Making Floating Point Calculations Less Cursed When Accuracy Matters

An unfortunate reality of trying to represent continuous real numbers in a fixed space (e.g. with a limited number of bits) is that this comes with an inevitable loss of both precision and accuracy…