i. __The machine epsilon is the size of the smallest increment that we can differentiate between two adjacent values in the mantissa (regardless of the exponent). Values with a difference smaller than machine epsilon are seen as equivalent in floating point. For our system, it is equivalent to the difference between your answer to parts c and d above: 1/32. If you ask R for `.Machine$double.eps`, you get 2.220446e-16. What power of 2 is that equal to?__

h. __With our 10 bit floating point system, the machine epsilon is 1/32. What will the following return? (Hint: can you represent the left-hand side with a sequence of bits that is different from the right-hand side?).__

(1 + 1/64) == 1

(1 - 1/64) == 1