Endless Runner World - Remixable world template

Skybox configured using GenAI
Player runner settings and track generation settings
Level Progression setup
Level defination configuration to sustain playability and freedom to Level designer to hand craft patterns
seamless binding of dynamic assets for gameplay and environment
seamless switching of environment skybox
Base framework

Inspiration

All Procedural generated games and user generated content has contributed to this highly modular and configurable open source template. In this project, we have driven deeper into game developing architecture and Meta Horizon's design patterns to articulate the best possible template for an endless runner game. Open world feeling and ability to setup all gameplay entities make this an unique framework to develop hundreds of runner games with the same responsive and high quality runner controls optimized for mobile gestures.

What it does

Using the framework, other developers can build any runner game, where the character the choose any world, be it under water worlds, where player can drive on a shark, or in an Chinese garden where can player can drive on a dragon back. The endless possibilities of driving any creature or flying object makes it a standout in this genre.

How we built it

we deep dived into various state of the art runner games, and tried to refactored all components that makes endless runners a can't put it down addictive and engaging gameplay. The key components that came out in our research are - procedural generation, fast loading, quick restart of next run, recognizable patterns, yet random sequence of level chunks. The one thing that stand out and well made in our project are the control and camera implementations, which are engineered to give best playability at the same time gives level designers great freedom to build engaging narrow misses consistently.

Challenges we ran into

Making the system modular was a bit of challenge, given our moto was not to compromise even 1% of player or remix developers experience. The random bugs and sometimes laggy internet connection put some hurdles in smoothing the camera, control bugs, yet we think we have accomplished the high quality target we aimed for.

Accomplishments that we're proud of

Manage to achieve our goal of all assets to be loaded dynamically and based on configuration from scripts is what we aimed for and proud of. The modularity design as visioned with the use of Object pooling and environment switching loaded to the Level layout maths on paper paid off very well at the end. We are confident, this is a state of art quality from usability point of view for other remix developer.

What we learned

Designing the framework and setting quality benchmark with no compromise goals at the beginning of the project helps a lot in smooth framework implementation. we also learnt various prompting techniques to test our framework for the GenAI based asset creation. (this was new to us, as we are in EU and AI in desktop app came in recently).