being thin or resonant, increase the delay time a tad. Another way to create wider sounds involves EQ. Split your mono signal into two channels, pan one channel left and one channel right, then give each channel its own EQ settings to create more of a spread. For example, if you use a high-shelf filter to boost 4dB starting at 2kHz in the right channel, cut the left channel by 4dB at 2kHz, using the same type of filter. As mentioned earlier, highs are more directional, so this will definitely tilt the channel toward the right of the stereo image. For a more extreme image, cut the bass in the right channel and boost it by an equivalent amount in the left channel. This will "push" the image further into both the right and left channels. Stereo graphic EQs can also create a wider image. As one example, set up one graphic for full cut on even-numbered channels, and pan that to the right. Set up the other graphic for full cut on odd-numbered channels, and pan it to the left. This won't work successfully on instruments with limited ranges, like voice or a lead synth part. But if you're exercising all the keys on your 88-note controller, or using a drum machine, this technique can give a very unusual type of stereo imaging. Well, that's probably enough about panning for now. The point of all this is that every available parameter is an important part of shaping your sound. Spatiality is crucial, so put as much thought into placement as you do into levels, EQ, and other elements of mixing.
CRAIG ANDERTON is creative director for MusicPlayer.com (stop by his forum and say hi), and guitarist/synthesist with the German group Rei$$dorf Force. He's also author of the books Home Recording for Musicians and Multieffects for Musicians (AMSCO).
|
