A technical question about cameras, film, and 30 Rock.

After a trying week of marking papers, teaching, and generally being run ragged, I spent far too much time last weekend watching 30 Rock.  As to the content of the show, all I can say is that I find it remarkable that a program dedicated to the inside baseball of running a network variety show is even intelligible, much less popular.  Then I remember The Muppet Show exists or watch episodes of Saturday Night (not yet Live) from the year I was born, realize how old the trope is and find it remarkable that 30 Rock managed to enliven it.  If pressed, I'd argue that its success has something to do with the meaningful inclusion of the network brass, which is in marked contrast from Lorne Michaels playing himself on SNL, but tonight I'm more interested in why the cast seems to have such huge heads:

In this frame, the camera and lighting conspired to make Tracy Morgan's character, Tracy Jordan, look like something the first 10,000 fans under the age of 13 receive when they come through the gate.  How did this happen?  I don't know for sure, but thanks to an article [.pdf] in the Spring 2009 issue of Exposure magazine, I think I can make an under-educated guess. 

In it, chief cinematographer Matthew Clark says he employs wide-angle cameras to create a shallow depth of field in order to "lend a sense of immediacy to what's going on," which is all well and good, except that wide-angle cameras do the opposite.  He also notes that because the sets are so small, they are warmly and dimly lit, so in order to create any contrast, he has to shoot with a wider iris to allow in more light, and doing so diminishes the depth of field. 

It would seem that balance is struck: wide-angle creates a greater depth of field, the low-lighting and wider iris shallow it out.  The only problem, as evidenced by the frame above, is that the balance is frequently out of whack, and because wide-angle lenses exaggerate the distance between objects, whenever a character leans forward, their heads appear much bigger than the bodies they're attached to.

I think. 

I welcome those of you who know more about cameras, lenses, lighting and film stock to correct me.

Comments

[Peter Erwin]

The "iris-in, iris-out" film technique that you link to has nothing to do with camera optics; it's an editing transition using a variable-sized mask in front of the camera (or post-processing that simulates the same thing).

It may help to think about it this way: an "iris" is a mechanical or biological mechanism with a hole in the center and a way of shrinking or growing the hole. If you put an iris-like device in front of a camera, then you mask out more or less of the field of view. You can get the same effect putting your hands in front of your eyes (e.g., using your hands to pretend to be looking through a telescope).

If you put an iris-like device inside a camera (or, you know, inside your eye), then you have a variable-size aperture. Making the aperture larger or smaller doesn't change your field of view, it just affects how much light comes in and, secondarily, the depth of field.

It's perfectly possible to have an optical device with a fixed aperture (in which case, no iris); that's what pinhole cameras and telescopes are.

Note that if you want a smaller ("shallower") depth of field -- meaning that more of the foreground and background are out of focus -- you need a larger aperture. (The smallest possible aperture is a pinhole, which gives you nearly infinite depth of field = everything in focus.) What the cinematographer in the article actually says is: "I usually want the lens open to about a T 2.8"; the latter is some alternate method of talking about f-stops that I'm not familiar with, but the sense seems to be "I want the aperture opened up to something fairly large" -- which makes sense if he does indeed want shallow depth of field.

[M. Bouffant]

For the Benny show & most others (radio & tee vee from the '30s through the '50s) the network executive was pretty much a non-entity, as sponsors bought the entire time period from the network, & put on what they wanted.

I have a $5.00 (Yes!) four DVD set of Jack Benny shows (doubtless slipped into public domain) & ephemera (War Bond Drive & USO appearances) from Target, & at least one of the shows has Jack going to the sponsor more or less begging to be renewed.

I think it was the quiz show scandals that led to the end of advertisers purchasing time & more or less running it themselves. The quiz scandals certainly caused network "Standards & Practices" departments to exercise more control over what the networks were broadcasting.

[Neill Brower]

As someone above already mentioned, aperture isn't going to affect the proportions of the objects on film--it only affects light transmission and depth of field/focus.

However, the focal length of the lens used can do this: as you said, longer focal lengths compress distances between objects; shorter focal lengths exaggerate them. So movement of the head toward the camera, relative to the body, will appear to enlarge it.

Further, if the cinematographer uses a wide-angle lens that does not have a flat field, barrel distortion (bowing out) will occur outside of the centermost portion of the frame. If the actors are framed such that their bodies are centered and their heads are at the edges, the heads will be distorted in a way that can make them appear larger.

