yeah - that's what i was referring to a couple of posts back. FITS is somewhat poorly defined, or maybe very broadly defined, so there's lots of legal ways to represent data. maxim is particularly guilty of this - rather than using 16-bit integer samples (for which there can be no ambiguity), they use floating point numbers in the range 0.0 - 65535.0 for images that have been pulled from the camera. PI expects 32-bit floating point fits files to have numbers in the range 0.0 to 1.0, so any f32 or f64 file that has out-of-range numbers needs to be dealt with specially. if you open the format explorer and then double click on the FITS handler, you will see a dropdown that gives instructions of what to do with data that is out of range. in this case you'd need to define the range as 0.0 - 65535.0 and then the files will be okay in PI.
but you are right, there's just no reason to use floats for the CCD data, and i16 is the foolproof way to go here.
probably because I have OCD
i know that feel bro, i know that feel.
rob
