Math_BigInteger provices support of base-2, base-10, base-16, and base-256 numbers. Uses the GMP or BCMath extensions, if available, and an internal implementation, otherwise.