When you convert the number to a string you're really transforming it to a decimal format. Which is the domain where you should be solving the problem. Otherwise you're doing some sort transformation in the binary domain and then hopping to pull the answer out of a hat when you do the final convertion to decimal.