My take on roguelikes have a long start... four years ago with a little experiment, Goblin Invasion. It's not a game and it's not working anymore, but is a start as player and developer to understand these games. What is going on under the hood ?
I've started playing more roguelikes, forget about graphics and think about mechanics: what is cool? How to implement that ? And then Notch (do you remember? Minecraft before Microsoft.. ) develop for LudumDare Prelude of the Chambered.It's incredible how much you cand do, if you want and in small amount of time!
Then finally some decent roguelike tutorials from Trystan!
So I've started to use my library, MarteEngine (see Drone Defense, Fuzzy) to develop a little roguelike, here last MarteEngine roguelike tutorial.
But.. I was missing the point: understand the basics. So I've created CryptoRl, with some limitations on my mind:
- small in scope: 3 type of monsters, 3 items
- few mechanics: attack, get item from ground, use it
- small iterations, develop quickly
- no graphics
What went right
- small scope: don't do too much, I have a lot of ideas, but .. keep it simple! (stupid, kiss :P)
- quick iterations: release somenthing, request for feedback, but as developer "see" your project working and growing is rewarding
- use a TODO list, note ideas, but keep organized: first bugs, then TODO and finally IDEAS
What went wrong
- interface: I don't like characters only interface, using mouse could help a lot to figure out what is going on!
- animation: many roguelikes demonstrate that today I can develop a game with simple graphics, but few animations can help a LOT
Now it's time to collect ideas and move to a new project!