Overview

Ludum Dare Submission Page (11th Overall)

INFINIWITCH is a top-down bullet hell shooter where you summon & strategize with copies of yourself in order beat the clock and defeat the boss.

The game revolves around a 15-second timeloop, wherein any actions are recorded and re-performed by a summon of yourself on the next loop. Different classes have different abilities that, while powerful on their own, can be combo'd together across loops in order to deal massive damage.

Development Role

INFINIWITCH was created in 72 hours for the Ludum Dare 55 game jam by a team of 3 people.
During the game jam, I accomplished the following:
  • Designed & implemented all gameplay features
  • Created a timeloop system wherein player actions are recorded & replayed
  • Designed 3 player classes with versatile abilities that can be combined across lives, opening deeper strategization between loops
  • Designed a 4-stage bullet hell boss with ramping difficulty
  • Created all Menus & UI

Abilities

Each character in INFINIWITCH has an Ability. Abilities require the player to interact with their past selves in order to be useful - for example, the Mage can throw out a bomb, but it must then be detonated by the Ranger or Warrior. This creates a deeper level of strategy where players aren't just blindly throwing bodies at the boss - they must form a plan with their past and future selves.

Post-Mortem

Something always goes wrong, but given what we ended up with, I'm pretty pleased with how we did - 11th place out of ~2500 games is not something to turn up your nose at! That being said, it was a bit of a rocky road to get there, and I definitely could have done some things better as the acting team leader.

Things that went well
  • The game's core mechanic was unique to the jam, which actually surprised me. The game loop of having a discrete, 15-second bullet hell, with your power and numbers slowly rising over time, proved to be a very successful formula that both casual and hardcore players could hook into - something further bolstered by the late inclusion of a scoring system.
  • Like our previous Ludum Dare, a lot of care was put into creating robust systems up front, which allowed us to create content at an extreme pace when the time came. This included an ability system, an enemy system, and a unified damage pipeline for all entities, which made hooking everything up a breeze.
  • Feature implementation ended up being finished far before anything else, which freed me up to polish the game with a scoring system and a better user interface, as well as fine-tuning audio.
Things that went poorly
  • The lack of playability by the end of Day 2 led to a major morale crash for my team, who felt like the game was a wash - this persisted for much of Day 3, where one of the team members temporarily quit. Despite the fact that I managed to win everyone back by the end of the day once the gameloop fell fully into place, I could have done a much better job managing morale as the active producer; such as talking the team through my thoughts behind the design instead of the rushed 'Just trust me' method that I fell into under the time crunch.
  • Day 4 (Submission Day) was a bit of a rush - We still submitted a few hours early, but the lack of work from the previous day cascaded into the final day. This led to unnecessary crunch for my team, as much of the game was still without assets less than 24 hours from submission.