3) Find the nearest floating-point representation in this 12-bit construction for the following real numbers: (i) 19.25 (ii) -0.3333 and (iii) 0.005. (They may or may not be exact.) If a real number does not posses an exact floating-point representation, then there is a roundoff error for its representation, i.e., Eroundoff = |number (real) number ( floating point ) decimal X 100%. What is the maximum roundoff error for a normal floating-point representation in the 12-bit construction for -0.3333? What is the maximum roundoff error for a subnormal floating-point representation in this 12-bit construction for 0.005? 3) Find the nearest floating-point representation in this 12-bit construction for the following real numbers: (i) 19.25 (ii) -0.3333 and (iii) 0.005. (They may or may not be exact.) If a real number does not posses an exact floating-point representation, then there is a roundoff error for its representation, i.e., Eroundoff = |number (real) number ( floating point ) decimal X 100%. What is the maximum roundoff error for a normal floating-point representation in the 12-bit construction for -0.3333? What is the maximum roundoff error for a subnormal floating-point representation in this 12-bit construction for 0.005