Thu, 27 Aug 2020
Despite PacBun being a fairly small and simple game, I'm using it as a dry-run for implementing something more advanced in the (near?) future. That means I'm heavily straying into the dangerous territory of over-engineering.
For example, I've been working over the last week to make the game less hard-coded and a lot more data-driven. The aim is to allow most of the game to be put together with content defined in data files read by the game and leave the "kernel" of it as a much more generic framework or, dare I say, "engine" (perhaps "motor" might be a more accurate term).
There's big reasons to do this (I hope):
- it allows me to create the game without rebuilding (or even restarting) the main part of the program
- content creation is defining rather than coding - should be simpler
- it forces modularisation so that changing one bit of the game is much less likely to break another part
- it means the "motor" should be more flexible
- it opens the possibility of others being able to add content to my game(s) - i.e. mods. I'll essentially be modding my own game and it's in my interest to make that experience as easy as possible
I've obviously had to provide a bunch of test or placeholder content to work with the framework and it's very rough just now. You could say that I'm deep down a rabbit hole right now...