My advice to anyone setting out making games would be to pick any major engine with a trackrecord (can be broadly construed), stick with it, and just start pump

My advice to anyone setting out making games would be to pick any major engine with a trackrecord (can be broadly construed), stick with it, and just start pumping out games. Start small, like your smallest idea then cut that in half. Complete the game and ship it to Steam. There's a lot you'll learn in the last 20%. Grow the scale for the next game slightly. Repeat. If you're dying to build a dream game, don't. Save that for the next decade. Yes, decade. That's when you're skills will match your ambition. Ultimately, the tech isn't that important. Project management, scope, persistence, and execution are more important. Flipping around will drain your time, and reduce the rate you get better.