I've been trying to keep this project quiet until I make some significant progress on it, but it looks like there's quite a lot of things that I'm learning about that's fun to talk about. There's been interesting design decisions, so I think there's a bit for you guys to learn by making a project page dedicated to how I make this thing 
Easy coding vs easy storytelling
One interesting decision I had to make early was between long code and long scripting. The problem with anything long is not the effort put in writing it.. it's the effort put in reading it. Making long code makes it harder to debug it later; long scripting makes it hard to read and write new events.
Here's an example: the array storing a value for a population growth from a crisis looks something like cri["dwa"][0][0]["PopG"]=20. This meaning that a certain decision makes the dwarf population grow by 20 units per year. It's not hard to type this, but it's hard to read in a rush. This format gives you a lot of control, like over which race it affects, which events, and how much.
But I didn't need control. What I needed was fast storytelling. There are probably a few dozen events to write and I wanted to write them quickly. Maintaining control helps in debugging and makes programming a lot faster, but it doesn't help in writing an event at 2 AM, or on the train on my cellphone.
So, I went and wrote over 200 lines of code (including comments) trying to simplify them to a few letters. I ended up with sol_popg(20), a sweet little function. Was it worth half a week and 200 lines of code? I think so, especially seeing how much faster I can make things later on.
It's not exactly a foolproof choice, though. If I had only two races, I wouldn't go to the trouble. It's like a movie set.. you just go through enough effort to make it look what you want. What goes on behind the scenes is irrelevant, you'll likely scrap it all later.
How it started off
Web of Time started off as an experimental sub-project to another experimental project. It's about trying to see what happens when you let the AI control itself. It's about how some races influence each other and stuff.
The idea is to keep it as simple as possible. I keep the races as cliche as possible so that it's easier to control. Strangely enough, it's a challenge to keep it cliche. There are points in the storyline where you can make dwarfs who live in woods or barbarian elves. This leads to a whole new way of playing. New plot threads are spawned. More threads, more possibility of things going wrong and more exploits.
Funny thing is that when you make a simulation game, there's no such thing as an exploit. The obvious choice is the right one. This helps greatly in game design. You no longer have to worry about the game being "fair", the focus is on being interesting, on exploring the solutions. Sometimes you may even choose the "hard" way, like bringing your tree-dwelling dwarves to greatness.
|
Advertisement
Muz 
Favourite