7) To add two floating point numbers a and b with equal exponents, we can simply add the mantissa together, and then adjust the exponent by 1 if there is a carry. When exponents are not equal, we should modify the mantissa of one of the two numbers a and b (by shifting it) such that the exponents become equal, and then we can add the mantissa together. In doing so, we should not add any_significant zeros to either a or b (See 'significant figure' on Wikipedia if you don't know what it is. You may also want to consult Google on floating point addition Consider a variation of IEEE 754 with 4 bits for exponent, 8 bits for fraction, and a bias of 7. Add the following numbers, once assuming to have unlimited precisions, and once under this variation of IEEE 754. Calculate the numerical error between the two cases. .4 A (1.01100000)2 x 2 9 , B (1.01100001)2 x 2