Ricochet is the best place on the internet to discuss the issues of the day, either through commenting on posts or writing your own for our active and dynamic community in a fully moderated environment. In addition, the Ricochet Audio Network offers over 50 original podcasts with new episodes released every day.
Doing Sound for Films
For movies and television, the image is king and always will be. After all, they call them movies, not soundies. But since 1928 or thereabouts, most films have been made with live sound. Audiences usually want some degree of real-life to mingle with cinematic fantasy. Doing the show right from a technical standpoint is a key element in maintaining a viewer’s willing suspension of disbelief. Whatever you think of Hollywood, the polish and expertise of our technical crafts have led the world’s screens for more than a century, a good part of the gloss of an American success story.
Doing sound for the movies is a little different than doing camera. On a movie set, the camera is treated like a sacramental altar, with attendants performing guild rituals, a technical priesthood, and its own nearly incomprehensible jargon. Superficially, on the other hand, sound looks like an afterthought that seems easy to do — just stick a mike in someone’s face, wear earphones, and run a tape recorder. Simple, right? But it’s surprisingly hard to do it well, especially on the cramped confines of a noisy film set.
An example: Consider the sheer difficulty of finding or creating a truly quiet room on location. Maybe you’re in one right now. How quiet is it, really? There’s a laptop in front of you, with a cooling fan and probably a hard disc. If you’re in the kitchen, the refrigerator motor quietly cuts in and out. A clock is ticking in the next room. A plane passes overhead. Outside, in the distance, a dog barks. A truck rumbles by. A heating or ventilation system whispers in the background. Upstairs, someone is taking a shower and the faint sound of running water runs through the pipes in the walls. You don’t normally hear all this, but the microphone does.
That’s why at the end of a day in a new location, the sound crew will ask for silence so they can record “room tone,” just what it sounds like: the faint sound of a specific room.
The right microphones make a difference. Controversial TV talk show host Les Crane, sort of a 1964 cross between Phil Donahue and Glenn Beck, used a photo of himself pointing a so-called shotgun mike as his signature image. They were a new, faddish thing then. They aren’t really what they look like, sound telescopes that can focus in as exactly as telephoto lenses can.
Les Crane is forgotten now, but his pointing the shotgun mike into the crowd was meant to symbolize his willingness to go farther than mainstream hosts to seek out politically ignored populist voices in his audience.
At the other end of the microphone “closeness” spectrum, lavalieres—neck microphones—can be very useful, and in some situations like outdoor recording they are real lifesavers. In a film like Robert Altman’s Nashville, as many as a dozen actors were miked up with lavalieres, so Altman’s cameras could roam everywhere in the scene without fear of filming an intrusive microphone boom. Lavalieres don’t have absolute top sound quality, though. For dialog, there’s still nothing better than a good mike on a fiberglass pole overhead, pointed right at the actor.
On some filming days, a crew won’t have to bother with location sound. The film industry has a tradition that goes back ninety years: of referring to shots filmed without sound as being “MOS” in camera report forms — “Mit Out Sound” after Josef von Sternberg’s accent. Examples are brief shots of a car driving up a trick ramp and overturning, a safecracker turning a dial, checking his watch, or jumping in a cab, a nun crossing the street to a phone booth, or close-up smiles of delighted kids filmed among other seated actors in an otherwise empty stadium. You don’t always need a sound crew.
Another time-honored exception to the difficulties of recording live sound goes back nearly to the dawn of the talkies. “Filming to playback” is what we’ve come to know as “lip-syncing.” No sound is recorded on the set because everyone is pretending that they’re speaking or singing what’s coming out of the loudspeakers. This is how nearly all musicals have been made since the earliest days of sound, but not quite all of them.
In At Long Last Love, director Peter Bogdanovich set himself and his actors the challenge of doing outdoor musical numbers with a live band riding alongside them. I like an original approach and appreciated the tribute to a brief, obscure moment in early film history. It wasn’t a disaster, but it didn’t really work out either.
