Talk Layout for SloC

layout.md

• Introduction (Ryan)
  • Personal Introductions
  • SloC is a procedurally generated dungeon crawler, themed around a developer preparing their project for release
  • It was created as a technical challenge to push our understanding of algorithms and data structures in a brand new language
  • Technologies were native JavaScript supported with Mocha as a testing framework
• Grid (Parker)
  • Abstraction of a traditional 2D array
  • Manages level state through individual cells
  • Replayability was a serious topic for us
  • Accomplished through managing cell state, procedurally changing values through certain semi-predictable patterns
• Generation (Ryan)
  • Biggest issue was optimization, which was solved with low-calculation algorithms
  • This begins with Room placement, which is accomplished through randomly generated sizes and a seeded number of attempts
  • Rooms are then manually connected with a tunneling algorithm, which connects two points from seperate rooms
  • GameObjects such as enemies, keys, and teleporters were then managed via random seeding and amount
  • These values were exposed through an easily usable internal interface
• Scaling (Parker)
  • Using the internal API, it became incredibly easy to create an infinite number of levels
  • We used a scaling value of player level in order to procedurally increase the difficulty of the game
  • While this approach worked well, we began to run into an issue with large map sizes, due to the number of redraws
• Camera (Ryan)
  • To solve this scaling issue, we implemented a dynamic camera
  • The camera finds the player position, and then draws a certain number of valid cells surrounding the player
• AI (Parker)
  • Our final challenge was creating real-time challenge for the player
  • We accomplished this via enemies that follow the player in real time
  • During development we discovered an unintentional feature, which was the ability for the enemies to track the user through walls and then dig to them
  • This actually resulted in an unintended improvement in game immersion, improving upon the game as whole and wrapping up the project
• Conclusion (Ryan)
  • This project presented a number of great technical and design-based challenges
  • A number of students continue to play, challenging one another to get farther
  • Going forward we'd like to add a backend to make it easier to compete with others
  • You can find the game and it's source code at the following links