Half Annual Sale: All Plugins $29.99* | Shop Now »

Backstage with Leon Rothenberg

Jan 15, 2009


Tell us about doing sound for theater.

I often find myself needing to do the same thing for a lot of playback cues, like a particular EQ problem or a particular volume leveling issue for which you need something like MaxxVolume. That’s one scenario where you want to be able to batch process a lot of things fairly quickly. The other scenario is that you have a specific cue that has a problem that needs to get fixed, and you have one chance to do it. So to have a utility that allows you to load up a sound, load up a plugin, fix it, spit it back out, and get it back onto your playback machine without a lot of fuss is extremely useful for workflow.

Often you will just have moments to get something fixed, and you always have to decide, "Do I want to make everyone stop and wait for me, or can I just go with the flow". The more you can get things done fast, the more you can save your stopping for when you really need it, when it’s really important. Otherwise, you've gotta take advantage of that "Let's do it one more time" time, and fast workflow is the key.

From show to show, how much adjusting will you do?

From show to show for prerecorded cues, none. The only reason you would adjust something after the show has opened is if there’s a problem or a change to the show itself. For example, on South Pacific which we just did, there is an understudy for the male lead opera singer because he’s going on vacation for a few weeks. So we needed to go back and lay in new voiceovers because there’s a number of voiceover cues where he’s speaking from a radio, and we have to make sure the voices match. That’s really the only sort of thing, unless someone changes there mind on a particular point and a dramatic moment has changed or needs to change, but otherwise, once the show starts, the mixer mixes the show but you don’t really change the cues.


How did you get into sound design?

I came to sound design through music. I studied as a composer in undergrad, went to Oberlin Conservatory, and started writing music, just me, alone, with a piano and a pencil. In my fourth year one of my composition teachers was writing an opera based on a play that was going to have a staged reading, and asked me if I wanted to sound design it. That was the first time that I had heard that phrase. So I and one of the electronic music composers, we got together and did it. My teacher?s interest was in a composed soundscape for the play, because the opera was based largely on soundscapes. We did this sound design and came up with a whole bunch of sounds and effects and did a whole lot of recording and banging on bottles and large coffee cans and things like that, and put the whole thing together in a way that made sense to us, which turned out to be very musical.

Often you will just have moments to get something fixed, and you always have to decide, "Do I want to make everyone stop and wait for me, or can I just go with the flow". The more you can get things done fast, the more you can save your stopping for when you really need it, when it’s really important. Otherwise, you've gotta take advantage of that "Let's do it one more time" time, and fast workflow is the key.

What kind of gear did you use back then?

We were using Pro Tools which we recorded to cassette tape, reel-to-reel and DAT, because that’s what we had available. We needed a lot of playback, hence the three different kinds of machines. After that experience I started writing music for theater. It was still music, but it was writing it as part of a whole, something larger, as opposed to its own free-standing event. I really liked that, mostly because it meant that I got to work with other people instead of working by myself all the time, which was fun, and because I liked being part of a whole entity and supporting a larger piece musically.

So I did that for some time and when I finished school I promptly started a job writing internet software, which I did for a couple of years. Then I decided to go back to grad school at CalArts and study Sound Design from ‘99 to 2002. The dot com bubble was fun and lucrative, but not what I wanted to do, sitting in the same office everyday and writing software instead of being in the theater and doing new things all the time. So I got out of that and got back into sound design and now here I am in New York City.


How do you prepare for a new show?

You start by speaking with the other members of the creative team and the director and reading the script, or whatever guiding text there is to work from. I just did a show that wasn't really a script, more of a series of vignettes that were to be ordered in some undetermined way. But whatever the material is, it could be anything from a series of paintings to a text to whatever, it all starts with creative discussions, bouncing ideas off each other, looking for a point of view.

Next thing is to design a sound system, and get it installed in the theater during the load-in period, which is anywhere from a couple of days to five or six weeks, depending on how large the show is. Then you have what’s called technical rehearsals for about ten days, where we go through the show, step by step, and add all the technical elements. The actors and the directors have been rehearsing the show separately in a rehearsal hall somewhere for the last 4-6 weeks, but this will be the first time they are interacting with the set, lights, costumes, sound, video and whatever else the show has.

Often the actors will wear wireless mics, and the engineer turns the mics on only when someone is talking. I don't have a lot of recent experience with Rock and Roll, but I think theatrical mixing is quite a different paradigm. The front of house engineer will have literally tens of thousands of cues during the course of a show, turning mics on and off. It’s literally as fast as: you say something, I say something, you say something, I say something--he’s going up and down, riding the faders. This requires the preparation of a show script with all the cues, and keeping it up to date. It's almost a full time job, just keeping track of the script.


How do you trigger sound effects?

Sound effects-wise, nowadays it’s all generally off a computer and it’s triggered by MIDI or a switch closure, but usually MIDI. Either some sort of big ‘Go’ button, or you can attach it to the automation in your console, so when you go to the next console scene, you can trigger a sound effect or music playback at the same time if you need to. And it’s always good to run two computers redundantly, in case one goes down.

What would happen if you blew a sound cue?

Generally, you fire a cue on a particular line or a particular word or a particular moment. Somebody steps their left foot downstage, or in the middle of the word "Mer-cutio", or something along those lines. In the cases where you have something like thunder and it’s coordinated with a light cue, you might have either the lighting console trigger the sound or vice versa, so that they’re in perfect sync. Otherwise there’s almost always somebody on the button, pressing ‘Go’ each time they get a cue from a stage manager, who follows the show with a script and tells everybody what to do when. This central command model is what keeps everything synchronized. But if you blow a cue, you just have to go on--hope you get it right tomorrow night.


