Being a greenhorn game developer in college is (big surprise) a lot harder than it sounds.

And it sounds pretty damn hard.

To recap, since this dev blog has been silent for quite a while, we are on about the fifth art team roster and the fourth team of programmers at the moment, not counting the core of Ductomaniac, Mix3d, and myself who have stuck around for most of the game. We have gone through Python prototyping, a switch over to XNA, and a slew of software in the pursuit of a process that works. This latest team is very skilled and passionate about games; save scheduling and the stressful life of summer, we are doing well.

As we approach a point where we need to pump out art to sate the engine's hunger for content, our process goes something like this:

1. We identify a game entity that we need
2. Usually I draw some rough concepts of the entity using a combination of Alchemy and MyPaint
3. We discuss it in meeting
4. We revise the concept and draw it again and discuss it again
5. A schematic is drawn up for the 3D artist, then scanned
6. Using Blender 3D 2.49, a model is produced and UV mapped
7. Using MapZone, the artist creates the textures that adorn the model in-game. We use MapZone because, with practice, you can create very good textures procedurally that can be tweaked later.
8. Export the model to FBX, then upload it and its accompanying assets to Dropbox where the programmers take over.

If you are interested in game art, I highly recommend all of those sites and programs. There are all free and absolutely splendid tools that make your life easier, cheaper, and free you from bug nests like Maya.

To demonstrate how this works, we have this progression of the initial idea to the full-fledged design of a game entity, in this case one of the IAF's cruisers. I did most of the detail work on this on a whim while settling in at home, so we did not get to put it through the meat grinder, but it shows how we go through an idea to bring it to a place where we can schematicize and then model.

That is how we do your typical enemy or set piece, but we have recently gotten to the stage where environments need consideration as a whole. With about a dozen planned levels in at least six settings, many of which will be more than a simple backdrop for all those lovely explosions, there is a lot of work to do to get the atmosphere consistent, engaging, and most importantly, cool.

The fist environment on the list is space, for the practical reason that it does not have a lot in it. In particular it does not have terrain, a system for which we have not implemented nor designed yet. We thought this would make things simpler, but it does nothing of the kind. As any geek will tell you, real space is extremely boring. Any vaguely reflective object nearby will drown out the stars, you will not notice a perspective change in the starscape unless you travel at stellar-hoping velocities, asteroids and debris are thousands of kilometers apart, nebulae are actually nearly invisible, and in orbit everything is traveling at hundreds of meters per second.

All of this means that all that typical sci-fi space combat we like is completely unreasonable, and usually impossible. It also means that in a realistic space situation, the player would have no sense of direction, speed, or location, not to mention that in order to maintain a safe distance from other craft, you would be so far away you would not be able to identify or anticipate them. Thus, in the tradition of all games before us, we have to pull all the typical sci-fi tropes out of the toolbox. That means big nebulae, scrolling stars with perspective, and impossibly concentrated debris and spacecraft. Oh yeah, and fiery explosions, too, though maybe I can convince everyone that colored plasma explosions are better ;).

Which brings us to this post's showcase: a time lapse of me painting one of our test nebula. This took around an hour or so, and I did it in MyPaint using CamStudio with a frame captured every two seconds. Watch out, this kind of thing is mesmerizing.