If ever I get to the point where I can be skilled enough to program a game, I should start off with a simpler single-screen game, to limit the amount of potential scope creep. It's important that it be a beginning project for which I can reasonably make design documents, make static and animated pixel art assets, compose music, make sound effects, repeatedly test and debug as thoroughly as I can, optimize as best as I can, and then at some point consider stable enough for release. Even a relatively simple game, where each playthrough might last a minute or two, can be a big undertaking, if I want to polish it up as much as I can. I also keep thinking: should I start with an overhead game, to not have to worry about implementing gravity, though I'll still have to make sure that a character will not clip past walls? Should I begin with a controllable character against stationary puzzle elements, so I don't have to worry about computer-controlled adversary routines my first try? The only game thus far that I can claim to have made, over 15 years ago, is an eight-second one in Wario-Ware DIY for DS (and I'm trying to figure out where my cartridge for it ended up). I was proud of how much I was able to cram into my small game -- though, I later realized it was also likely fairly unintuitive for newcomers, likely only being solvable after repeated playthroughs. That was with a pre-existing engine, too. I'm likely to be working at a lower level for any sort of next project.