The Making of “Titan”

As a sci-fi nerd, I often have some vision or concept knocking around in my head – snapshots of amazing vistas or interesting vehicles that never leave the margins of my notebook. But even simple sketches are a representation of something that could be, even if it takes a while to understand how to realize the vision.

Sketching answers a lot of initial questions, such as: what’s the orientation of the scene? What’s the general placement of objects? What’s the mood or atmosphere? By answering these questions, the sketch proves its value in the form of constraints. The sketch challenges you to stick to your vision. Of course, sometimes your initial concept needs to change in order to produce the best work, but it’s a useful touchstone. My sketch for this project shows what I had in mind: a Jovian-like planet with sweeping rings and satellites, visited by a tiny spacecraft.

Settling on a spacecraft design was my next focus, since it would be the focus of my efforts to learn more about low-poly modeling, unwrapping and texturing.

From the beginning, it seemed to be a a kind of cargo ship, with the sketch even playing with the idea of multiple craft in a sort of shipping lane. So I decided to go for a blockier, more utilitarian look. One design, bristling with antennae, had vertical struts that looked like they could accommodate a big cargo container between them. It was neat from the front, but I didn’t have a clear idea of how it would look from behind, or how the container would look and function. My original sketch has the ship moving away from the viewer. I wanted to keep that sense of the ship going into the scene.

After some messing around, I ended up with a kinda blocky retro shape. The simple shape would make it easy to experiment with its flat surfaces. Also, since I knew the ship would appear pretty small in the final composition, I felt it better to go with the simpler silhouette.

With the sketching out of the way, I cranked out a model of the ship and moved on to unwrapping and texturing, which was pretty new to me (though I knew the basics).

This was by far the most complete effort I’ve made to paint a model, and I’m really happy with the results. This all went pretty smoothly and I was able to finish the maps in about 6 hours total. 3D Studio Max automatically updates the textures when they’re re-saved, so it was a simple matter to just render things out continually to see if it was meeting my satisfaction.

Having never built a space scene, I found Hoevelkamp’s tutorials (the first 5 featured on that page) to be extremely helpful. They were very clear and allowed for a lot of experimentation. After the initial setup, I spent several hours just tweaking things, which got me to a good starting point. While it definitely needed more work, this gave shape to the sketch and achieved the true perspective of the scene that I couldn’t capture with my simple, quick drawing.

I knew I wanted the background of the scene to have a nebula and starfield, to increase the sense of awe and make the ship really look small and insignificant – almost stunned by this stellar vista. I decided the best way to do this was to composite the background in after the fact. As I researched how to do this in Photoshop, I asked a friend of mine for input on the scene composition. His feedback was invaluable:

1. Try to unify the color scheme (most space scenes concentrate on two colors)
2. Make the ship more dynamic and the composition more balanced by re-positioning the ship off center and changing its orientation
3. For the nebula/starfield, you can start by looking for images of clouds or nebula online and then modifying them for blending with your composition
4. Consider using actual 3D models of asteroids to give the rings a more realistic appearance

As you can see, his input really made the difference for this image. For the planet’s rings, I adjusted the granularity of the map and then tweaked its opacity so that the rings would appear solid and smooth farther away from the camera, but more translucent and filled with particles closer to the viewer. I felt this was a good compromise as opposed to trying to model actual rocks.

Overall, I’m happy with how the original concept translated into the final product, which I credit largely due to the constraints I introduced early on.