I'd point out everything in your example is a characteristic of the program, not the runtime. There isn't any other runtime or compiler than is going to do significantly better, or at least I'm not aware of any. Regardless of language, if you want to be high-performance on that task, you're going to want to chunk your game world into tiles and dispatch tiles to workers, for all kinds of reasons, or something similar. (Assuming you even win from concurrency at all since this is on the borderline of where a single CPU may be able to keep up with RAM streaming speeds, assuming modestly clever encoding. And assuming a straightforward implementation rather than HashLife, etc.)

It's true that it's hard for a scheduler to "know" that useful work is going to be done, but that's just a fact of life, just like the CPU doesn't "know" whether or not you're going to use the next page of RAM or go flying off in an effectively random direction.