Outdoors, even a peaceful breeze that just ruffles leaves can make it hard to record acceptable sound, let alone crowds of spectators, car alarms, or aircraft. Period films have special problems with anachronistic sounds.
A film crew can work much faster if sound isn’t a consideration. In Europe, dubbing has always been much more popular than it is in the States. Many or most of their golden age films were filmed on the streets without live sound, to be dubbed later even in their own language. Of course, it means lengthy sessions in the dubbing studio later, something actors normally dislike. It’s harder than it looks to match your own speech rhythms and lip movements, and harder still to do it with anything like the dramatic effect it had on the set when it was filmed.
When you do have a strong, clear signal from the microphone, what you record it on has changed greatly over the years. In the beginning, it was phonograph records and then a separate “sound camera” flashing a fluttering signal onto a 35mm soundtrack. The great big camera and the great big sound recorder were linked with a mechanical cable, like the brakes of a bicycle. During the Thirties, selsyn motors or synchros started to replace the mechanical connection with a multiphase, high amps electrical one.
That’s the meaning of the zebra stripes on a classic era film slate; the “clacker” gives an exact moment of synchronization between the picture and soundtracks. Editors marked a grease pencil X on each spot and spliced away excess picture and sound film. That was called “syncing up the rushes.” From this point until the final stage of the film production process, they will be handled separately but in sync with each other.
A top-quality Swiss tape recorder called the Nagra became the industry standard of sound recording nearly everywhere in the non-Communist world. Virtually every movie or TV show you ever saw between about 1960 and 2000 was recorded on one. Today, many of the problems of isolating good location sound are the same as in the past, but the equipment used to record it has changed. The extreme mechanical precision that led to nearly perfect recording isn’t needed anymore. A modest lump of solid-state digital technology can do what a Kudelski Nagra could twenty to sixty years ago, and at a twentieth of the price.
Today’s digitized soundtracks are vastly easier to clean up, copy, and shift around. Many film industry procedures of the 21st century still echo those of the film era of analog sound and photochemical images.
Electronic filters can reduce extraneous noises, like faint hums or buzzes, and can reshape sounds to make them more top or bottom heavy. But they can’t accomplish the miracles that they can in fiction, eliminating specific people’s voices, or stripping away an orchestra so you can hear the singer, solo.
In post-production, once a particular section of the film is declared “locked,” picture and dialog editing are considered over and the timings unchangeable. That means the musical score can now be recorded with some confidence that it will match the picture. One whole subunit of music editors works with the composer, conductor, and film director, under the supervision of the chief editors, to determine, to a fraction of a second, where to place the music once it’s recorded.
While that goes on, a different set of small editing teams are working over the sound effects on those same “locked” reels. You’ve probably heard of “Foley artists,” a fancy name for people who make sound effects, often out of seemingly outlandish materials that sound terrific.
It all comes together in the sound mix. The goal is to leave with a fantastic soundtrack, but more specifically, for a feedstock mix that is as final as the one in theaters, but is separated into DME—Dialog, Music, and Effects. The foreign market can dub a version in their own language that will still have the complete multichannel wraparound music and sound experience.
For nearly fifty years mixing boards have shifted over to linear volume controls; sliders, rather than knobs. It’s easier to see at a glance and manipulate as groups of tracks. But plenty of us remember those big, solid RCA dials, and the flickering needles of analog gauges. On a mixing panel, each individual track can be steered along a left-right stereophonic sound field with “Pan Pots,” panoramic potentiometers.
It was noted in the recent post about editing Star Wars that in May 1977, George Lucas was still re-dubbing and re-mixing the film for later monophonic release even as the stereo version was premiering in theaters across the country. You might ask why mono required a separate mix.
