hn.today

How I made a custom game engine for Satelital

spectralorbit.com4 points0 comments
Screenshot of How I made a custom game engine for Satelital

Satelital is a solo-developed puzzle game about discovering alien rules, built over seven years alongside a custom engine and development tools. Its creator chose to work from scratch to understand how games function beneath the surface, moving from early SDL2 prototypes to OpenGL and later Vulkan. Rather than build a general-purpose engine, they added only what the game required, replacing SDL’s platform support with a small macOS-and-Windows layer and writing a compact base library to reduce compile times and learn how standard programming tools work. Shader hot reloading proved especially valuable: instant visual feedback made graphics experimentation faster and avoided the cumbersome configuration system previously used to adjust effects.

The game aims for a dimensional appearance without rendering 3D geometry at runtime. Assets are prerendered from a fixed, top-down orthographic view, with albedo, normal, and height maps providing materials and depth; shaders generate animations and large planetary terrain. Ray marching over height maps creates dynamic shadows, while the same depth data lets objects sink into water and waves meet shorelines convincingly. Custom in-game editing made it possible to design and inspect puzzles in context, while a Blender plugin offered a less useful alternative. The account also emphasizes the costs of bespoke tools: Vulkan’s complexity, hardware-specific graphics bugs, inconsistent shader precision, and hard-to-reproduce player problems made validation, broad testing, and detailed logs essential. Limiting the engine to immediate needs - and refactoring freely - helped keep the long project manageable.

Read on spectralorbit.com0 comments on Hacker News

Summary generated by AI from the linked article. hn.today is not affiliated with Hacker News or Y Combinator.

More in Programming

Big Data Is Dead

Big Data Is Dead

Big data is not the main obstacle to useful insights: hardware has kept pace with data growth, while teams still struggle to make sense of their information. The author argues that organizations should focus on making data easier to understand and use, rather than buying tools built for ever-larger datasets. (motherduck.com)

The daily digest

Today's best Hacker News stories, summarized and screenshotted, one email a day.