At what point do you process the sound?

Waves comes in during the technical rehearsals, when you’re in the theater as well as beforehand, when you’re building sound effects. For example, in Coast of Utopia from a few years ago, there was a long, complicated kind of ocean-ambience-opening-sequence where the ocean starts in from a distance with a buoy bell, then the music starts and swells as the ocean crashes and moves around the room. You’re in the middle of this James Cameron-esque, ‘perfect storm at sea’ kind of thing. As you’re building a cue like that, first of all, you build it in something like Pro Tools in the studio and bounce it out to multiple tracks so that you can then remix it in the theater. It’s always going to sound different in the theater than in the studio, because it’s a different space. You mix it all up and do all the sequencing and volume automation and everything and you have a problem: You can’t hear the music because there’s too much high end in the waves. So you very quickly want to pull one of those wave tracks out, the offending ocean track, run it through a low pass, get it back into the sequence, and be able to run it again in the 3 or 4 minutes that it takes for everyone else to reset what they’re doing, so that you get another pass at refining your work.

How much time do you have to perfect the sound?

In most cases, you only have a couple of shots to get it right. But perfection comes in later, during preview performances, which last from 2 to 4 weeks before the official opening. The whole show goes through this process of refining and refining and refining in front of an audience during the preview period, where you can make corrections and then see how they work out.


Which tools do you find the most useful?

EQs, compressors and limiters, volume leveling, things like that when I'm in the theater. In the studio, that’s when you sort of use the more creative tools like Enigma or MetaFlanger. When you’re in tech and fixing and refining, it’s a bit like the mastering process, where you’re going to be reaching for MaxxVolume or C4 or L1, L2, L3 or MaxxBass to tune the low end. In the studio you have more time to be creative, and that's when you look to the fun stuff.

Do you use convolution reverbs?

I use the IR1 a lot on Utopia, actually. I use the sort of the large and medium hall settings and came up with one that worked well in the theater for the music tracks and then sent that around to the surround system. So you’d have the music tracks coming out of the main left-right system, and then the reverb track using IR1 coming out of the surround speakers in the theater, giving a big space to the music. It’s a bit like sending the music to a live reverb except that it’s prerecorded and controllable from within the sound effects playback engine. That way I was able to use a convolution reverb live.

How can you do automation in a theater environment?

Well, you have an automated console, you take a cue and it moves your faders. You might have a cue at the beginning of a song which sets up your orchestra for that number: more harp, less drums, etc, and when you get to the next song or section, you take another cue and it adjusts. It’s a bit like mixing with volume envelopes on a DAW, except that you’re doing it live. You’re doing it in this sort of cue-based way and every once in a while, when you hit that ‘Go’ button, it’s going to trigger sound effects as well, which will be completely automated in volume and routing and whatnot. But automation is for the things you don't want your operator to have to worry about. Theater is about space, not a two-dimensional two-speaker or even five-speaker space. You’re dealing with a three-dimensional higher-lower-further-closer image. There’s depth and there’s width and even sort of beyond what you’d think of as 5.1 which makes it quite different from film sound. Anyway, automation let's you take advantage of all that space.

How about outboard processing?

Absolutely, yeah. I used the MaxxBCL with Jonathan [Deans, sound designer] on Cirque Du Soleil's KOOZA. We also used it on Coast of Utopia, not so much to juice up the bass but just to make this sort of nice warm lovely image; it’s very useful for that. I have put one on my next show as well.

For LOVE, JD [Deans] had four to six BCLs and we routed them to various different outputs in the theater. There’s one for the main left/right, there’s one for the subs, one for the seat speaker systems and a few more.


What’s it like to mix Cirque du Soleil? It seems pretty intense.

In structure, it’s sort of the same thing as Broadway, but in practice, it tends to be a little bit different; the way in which you mix is different. There’s not a lot of dialogue, so you get to have more of your fingers on individual elements of the music, rather than a Vocal/Band/Reverb model. Cirque always has some sort of story, there are always some through-lines, but it is told through the acrobatics, the music and the imagery. This allows a lot more room to be musically expressive with the mix.

The music is ever-changing and depends on what’s going on with the act. And the sound effects are always mixed like another musical element. There’s a general automation setting but the operator will have control over how it fits into whatever else is going on.

What challenges does vocal mixing present in a stage show?

Just like in the studio, the more control you have, the better able you are to translate what you’re hearing in your head to what you’re mixing, quickly and efficiently. Any of the EQs, like the Q10, or SoundShifter which always do exactly what you think they are going to do, are incredibly useful from the creative end. In terms of making something work when you go into the theater, something like C4 which gives you such tight control over the band and has such a sweet kind of natural sound, is incredibly useful when you have to really carve out the midrange of a sound to make room for the music, or room for vocals, or dialog, or whatever. Those are the sort of things you find your self coming up against.

Another situation is where you have a LONG soundscape or texture that is not constant in volume--it’s fine for a long time and you get to a certain point and it’s too loud, and you get to another point and it’s too quiet. To be able to even that out with something like MaxxVolume or L1 or L2 or L3 is really nice. MaxxVolume worked brilliantly on South Pacific. Say you have a bird track that needs to run for eight minutes, and there’s one annoying bird that keeps popping out every two minutes, it just doesn’t pop out anymore. I also use Renaissance Reverb a lot because it’s simple, it’s trustworthy, and it’s fast. I know what it’s going to do. I can just put a little bit of space on something and be done with it.