Fixed-point real computations to 8 decimals
