How to make a game

There are two dirty little secrets to making games. The first one is: Don't make a game. Seriously. You're making a series of steps. You're building up a group of subsystems. But making "a game" is too big a task for anyone to handle.

The second secret, is especially important: take crappy, easy steps. Go for the low hanging fruit. Monkey steals the easiest peach. Frequently, you can make 3 or 4 cheap 'n easy steps and discard one or two of them in the time that it would take you to make 1 step well. Don't make great stuff for an experiment when bad stuff would either prove or disprove your theory. And don't build up a "perfect" system when the player would never notice the difference.

The first step down the road of making your first game is to find a unique seed of an idea. A story is not a unique game idea. "Let's make a warcraft game" is not a unique game idea. "Let's make a game where you try to keep store shelves stocked with stuff" is a unique game idea, and should be pondered.Gather groups of people together... friends, associates, random strangers off the street... and brainstorm. Brainstorm a lot. People may not have the same ideas or same vision, but thinking through all of this will help your game cohese in your mind.

To visualize your idea, create a cheap and crappy paper game. Yes, I'm serious. Draw some stick figures, then cut them out and place them on the board. These are your high-resolution normal mapped models, with ultra-realistic fiberous texturing. Careful, they cut. Now, work out how it actually plays. Part of this is a humility step, but part of it is to make you as a designer sit down and figure out all of the mechanics each step of the way, and ways to make it work in reality. How does the player restock the beans'n'cheese aisle if they're in groceries? How does stopping to mop the floor effect checkout times? Put this in front of your friends, family, and associates for feedback.

While you're doing this, get a coder on board... preferably you. If you don't know how to code anything, start studying now.

Now that you have a team, you want to knock out 3 - 5 super cheap-'n-fast internal demos. At this step, there should be no animation, no music (unless integral) only white stick figures on a black background. Again, show these off to people that you know for feedback.

While you're doing that, get an artist on board. Now do the above again. 3 - 5 more super cheap-'n-fast internal demos. By the end of this process, you should have more or less proven the gameplay that you want to make. Or gotten sick of making games. Let's hope it is the former. Now you need something to demo to the public, and possible investors / artists / musicians. Pick the one crappy internal demo that worked best, and polish it up to be a not-quite crappy external demo. Add some art, stand-in music, and gameplay.

Demo this a few times to people who you don't care about.

Polish up the crappy external demo to an awesome and shiny (but super short) external demo, based upon the feedback from the people you didn't really want to impress anyway. This should only be 1 - 5 minutes maximum... the shorter the better. Only show the ready stuff, imply the rest. This might consist of one screen full of playable game while an unlicensed but appropriate MP3 plays in the background. Finish with some images sliding by to give a feel for the scope.

Hopefully you've been flexing your contacts all this time. Chances are, you won't get buy-in from a major company, simply by the nature of game development politics. That's OK... Do it yourself. Small teams frequently make great shareware titles on shoestring budgets. There are whole infrastructres out there to support shareware authors, from payment processing to distribution. You should be able to find a few programmers, artists, and musicians that are willing to work for the experience and a percent of the final game.

From here, the development process is pretty straightforward, if painful, so I leave the rest in your capable hands.

Good luck and Rock On!