Suppose your floating point numbers had only 8 bits with a 3 bit exponent and a 4 bit mantissa, using all the same conventions as the IEEE standards (biased exponent with all 0's for subnormal numbers and all 1's for +/- infinity and NaN). Show your calculations and/or thought process for each of the following.

What is the smallest possible normalized positive number?