Yes! That happens when you exhaust the limits of double-precision floating point, at zoom level 48 (that is, 2^48 x magnification). One could do better with ar

Yes! That happens when you exhaust the limits of double-precision floating point, at zoom level 48 (that is, 2^48 x magnification). One could do better with arbitrary-precision arithmetic, but I wanted to keep the code small and simple - and also thought this is a neat illustration of why double-precision numbers are really not the same as reals.