Question

How can I convert SInt16 to double or Float32 to double? I know I can convert SInt16 to Float32 like this:

double *buffer;
SInt16* frames;

for (i = 0; i < number_of_frames; i++) {
  Float32 currentFrame = frames[i] / 32768.0f;
}

but I’m using many C functions and I want to pass data to them in double.

Now I use this code:

for (i = 0; i < number_of_frames; i++) {

  Float32 currentFrame = frames[i] / 32768.0f;

  buffer[i] = currentFrame;
}

but I’m not sure the result is correct.

Answer 1

Primitive C types (integers, chars, floating point types) can be inter-converted with casts. So given a signed integer value my_signed_integer, you can convert it to a double with a simple cast:

((double) my_signed_integer)

I’m guessing your Sint16 and Float32 are typedefs, so you can still do the same thing:

((double) my_SInt16_value) /* Just use this expression as a double
                              because it IS a double.  */

Eventually, this means:

for (i=0; i<number_of_frames; i++ )
  {
    buffer[i] = ((double) frames[i]) / 32768.0;
  }

^{Actually, you don’t need this cast. frames[i] will be automatically converted to double because of an implicit conversion.}