Scope Creep

Help!

I used to be able to see the end of the project (MMMarbles)
Now it seems fuzzy.
Initially I limited the number of enemy/gadgets to 64 - hoping to never go beyond. Now I've gone and doubled that to 128 (stupid binary system)

It was just getting crazy - too many things we want to throw in so I've had to put out a call for final enemy/level design -
Design freeze should help - if we can stick to it :)


So how does everyone else deal with non-structured projects?

I just keep going, implementing new ideas off the top of my head (and figuring out how to leave the door open for more later) while trying to remember my original (modicum of a) plan, until the goal gets fuzzier and fuzzier or until I get stuck and I start another project. Sometimes I just start another project altogether!

(Sorry - I didn't say I dealt with it *well*...)

=)

"So how does everyone else deal with non-structured projects?"

I structure them...

I plan ahead. I rarely make hard limits on values.

If I have an array, I'll have a constant which says MAX_ENEMIES, use that to initialize the array, and then use it whenever I need to loop through all items in that array, or perform some other operation on it which requires knowing how many potential enemies there are.

Of course, I do not use arrays for storing enemies, for that I use types, which are better for that. This is just an example. I use arrays more for things like storing height data for terrains.

You normally stop when the features you wish to add slow the project down waaay to much.

Ideally I get the basics in first that the game is unplayable without. Then after that (when the game is basically finished) I start to add the extra features in my to do list until I get bored, stuck, or like EdzUp said, the project starts running too slowly.