Al,
>Fabio's method is common usage.
The way I expressed the difference might not have been the norm (at least from your and Fabio's POV), but it was not incorrect.
>English may be imprecise but your meaning wasn't. Although the error in this case was small because the relative sizes are small, your calculation method produces results most would consider absurd when relative sizes are larger.
There's a situational dependency, given other sets of numbers I'd probably have done the calculation using one of the other 3 choices.