How can floating point calculations be made deterministic?

Question

Floating point calculation is neither associative nor distributive on processors. So,

(a + b) + c is not equal to a + (b + c)

and <

Answer 1

Use a decimal type or library supporting such a type.