> arbitrary precision libraries often come with their own optimised Fibonacci functions. I will check out some of these implementations. http://cs.stackexchan

> arbitrary precision libraries often come with their own optimised Fibonacci functions. I will check out some of these implementations. http://cs.stackexchange.com/questions/7145/calculating-binets-formula-for-fibonacci-numbers-with-arbitrary-precision http://cs.stackexchange.com/questions/7145/calculating-binet... This post suggests that the matrix solution is desirable. I wonder if the complexity of performing exponentiation on high precision values would ever overtake a logarithmic solution.