Magazine Archive

Home -> Magazines -> Issues -> Articles in this issue -> View

Article Group:
In Session

Stripes & Stars

Todd AO Studios

Article from The Mix, June 1995

Simon Braund visits New York's busiest movie post-production facility

In the age of Dolby surround and fantastic sound effects, film sound recording has become ever more sophisticated. Todd-AO is the New York post-production studio whose name is most often seen in the credits of Hollywood movies. Simon Braund untangles the dollies from the dailies...

If the experience of entering a recording studio and I seeing the walls adorned with gold discs has worn a little thin, take a trip to Todd-AO studios in Manhattan. Here, their trophies come in the form of autographed movie posters, bearing the names of Ron Howard, Mike Nicholls, Tim Robbins and Sidney Lumet, amongst others. It's an interesting place, and you don't need to be a movie buff to get a thrill from entering one of their vast studios, or glimpsing the names on the booking chart in the office of Steve Castellano, Vice-President of operations.

Studio C, for example, is an evocative place, if you've ever witnessed a movie scene set in a Hollywood screening room. There are the plush, sofa-size seats, bearing the imprints of some very famous rear ends, and there, perhaps looking a little incongruous in the age of video, is the enormous cinema screen. What you won't see in the screening room, however, are the racks of outboard effects that are used to create Hollywood soundtracks, or the mixing desk the size of several single divans laid end-to-end. It's a venerable custom Quad Eight console, while the effects comprise a collection of dialogue processing filters, compressors and parametric EQ, plus extensive reverb and delay.

Used for isolating certain frequencies of background noise or dialogue, and enhancing or subduing them as required, this is the post-production sound editing gear that marries location recording, music, dialogue and effects to the visual images of a motion picture. It's a highly creative and exacting operation, and one that involves infinitely more than simply editing a recorded audio track to fit a celluloid cut, and then sticking the music on. This is a very different environment to a music recording studio.

My guide is Bob Chefalas, Todd-AO's chief engineer. Standing behind the acres of mixing desk in Studio C, Bob offers a basic run-down of the procedure. Most recently, he has been working with director Ron Howard, on the forthcoming blockbuster Apollo 13. It's due to open in the States in late June.

"Apollo is filmed on location", he explains, "using a time code DAT machine, which is recording the location sound. It's also recording time code: 30-frame track time code. When they film on location, the cameras are filming the picture, and the DAT is recording the sound; two separate elements. That's why when you see them filming, you'll hear, 'Scene 1, take 2', or whatever, and you'll see the sticks go (the clapperboard, in incorrect, but more familiar jargon). The camera shoots the sticks, and the sound records the sound of them clapping together. Then, when the dailies of the picture (film shot during a day that is likely to become part of the finished movie) are sent to a lab, the lab takes the negative, and makes a print of it for the editor to work with, so he can see the day's takes. We transfer the sound from DAT to something called stripe. The editor then takes the 35mm picture and the 35mm stripe, looks at the picture, and finds where the sticks come down, finds where the clap is on the sound, lines it up, and from then on he's got two separate elements in sync. And that's how they do the picture editing; sound is always separate from the picture."

All well and good. But just to explain: 35mm stripe is the medium onto which sound recorded on location is initially transferred. It is sprocketed like 35mm film stock (for synchronisation). On one side, it carries the sound transferred from the location DAT, and on the other, the time code.

"The picture editor only works with the sound," explains Bob, "he has nothing to do with the time code, we use this later on in the project."

Basically, with picture and sound editing kept separate, a movie arrives at Todd-AO in the form of a working cut: An edited version which has a rudimentary soundtrack on stripe, 'locked' to the visual image, which will (hopefully) require no further picture changes. Says Bob:

"The picture editors have been working on the film, and once they've finished, it's handed over to the sound editors. The sound editors can now start working on it. That's when the dialogue editors come on staff, the effects editors come on staff, and they give a copy of the film in its embryonic state to the composer who's going to score it. So now, everyone has a copy, and it becomes the reference master for them to do all their work to. The dialogue editors have to listen to it, clean it up, and so on."

