Honestly, if you're asking this question then you're not ready to write an engine. Don't worry about engines; write games. Once you get a few under your belt you'll begin to see parts that can be reused, bits that could be reused if they were coded a bit more generically, etc. Essentially your engine will grow and evolve over time from these bits. Without doing this you simply lack necessary experience to know how to approach things and what to avoid.

I gues I could give some more immediately helpful advice ;)

C++ is rarely, if ever, a good first language to learn. It's possible, and many have gone that route, but people tend to recommend Python or C#, and you can get some decent early results with either language.

People often suggest starting with some simple text-mode games before moving on to shiny graphics: guess-the-number, is probably easiest and most common. Some people like to work their way up to a simple text-mode RPG. These programs help ensure you know the fundamentals of the language before tackling more complex problems.

Once you make the jump to graphics, the recommended path is: Pong, Tetris, Pac-Man, then some knd of platformer. Pong is about as simple a graphical game can get, but still offers plenty of opportuniy to learn. Each game builds on concepts learned from the previous one and introduces new problems to solve and game elements to consider.

Don't worry so much about making it "the right way." your focus should be on making it work. Nearly any chunk of code can be improved, and it's all too easy to fall into the trap of improving existing, working code rather than finishing the damn game. If you don't like the way you coded something or think there must be a better way, comment the hell out of it so you can find it again after the game is finished and playable, and ask about it here. Apply what you learn to your next game. Applying it to your last game rarely grants any obvious reward.

I would recommend sticking with 2D for a while. There is little if any real difference in the code, but 3D assets are more complicated than 2D assets (and generally take additional skills and software to create, although there are plenty available online if you can find something that suits your needs) and dealing with that added ccomplexity can distract you from where your focus should probably be.

Elements of a game engine include audio, graphics, user input, AI, asset management, game logic, save/load system, physics, networking… There are other possible systems and not all those listed are needed in every game. Each of those is a complex subject in it's own right, as well as the task of integrating them into a cohesive whole.

That should give you some stuff to think about and some direction, at least. Good luck and enjoy the ride. This hobby is both among the most frustrating and most satisfying experiences I've found ;)

I gues I could give some more immediately helpful advice ;)

C++ is rarely, if ever, a good first language to learn. It's possible, and many have gone that route, but people tend to recommend Python or C#, and you can get some decent early results with either language.

People often suggest starting with some simple text-mode games before moving on to shiny graphics: guess-the-number, is probably easiest and most common. Some people like to work their way up to a simple text-mode RPG. These programs help ensure you know the fundamentals of the language before tackling more complex problems.

Once you make the jump to graphics, the recommended path is: Pong, Tetris, Pac-Man, then some knd of platformer. Pong is about as simple a graphical game can get, but still offers plenty of opportuniy to learn. Each game builds on concepts learned from the previous one and introduces new problems to solve and game elements to consider.

Don't worry so much about making it "the right way." your focus should be on making it work. Nearly any chunk of code can be improved, and it's all too easy to fall into the trap of improving existing, working code rather than finishing the damn game. If you don't like the way you coded something or think there must be a better way, comment he he'll out of it so you can find it again after the game is finished and playable, and ask about it here. Apply what you learn to your next game. Applying it to your last game rarely grants any obvious reward.

I would recommend sticking with 2D for a while. There is little if any real difference in the code, but 3D assets are more complicated than 2D assets (and generally take additional skills and software to create, although there are plenty available online if you can find something that suits your needs) and dealing with that added ccomplexity can distract you from where your focus should probably be.

Elements of a game engine include audio, graphics, user input, AI, asset management, game logic, save/load system, physics, networking… There are other possible systems and not all those listed are needed in every game. Each of those is a complex subject in it's own right, as well as the task of integrating them into a cohesive whole.

That should give you some stuff to think about and some direction, at least. Good luck and enjoy the ride. This hobby is both among the most frustrating and most satisfying experiences I've found ;)

To be honest I'm not yet familiar enough with C# and Python to recommend one over the other. C# does have a more similar syntax to C++, but once you really have enough experience syntax tends to become less of a concern.

I strongly encourage you to begin a programming blog. You can have one right here at GD.net, and many people do. Documenting your progress helps in many ways: putting what you learn into words helps to reinforce the concepts in your mind; discussing your ideas for games helps you to develop them, and gives others the opportunity to provide input, which may lead to improved ideas; it gives you something you can look back on over time and see just how far you've come, giving you a sense of accomplishment; and it allows others to see what you're doing, which can inspire them to begin or resume game programming, and possibly have their own blog, which serves to expand and improve the entire community.