I think it's fair to say that Particulars has had a particularly long development history: the project started in mid 2011, was put on hiatus while funding was found, continued hiatus while other projects completed and is finally rearing its physics-soaked head again in 2013.
I think it's also fair to say that I'm both a little addicted and a little over that particular pun.
So I'm pretty excited to say that this game is almost out of Pre-Production and will be, as of next Monday, in Production.
This is an overview of the preproduction of the game, and at some point the ordering becomes a bit moot as we do many of these simultaneously. Feel free to ask questions in the comments.
STEP 1: Make a Prototype. Or 3. You probably only need 1 though.
STEP 2: Prove that there's a market. Or, in our case, apply for and get funding over a 10 month period (in general, this step should take less than 10 months. But if it does...)
STEP 3: Figure out the Pillars of the game. These are the things that your game couldn't do without, and they form the basis of everything else you do. Note that they can, and likely will, change during development.
Particulars is a Minimalist Meditation on Chaotic Flows
(importantly, each of those pillars have a separate, 2 sentence definition that explains what they mean in the context of this game)
STEP 4: Figure out a plan for the rest of Pre-Production. This includes timelines and deliverables, coz the pre-production deliverables for each game are different. The first time you do this you'll miss things and/or put too much in. But that's how you learn.
STEP 5: Build the basic game engine. Hooray! You should also draw a UML diagram at this point, rather than waiting till you're doing a final pre-production code review (whoops).
STEP 6: Make some Technical Prototypes. Oooh, pretty... (yes, all our tech prototypes were around graphics)
STEP 7: Make a game development timeline. Ours is full of "Holy crap that's not a lot of time" bits.
STEP 8: Budget. This is a 'set up once, tweak forever' thing. Where possible, make sure you take into account when money comes in so you don't run out of cash.
STEP 9: Realise you should have made design doc already. This project has made me realise the incredible value of a design doc. It's just so much easier to get people up to speed. Our one opens with your pillars, and shortly afterwards has 2-3 paragraph descriptions of the Core Mechanics, Designed Progression, and Look/Feel/Theme of the game.
STEP 10: Oh, there's a marketing plan as well. Really important. We made a fair effort to outline what assumptions we were making here, so that we can actually compare that to reality once the game releases.
STEP 11: Make your level design tools. Love me some good level design tools.
STEP 12: Get some Concept Art, Audio and Narrative. Make sure you cover all major styles. Also get some UI design done as well.
STEP 13: Make some sample levels. This part can be lots of fun. Ideally it tests your ability to teach the game as well. Get people to play it. You know, that playtesting thing.
STEP 14: Figure out the Progressions of the game. Narrative, Level, Chapter, whatever you have. Just have a draft of them.
STEP 15: Pipes. Lines. Procedures. Asset/Feature Lists. Have them. All of them will evolve, so its mostly about having a starting point. Don't forget to have a testing plan, and a procedure around versioning.
STEP 16: Realise that some of this is going to bleed into production. Oh noes! Some of the audio and art is still being worked out. You have enough to go into full-on 'code/make levels/make assets' mode for everything else though, so you can still be in production. And the audio for this project is being deliberately simple, so you can afford to lose a week on that. It's not a blocker. (Note: Audio may be a blocker for your project).
STEP 17: Write an article about your pre-production process. Hopefully it's in a nice, simple, stepwise format, and it doesn't lie about how many steps there are.