I have seen floating point bit hacks to produce the square root as seen here fast floating point square root, but this method works for floats.
Is there a similar method