A collection of identical 3D scenes implemented across 20+ graphics APIs, engines, and platforms for rendering comparison.
Here be dragons is a code repository that implements the same 3D scene across over 20 different graphics APIs, game engines, and hardware platforms. It provides a direct comparison of rendering techniques, performance, and implementation details, helping developers understand the strengths and trade-offs of each method. The project includes versions for modern APIs like Vulkan and WebGPU, game engines like Unity and Unreal, and even legacy consoles like the Nintendo DS and PlayStation 2.
Graphics programmers, game developers, and computer graphics students who want to compare rendering implementations across different technologies and understand low-level vs. high-level approaches.
It offers a unique, hands-on reference for comparing rendering pipelines side-by-side with identical scene content, saving developers time from building their own test scenes. The inclusion of both cutting-edge and legacy platforms makes it valuable for both learning and historical perspective.
A basic 3D scene implemented with various engines, frameworks or APIs.
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.
Covers over 20 implementations from modern APIs like Vulkan and WebGPU to legacy ones like Glide and retro consoles, as listed in the README, offering broad comparative insights.
All versions render the same core elements—ground plane, Stanford dragon, rotating Suzanne—enabling direct apples-to-apples comparison of rendering techniques.
Provides tangible code examples across high-level engines (Unity, Unreal) and low-level APIs, helping developers understand trade-offs in rendering pipelines.
Includes ports to constrained platforms like Nintendo DS and Game Boy Advance, showcasing practical rendering challenges on legacy consoles.
The README openly admits that comparisons are inherently biased due to varied algorithms and hardware, limiting objective performance analysis.
Each implementation has a brief readme but lacks in-depth tutorials or error troubleshooting, requiring significant prior knowledge to navigate effectively.
Some versions like OpenGL ES are only tentatively planned, indicating gaps in the implementation set that may not meet all learning or comparison needs.