Sounding out The Simpsons

If The Simpsons is any indication, modern animation is ready to recognise the value of a soundtrack where 'live' motion pictures often are not. Richard Buskin enters the strange world of the cartoon


AMERICA'S FAVOURITE animated family is now enjoying its eighth season on the Fox Network. In this time, Homer, Marge, Bart and siblings have given Occidental animation a new facet--delivering satire and social comment through the familiar medium of cartoon. Keeping company with the likes of Doug and Rugrats, The Simpsons also takes particular pride in the construction of its soundtrack, for which Rusty Smith is the rerecording, dialogue and music mixer (in addition to Chicago Hope and 'whatever the heck else walks in through the door' at Sony Pictures' Stage 7). Bill Freesh, an effects mixer at Stage 7, works on the sound effects, backgrounds and foley for the aforementioned shows. Bobby Mackston takes care of the editorial work while overseeing the entire sound process. Travis Powers, who works mainly from home, creates and edits The Simpsons' sound effects. Chris Ledesma is the music editor.

In a nutshell: The Simpsons' dialogue is recorded at Twentieth Century Fox, in what is known as its 'Basement' facility, using eight U87 mics that are summed together and run to time-coded Sony 7030 2-track DAT. (Up to the end of last season the console being used was an old Quad 8--This has now been replaced with an SSL.) Each show, from the first story conference to the broadcast, spends about nine months in production, six of those elapsing between the recording and the screening of the first colour pictures--the bulk of the animation is done in Korea, where labour comes cheaper.

'This is the show that needs 16 months a year,' asserts Chris Ledesma. 'We are always in some phase of making the show, and twice a year we have a two-month overlap. In February, when they start up, we are coming to the end of the post season, and then we overlap again in late-summer/early-fall as production is winding down and we're just starting to gear up.'

Still, the recording of the dialogue only takes a few hours. First there's a read-through in the morning, followed by taping sessions in the afternoon. The DATs pertaining to the different scenes are then loaded into a DAW and assembled, and the select takes are chosen ('and re-chosen and re-chosen,' according to Bobby Mackston). At this point about 70% of the dialogue that will end up in the finished show is already in the can.

Following the first series of edits the DAT is transferred to a 'mag final' (35mm magnetic stripe) and sent to the animation people. Later, when the pencil test has been completed, it is viewed with the audio and decisions are made concerning additions and deletions with regard to the action and the jokes. The result of all these revisions is the 'post animatic final', the last version of the show before it actually goes into postproduction.

Once the animation has been completed, the original audio recordings are toyed with in order to match the new picture. Replacements or rewrites require additional ADR, prior to spotting taking place for the music and the sound effects. For his part, Travis Powers will receive a version of the show as early as possible so that he can set about creating the effects, and he will be continually modifying these up to and including the final dubbing process.

The Simpsons, after all, is different from most other cartoons. Lampooning life and human behaviour, the show is written for an adult audience and, as such, it doesn't fall back on slapstick and 'bam-splat' Hanna-Barbera-type sound effects to support the story-line. Just as the music underscores emotion rather than comedy, so the sound effects need to be real, and this can be quite demanding when considering some of the unconventional situations and locations in which the characters find themselves.

'I've saved every sound that I've ever created,' says Powers, who has been with The Simpsons ever since it debuted on The Tracey Ullman Show. 'That means that after the first three or four years I had a substantial library, and so I was no longer going out and getting all of the basic stuff that pertains to each show. Gags recur, and that's why the consistency is good; when a gag does recur we're not building it from the beginning. We just go back to that exact sound.'

Powers starts his work on each episode by watching every frame of the time-coded colour animation and making notes, before then going out and about with his Sennheiser 415 shotgun mic and Sony portable DAT recorder. 'I'll go out into the field and record footsteps in numerous ways and proper locations, before cutting them up into individual footsteps and playing them back off of a MIDI guitar through a Synclavier,' he says. 'That's standard stuff, but then there are the sounds that really relate to a specific show. Instead of overplaying things we often just underplay them and make them sound true to life. That makes the situations even funnier--like if Homer gets hit in the head it'll sound exactly like that, because the pain comes from the realness of it

'When I'm recording indoors I'll even let some of the room get into it, because I'm going for a grainy old mag-style sound rather than pristine studio material. The real small sound of a bathroom, a kitchen or a living room is what I'm after, and so, whenever I'm doing prop work, I'll try to do so in an actual room.'

Powers gets between five to seven days to work on each episode, and this occurs right at the show's deadline. 'The day after I finish it, it's gone,' he says. 'Usually we finish it on a Tuesday, when I show up with my Synclavier and the whole library, and it airs on Sunday. Often, however, due to last-minute changes, we're later than that and it's just satellited up.'

