Musician, October 1998, Issue No. 239, by Howard Massey

The technical and aesthetic basis for setting up your home studio to mix in 5.1.

Last month we described the requirements for surround sound mixing in a home studio. This month, we'll talk about wiring interconnections and speaker placement and then join veteran producers Elliot Scheiner and Bob Margouleff for a discussion about artistic approaches to surround mixing.

The Setup

The interconnections for a surround-sound rig aren't particularly complicated if you keep in mind the main goal: to route signal freely to any or all of the six channels and to route each of the six channels to its own dedicated speaker. There is no set bussing scheme, but the one used by the 02R is: Bus 1 = left front; Bus 2 = right front; Bus 3 = left rear; Bus 4 = right rear; Bus 5 = center; Bus 6 = subwoofer. (In film surround mixing, this is called the LFE channel, for "low-frequency effects".)

Accordingly, you'll be sending line-level signal from each of these six busses to both the associated track on your recorder and to the power amplifier that drives the equivalent speaker. Assuming you're going to use the 02R scheme, the bus 1 OUT of your mixer therefore needs to be connected to both the track 1 IN of your MDM or hard-disk recorder and to the power amplifier channel that drives the left front speaker.

There are several ways to accomplish this dual routing. First, you can use Y-cables, signal splitters, or patch bay mults to send the signal from each bus to both the recorder and power amplifier. Alternately, you can connect the bus solely to the recorder and then route the recorder's outputs to the appropriate speakers (i.e., bus 1 OUT to track 1 IN and track 1 OUT to the input of the first power amplifier). With this scheme, during mixdown you'll have to set the recorder to monitor the input signal arriving at each of the six tracks.

An even better solution is provided by the 02R and other digital mixers that have provision for expansion cards. These cards allow a single bus OUT signal to be routed to multiple destinations. For my rig, I installed two single-slot ADAT interface cards and one dual-slot analog interface card into an 02R. I then made fiber optic connections between the ADAT cards and two ADAT-XTs. This allowed signal from tracks 1-8 of the first machine and tracks 7-8 of the second machine to be input to the 02R, and at the same time allowed my surround mix to be output to the remaining six tracks (tracks 1-6 of the second machine). Next, l routed the analog card bus 1-5 OUTs (which are balanced line-level analog) to the five Genelec 1029As. Finally, I plugged a balanced Y-cable (actually just a two-into-one stereo headphone adapter purchased at Radio Shack for $7.95) into the analog card's bus 6 OUT and ran cables from the adapter to both Genelec 1091A subwoofers. With this setup, channels assigned to bus 1 would be sent simultaneously to track 1 of the second ADAT-XT (in the digital domain) as well as to the left front speaker (in the analog domain); channels assigned to bus 2 would go simultaneously to track 2 and to the right front speaker, and so on.

This makes, however, for an interesting dilemma: how to set the recording levels and monitoring levels independently. Fortunately, the 02R provides an elegant solution to this problem, even though it meant losing my external aux sends. (Other mixers may provide alternate routing schemes to the one described below; if all else fails, you can simply use the volume controls on your power amps or self-powered speakers to set monitoring levels.) To set recording levels, I used the 02R's onscreen bus master faders (Meter 3/3). To set independent monitoring levels of those tracks, I went to the "Slot Output Select" screen (Digital l/O 3/5) and switched analog card outputs 1-6 from bus 1-6 to aux sends 1-6. Then I assigned the tape returns from the target recorder (in my case, the second ADAT-XT) to the aux sends instead of the stereo or bus outs-track 1 to Aux 1, track 2 to Aux 2, etc., all at +0.0 d B. Group all six faders together for master monitor volume control, and you're in business.

What's particularly cool about the Genelecs is that each of them has dual inputs. This allowed me to use the left and right front speakers--along with both subwoofers--for stereo monitoring as well. I simply connected the 02R's master stereo line OUTs to the second input of each of the four speakers, again using Y-cables to split the signal so that the left master stereo OUT ran to both the left front 1029A and left 1019A subwoofer and the right master stereo OUT ran to both the right front 1029A and right 1091A subwoofer. To avoid speaker overload, you have to take one precaution with this particular setup: Make sure that all input channels are assigned either to the stereo bus or to individual bus outputs, never to both.

