Issue link: https://digital.copcomm.com/i/426320
10 cgw n o v e m b e r . d e c e m b e r 2 0 1 4 you have a large, gravitational body like a black hole, light moves in curves. The gravity is so strong, it warps space-time and acts like a giant lens that bends the light rays around. We wanted to see the starlight warping and refl ecting around the black hole from a spaceship moving toward it." James knew that if he could tell Thorne precisely what equations he needed, Thorne would be able to provide them. They both realized they were speaking the same language. "Oliver has superb training in optics, better than mine," Thorne says. "But his training is not as good as mine in relativity. So, it was a great match. And, he had fabulous so ware already for dealing with optics, raytracing, and light beam propagation in fl at space-time. We needed to carry this over to curved space- time. The bending of light beams produces the eff ects." To generate equations that governed the propagation of light beams and the orienta- tion of their elliptical shapes through curved space-time, Thorne discovered and adapted work from an obscure pair of papers written in the 1970s by Physicists Rob Roeder and Serge Pineault. He checked the equations using Mathematica and sent them to James in the form of documents. The process of refi ning the equations and implement- ing them extended over several months. "I needed a document control system," Thorne says, "because there were a lot of documents and a lot of iter- ations. The documents were like highly technical papers in a journal but aimed only for me and Oliver to communicate." "Kip is such an expert," James says. "When he explains some- thing, it's simple and clean, so I had to do the same thing – to ask very clearly what I wanted. He'd think about it and a week later a paper would appear in my e-mail box, maybe 10 pages long, detailing exactly the answers to the questions I had asked. He must have written a dozen of these papers." "Once we had traced the path of light rays," Thorne recalls, "and were getting nice images, Oliver said, 'Now we have to do light beams.' That's what was necessary to get the high-resolution images he needed. Computing how a light beam changes, including its gradually changing shape and orientation, is a far more com- plex thing to do. I don't think physicists have done that. He had warned me that was where we were going, but it was still a surprise and a challenge. I knew how in principle, but in practice, the equations were complex." "Kip put in an amazing amount of work and time to develop the equations," James says. "From his equations, I could generate a proof of con- cept. I wrote a bit of C++ code given a camera position, the position of the black hole, and the texture of the stars. It was crude, but it showed the idea could work. Our team at DNeg then implemented the equa- tions in a new suite of rendering tools that accurately traced light rays around the black hole." They created so ware capable of producing the ultra-high-res- olution, rapidly changing images needed for the fi lm. One surprising result of their work was a new discovery, a new structure on the edge of the black hole (see "Revealing a Mystery," page 12). The other was that once James showed Director Nolan the work, Nolan wanted the actors to see the images on stage. "The images were so compel- VFX ARTISTS PROJECTED OUTER SPACE IMAGES ON SCREENS DURING FILMING FOR THE ACTORS, DIRECTOR, CINEMATOGRAPHER, AND CREW. VIDEO: GO TO EXTRAS IN THE NOVEMBER.DECEMBER 2014 ISSUE BOX C G W. C O M