[bitwize]

I've found a good strategy is to focus on the game, not the engine. That is, don't go in hoping to write the be-all and end-all of game engines. Instead, sketch out a game that you would like to play and build as much engine as you need to support just that game.

It's easy to build an engine and then spend an eternity bikeshedding it to accommodate all sorts of games, while not actually writing any games. What you should be doing is focusing on getting a game out there, and letting the engine serve this goal and only this goal. Any engine work that doesn't directly contribute to a specific game reaching completion should be postponed or cancelled altogether.

I know because I used to be That Guy, the one who was going to write the awesomest game engine ever and use it to build awesome games. But the engine code seemed to get in the way of building games, and changing it took up a lot of time. So I pivoted to just writing a damn game, from scratch, and that so much better that the engine I wrote to run this one game was even awesomer than my universal engine.