Spinnerets --- A Post-Mortem
Hey everyone! Now that rating has kicked into gear, I thought I'd share some insights and takeaways from my eighth Ludum Dare Compo.
Overview
For this event, I ended up going in a different direction than usual - local multiplayer. While I don't know as I'd necessarily recommend it, since it places more restrictions on who can accurately review your game, I think the experience was positive overall.
Of course, game design doesn't happen in a vacuum. These are some of my favorite multiplayer games that influenced my design decisions:
0Space (Teknopants): Also a game jam game, 0Space is one of my go-to examples of great game design in a multiplayer setting. One thing worth noting is that the gameplay isn't particularly fast... it's strategic, and relies on precise timing. (It's also now free on Itch! Highly recommend!)
12 Orbits (Roman Uhlig): The pinnacle of one-button multiplayer games, 12 Orbits lets you play with up to twelve people on a single keyboard. When I decided to make a game with one-button controls, I took a hint from their key select screen and let players bind to arbitrary buttons.
TowerFall Ascension (Maddy Thorson): Another excellent couch co-op/competitive title, TowerFall does a lot of things right in terms of replay-ability. Notably, it has a variety of maps and powerups that change how players interact with the game at any given time.
What went well
Overall, I think I did a pretty good job this event.
The Pivot: I originally was going to make this game a dungeon-crawler rogue-lite with the same theme and controls. I realized the scope for this was likely too large early on, and pivoted to local versus, which was definitely for the better.
The Polish: With this reduced scope, I was able to spend more time on polishing the UI, character select screen, animations, and effects.
The Feedback So Far: People seem to like the game so far! Which comes as a little bit of a shock, because I never actually playtested it with multiple people, just me versus empty players or some of the very early enemy code I had written.
The Sleep: I actually took care of myself this weekend!
What went poorly
Still room for improvement!
Rushing the Music: I think the music turned out decent, given time constraints, but it was really rushed. My usual workflow for music was changed after my student Sibelius license expired, so I ended up actually directly recording me playing my Casio keyboard (after running to Office Depot for a quarter inch adapter). I only had time for two or three takes, and was mostly improvising, so you can hear a few spots where I hesitate or mess up.
Scope of Art: I think I took things a bit easy in terms of visual assets. The game looks clean, but there's not much to it. I actually only did one frame-by-frame animation --- the bullet exploding --- and everything else was with modular animation or no animation at all. I would also have liked to have more interesting map components, like hazards or conveyor belts, to spice up the gameplay further.
Collision Code: My collisions are suspect at best. Occasionally, players will pop through walls. In 48 hours, it's tough to write robust collision code from scratch.
Going forward
While I probably won't be making another local multiplayer game for Ludum Dare anytime soon, it was definitely a fun experience. In the future, I'll probably want to start with some base code for 2D collisions and tile system, since those things took a lot of time this event and are very generalizable.
