If our reality is simulated, or if the idealists are right, then physics might well have temporal and spatial anomalities dependent on observability.
Ironically I've just elsewhere on this thread argued that materialism makes sense as long as we can not observe anything to violate it, and as such idealism is largely pointless. I should perhaps have added a qualification that this is true in terms of practical application.
That is, we have every reason to assume that the laws of nature are universal, but if they are 'just' universal within the constraints of what we can observe, then while we can continue to act as if they are universal, we may have no way of determining if they are.
As such I do think the question has no practical bearing. Yet we can not dismiss the possibility that what we observe as reality takes shortcuts when we're 'not looking'
Yes, you were actually replying to me above. We do already observe plenty of things that violate our current understanding, quantum entanglement being the first that comes to mind, but we definitely do not lack paradoxes as of now. Sure, as I stated earlier, they might just be knowledge holes which we might fill once our experiments become more sophisticated, but then I can't help but sensing faith.
If the simulation only aims to be convincing, it can take all kinds of shortcuts.
In the latter case it only matters if it is directly or indirectly observed. A simulation might choose to resolve that lazily in the case of indirect observations, in which case it'd only need to determine the proximity of entities deemed to be observers.
We have plenty of cases of that in games already, where events only take place if a player entity is present.