Is any precision difference between (float int) and (float (float)int) in C ?-CodePudding

For example,

If I have int A = 106 and float B = 10.345f, which operation has better precision, A B or (float)A B? Or do they actually have the same precision?

CodePudding user response：

No there is no difference. When adding two arithmetic types, there is a set of implicit conversions that are applied by the compiler. You can find a list of these rules for example on cppreference. In your case rule number 3) applies:

Otherwise, if one operand is float, float complex, or float imaginary, the other operand is implicitly converted as follows:

integer type to float (the only real type possible is float, which remains as-is)

[...]

So if you do not explicitely state the conversion using a cast, the compiler implicitely does exactly the same conversion for you. And because the expressions are actually the same, there is also no difference in precision.

CodePudding user response：

If you look at the compiler results https://godbolt.org/z/x5174j8n9, then you see that functions

float add1(int A, float B)
{
    return A   B;
}

float add2(int A, float B)
{
    return (float)A   B;
}

give identical compiler output with O2 or O3 optimization enabled:

add1(int, float):
        movaps  xmm1, xmm0
        pxor    xmm0, xmm0
        cvtsi2ss        xmm0, edi
        addss   xmm0, xmm1
        ret
add2(int, float):
        movaps  xmm1, xmm0
        pxor    xmm0, xmm0
        cvtsi2ss        xmm0, edi
        addss   xmm0, xmm1
        ret