My creative process seems to fall into two general categories of activity. The first happens so quickly that it hardly deserves to be called a process: the image just pops into my head fully formed. Making the image a reality takes time and some details may change along the way, but I am never in doubt of the goal. Double Dynamo's core gameplay (based on dual N-back) is a good example of this first sort of creativity. After four months of work the game itself still plays nearly identically to my 2-week prototype.
On the other hand, I find myself immersed in the second sort of activity when I am not at all certain what the final result will look like. Sometimes I'm not ever sure what problem I'm trying to solve, if any. This process remains mostly inexplicable, but as far as I can tell it amounts to a willingness to rabbit-hole – to spend time and energy on avenues that may or may not bear fruit – but to do so in a controlled way. And somewhere along the way, regardless of whether I end up using the results, I usually gain new skills. Even more importantly, I gain a better understanding of what I am capable of accomplishing and in what amount of time.
The Double Starburst
The game's background is another example of the first sort of creativity, the immediate inspirational kind. I've always been drawn to starburst patterns and I needed a way to visually link two locations on the screen in a pleasing and hopefully intuitive way. The solution, a double starburst, seemed obvious.
Figuring out how to actually draw an animating double starburst was not so obvious, and I spent three full days working on a fragment-shader-based solution before scrapping it due to severe performance problems. I did learn a little bit about polynomial approximations to trig functions in the process, and I also ended up with some pretty cool visuals that were not at all what I was going for. (Alas, I never even committed these changes to a branch, so the pretty pictures I was planning on showing you are beyond the reach of git.)
Then a friend more experienced in the ways of OpenGL suggested I just draw it with polygons, so I did, and ended up with pretty much exactly what I'd envisioned.
Towards a Cohesive Visual Style
About a month ago I was trying to come up with some better-looking buttons and generally tighten up the UI. And so naturally I started playing Go. My working notes for those days simply read "playing with go stones" and "textures for stones". What was I doing? I think what was going on was that I wasn't satisfied with the way the game looked and here was something new and different and completely unplanned that seemed promising in a way that wasn't immediately clear. The early results were encouraging, though, and I spent the next few days walking around the neighborhood and taking photos of every interesting texture I saw.
This kind of image manipulation wasn't in my toolbox before I went through this exercise, but now if I need a similar effect I won't even have to think about it. I did not get a shiny new set of buttons out of this process, nor a cohesive UI. (Those came later.) I did get something I wasn't expecting, something I didn't imagine at the outset, and I think the new symbols have a pleasingly tactile and approachable look. Don't you agree?
Like how Double Dynamo looks? Sign up for the mailing list to keep up to date!