Crowdjump: Investigating a Player-Driven Platform Game
summary
Player-run platformer Crowdjump runs for 23 days with 25 participants and targets two feature releases per day at 7:00 PM; players submit and up/down-vote textual ideas (138 suggestions) and 43 selected ideas get implemented. The team estimates 137 development hours but completes chosen features in 106 hours, ships on 19 days two features, logs ~13,000 plays (median 400, one player 6,000+), and finds only 6% of runs beat the level as it grows harder. Players favor game mechanics and aesthetics; community-building features are rare, and idea voting behavior stays stable.
AI-written summary · about
chapters (14)
- 0:00Introduction to Crowdjump
- 1:08Game design and player-driven features
- 2:55Study setup and participant details
- 3:57Game evolution and feature additions
- 5:05Estimation accuracy and release cycles
- 6:10Player feedback and challenges
- 7:14Community dynamics and voting behavior
- 8:38Future research directions
- 9:45Q&A session begins
- 10:05Question on voting and accounts
- 11:10Question on game difficulty changes
- 12:23Question on player influence distribution
- 13:25Question on surprising features
- 14:53Conclusion and thanks
related talks
27mStart with a Game Jam
45mHow to Fall Fast
47mGameJams or the art of fusing a week-end & coffee to create a video game
08mCreative Console: A Player-driven Game Based on A Modular Fast-Evolving-And-Verifying Framework
18mRapid prototyping for mobile games development
15mWhat Gamee learned about hyper-casual games from 3 billion gameplays across 70 games
35mBuilding a Local Community While Building Oliver and Spike
53mPlanning For The Win