3 ms·
About 1: There are some recursions that after some iterations approach a stable cycle, in the most simple cases it can be a cycle of length 2, like A -> B -> A
by gus_massa 6y ago
About 1: There are some recursions that after some iterations approach a stable cycle, in the most simple cases it can be a cycle of length 2, like A -> B -> A -> B -> ... but there are longer cycles, and in some cases no cycles at all. This example has a lot of possible behaviors changing the parameter r https://en.wikipedia.org/wiki/Logistic_map https://en.wikipedia.org/wiki/Logistic_map
- mannykannot 6y agoI don't think there are any causal cycles here - I don't think a simulation can have a causal effect in the sumulation (or real world) which launched it. Even though the simulator can observe the simulated, which might seem to provide a channel for communication, the latter cannot do anything that was not determined by the rules set up by the simulator.
- JadeNB 6y agoEvents in the physical world cannot do anything that was not determined by the laws of physics, and yet they affect the rest of the physical world.
- mannykannot 6y agoFair point - I suppose that, for example, a simulation runner could make a note of something that occurred in her simulation, and that would count as a causal effect on the simulating world from the simulated one. I still feel, however, that the story's requirement, that the simulated world mirrors the simulating world (except when the latter is the real world), depends on the physics of the real world (and its simulations) being maximally deterministic (in that everything that is possible happens with probability 1) or else the butterfly effect will cause them all to diverge quite rapidly. But with that level of determinancy, there is no need for a feedback mechanism to achieve "midpoint stability" - there is no possibility of divergence from the already-determined future.