3 ms·

Depending on how a simulation works, it could introduce phenomena or artifacts like results of numeric approximations into the universe that can be tested for.

by _yb2s 3y ago

Depending on how a simulation works, it could introduce phenomena or artifacts like results of numeric approximations into the universe that can be tested for. I would argue that unless the universe is using some sort of compute saving heuristic, then it is not in fact a simulation, even if it can be, and is run on a computer of some kind, because in that case it has its own independent “existence” in a mathematical sense. For example, if we could “run” an exactly identical copy of our own universe on a computer without errors, it is not really a second universe, but the same one, in the same sense that I could write out an equation and solve it on paper, or solve it with a computer algebra system, but they are still the same equation not two equations. This is based on the assumption or idea that mathematical existence is identical to physical existence.

staticman2 3y ago
Such a test wouldn't show we are in a simulation. It would show the universe follows weird rules, which is already what we know from quantum physics.
- _yb2s 3y ago
  If you can understand the nature of the simulation and then make and confirm different otherwise surprising and unexplainable experimental predictions consistent with it, this is a good as evidence gets scientifically. Sure there could always be some more complex or unimagined explanation for the same results, but that is reasonable to dismiss with Occam’s razor. If you can’t accept this, you would need to reject all past scientific discoveries as well, as this is how they were all confirmed. “Weird rules” to me just sounds like observing unexplainable things. It’s very different from being able to say before hand “if the universe is simulated using x bit floating point computation, we will get result 1 in this experiment, and if it has infinite precision as we’d expect from a non heuristic physical universe, we would get result 2.”
  - staticman2 3y ago
    >It’s very different from being able to say before hand “if the universe is simulated using x bit floating point computation, we will get result 1 in this experiment, and if it has infinite precision as we’d expect from a non heuristic physical universe, we would get result 2.” If you're in the physical universe, every experimental result, no matter how surprising, simply shows you what a physical universe is like. The physical universe not conforming to expectations is just a psychological problem that presumably comes from being a descendent of apes who survived and evolved when they expected certain things. If you're in a simulation, you have no idea what to "expect" because you're never seen a physical universe.
    - _yb2s 3y ago
      We can formulate and test hypotheses about the nature of our universe via experiment. I'm not talking about the universe "not conforming to expectations" or being "surprising" as evidence of a simulation, but of our ability to form and test very specific hypotheses, where the outcome is expected to be a certain way due to artifacts of certain types of simulation, thereby providing evidence for it. This requires a specific hypothesis about the nature of the simulation, and testing a corresponding prediction that we would never have thought to look for, or expected otherwise. The experimental confirmation of General Relativity by confirming that the perihelion precession of Mercury is almost exactly as predicted, is a good model of what I am talking about. Although the fact of perihelion precession was already known, the magnitude of it could not yet be accurately measured at the time, yet General Relativity was able to precisely predict the magnitude, which was confirmed much later when the measurement became possible. There are a large number of papers on arXiv outlining paths towards such experiments to test for specific types of simulation artifacts.
      - staticman2 3y ago
        In the sense an intelligent design advocate would also claim they can test "specific hypothesis" about god, perhaps.
        _yb2s 3y ago

by _yb2s 3y ago

staticman2 3y ago

Such a test wouldn't show we are in a simulation. It would show the universe follows weird rules, which is already what we know from quantum physics.

_yb2s 3y ago
If you can understand the nature of the simulation and then make and confirm different otherwise surprising and unexplainable experimental predictions consistent with it, this is a good as evidence gets scientifically. Sure there could always be some more complex or unimagined explanation for the same results, but that is reasonable to dismiss with Occam’s razor. If you can’t accept this, you would need to reject all past scientific discoveries as well, as this is how they were all confirmed. “Weird rules” to me just sounds like observing unexplainable things. It’s very different from being able to say before hand “if the universe is simulated using x bit floating point computation, we will get result 1 in this experiment, and if it has infinite precision as we’d expect from a non heuristic physical universe, we would get result 2.”
- staticman2 3y ago
  >It’s very different from being able to say before hand “if the universe is simulated using x bit floating point computation, we will get result 1 in this experiment, and if it has infinite precision as we’d expect from a non heuristic physical universe, we would get result 2.” If you're in the physical universe, every experimental result, no matter how surprising, simply shows you what a physical universe is like. The physical universe not conforming to expectations is just a psychological problem that presumably comes from being a descendent of apes who survived and evolved when they expected certain things. If you're in a simulation, you have no idea what to "expect" because you're never seen a physical universe.
  - _yb2s 3y ago
    We can formulate and test hypotheses about the nature of our universe via experiment. I'm not talking about the universe "not conforming to expectations" or being "surprising" as evidence of a simulation, but of our ability to form and test very specific hypotheses, where the outcome is expected to be a certain way due to artifacts of certain types of simulation, thereby providing evidence for it. This requires a specific hypothesis about the nature of the simulation, and testing a corresponding prediction that we would never have thought to look for, or expected otherwise. The experimental confirmation of General Relativity by confirming that the perihelion precession of Mercury is almost exactly as predicted, is a good model of what I am talking about. Although the fact of perihelion precession was already known, the magnitude of it could not yet be accurately measured at the time, yet General Relativity was able to precisely predict the magnitude, which was confirmed much later when the measurement became possible. There are a large number of papers on arXiv outlining paths towards such experiments to test for specific types of simulation artifacts.
    - staticman2 3y ago
      In the sense an intelligent design advocate would also claim they can test "specific hypothesis" about god, perhaps.
      - _yb2s 3y ago