Game Development Reference
This loop can repeat a few times or thousands of times, depending
on the project. Sometimes developers will plan the number of cycles they
want before release. Other times, they'll just continue looping until the
game hits a target level of quality, or until they run out of money.
We don't only iterate on an entire game. We can iterate on a level, a
tool, or an interface. On larger teams, there should be many different itera-
tion loops running at the same time.
Since every design challenge is different, each iterative process must be
tailored to the challenge at hand. Here's an example of a simple iterative
process that I've used to develop combat scenarios in first-person shooters.
This process wouldn't be appropriate for other challenges or developers—
it's just one possible example of iteration.
I start by roughing in a basic fight as quickly as I can. I throw ele-
ments in as I think of them, not pausing to analyze. I might have an idea
of where I'm going, but I don't have to. My only goal is to play the fight as
soon as possible.
Within an hour, I've got the fight running—and as always, it's awful.
It plays like an amateur mod by an especially unmotivated first-time de-
signer. Gray blocks of cover are scattered haphazardly about, the world
geometry is a handful of poorly scaled cubes, and enemies appear in giant
clumps. And since I usually forget to give the player a weapon, he always
But despite its poor quality, this first version fulfills its purpose. It has
closed the iteration loop. The fight is no longer a mental movie. It's real.
And playing a real fight, with hands on the controls and the real seconds
ticking by, sparks thought processes that can't be replicated in any other
way. This first attempt was never meant to be anything like the final prod-
uct. Its only purpose is as a platform from which to jump to something
And in that, it's successful. As I play it, the ideas flow, and they're
more specific, and more concrete than anything I could have thought of
cold. I get excited about them, and I don't have to wait, analyze, or docu-
ment. My inspiration doesn't get time to degrade. After one test, I'm back
in the editor, ripping out pieces that didn't work, shuffling cover around,
rearranging weapon pickups and enemies. Maybe I even remember to give
the player a weapon this time.
Search Nedrilad ::