Earth From Another Sun

Freeman Fan

Freeman Fan

Shenzhen, Guangdong Province

1 0
  • 0 Collaborators

Earth From Another Sun is a first-person shooter game developed with Unity 3D engine. It applies the most advanced High Definition Render Pipeline (HDRP) of Unity. The gameplay integrates Roguelike, RPG, and FPS elements, breaking the traditional gameplay limitations. ...learn more

Project status: Under Development

Game Development

Groups
Games China

Intel Technologies
Intel GPA

Links [7]

Overview / Usage

Earth From Another Sun integrates Roguelike, RPG, and FPS elements to provide players with brand-new game experience. It breaks the conventional idea of pixel art graphics for Rogue-like games, and uses first-person to enhance player engagement.

The game supports different computer configurations, and makes the most of hardware performance at different levels. The interactive computing and culling computing of most of the particles, special effects, and bullets etc. in the game are executed on the CPU. To maximize the computing resources of high-end platform, we also added some physical collision effects which are more realistic and complex.

Methodology / Approach

We hope that players can experience a very chaotic FPS journey. Therefore, we added a lot of special effects in the game. In order to ensure game performance and maximize the use of computing resources, we used CPU to compute the culling of special effects, which is attributed to avx and avx2 instruction sets. Resources like GPU rendering position, speed, and life cycle were also computed and submitted on the CPU.

To improve frame rate and reduce pressure on the main thread, main computations in the game are all executed asynchronously on the CPU.

Technologies Used

We used Unity Burst to optimize most CPU-side codes. (Unity Burst is a compiler very similar to INTEL ISPC.)

We encountered many times of game lags and unstable frames during the development. To solve this problem, we used Intel GPA to test unstable frames in Release Build and break the bottleneck of low-end platform.

https://software.intel.com/en-us/blogs/2019/06/27/2019-R2-intel-gpa-product-announcement

Comments (0)