The frame you show doesn't really look like a barrel distortion effect to me (though it's a bit small for me to get a better read); it looks more like exaggerated depth, if anything.

[HP]

Ahistoricality: Regarding OTR, Arch Oboler on Lights Out did this several times. Of course, Lights Out was a horror show, not a comedy, but the principal is the same. These are mostly stories in which Arch Oboler plays himself, and the stories are set "behind the scenes" at Lights Out. "Murder in the Script Department" is about two typists for "Lights Out" staying late to finish typing up rewrites before the broadcast day. Let's see: There's "The Coffin in Studio B," "The Author and the Thing," and I can't find it, but there's also an episode where Al "Jazzbo" Collins, playing himself, meets with Arch Oboler, playing himself, and tells him a story that he thinks Arch should put on his show.

Of course, this being Arch Oboler, the writing is insanely loopy and hamfisted, which is either a feature or a bug.

[HP]

Oh, I forgot to mention: "Lights Out" aired from 1936 to 1943, with "The Author and the Thing" dated to 1936.

Of course, "Gold Diggers of 1933" is a film about making a stage revue. Oh, and there's "Ruggles of Red Gap" (1918), a silent movie about making silent movies. (I have the novel!)

[Ahistoricality]

HP, that's great stuff!

[Sisyphus]

So when people are talking about Gregg Tolund and his development of deep focus cinematography in, say, Citizen Kane, they mean he was using a wide-angle lens? Or is this a case of using a smaller aperture?

I keep getting confused between the terms "depth of field" and "focal length." This may be one of the things, like the phases of the moon, that no matter how many times you explain it to me, doesn't stick from one minute to the next.

[Gary Farber]

Incidentally, I notice this.

[Neill Brower]

@Sisyphus: Deep focus requires a smaller aperture for greater depth of field *and* a wide-angle lens for exaggerated depth cues.

Focal length is a physical property of the lens assembly itself. A wide-angle lens refers to a lens with a "short" focal length, defined relative to the diagonal measurement of the film or sensor frame. For example, a 35mm film camera's diagonal is about 43mm, so a wide-angle lens would be anything with a focal length less than that: 35mm, 28mm, 24mm, etc.

Depth of field refers to the portion of an image (front to back) that appears sharp. Aperture and focal length (also film/sensor size) determine the depth of field of a shot.

[nutellaontoast]

Sisuphus, remember it this way: Depth of field refers to how much of the frame is in focus, so it is a term regarding your "field of vision" A larger depth of field gives you a larger "field" to play with (ie, more stuff is in focus) A shallower depth of field gives you a smaller "field" to play on (ie more stuff out of focus).

If you think about the way light passes through a lens, with the rays converging, focal length will make sense regarding how the image looks. Something with a shorter length will have the light rays converging at a steeper angle and so it gets more "pancaked out." I'm not 100% sure if that is an accurate description of the phyiscs but it might work as a sort of pneumonic. Here's a pciture:

http://media-2.web.britannica.com/eb-media/27/4627-004-E3B847C3.gif

See how if the focal length gets shorter, the rays converge more steeply on the focal plane (which is where your picture is captured by the film)

[Robert M.]

Physics: something I know something about!

Basically, focal length is the distance between your lens and the point of convergence for the light rays which pass through it. As Nutella&c. says, a short focal length means the convergence point is close to the lens, so the light passing through it is bent further. That steeper angle means more of the world will "fit" through the lens onto the film or CCD, which gives you a broader field of view; because you're fitting a larger image on the same size film, the level of magnification is also lower. In the same way, a longer focal length gives you a narrower field of view and higher magnification; here, the light is bent less drastically, so less of the world fits through the lens.

In what way this interacts with the depth-of-field question is a topic for people more photographically inclined than I.

(To correct a minor point, you can't capture an image exactly at the focal plane, because your image would then be an indistinguishable point of bright light. At least in the eye, the retina is located behind the focal plane, producing an image which is rotated 180 degrees with respect to the world.)

[nutellaontoast]

My understanding is that the focal plane is where things are in focus. Not all light goes the same point, only colinear light. If light coming from one point was recorded past the focal plane, it would "spread" and crossover onto the other points. Again, I could very easily be confused on the subject.