There exists a problem in floating point precision for floating numbers.