Should I use `strictfp` modifier when converting `float` to `double`?

Question

If I want to write method converting from float to double.

Is there any case the two would give back different result?

Which version should I use if I don't care about platform consistency?

What if I need it to behave consistently in different platforms?

  double f2d(float x) {
    return (double) x;
  }

or:

  strictfp double f2d(float x) {
    return (double) x;
  }

JLS 5.1.2:

A widening primitive conversion from float to double that is not strictfp may lose information about the overall magnitude of the converted value.

why will this conversion lose info?

Answer 1

strictfp will make no difference for float to double conversion. The reason is in Java Language Specification, 4.2.3. Floating-Point Types, Formats, and Values. In particular:

Note that the constraints in Table 4.2.3-A are designed so that every element of the float value set is necessarily also an element of the float-extended-exponent value set, the double value set, and the double-extended-exponent value set.

Since every element of the float value set is also an element of the double value set a float to double conversion will be exact regardless of whether it is FP-strict or not.