Here’s an example: Suppose Ricochet member @Arahant is walking along a factory floor talking with someone. In stereo, you can toss the sounds of machinery all the way to the left and right, at maybe 20% of the total volume on each, with Arahant’s voice right down the middle at 60%, three times as loud as either extreme. He’s perfectly audible and every word is clear. Now take the same mix and play it all through one speaker: it’s muddled. His words are barely louder than the noisy machinery. In mono, you don’t have left-right position to differentiate sounds, so you simply have to give the dialog track priority, fading back the surroundings of the factory.
A final word on the value of doing sound. Young directors are able to attract free acting talent to their early films with the promise of showcasing them. They are also able, in many cases, to get ambitious young cinematographers to work for almost nothing, or for shares in the finished film, because a great camera job on a low budget independent film can launch a Hollywood career. But doing sound, vital as it is to the film, is merely down to Earth, hard, unpretentious work. As a result, soundmen always get paid, because there’s no dream of yours that you can fob off on them and yet you need them anyway. That’s pretty close to a bedrock capitalist proof of their necessity
Published in General
Excellent again, Gary.
There’s nothing like getting the endorsement of a knight in shining armor.
Or is it armour? Anyway, thanks, Percival!
“What did he say? Speak up, sonny!”
Al Jolson, 1927: “Oh, Mama, we’ll move to the Bronx. You’ll be with the Ginsbergs and the Goldbergs and all the other Bergs”.
I first heard of “room tone” from a friend who was a “sound guy.” It made no sense to me, but he assured me that “recording the room” was essential to making it realistic.
A question: I very often find footsteps very unrealistic sounding. The are usually too even and symmetrical. People just don’t walk like that, and the slightest shift in weight or surface can change the sound. My wife doesn’t hear it (though she’s started to), but for me it breaks the immersion in a big way. Are footsteps typically dubbed, or am I just imagining things, and my wife humoring me?
Great post.
If you’ve ever seen-heard the first three years of Hollywood talkies, you know how some sounds tend to intrude far more than they should, while voices on some actors are too muted, because they’re not directly in front of the camera (and the cameras themselves could be really static compared to the silent era right before, because they had to muffle the noise coming from it so the mics didn’t pick that up).
West Coast productions seemed to obtain technical proficiency in sound well before their East Coast counterparts, in terms of not making the sound limitations intrusive and annoying. Laurel & Hardy’s second sound film, “Berth Marks”, shot in the spring of 1929, has a pretty impressive use of on-location sound, at the old Santa Fe Railroad depot in Los Angeles, though apparently it was decided on re-release seven years later to extend the duo’s opening theme into the first minute or so of the movie to obscure some of the clunkiness of the non-dialogue station shot, with Stan & Ollie managing to miss seeing each other in a 10-foot area. The conductor bit that followed, though, was a really good example of a gag that still rings true today, but only works in a sound film:
Footsteps usually are dubbed, for the purpose of keeping every mixable element as separate as possible. Sometimes that leads to a marionette-like artificiality.
And thanks.
Very informative, Gary. You sound so knowledgeable that you must have some kind of intimate connection with sound work in movies. That is, your post sounds like it’s more than mere research through Google.
Now I wish the sound men would stop with the loud music that obscures the spoken word.
And isn’t there something to be done in sports when crowd noise obscures the words of the play-by-play man? Can’t his words be isolated somehow from the crowd?
I didn’t know I’d learn something today when I got out of bed. Great job!
TV shows are terrible about this. I keep the remote handy because I’m frequently adjusting the sound.
Thanks for this excellent piece, Gary. Bad sound editing can make an otherwise good film almost unwatchable. Bird by Eastwood is the worst example I think.
Yep. These scenes in movies always cause me to wanna scream – “They can’t do that!” The human voice creates sound anywhere from 100 cycles (or even lower for some manly men) to 4 thousand cycles (where our “esses” and “tees” become clear). If you remove or duck a noisy frequency, chances are you will change the quality of the actor’s voices dramatically.
I think I may have relayed this story in another thread so forgive me if it’s a rerun for you…
Sound pickup for musicals used to mean a lot of pre-record and a lot of re-record. Vocals were all pre-recorded and played back on the set for the performers to lip-sync to. The dances were then re-recorded as they couldn’t get the microphones close enough without being visible to the camera.
