This is my post mortem of Doughnut Factory
The requirements for this project were to do a complete game in 3 days time. I have always been wondering how people make like games in a week or a day or something like that and thought that the AGK2 would be the right environment to try this myself. Of course I am still wondering how one does it without such a framework.
Here is what I have learned:
I started of by looking for an idea, but I couldn't find a suitable one. Finding an idea took all of the first day, this is 33 percent of the whole day. When I started writing out the design document I had so many ideas, I saved most of them for later because of the time constraints, but even then I just gave up on writing them down.
So I had an idea, the idea was so simulate an assembly line in a doughnut factory and the player has to sort out doughnuts which are not up to par.
33 precent just to come up with an idea, was totally worth it.
Kicking out things is a constant to meet a deadline.
Suddenly you implement a feature not planned because it's just 5 lines of reused code.
Code quality does not exist, this code is horrible.
I have no optimization at all. (The performance did not suffer)
A final bug remains.
You implement lots of things just because you can do so without effort.
Sound comes last.
Tools and framework are huge timesavers
I ditched atlas images, because I had no good tools that automated the process.
Resolution is not everything
Resolution is everything
Doing Graphics does take as much time as coding, and I used a lot of stock too.
When time runs out and you have to finish, but you have a hard time doing so, just lay down and sleep.
You need to throw out stuff constantly
Concentrate on the core
I had to relearn the AGK2 interface, because I had forgotten all of it, and people changed some things in between
Complexity gets constantly reduced
I used stock graphics
1 Day finding an idea
1.5 Days doing graphics and coding.
0.5 Days of finishing early.
