Institutionen för datavetenskap Umeå Universitet

Ray-tracing

Ray-tracing is a technique for rendering an image. The method works by tracing a ray of light through the objects making up the scene, calculating lights and shadows as the ray reflects.

To obtain a embarrassingly parallel implementation of ray-tracing, the complete scene is duplicated at each Grid resource. Each subjob ray-traces its part of the resulting image. The problem solver generates the complete scene by merging the images from each subproblem.

Alternatively, one could ray-trace a sequence of images, creating a short animation. In this case, a subjob ray-traces a whole image (one frame in the animation). The problem solver constructs the animation from the completed frames once all subproblems are solved. Suitable datasets (POVray format) can be found here (small) and here (large).

Use some existing ray-tracer for the tracing, e.g., POVray.

Additional links:

POVray

http://www.cs.umu.se/kurser/TDBD20/VT07/lab/raytracing.html
Ansvarig för sidan: Erik Elmroth, P-O Östberg, Johan Tordsson
Senast ändrad 2007-01-29