Repo for my LD38 Compo entry
ALRIGHT. At this point I've got something like 45-ish hours left. Let's say I'm sensible and sleep and eat for a bit in there too, what that means is it's not worth breaking up that time into "by time X, I should have done this". OK. Fine. I can work with that. Regardless of anything else, I've got a limited amount of time, and I want to leave time for debugging too.
At each top level, I want to do at least the first thing for each one (some game, some art, some music), and delve deeper into the trees as time permits.
LAST MINUTE LIST
-
Finish level 1
-
Fix janky animations
-
Splash screen/menu
-
Make the game
- Make the unity project
- Top down framework
- Controls - DONE
- Simple level - DONE
- Triggers (level completion)
- Level generation - NOT DOING (not enough time to learn and do)
- More complicated levels, obstacles (add as needed?)
- Procedural generation?
- AI
- Make some simple AI - DONE
- Add damage/health (damage is to key? Maybe you get a bunch of 'lives' at that level, until the key is gone?)
- Generate some more complex AI behaviours - NOT DOING
- Pathfinding?
- Level transitions
- "Zoom out" effect - kind of important for the theme? Maybe context is enough - you become the enemy you faced in the previous level.
- Level theming
- What makes each level interesting and unique?
- Different enemies/weapons
- Create the different levels
- Mouse
- Cat
- Person
- ?Car
- Mech
- Ship
- ?Bigger ship
- Spacestation
- Moon
- What makes each level interesting and unique?
-
Make the arts
- Make the programmer arts
- Simple sprites, no animation - DONE
- Make the animations
- Animate main sprites
- Animate tertiary sprites
- Make the splash screens
- Do the thing
- Do a logo?
- Make the programmer arts
-
Make the musics
- Make a simple level loop
- Make the menu music
- Make different level music for each level
- +/-1 effect
- There's going to be a level above and a level below you, after the first level.
- Will you affect the level below?
- Does killing the things below you hurt your chances on the current level? * If you lose at a level, do you go back to the previous one?
- Refactorings!
- At the moment the AI has several different behaviours, that trigger different things based on the context, so a given 'behaviour' is spread out among many contexts. Would be good to group all the behaviour's behaviours together in one class, and register callbacks or something?
- Try AddForce instead of modifying velocity directly