Can you tell us a little about yourself and your history with 3D?

I've been interested in 3D graphics since I was very young. I learned various 3D packages on my own during high school, then spent a summer and winter vacation in college interning for studios in LA (Foundation Imaging and Station X). After that, I started up a small freelance animation business, which has now grown into Maas Digital LLC.

My best-known work has been the series of animations I created for NASA's Mars Exploration Rover mission. The original piece (as featured on my website) led to several follow-on projects, including original animation for a PBS/NOVA documentary that earned me an Emmy nomination in 2005.

How were you first introduced to LightWave 3D?

I've been using LightWave since version 2 back on the Amiga Video Toaster!

What feature of LightWave do you like the most?

I love Modeler... LightWave is by far my first choice for polygon and Sub-D modeling. Keyframe animation is fast and easy.

Are there any plug-ins you use on a regular basis?

Maas Digital runs a pretty complex internal pipeline. I wrote a custom application that imports objects and animation data from LightWave, applies RenderMan shaders, then sends the scene to Pixar's RenderMan. This way we couple LightWave's excellent modeling and animation tools to a customizable high-end renderer. Additionally, the system helps us create procedural effects like high-resolution planetary terrain.

Of course, all of our beautiful star fields are rendered with StarPro™, which is a plug-in we sell :).

What was your primary role in the production of Roving Mars, and what did it entail?

I was visual effects supervisor and lead artist. My team consisted of three other artists who focused on modeling, animation, and texture painting, plus one programmer who helped interpret all the data we obtained from NASA. I did all of the shading, lighting, compositing, and render management.

The two main challenges of Roving Mars were to "up-res" the old animation to look good in 4K format, and to replace the generic terrain with data from the actual Mars Rovers. We put lots of extra detail into the Mars Rover model and made extensive use of camera-projection mapping and RenderMan displacement mapping to enhance the appearance of the terrain. I developed some special algorithms for importing 3D terrain data from NASA as well as regular 2D images.

How scientifically accurate did you need to be for Roving Mars?

Our goal was to be as accurate as possible within the constraints of acceptable image quality and the production schedule. We put as much detail into the rover and spacecraft models as we had time for. We made extensive use of Mars Rover images and 3D data sets to reconstruct the Martian terrain. In most shots the environments are accurate down to individual rocks, craters, and sand dunes. The rover motions and track paths are all based on actual operations that took place on Mars.

We made a small concession and showed the rovers driving a bit faster than they do in real life - otherwise some of the shots would have been hours long! Also, for parts of the terrain where data was not available, like the back sides of rocks the rovers couldn't see, we constructed plausible "filler" data, usually by cloning pieces from other parts of the terrain.

Naturally we've been asked whether there should have been sound effects during the outer-space shots. Just imagine we bolted a microphone to the spacecraft to pick up sonic vibrations through its metal structure!

Do you prefer more technically constrained projects like Roving Mars, or more creatively open traditional entertainment projects?

Both kinds of projects can provide interesting challenges. I like learning about the nuts and bolts of real missions in order to depict them accurately. Sometimes truth is far more interesting than fiction - for instance, I think the Rovers' complicated, hair-raising landing process is more spectacular than anything you'd see in a sci-fi film. On the other hand, I would certainly be happy to work on a project that allowed more creative freedom.

How do the requirements of producing for the IMAX format differ from projects for more conventional screens?

The chief issue was resolution. Our CGI had to be created at 4K (4096x3072 pixels) in order to hold up on the giant screen -- that's four times as many pixels as a 2K film frame. We had to be very careful to avoid letting render times get out of hand. For instance, use of ray tracing was kept to an absolute minimum.

The rules for composing shots are slightly different in IMAX. The field of view is very large, which led us to use wider lenses than we were accustomed to (zoom factors of 1.6-2.0 were typical). Also, IMAX screens extend quite far above the audience's normal line of sight, so action must be framed towards the lower third of the image.

The post-render pipeline was similar to CGI for 35mm film. We rendered elements to floating-point OpenEXR format, composited them in linear light, and delivered the final frames as 10-bit Cineon files.

Are there any particular challenges you faced in the production of Roving Mars?

Roving Mars was the largest animation project I had faced so far and my first film project. It was also my first experience managing a team rather than working on my own.

The main technical challenge was putting enough detail into shots to look good in IMAX format. After a few days of work on a shot, we usually reached a point where it would be "final" in NTSC or HDTV format, but at 4K you could still see fuzzy textures or missing details. Our most powerful solution was camera-projection mapping.