Please Explain

Question 1: Relative Error in Floating Point Consider a floating point representation with machine epsilon e. Let x be any real number with magnitude between the smallest and largest representable normal floating point numbers. What is the best estimation of the maximum relative error of x when it is represented in floating point? O (a) (b) 0 O (c) e/2 (d) e/lx O (e) 18 Save & Grade Save only