[jlees]

"The best choice" is a woolly phrase. Best for whom? Stopping all development to totally rewrite the core game engine is best from a technical perspective, but that's not necessarily the optimal solution for player enjoyment -- and would have impact on other things, e.g. the competitive aspect of the game.

Continuing to develop while totally rewriting the core leads to wasted effort and a "running to keep up" effect, though allows for a strong technical foundation, provided you actually finish it. Incrementally rewriting the codebase takes longer, but is easier to do in flight, and easy to test in a modular fashion -- and that's already happening.