Cimmerian is a pseudo 3D game written in C89 and similar to the likes of Wolfenstein 3D.
The libraries used are:
- naturally the C standard library,
- OpenGL (Open Graphics Library) for the graphics,
- GLFW (OpenGL Framework) for the windowing (= GUI & input handling),
- and OpenAL (Open Audio Library) for the audio.
I've decided to specialize in Unreal Engine, so why am I even working on a low level project? Because it's fun. Moving on... No seriously, I've always wanted to work on such a project, so I'm scratching this itch before going full throttle into Unreal.
I love the C programming language, and specifically C89 which is the version from 1989, but it also fits nicely into the idea of going retro. Being limited is also a good challenge for me, and there are different aspects to this project which limit my range of actions. For starters, C89 being an old version, it has limited features and a slightly stricter writing style.
Language aside, I've gone with OpenGL for the graphics library, and it's not for a technical reason and simply because I've always wanted to, for some reason. This is my first dive into low level graphics programming. From the beginning, I intended for Cimmerian to be a pseudo 3D game and not actual 3D, but if you look into the commit history you'll see that I've dabbled into real 3D before going back to pseudo. It was a great learning experience, and gave me an introduction to modern graphics programming. However, this was too difficult for me to figure out at this time so I couldn't go on this way, and went back to pseudo, which as previously stated was my intent anyway.
What even is "pseudo 3D"? Any graphics library, even LÖVE (or Love2D) which is a Lua framework, can be used to develop a pseudo 3D game. It simply requires to draw pixels on the screen. Basically, we take a 2D map which indicates where walls and objects are, and we translate it into 3D. The walls are cubes, the objects are sprites (= flat images) instead of 3D... Through the power of math (mostly trigonometry), these pixels can emulate 3D.
Going back to OpenGL, the "real 3D" detour taught me precious concepts. Precisely, when looking up raycasting tutorials which use OpenGL as an example, deprecated features are likely to be used, because the purpose is not to teach the library but the concept of raycasting. Thanks to this detour, I know that the "fixed function pipeline"/"immediate rendering mode" is deprecated, and that the "programmable pipeline"/"retained rendering mode" is the current way of doing things. It's about using shaders, and it's not only more efficient for the machine, it's the basics of modern graphics programming.