Scoping the lobe - What Every Computer Scientist Should Know About Floating-Point Arithmetic