For the Fred Astaire-Ginger Rogers classics of the 1930s at RKO, this presented a particular problem. Rogers made twice as many movies as Astaire because she was also a gifted light comedian and would move on to other projects as soon as her turns with Fred were done filming. So for postproduction when they needed to record the tracks for the taps, Ginger’s role would be taken over by a man, Hermes Pan, Astaire’s choreographer and “idea man.”
This re-record took precision. After seeing the premiere of one film Astaire complained to the studio heads that the sync on a number was off. Before release they checked it and, yes, Astaire was correct. It was off a single frame, or 1/24th of a second.
Excellent insight. Loved the post, as a former aspiring audio engineer, I found this quite interesting. I’m new to Ricochet and look forward to more from you.
I’m highly sensitive to that myself, which is unfortunate in the age of Youtube.
Welcome.
Well, you could always put him in an isolation booth and let him call the game off a monitor. (Don’t laugh, as a way to save money that’s being done more and more.)
There are so many variables to your question I don’t know where to begin and that’s why a good sports mixer is never without work. For every camera there is a microphone. And then there are “effects mics,” strategically placed parabolics that are used to pick up sounds on the field of play: the crack of the bat, the clash of shoulder pads, the squeak of shoes on hardwood. In the case of basketball there are small mics placed on the backboards to pick up the “swoosh” of nothing but net. Then add the booth mics and sideline reporters.
The number of sources to track can get unwieldy and for the bigger events submixers will be employed.
It’s also why it’s the one job in the truck I have deliberately avoided for almost four decades.
Les Crane……was that the same Les Crane who had the hit “song” Desiderata?
Another fantastic post! I love learning things I did not know – and Gary is a superb teacher!
If the basketball announcer is Dick Vitale, be thankful for crowd noise that obscures his blathering. There are also a few others who have no information content to add to the game. But Vitale is the worst.
Thanks for an excellent article. It was informative from beginning to end.
Regarding dubbing of foreign languages: I hate it. If I can’t have subtitles I’d much rather have voiceover. I don’t think voiceover was ever done much in the U.S., but is it dying out elsewhere, too?
In my opinion, one of the most impressive achievements in live (on-set) sound recording in a film was for Tom Hooper’s Les Miserables. All of the singing performances were recorded live, which is almost never done in musicals; it’s just too difficult to capture a good vocal recording with an actor who’s moving around, and it’s also difficult for an actor to act and sing well at the same time. (Stage actors do it all the time, of course, but film is a bit less forgiving.)
It makes a difference, though. The performances come across as much more intimate and emotive, with every little intonation and inflection visible and audible at the same time. But it was incredibly difficult to do, which is why most musicals don’t bother.
https://www.youtube.com/watch?v=ulJXiB5i_q0
Tremendous post!
The room noise must be the equivalent of a flat field in astrophotography. It is an image of just plain light, no stars, passing through the optical train. You use it to remove vignetting and dust. Seems very similar.
And wasn’t Singing in the Rain built around the advent of sound in film?
It works onstage because from the audience perspective, all the sound is coming from the same direction.
Yes, it was.
South Pacific’s Emile de Becque: looks like Rossano Brazzi, sounds like Giorgio Tozzi, both of whom are as French as osso buco.
Yes, that was highly informative.
The Germans are amazingly good at it. You rarely notice it at all. However, like all professional arts, it took time to develop. The German dubbing of American and British movies from the 60s and early 70s is often almost as bad as the American dubbing for Asian films from the same era, but by the time you reach the 90s, my description in the first two sentences applies.
Such an interesting post. Thanks.
But I wonder if I could start a movement to change “sync” to “synch”? “Syncing” just doesn’t work for me. And don’t get me started on “mic.”
I’m afraid you’re swimming against the tide, Suspira! “Synch” is used less and less. I still call ’em “mikes”, because I don’t like to mice up the drum booth.