Which, simply put, is Todd-AO's stock in trade. One thing I've always been curious about (and I'm sure you have too), is exactly how much of the location sound is useful to the sound editors. Obviously, when you're filming out in the open, background noise is going to be a problem, so I wondered how much of the dialogue would have to be replaced.

"We like to use as much as we can of the actual location dialogue," says Bob. "But there are cases of technical problems, or just sound quality problems, where they might be shooting near a waterfall, or a plane goes overhead. That distorts the sound, or overpowers the dialogue so we can't use it. Then we have to replace it, and that's when an actor goes onto a stage that we have downstairs, sees him or herself up on the screen, gets a cue through the headphones, and re-voices the dialogue."

This is known as ADR (Automated Dialogue Replacement), and we'll be hearing more about it later.

"What we do now," continues Bob, bringing us back to Studio C, "is read the time code that's been put onto the stripe, back through a computer. From this cut-up time code, we'll generate an EDL, an Edit Decision List, because all the dialogue is going to be edited on a workstation now."

In essence, the EDL is the roadmap that will tell the sound editors where to find relevant sound on the location DATs, and enable them to load it into the workstation. It's a way to get the old technology (the 35mm stripe) to work with the new (DAT and digital workstations). The luxury of editing film sound digitally is a fairly new phenomenon, particularly for dialogue editors. Previously, they worked exclusively with stripe which, as you can imagine, is not only cumbersome and bulky, but offers nowhere near the flexibility afforded by a modern workstation.

"Once you load the sound takes," says Bob, "it all stays in the computer, and we can play right off of the workstation, and transfer to Tascam DA88s. The editors' work has been simplified, and they're able to be more creative with the time they have. They can now zero in on little segments, make extensions, make loops, things that they could never do before. Well, they could have done those things on stripe, but maybe they wouldn't have bothered, because it was so time-consuming and tedious."

Bar the video screen, Studio E could be a standard-issue music recording studio

Bob describes this capability as, "coming in almost overnight", which would indicate that in this area at least, film sound studios have been lagging behind music studios. Would he agree with that assessment?

"I guess in terms of introducing workstations, yes. Music studios have been using workstations for years, but I think it was a question of the time it took until somebody built a workstation that was able to accommodate what film editors need. One of the most important things is that a workstation has to be able to work in footage, not time. Film is all footage, and simple things like that just didn't exist when it was introduced into music studios; it was built as just a stereo music-editing system. To edit dialogue, I guess you need a lot more bells and whistles."

So, is working with film sound essentially a lot more complicated than, say, producing a record?

"I would say so, yeah," he replies, without much hesitation. And when you consider that for a two-hour feature film, the sound editors are dealing with, on average, a million feet of recorded footage from which they have to extract what they need, it's not hard to believe him. What mixing a movie soundtrack does have in common with mixing an album, is that the whole is comprised of many different elements (dialogue, backgrounds, effects, music), all interwoven and layered to create the finished effect. A further similarity is that a soundtrack will often be released in a number of different formats, which correspond to the playback systems installed in movie theatres. There are several systems that allow a full-level digital mix to be played back digitally, but the number of theatres that possess this capability still represents a surprisingly small minority.

What you're usually hearing (in about 90 per cent of theatres, according to Bob's estimate) is a playback via the movie's optical track. This is the oldest medium for recording film sound, and it's easy to appreciate Bob's frustration that so many of the magnificent mixes created at Todd-AO are hampered by its continued widespread use. The problem is, of course, that re-equipping a theatre with digital playback is enormously expensive, and since all films employ an optical track regardless of any other formats they might be carrying, it's unlikely that their use will be discontinued for the forseeable future. But before we get ahead of ourselves, an explanation of exactly what an optical track is might be useful.

"Optical has been in service for film sound for 50 or 60 years," Bob explains, "but we're still using it now. When we shoot an optical track we have an optical camera, and what we shoot is a 35mm negative. The sound is played through the camera, and to shoot a stereo optical track, you've got four ribbons in the camera: two make up the left channel, and two ribbons make up a right channel. Light is sent through the ribbons (it's kind of like a ribbon mic), and as the sound modulates, the ribbon moves and changes the waveform of the light, with the negative behind it.

