My second attempt at a volumetric nebula was a bit more successful. I think I'm getting the hang of how to use the technique (and that VIPER viewport mode is an absolute godsend). The only wrinkle is that it can be tough to figure out how the fine details are going to come out before doing a large render, but I think judicious tests using the limited-region camera will solve that. I think this technique has legs.
So I took the animation I did earlier, the shield world, and this new nebula and put them together into a thumbnail-sized test render. For the record, I rendered an image of the nebula as a large EXR image and mapped it to a sphere in the background.
I'm satisfied enough to jump to working on another shot. Still on the to-do list for this one:
I'm going to change the flight path of the shuttlecraft going over the top of the ship's saucer. I expect it's right around where the viewer's eye will be left by the previous shot, and it doesn't draw you anywhere interesting, so you'll miss the ship and the environment. I think the new flight path will have to have it pass over the ship's neck, which should serve as a jumping-off point to scan the rest of the frame. Maybe I should look up more about visual paths and composition. It's possible I'm framing a still when I need to frame an animation.
More volumetrics! I want to have dust clouds around the various chunks of the shell, as well as a ring of dust and debris.
More complex lighting.
I've decided I'm not going to settle on a color scheme for the shots yet. Color theory isn't a language I'm comfortable speaking yet, but I know that just doing whatever feels right for each shot without regard for the whole is a terrible idea, especially for shots like this where there's no motivation for any color over any other in the set-up. Once I have more rendered tests with some environment in them, I'll start playing with colors.