"you can use integer calculations but premultiply the numbers by a factor -say 100 if you want two decimal places."

Definitely the way to go - no need for floating point at all!

The technique is known as "Fixed-Point Arithmetic" - a search here & with your favourite internet search engine will reveal vast numbers of references!

Note that you don't necessarily need to stick with power-of-ten scale factors; use 50, or powers-of-two, or whatever is most convenient in your system!

It's just like working in pennies instead of pounds, or millimetres instead of metres, or inches instead of feet (to give a non power-of-ten example).