This project was chance for me to practice my newly learned optimization skills from the Nanodegee program. Udacity presented me with a VR game that was extremely unperformant, my goal was to use my talents to optimize it to a consistent 90fps.
Performance optimization is one of the most important skillsets for a VR developer, since apps running below the target framerate are essentially unplayable, causing nausea and discomfort. This project simulates a real-world scenario where a VR game has been hastily built with a focus on functionality rather than performance (a common approach while prototyping). You’ve been hired to take the game and make it a performant app ready for release.
In this project, I learned that utilizing lightprobes properly is just as much art as it is science. I was a bit nervous about my implementation here but the reviewers reassured me that I took a great approach.
This project was about combining various ideas and skills I've been practicing throughout the course. They include:
- Using static and dynamic batching
- Optimizing physics
- Creating object pools
- Caching variables and optimizing code
- Setting up lighting with MSAA
- Baking lighting
- Using light probes
- Using the profiler for performance optimization
-SteamVR
