This universe is a simulation? What a tangled web we weave…
Problem #1 — What is the number base? The sim cannot use any known number base because there are impossible (e.g. irrational) numbers in all bases, but these are fundamental to this universe. For instance, π is the ratio of radius to circumference.
Problem #2 — Turtles, all the way down? If we say that this universe is a sim, then this requires a different universe in which the simulation computer & software are running. That means that there is a “real” universe in which the computer and its software exist, or you have a “Turtles! All the way down!” problem. But if there is a “real” universe in which all those simulations are running, then what is the nature of that universe — the “real” universe exists in… what?
Problem #3 —Don’t look behind the curtain? If we decide to accept that some real universe exists, and say, “We won’t look behind that curtain!,” then we have a universe requires that requires energy & computational hardware, and has real physics limits. That puts limits on our universe’s simulation which would prevent such a simulation from computing the universe in which we find ourselves.
Problem #3A — Inevitable synchrony issues: Early in my career I was presented with problems of a massive simulation for military war-games. Bullets and lasers travel very fast — so fast that miniscule differences in clock rate and synchrony between computing centers create paradoxes and impossible outcomes. For instance, both simulation centers may calculate that red team and blue team scored kills, when a synchronized simulation can only have one kill.
Solution 3A1 — Cleanup after the fact. A solution of sorts is to do cleanup. But this leads to weird behavior, where both red team and blue team’s players “die” but then one team’s kill gets reversed. There may be an “argument” where an algorithm uses Markov modeling to decide, and the decision might even switch back and forth.
Solution 3A2 — Minimize cleanup. The solution that was best, but still not perfect, was to synchronize computing hardware clocks using GPS signals. This doesn’t entirely get rid of the problem, but it does cut the rate of such down tremendously.
In a huge simulation such as ours it would be inevitable that we would spot inconsistencies between different parts of our simulation.
Problem #3B — Simulation of endless Np hard solutions = non-polynomial time. From sims we know the time ratio between simple molecular docking & a sim of molecular docking is astronomical. Go ahead — check out Scripps’ Autodock. They don’t do actual complete simulations, and yet these can run for a while — and the results are guidelines.
The problem is that computational/energy requirements are non-linear. These are called Np hard problems, and a simulation of our universe would be filled with endless Np hard problems that must be solved.
Our conclusion is that simulations of universes end up having to face the same issues we do in this one.
