Recording Dialogue On-Set: A conversation
 

I've been hitting the books, trying to saturate my mind with all things 'sound design'. The following excerpt from one source has got me thinking, and I'd love to know what others think:

"Ideally, everything the audience hears except human speech will be created in post-production...a major departure from run-and-gun video/indie guerrilla film-making. Even sounds that result from on-screen actions are to be avoided. For example:

-when actors set cups on saucers, the prop department should have it lined with a damn napkin to avoid the clink
-when actors read a newspaper, the prop department should have it sprayed with water until its damp enough to avoid capturing the annoying rustle of the paper.

To runners-and-gunners, this can be a difficult mind-set to acquire, but it will pay dividends in post-production, as veteran movie sound engineers well know."

-Would an audience really perceive the rustle of a newspaper during dialogue as 'annoying'? Would an audience perceive the clink of a cup in any negative way?

-'Ideally', would not most indie film-makers try and avoid creating everything but human speech in post?

Cheers,
Miggy

I think it's ultimately all about having maximum control over every aspect of the movie, including sound. In real life the paper might be as loud or louder than the spoken words. . .not good. If you add the rustle later, you can eq it, leave it very low, etc, while boosting the voice. Same with saucers. Or you can leave those sounds out completely if they're distracting from the scene. But not if they're burned into the audio.

You're asking the wrong question. It's not the audience that will find those sounds annoying; it's the editor. Imagine multiple takes that may have to be cut together, but the rustle of the newspaper or the clink of the cup isn't in the same place each time. Or it covers some dialogue and can't be cut out.

That's why you avoid these sounds in production and add them later.

Bingo! People have NO IDEA how big a deal this is until they have to cut... and you end up not using the best performance, or even the second best, because of some other technical issue. Kitchens are the WORST because there are so many hard surfaces. And newspapers are LOUD.

I am currently cutting a film in which ALL sounds were captured in production, and let me tell you, it's an absolute cutting NIGHTMARE.

Maybe so, but I've got some great answers :) And a major pointer, so far:

BE MORE/VERY ALERT TO THINGS ON-SET THAT WILL CONTAMINATE DIALOGUE

The way it has been explained to me and it makes perfect sense (at least to me), think of your Post Production Sound Designer as the person who creates what your movie will sound like. In a sense he is the equivalent to the editor but for your sound. With that mentality I approach sound as I would approach coverage as a DP or Director. I want to give my editor alternatives to be able to build the story. They need all of the blocks so that we can create the way we want to tell the story.

For sound, I look to my production sound mixer to capture what is needed to give the post sound department what they need. That usually means they have the separate pieces that are needed to mix a solid soundtrack. It's odd because just like with the visual side, how we hear things is not what is actually present. In real time, we are constantly filtering what we hear to create the scene that we perceive. We key into certain sounds that emphasize or reinforce what we experience.

However, if we presented a scene with all the visuals and sounds that are actually present when it occurred, we'd drive our audiences screaming. People can't do the same thing when watching a movie. I've been told that with sound, this is true more so than with the visuals. So as movie makers, one of the responsibilities we have, is to make the choices and act as the filter. By doing so we control what our audience focuses on and we determine the emotions that the audience will feel.

I'm constantly reminded that to even come close to achieving that, you have to do some pretty crazy things. Shooting barefoot, stabbing apples to get the sound of a creature eating a brain, having a guy walk around an empty studio in high heals to get just the right sounding foot steps, or having extras have silent conversations are all things I've done or had done on movies I've worked on. I suppose it could be desirable to have all sounds involved in your movie done in post. Even so far as having all the dialogue ADR'd.

But remember one of the truisms in any venture. There are three elements involved in producing; you can have it be done quickly, it could be done cheaply, or it could be good. However in a small productions, you only get to choose two. For an indie production you don't get to choose cheap, you're stuck with it. And, there are times when you have to choose quick. But given a choice I choose good as my second piece.

So, as an indie filmmaker I'm actually striving to control everything which usually means having as many separate pieces as possible so that I can create what I want. And that does usually mean that the only audio I'm capturing during the actual take is the dialogue.

-Garrett

Garrett, it makes a lot of sense to me too. Awesome info!

I just had to steal these Steve House comments from another thread, as it is so relevant here:

Explosions etc are not ambience, they're practical effects.

Each of them should be recorded separately from the dialog scene so that each element - dialog and the various FX - is recorded at the best fidelity possible, NOT recorded all together with the dialog as part of the scene.

The separate files for the FX can be precisely positioned against picture to within a fraction of a frame during the editing process so they can occur at exactly the right moment - they do not have to be recorded at the same time in order to get the timing right. If you do, you'll have the explosion sounds bleeding into the talent's dialog mics which makes it very difficult to adjust timing in post.

Also, if you record both at once on two stereo tracks, you'll need to split the tracks into two mono files when you get into post in order to edit properly.

A thoughtful sound person will concentrate on dialog, and during quiet moments on set they'll record "wild" sounds to help post. So during the dialog scenes there will be damp napkins keeping cups and saucers from clinking, and when it's quiet they'll record a wild track of the cup clinking on the saucer without damp napkins. This way post will have the choice of using the sounds of the actual props on set, or not.

Capturing clean dialog - Prime objective.
Capturing wild sound - Secondary objective,
Capturing ROOM TONE - Priceless.

Room Tone, never leave without it.

-Garrett

Super-practical advice. Cheers!

When recording "wild" sounds, like the cup clinking etc., is it done the same way as recording dialog?
For example, the strong advice here has been to keep shotgun mics 20-24 inches from the throat/chest. Does that advice still apply for capturing the cup clinking?

Cheers,
Mig

IMHO, yes.

Get too far away from the cup and you get too little signal, too much noise, and too much room. Get too close to the cup and your coverage gets to narrow and any small variation in location changes the response.

I like to get a bit closer, like 14-18 inches for dialog or whatever, but the same principles apply.

Are there any editing conventions when it comes to logging wild tracks/room tone etc.?
Without someone slating, how does the sound dept. log such audio takes for the editor in post?

Thanks Jon!

The same principles apply, got it. But do the same mics apply? Should the sound recordist have a mic dedicated to capturing wild sound?