Needless to say, speaker placement for surround monitoring is important, though not nearly as critical as some audiophiles would have you believe. Let's deal with the five "main" (i.e., nonsubwoofer) speakers first. Ideally, you want each of them to be at the same height (ear level is best) and at an equal distance from your normal mixing position, although if this isn't possible you can compensate by making distant speakers slightly louder (by turning up their power amp) and/or by slightly delaying the signal arriving at closer speakers. The basic rule for this delay is about one millisecond-that's one thousandth of a second-per foot. This means that if your rear speakers are six feet away and your front speakers are four feet away, you should delay the signal to all three front speakers by about 2 ms. This is easily accomplished in digital devices like the 02R or ADAT-XT, where individual channels or tracks can be delayed precisely.

In practice, I found that slight distance variations like this are virtually imperceptible. Basically, as long as all five main speakers are roughly the same distance away, you're not going to have any audible problems. The vagaries of my home studio meant that my rear speakers were about three feet further away than my front speakers and slightly off center as well. By simply making them a little bit louder than the front speakers, I was able to achieve a good listening balance without having to delay the front speaker signals at all.

The front three speakers should be placed in an arc, with the left and right speakers toed in at a 30-35° angle. (Finally, a use for that $1.98 plastic protractor you were forced to buy in high school!) If possible, the rear speakers should be angled in as well, at about 110ø relative to the front center speaker (see diagram below). This is less critical if you have reasonable acoustic absorption treatment in your room.

Subwoofer placement is a bit trickier. Bob Margouleff comments, "The low frequency information needs to fill the room. It's like liquid: It fills the space and oozes everywhere, so you want to make sure the subwoofer is placed where the room itself helps support the resonant qualify of the low end." Margouleff has devised a unique technique for doing so. "I start by identifying the sweetest spot in the room. Then I put the subwoofer on its back right in the middle of the sweet spot. I put pink noise or an 80 cycle sine wave through it, and then I walk around the edge of the room until I hear where the subwoofer sounds loudest. That's the best spot for the sub."

If you're using two subwoofers, Margouleff cautions against facing them towards each other--a "sure way" of getting phase cancellation problems, he tells us. "In a modest space, you can place two subs right next to each other. If you're in a bigger space, they can be on different walls on the edges of the room. You want to get the bass energy as dispersive as possible, but I don't believe in firing them into the wall or floor since, in pop music, they provide considerable low-frequency musical information, not just rumble."

The Mix

The best way to learn about surround-sound mixing is to listen to some of the work that's out there. The least expensive way to do that is to buy, borrow, or rent a CD player that has a digital output and then connect it to a DTS/AC-3 decoder. (Basic models from Yamaha and other manufacturers have a list price of under $400; if budget permits, you can spring for more expensive but considerably more full-featured audiophile products such as the Lexicon DC-1.) There are dozens of DTS-encoded CDs available; they're a bit more expensive than standard CDs (about $20 each), but some hip hi-fi stores are now renting them for just a couple of bucks a night. Among the selection are Steely Dan's Gaucho and the Eagles' Hell Freezes Over (both remixed by Elliot Scheiner), Boyz II Men II (a Margouleff remix), and classic titles from artists as diverse as Clapton, McCartney, Vince Gill, Chick Corea, and Patrick Leonard. If you own a DVD-video player (which can also be rented on a nightly basis from many video stores), you can connect it to the same decoder, set it to AC3 mode, and listen to movie soundtracks galore as well as the few all-music DVD titles starting to trickle out.

What you'll hear is not only a new degree of spaciousness-which is to be expected- but also a great deal of clarity. The explanation is simple: By spreading a complex musical signal among six speakers instead of just two, you're not asking each individual speaker to work as hard. As a result, each sound wave is generated with fewer frequency components and correspondingly fewer phase cancellations.

