I'm rapidly coming to the conclusion that while the infinity machine is a cute idea, only certain kinds of ideas are meaningful to contemplate in the context of a working IM. Some ideas press up uncomfortably against the (patently absurd) assumptions that we've made in order for a IM to really exist.
I think that the idea of a full-universe simulation is one of those ideas. You suddenly hit a whole shedload of horrible violations of information theory the moment you try and put information into or get information out of that simulated universe. Remember - observe something, change that something.
That's why I think my simulated universe is more practical than yours. ;) I don't need mine to be accurate. I just need it to serve as an incubator.
Yours will wildly diverge from whatever reference point (and checkpoints) you set up, thanks to the fact that your simulation will be imperfect (your reference data can never be 100% precise) and that you are also observing it.
no subject
I think that the idea of a full-universe simulation is one of those ideas. You suddenly hit a whole shedload of horrible violations of information theory the moment you try and put information into or get information out of that simulated universe. Remember - observe something, change that something.
That's why I think my simulated universe is more practical than yours. ;) I don't need mine to be accurate. I just need it to serve as an incubator.
Yours will wildly diverge from whatever reference point (and checkpoints) you set up, thanks to the fact that your simulation will be imperfect (your reference data can never be 100% precise) and that you are also observing it.