American Cinematographer - February 2009 - (Page 29)

horizontally to achieve the desired IO and back-panning to converge on objects. Lenny advised leaving out the convergence axis and aligning in post by sliding one image over the other. We needed extra picture width for that maneuver, which a 3K crop of our 4K sensors allowed. Armed with a couple of prototype 3-D sliders, John Ashlee began experimenting with using forced-perspective sets without tipping our hand. We learned that a 1Ú2-scale background looked natural in normal stereo, and a ¼-scale background would work in weaker stereo. We made a composite of several elements built at different scales, scaling a camera move to match, to see if a composite in 3-D would hold together. It sounded promising, and it works just fine as long as you carefully set up each element, scaling everything, especially distance to camera and the IO distance. Paul Gentry set out to empirically determine benchmarks for IO distance. He shot puppets in a matrix of close-up, medium and full-body shots at different focal lengths and IOs. We projected each frame in 3-D and rated puppet heads for normal, extreme and reduced roundness. Not surprisingly, we found that the closer you get to the subject, the smaller the IO you need. And we quickly found out how painful excessive IO can be Ñ painful enough to pull an audience right out of the narrative, if not the theater. The big surprise was how little it takes to create a normal sense of roundness. We reasoned that puppets would look natural by setting IO as measured between CoralineÕs puppet eyes vs. the distance between a pair of human eyes Ñ 19mm puppet vs. 64mm human. But to our surprise, normalfeeling roundness in puppet closeups ranged from 1-3mm IO, and in wide shots from 3-10mm IO. Left: Animator Travis Knight works on a scene in Dad’s “Real” study. Below: Coraline’s “Other” father puts an entertaining spin on things in the “Other” version of the same space. We had simplified by limiting the test to a single subject, a good starting point in setting up shots and helpful for newcomers to the show. But things quickly got more complex in deep sets that featured objects both close to and far from camera. At that point, we needed more than an IO cheat sheet; we had to rack up enough experience to make informed judgment calls. As with any other aspect of cinematography, with experience, we gained confidence and a more instinctive approach. We all agreed 3-D had to be used to enhance story and mood, like any other photo technique. Along with the story arc, lighting arc and color script, we decided to impose a complementary Òstereo arcÓ on the show. Henry wanted 3-D depth to differentiate the Real World from the Other World specifically in sync with what Coraline is feeling. To do that, we kept the Real World at a reduced stereo depth, suggesting CoralineÕs flat outlook, and used full 3-D in the Other World. At first, full 3-D opens up a better world for Coraline, but when things go bad, we carefully exaggerate stereo depth to match her distress. 3-D adviser Brian Gardner pointed out the emotional effects of placing a subject behind or in front of the screen. Similar to shooting up American Cinematographer 29

Table of Contents Feed for the Digital Edition of American Cinematographer - February 2009

American Cinematographer - February 2009 Contents Editor's Note Short Takes: Circus Production Slate: The International Reverte 2 Worlds in 3 Dimensions Dead Reckoning Embracing Anamorphic Citizen of the World Post Focus: Restoring Manhatta New Products & Services International Marketplace Classified Ads Ad Index Clubhouse News ASC Close-Up: Peter Suschitzky

American Cinematographer - February 2009

For optimal viewing of this digital publication, please enable JavaScript and then refresh the page. If you would like to try to load the digital publication without using Flash Player detection, please click here.