I started doing game dev in DOS after reading Tricks of the Game Programming Gurus by Andre LaMothe in the 90s, so keep that in mind for the following advice.

I started doing game dev in DOS after reading Tricks of the Game Programming Gurus by Andre LaMothe in the 90s, so keep that in mind for the following advice. Depends on what you want to do. If you want to code games because you find the programming aspect interesting, start small by writing your own versions of some simple games. I personally wouldn't recommend any frameworks or libraries other than (maybe) SDL. Implement everything in the simplest way you can think of that will actually work and only go back and refactor if you need to, that's the time to look up how other people have solved that problem [0]. Resist the urge to over engineer. I might be biased but I say target PC first. Windows specifically, but Linux isn't much worse as long as you never plan to deploy the thing. This is because these platforms are incredibly open and there is a lot of information and tooling available. After you get a feel for it, and have a good idea of what you want to make next, start incrementally branching out in directions that interest you. If you have a good idea of the kind of game you want to make and want to start making it with as little friction as possible, then your best bet is to find an engine that is already well suited to that kind of game and learn just the things you need to in order to make it happen. Again, you'll want to start small regardless of what it is you actually want to make, just ensure that you're always moving toward that goal. That is very much not my path, so I have little other advice. [0] If you look it up first without trying it yourself, you won't have a good understanding of the problem space. You'll end up believing in the commonly accepted answer as dogma and severely limit yourself.