"So as the sound is changing, the negative is recording light. Then that negative is put on to a print, and the print is put up on a projector. The projector has a light source, with an optical cell on the other side, and the light source is sent through the optical track. The same modulations that we recorded are there, and the optical cell picks them up, reads them and, just like a record needle, makes sound out of them."

A fascinating method of recording and transmitting sound, and no doubt hailed as revolutionary in Al Jolson's day. But it must seem far from ideal, 60 years after The Jazz Singer assured us that, "we ain't heard nothing yet."

"The problem is that there are so many variables," confirms Bob. "The recording is only, maybe, 50 Hz to 12 kHz. The optical cell, the light source, has to be in focus, and you can only get so much level on it, before everything clashes. It's also susceptible to any handling noise, any noise on the print. You get popcorn noise and you get dirt, and it's just..." he trails off, shaking his head. "But we can make it sound good," he emphasises, perhaps anxious to dissuade cinemagoers from boycotting theatres simply because they employ technology with a history almost as old as the wax cylinder.

"We use Dolby noise reduction, and we've added special electronics to cameras, so they can make a good optical negative," he says. "But it's still an archaic medium. With the digital technology, we still have to record onto that negative, but we put on a lot of different information. Instead of just having two tracks of optical, we'll add a third channel of time code, or Dolby SR* puts their digital information between the sprockets, and SDDS** does the same thing. So we're adding a lot more information to the negative, and when it's delivered to the theatre, whatever equipment they have to play it back, they'll use."

So everything is crammed on at some point across the width of the film?

"Well, not everything, not every format," qualifies Bob. "If it's a DTS*** release, it'll have its format, if it's a Dolby release it'll have that format. There have been formats that have both, so if a theatre can't play one of the special digital formats, it can at least play the optical track, which is why it's still in existence."

It seems odd to me that given the extremely advanced facilities available for sound mixing and post-production at Todd-AO, the finished product is heard by so few people in its full glory. On the other hand, it's not so different from a track produced in a state-of-the-art recording studio being heard by most people on their radio or TV. The difference is, of course, that you have an easier choice of the medium you listen to your music on, than the movie theatre where you see a film.

It's arguable that exhibitors are short-changing a significant proportion of their customers. But with the increasingly high standards of sound to which Nicam is accustoming consumers, optical soundtracks might be phased out sooner rather than later. Bob Chefalas, though, is not so optimistic.

"I don't think we'll see it phased out in this century," he says. "Maybe five, ten years. It's going to be a long time before you can get all the theatres to throw out an optical track, and be able to play from something else."

Studio C contains a Quad Eight console, viewing screen, and...

"It's an interesting world down here. Every day is different, and every day is full of surprises"

Having said all that, Bob is quick to point out that 'digital' should not be perceived as a magical buzzword, or the saviour of an industry mired in the past.

"I've worked with music mixers who will not record digitally," he says, "they'll record with analogue Dolby SR. Digital gives you certain advantages, but sound-wise, I think you can still get as good, or better results with certain analogue equipment. But of course," he concedes, "it is great for theatre playbacks."

Okay, that's Studio C. Now let's go downstairs and delve into the weird and wonderful world of Automated Dialogue Replacement, and Studio E, the ADR/Foley studio (another bit of movie jargon). It's more reminiscent of a music recording studio than the cavernous mixing rooms. That's because this is the only place in the building where any actual recording takes place. It consists of a control room and the studio itself, which, apart from a large video screen at the far end, wouldn't look out of place at Metropolis.

However, the main installation in this room is a reading lecturn and a single vocal mike. Paul Zydel, the ADR/Foley mixer, explains what goes on here.

"Every film that's shot has location sound that's either unusable, or that will need lines added. Or sometimes a director will want to change a performance. What we do here is put the actor in front of the lecturn, with notes on those changes and the script. We re-record their dialogue, and try to get as close a match as possible to the production track. So when they get to the final mix, and they get to a part that they can't clean up, they can go to this.

