Carrol calls it the "Resolution Conundrum" [0]. If you believe in the underlying logic that there is a kind of runaway effect, of simulations within simulations, etc, it's inevitably going to be the case that the most probable simulated universe is the one with the least available computing power. So whatever trickery you use to get around the problem of low resolution just substitutes one problem for another, and then the new thing runs up against limits of computing power. We would be most likely to experience the simulation that's most likely to generate noticeable artifacts as the simulation runs up against limits.

0: https://www.preposterousuniverse.com/blog/2016/08/22/maybe-w...