> I really appreciate you taking the time to reply to this, because the argument is crystal clear in my head, I just struggle to convey it to others. No worrie

> I really appreciate you taking the time to reply to this, because the argument is crystal clear in my head, I just struggle to convey it to others. No worries! I'm interested in the discussion and really this is meant to be about healthy exchange of ideas to better understand different view points anyway. Apologies for taking so long to reply, hopefully you still see this. > Being "manifested with substance" is something you experience from the inside of the universe. How do you know there is substance? Because you experience it with your senses, that's all.v I don't necessarily disagree, but this isn't refuting my point. I'm not denying that if we are a simulation we have a mathematical description, i'm saying most of the substance of this simulation, whether it be experienced by us or described as math, came about as a result of the simulation, and was not described in the initial model. > You don't have to. Nobody needs to give any input. It's enough that if a given input were given then a given output would arise. The output would arise from the input regardless of whether anyone ran the simulation. This is only relevant if you can predict all possible outputs for all possible inputs, and you can't. No one can. > I'm saying you don't need to compute it, the outcomes were already there. That's the problem though. You do need to actually compete and run it to generate any meaningful data i.e. substance. Your argument, if I understand it correctly, is analogous to saying you don't need to play a game of CS to know how it will turn out, because you can generate all the possible scenarios. CS scenarios don't have an infinite space of possibilities, just a very large one, but even then it's more than we could compute, and it may still not capture everything (while everything in the game could be mathematically modeled it doesn't mean every aspect would be being recorded or considered significant). When you take that to a simulation with basically infinite inputs, it becomes unsolvable and impractical at least in any way we can understand. > You don't know how any individual match will go because it is non-deterministic (like the coin-flipping example), but the set of all possible outcomes can be inferred from the starting state based on the rules. They're all already there in the vast structure of all that is possible in the game. You just only simuulate one path. So you're saying our simulation is more like a match being replayed where everything is known? That's not really a simulation then, is it, but a replay. The point of running a simulation is generally to gather data. If, as in your scenario they (whoever 'they' may be) already have all the data, there would be no reason to run the simulation. > (The main caveat in my argument is that it begins with "IF the universe can be simulated" -- if you don't think it is possible to define a computable function that would simulate the universe given arbitrary time and space etc., then the rest of my argument is irrelevant). I certainly think a simulation is possible, I disagree on how predictable or known ahead of time the outcomes of that simulation would be.