"In addition to replacing dialogue, they may want to change lines, or the reading of a line, or even a whole scene. Sometimes they may want to add lines on the back of an actor's head (like at a point where you can't tell that the actor is not speaking), perhaps to explain something to the audience that isn't clear. There's a lot of different reasons for doing this kind of work," he concludes.

In many ways, the ADR system (for which Todd-AO are to receive an Academy Award certificate this year), is quite a straightforward premise. But, naturally, there is more to this than meets the eye. Today, Zydel is waiting for Bridget Fonda to come in and record dialogue for a film called Rough Magic. He has a list, compiled by the film's editor, of the changes to be made, relevant notes on those changes, and the time code location on the film where they will occur. When Fonda arrives (and no, unfortunately she didn't, while I was there), she will stand at the lecturn while time-coded footage of the movie is run on the screen in front of her.

"We'll run the production track, or a guide track, or sometimes a premix on a cue track, and feed it to the actor through headphones," explains Paul. "We record in sync, with one or two back-up recordings. Lately, we've been running DATs and 1/4" inch tape, and we're mastering either on 24-track or six-track sprockets, and soon we're going to be doing it digitally. There's different ways, and each job is a different format. Everybody wants something different, so we have to be able to accommodate them.

"What happens then, is we'll go to the cue that's given to us by the editor, run the footage down to there, and cycle that section back and forth. At the head of the section, the computer automatically superimposes three beeps. On the fourth, imaginary beep, the actor starts."

In this case, Fonda will watch several run-throughs of the scene, familiarise herself with the changes to be made, then, on cue, lip-sync her dialogue to match the action on-screen. Either that, or supply new dialogue from an amended script.

"On this job, I'm going to use a six-track master, so we can do six takes that we can save," says Paul. "And they'll have everything on the DAT, it'll all be slated and logged, and the 1/4" will have the same time code that you see here, so that when the dialogue editors get it, they can use the 1/4", and regenerate it at the same time code. They assemble it, and then they use it as a mix element."

However, that's not all that happens in Studio E by any means; it's also where the foley work takes place.

"Foley is things like synchronised footsteps," explains Paul. "We have an artist, who reproduces actors' movements, the rustle of their clothing, and so on. It's basically another layer of sound that we add to every mix."

In this respect, foley is similar to ADR, in that it involves replacing sound from a location shoot that's inaudible, unusable or, occasionally, just plain unconvincing.

"The boom person on location is mainly concerned with dialogue, so they're booming for that," says Zydel. But if a person is walking and talking at the same time, they may not pick up the footsteps."

...some serious items of rack gear

Underneath the neat carpeting of Studio E are 18 foley pits; different surfaces, from gravel paths to tarmac to floorboards, which are used to recreate the background noise that accompanies the actors' physical movements.

"And for a foreign release, where all the dialogue has to be replaced in a different language," comments Paul, "every time a word is spoken, anything that's mixed with it has to be replaced."

Does one person handle all the foley work?

"Yes, she's a dancer, and she's very fluid and knows how to reproduce peoples' movements. She'll do the whole picture, it takes two weeks. It's all pre-cued, because the editors know what they need, so they'll give me cue-sheets, but it's still time-consuming, because it has to be right. The sync has to be right, and the sound has to be right."

It sounds obvious, but it's not as straightforward as you might imagine. It's more than just footsteps that often have to be replaced.

"Part of the problem is that a lot of times we'll find that taking the actual object used on location doesn't give the right result," explains Paul. "So we wind up using a can-opener for a gun, or something like that. We'll use a telephone for a computer keyboard; you use your imagination, and you create as you go. We have a sandbag that we use for bodyfalls, if somebody gets shot in this room, or they take a tumble, or get hit by a car. And it's fun, because you get to break chicken bones, and stab roast beef, and smash cabbages, when somebody's head gets crushed."

Paul shows me into a small storeroom where he and his staff have amassed a huge collection of found objects: Racks of boots, shoes and sneakers, a car bonnet, various chairs, buckets, pots and pans, balls of every description, shelves full of bottles and crockery, even a wheelchair, which is used primarily for the sound of bicycle wheels. It looks like the inventory of a particularly weird car-boot sale. Everything here, though, is regularly employed to generate the myriad sounds which might only be perceived subliminally, but whose absence would rob the soundtrack of its life. Imagine, for instance, a western without the jangling spurs, the clopping of the horses or the plink of the spitoon.

