When I composed a raytracer, I output it as ASCII initially even if it was quicker/easier. It’s great to eliminate any blockers/excuses for beginning a task (like requiring to open a window, produce a surface area, output to it. choose which graphics library to utilize, and so on). When it initially revealed a ball shape over an aircraft, that was a great indication. After it was operating in ASCII, I output high-res with colors to RAW gfx files instead of the screen (once again, to be lazy). That had a good adverse effects of making it simple for me to pick HDR output (with drifting point RGB worths) which I saw with an external audience with good controls and might import from other software application, so it in fact ended up much better than if I had actually pursued my preliminary vision. As an additional bonus offer, I might render to a greater res than my screen. and the audience looked after perfectly tasting it when scaling it down to my screen resolution (so I generally got antialiasing free of charge).