You’re excluding a possible middle: Different conservation laws. They don’t have to be “magic,” just different enough that we don’t understand them, at our level of simulated cognition.
You also have dismissed the possibility that we, being simulated, work at a VASTLY slower clock-rate than the time-rate of the simulators. Maybe every microsecond we experience took them two of their years to compile. We wouldn’t be able to tell the difference, but it rebuts your argument regarding energy conservation.
(Real world climate simulations work much, much slower than “real time.” This isn’t merely an hypothetical suggestion.)
Voyager:
We just had a big thread about this nonsense in GD. Anyone who believe this doesn’t understand simulation, or the laws of physics for that matter.
When you write a simulator you concentrate on the level you want to learn something about. I’ve written several and have had a couple of papers published, but none of what I’m saying would be surprising to any simulator writer.
Say you are interested in simulating a computer. You can work at the transistor level, the gate level, a high functional level, or an instruction set level. But if you work at the transistor level you can only simulate a tiny bit of your computer, since it runs too slowly for the entire computer, and takes up too much memory. Gate level simulations run thousands of times slower than the computer does, transistor level ones run thousands of time slower still.
If we live in a simulation, we live in one where we can see the finest level of detail of the universe, so if our universe is 14 billion years old the simulating universe has been doing it for a couple of hundred billion years - so the comment about about it taking longer than the heat death of the universe is right on.
Forget about starting it last Thursday - if you knew how to set the state of the universe that precisely, you wouldn’t have to do the simulation.
Maybe you are thinking these aliens can invent really fast computers. Not good enough - by information theory processing information takes energy. I’ll leave it as an exercise for the reader to figure out how much would be required to simulate an entire universe, but I suspect you’d be lighting off a few supernovas per second to do it.
So - if one simulation per universe is even possible, which I doubt, many are definitely not possible. Which kinds of wipes out the statistical argument.
What could be possible is that the entire universe revolves around you or me or Musk, and we only see a Potemkin universe where just enough is created to convince us of its reality. But that isn’t a simulation.
What if you had ‘pocket battery’ dimensions filled with nothing but supernovae that the alien intelligence could simply switch in and out?
Or maybe that’s just what they want you to think. [Baa-baa-baaaaaaaaaaaah]