"I once put Hume Cronyn's head in that," says Paul Zydel, proudly displaying a very ordinary-looking plastic bucket. "It gets pretty difficult sometimes, because it has to be right, and often you have to duplicate things like actors walking on snow," he adds, describing what, to me, sounds like a practical impossibility. "Actually snow is pretty easy; you use corn starch," he confides. "A lot of times you'll get an actor in a car trunk, and how do you re-record their dialogue without actually putting them in a car trunk?

I think you'd better tell us that, Paul.

"Well, we'll take a film box or something, and put it over the actor's mouth, until it muffles the microphone."

Easy when you know how. The objective of ADR and foley work is to ensure that things sound authentic, and to that end, Paul Zydel has not only dunked Hume Cronyn in a bucket of water, but had Billy Baldwin looping dialogue in full fireman's breathing apparatus (for Backdraft), and required Brook Shields to speak into a paper cup, for a TV movie about scuba diving. Another of his duties is to bowdlerise (taking the naughty bits out — Ed) movies bound for television or for showing on airlines. This is an exacting science and something that, as you'll probably be aware, is often done quite ineptly (the word "bulldish", is a particular favourite of mine). Not, I hasten to add, in Paul Zydel's studio, however. Even so, I don't doubt his secret relief that Eddie Murphy's career is on the decline.

"It's an interesting world down here," he says. "Every day is different, and every day is full of surprises."

Finally, all the elements are brought together in Studio C, or one of the other mixing rooms, and a final mix is created. Background pre-mixes are usually done first, then the dialogue, and finally the score is added. This is an exciting time for everyone involved.

"When a mix is going on," says Bob Chefalas, "you'll have the mixers up here working at the desk, and the director will be sitting up here, listening, and making comments. You'll have ten or fifteen people all putting creative input into how it's going to sound. Towards the end of a mix, you're working to a certain timeframe. You've got to quickly put up a reel, get to the change, and change over to another reel. When you're working to a deadline, it can get chaotic in here.

"But the Ron Howard crew has been great to work with, they're always very focused on their work. They're always involved in a project all the way through."

And does Bob have any particularly exciting memories of the movies he has worked on?

"The last feature, Wolf, was pretty exciting, because it was a new Sony eight-track digital format. But all the films have a certain amount of excitement, and you usually end up doing something a little different for each one. But the first day of any mix is always exciting for me."

FOOTNOTES: * Spectral Recording © Digital** Sony Dynamic Digital Sound*** Digital Theatre Systems (Unlike SR© and SDDS, which allow playback straight from information on the print, DTS carries a time code on the print while the sound is supplied via an interlocking CD-ROM).

More from these topics

Browse by Topic:

Recording Studios

Video / Film / Picture

Previous Article in this issue

Mixing metaphors

Next article in this issue

Hidden persuaders

Publisher: The Mix - Music Maker Publications (UK), Future Publishing.

The current copyright owner/s of this content may differ from the originally published copyright notice.
More details on copyright ownership...


The Mix - Jun 1995

Donated by: Colin Potter

Coverdisc: Mike Gorman

In Session

Feature by Simon Braund

Previous article in this issue:

> Mixing metaphors

Next article in this issue:

> Hidden persuaders

Help Support The Things You Love

mu:zines is the result of thousands of hours of effort, and will require many thousands more going forward to reach our goals of getting all this content online.

If you value this resource, you can support this project - it really helps!

Donations for June 2024
Issues donated this month: 0

New issues that have been donated or scanned for us this month.

Funds donated this month: £0.00

All donations and support are gratefully appreciated - thank you.

Magazines Needed - Can You Help?

Do you have any of these magazine issues?

> See all issues we need

If so, and you can donate, lend or scan them to help complete our archive, please get in touch via the Contribute page - thanks!

Please Contribute to mu:zines by supplying magazines, scanning or donating funds. Thanks!

Monetary donations go towards site running costs, and the occasional coffee for me if there's anything left over!

Small Print

Terms of usePrivacy