You'll also hear a lot of different artistic approaches. In some mixes-especially those of live performances--the rear channels carry ambient room information only, which makes you feel like you're sitting in the audience and watching a performance onstage, with only some back-wall reflections coming from behind you. In others, the instruments and vocals are panned all around you, making you feel like you're sitting onstage in the middle of the band. Some remixes of all-studio recordings take a very clinical, static approach, where each instrument sits stubbornly in its own space from start to finish. Others--particularly more experimental recordings such as Alan Parsons' On Air--feature audio which is fluid and moving, with constant sonic surprises springing up all around you.

Another advantage of listening to surround-sound releases is that it gives you the benefit of 20:20 hindsight. Veteran producer/engineer Elliot Scheiner tells us, "The Eagles' Hell Freezes Over surround mix was my first one, and I experimented tremendously at their expense. I realized after the fact that I made a couple of mistakes in there that I haven't done since, like dedicating the vocal to the center speaker; that's a bad thing." Most surround veterans agree that the center channel is the hardest one to deal with aesthetically. Very often, the ghost image from signals panned dead center in the left-right speakers is stronger than what's coming out of the center speaker. Scheiner again: "In the Gaucho 5.1 remix, I sent the vocal to the center speaker as well as left and right. What I'm doing now is sending the vocal to left and right, and just bare minimum to center. I use the center channel for just a little bit of vocal, a little bit of snare drum, sometimes just a little bit of kick drum, and that's about it."

The subwoofer channel can also be problematic, especially if you send signal to it from too many sources. While the crossover in a good subwoofer will effectively eliminate all high- and mid-frequency information from the input signal, you're still giving it extra work to do--and thus risking the possibility of time-smearing and other phase-cancellation anomalies--if you send it gobs of full-range signal. The onboard surround mix capabilities of the 02R (added in version 2.0 software) allow you to independently set the subwoofer level of each channel; this is really helpful in taming over-aggressive mixes. For example, you can get greater separation between the kick drum and the bass if you send one mostly to the subwoofer and the other mostly to one or more of the main speakers.

These considerations aside, there really are no rules, so your creativity can--and should--run rampant. For example, try assigning a short reverb with a slight predelay to the front left-right speakers and a longer reverb with a longer predelay to the rear speakers. Then drive it with a snare drum sample assigned solely to the center speaker. The crack of the snare will seem to come from between your eyes, then quickly spread out, first in front of you, then behind you. Or send one guitar track to the front left and rear right speakers at slightly different levels, and a doubling track to the front right and rear left speakers.

Of course, you can always use your mixer's panpots to place signal half-left or half-right, but if your mixer is fully equipped for surround mixing (like, for example, the 02R), it will allow independent panning between any of the busses. This will let you do things like placing a sound halfway between the left front and left rear speakers so that it appears from alongside you. The 02R's surround controls also let you move a sound dynamically through a straight line, an arc, or a circle over time; this information can even be captured and included in your automated mix. There are also computer programs for the 02R (such as Zeep's Localizer) that enable you to draw your own paths freehand and then move any signal along that path with the use of a mouse (or triggered by timecode). Affordable joystick controllers for the 02R and other digital mixers should be available very soon; these will let you dynamically pan a sound within the surround space in real time with simple hand gestures.

This is heady stuff, but it's about time recorded music got a major facelift. If there ever was an opportunity for today's musician to get in on the ground floor and take total control of his or her music, this is it. In the words of Elliot Scheiner, "Surround sound is the future. When I did the Fleetwood Mac surround remix, none of those guys came down to the studio to hear what I was doing. On the very last day, I begged them to come down after rehearsal, so they did. I put up one of the songs, and they sat there listening. And their mouths dropped. After it was over, Lindsey [Buckingham] said, 'I don't think I'll ever be able to listen in stereo again."' Neither will you.

Special thanks to Bob Margouleff, Elliot Scheiner, Will Eggleston, Peter Chaikin, Jim Mack, Buzz Goddard, Greg Braithwaite, Michael DiCosimo, Dave DelGrosso, Robbie Clyne, Lisa Vogl, and Hal at Zeep.