Meanwhile, the edited dialogue is mixed by Rusty Smith over the course of a couple of days. 'There's a 15-hour budget to mix the show and most of the time we get it done in around that time,' he says. 'On the other hand, if it's a very complex show with lots of sound effects it will take a little longer.

It takes about six hours to dub the sound for a 22-minute episode--it usually takes around four hours for a regular cartoon--and we go from 1.00 in the afternoon until 7.00 in the evening. Starting at 1.00 in the afternoon gives Travis just a little more time to get everything ready and delivered to the stage--he's basically a one-man operation and he's got a lot of effects editing to do.'

'Often, when Travis gets the tapes, pencil-testing is still going on,' says Bill Freesh. 'So, we'll go ahead and dub the show, and we'll come up with a list of whatever we need additional pieces for. During the second day of dubbing Travis brings his Synclavier to the stage, and within a couple of hours we'll be doing playback for the associate producers. Then, later on in the afternoon, we do playback for the rest of the group, and Travis will be able to add in anything on the spot. He brings all of his library on disk and we'll just add things as we go, as we need them.'

ss0298yy01'The interesting part of dubbing The Simpsons is on the second day,' confirms Rusty Smith. 'That's because the executive producer and director will be there and they'll listen to what we've done. Anything that we've done prior to that is based on the notes given during spotting, and it's all fairly obvious stuff. You know, getting things in the right place at the right level and making them work as well as possible. Now, however, when they come and look at it, you see why the show's successful. That's because these guys look at the timing of these things, they look at the gags and see how they're working, they see what the sound is doing to help or hurt these gags, and their notes are based on just turning something up or down. Comedy is based on timing, and the techniques that these guys have learned themselves or from others really come into their own. I've seen it happen over and over again.

'We'll be working on a section and it won't strike us as being all that funny,' adds Bill Freesh. 'Nevertheless, the next day these guys will come in, take a look at it and say, "This isn't quite right. What we need is this". They've known all along what they needed; they gave spotting notes, editors put it together and we mixed it, but they'll look at it and make two tiny little suggestions and it'll make it hilarious. The whole idea of this show is to make people laugh, and these guys know how to utilise sound to do that. It's a ball to watch the comedy get punched up in the show with the stuff that they put in front of you.'

The Simpsons is mixed in surround at Sony on a sound stage measuring about 22ft x 45ft and housing a Harrison MPC console. 'We create matrixed 2-channel Dolby and TV encoders, and we have a number of mix stems,' says Rusty Smith. 'I use one LT-RT stem with dialogue for imaging, and then we have two mono dialogue stems in order to provide enough separation to slip things around, because all of the stuff is married together and the source material is on 24-track. We also have four effects stems including the music stem, and so everything except for the mono dialogue stems comprises LT-RT matrixed 2-channel stems. In terms of processing, all that I use is dbx 160X compression. I don't de-ess it or do anything else, because it's pretty much handed to me in good shape.

'The layout that we receive from Bobby [Mackston] and Travis is pretty formulaic. Bobby will deliver the dialogue split up on a character-by-character basis.'

'It's delivered to me in mono,' interjects Mackston. 'It's like a radio script, in segments, and I divide the characters up and put them on separate tracks. The actors do multiple voices; one does 15 voices, another does 20...'

'So all of the characters are split out onto separate faders,' continues Smith, 'and they also bring in a walla group that shoots specific background crowd noises.'

For the uninitiated, a 'walla' group comprises those actors who make assorted mumbling sounds for background effect, as in 'walla, walla, walla,' or, as British crowd-scene extras often prefer to say, 'rhubarb, rhubarb'.

Back to Rusty Smith: 'The dialogue is on 2-inch 24-track tape with SR noise reduction--I wouldn't have it any other way--and then the music shows up on an 8-track DA-88, custom-scored every week. It's recorded 3-track--left, centre and right--and the centre track is usually a mono split of whatever it is that they want to keep as separate as possible. There'll also be music source material in there, including bands such as Aerosmith or Smashing Pumpkins, and in fact I have to say that one of the best things of working on the show and mixing the dialogue is that you get to mix in the voices of people who you never thought you'd get your hands on.'

'U2, Sting...' adds Bobby Mackston by way of example

'And I've had all of The Beatles on faders, with the exception of John Lennon unfortunately,' says Smith. 'So, that's one of the fun aspects to working on The Simpsons--you never know who's going to show up on a fader.'

'These guys all come in and record their parts for scale,' says Mackston. 'They're invited to do the show and they want to do the show...'

