I’ve spent years developing and tweaking my game engine. It is still not done. While I am happy with the progress I’ve made, I still have a list of additions for the engine a mile long. Unfortunately they will have to wait. I have to start a business for an entrepreneurship class I am taking for my MBA. I have $20 with which to start this business (and no free time). So ready or not I’m making a game with that engine. Oh, and I essentially have two weeks to finish the game.
There is some value to this activity though. It gives the engine the good shakedown that all my unit tests and demos try to emulate but ultimately fall short of. While they can ensure that my code is correct, the only way to really be certain that it makes it easy (well, easier) to make games is to well, make a game. I’ve already discovered a number of things which need to be changed to make the engine more developer friendly. I highly recommend that if you feel yourself getting burnt out while working on a game engine to try to make a game with it. Even if it is just an Arkanoid (Breakout) clone like I’m doing. It will probably reveal more about what your engine should be than you think.