'Their kids and their grandkids want them to do it,' adds Smith

'Most of the time they'll do it at Fox, or sometimes we'll have to go to where they are,' continues Mackston. 'With Paul McCartney, for instance, they just went over to his house in England. The musical part of the show can be very involved. It's not just stuff that they think about during post. A lot of times they'll already have things written into the script that will require Alf [Clausen] to come up with the music ahead of time. There will be big production numbers with people in the cast singing; a musical version of Planet of the Apes, you name it.'

ss0298yy02'There are two ways in which we handle the music," explains Chris Ledesma. 'There are those shows that have songs sung by the cast or guest stars, and then we also have our week-to-week scoring sessions. The score's written by Alf Clausen and we record between 30 and 35 original pieces of music for the show every week. These range in length from a second and a half up to a minute and a half or more, and they cover practically every style of music possible. We've done reggae, we've done organ at the baseball stadium, we've done country & western, we've done military... We've been talking about it lately and have wondered if there is a style that we haven't covered.

'The songs are written way in advance of when you hear them on the air, and they're sometimes hopefully written a month or so before we even record the cast vocals. This then gives Alf a chance to work with whoever the songwriter is--because they each take turns writing the lyrics--and they work out all of the metric details, and so on, to make sure that it's going to work well as a song. Once it's nailed down we'll also set aside time at one of our regular scoring sessions for a current show in order to prerecord this music. It's usually done with a trio--keyboard, bass and drums and we bring in professional studio singers who will sing the various roles. Then we take the prerecord, mix it down and send it out on cassette to the cast, while, if time is pressured--which it usually is--we also send a version off to animation so that they can animate the sequence including lip-synch for the voices to later be replaced by the cast.'

Chris Ledesma next coaches said cast with regard to a new song. As only one among them actually reads music the rest just listen to the song repeatedly until they learn it, at which point Ledesma helps out with phrasing and pitch.

'I conduct those sessions and then I take the tracks back to my editing room, where Ireplace the studio voices with the cast voices. I finesse and edit their performance a little bit to get it absolutely right, and off we go with that. Then, six months later, when we're finally ready to post it, we get rid of the little trio and replace it with the big orchestra. That's the basic way in which a song is done.

'The score, meanwhile, is done week in, week out. Doing the underscore, we take a look at the show less than two weeks before it goes on the air. As a matter of fact, we looked at a show last Friday--which is six days ago--which will score tomorrow [Friday], mix on the dubbing stage this coming Monday and Tuesday, and air on the following Sunday. So, there you go. That's the amount of time that we have to spot, write, orchestrate, record and mix 30 to 35 music cues. Plus, there are also weeks when we have to throw in a song that we did six months before.'

Not that the music and the effects are the only elements to undergo short-notice revisions. The dialogue may well be accorded this treatment as well

'A topical joke that was thrown in six months ago may no longer be relevant,' says Bobby Mackston, 'and something that is now hot news may be added instead. The Simpsons is like a runaway train. You know, you're just trying to drop things on as it's going, and eventually it'll get too far ahead of you and it's gone.' THE SIMPSONSTM AND © TWENTIETH CENTURY FOX FILM CORPORATION. ALL RIGHTS RESERVED 

The Fox angle

SINCE JANUARY 1998, Twentieth Century Fox has been taking care of the music recording for The Simpsons at its own brand new scoring stage. In the past, a number of different LA facilities have been used, and, according to Fox scoring engineer, John Rotundi, some of them may still be used in the future when feature film work fills the in-house schedule.

'For us, The Simpsons is pretty much a smaller orchestral setup,' he explains. 'They have a lot of options as to where they can gowhereas the bigger orchestras don't, and also there's an invoicing difference between television and feature projects. You see, the TV date takes up one to two sessions--it's a one-day affair--while a feature may take anywhere from two to four weeks. Preferably we'd like to accommodate all of the Fox productions, but the TV material is subservient to the features and so it's likely to get bumped when the schedule demands this.'

Still, when scoring for The Simpsons does take place at Fox, Rick Riccio works alongside John Rotundi and handles the mix. Everything is recorded to 3-channel LCR stereo simultaneously using the DA-88 and 4-track Studer A80 Mk.IV with Dolby SR (time code going to Track 4). There's also a 2-track 15ips 1/4-inch for the composer, as well as DATs that are run during the takes. Only the master takes are retained. Meanwhile, for the songs, there's a 24-track Studer A820 running at 30ips.

With an SSL 9000 at the heart of the proceedings, few outboard effects are used beyond a couple of Lexicon 480Ls. The one modification to the console is a split of the LCR buses at every 24 inputs, effectively making it a 4-man dubbing board.

'We can select either discreet section buses or a global monitor bus,' says Rotundi. 'In fact, we can select which sections we're listening to through the film monitor section... In between all of the split LCR buses and the monitors is a routeing matrix, and this enables us to monitor any configuration of mixdown busses and route them to the monitors independently. It also does a stereo folddown which we feed to the DATs and to the composer's 1/4-inch.

'The SSL is awesome,' asserts Rotundi. 'It's been flawless in terms of its operation --no hums, buzzes, clicks, pops or burps--and so we've been very pleased with it.'   
         Free Learning Space for Film Sound