[fitsbits] Re: FITS vs. TIFF (or other image formats)

[Lucio Chiappetti]

On 19 Nov 2001, Paul Schlyter wrote:

> Rick Armstrong <careful at times.com> wrote:
>
> > 1) Why do astronomers use FITS, instead of just writing huge TIFFs or files
> > of some other format?
>
> FITS is a general data storage format, not just an image format.  A TIFF
> file can only store images -- a FITS file can store any kind of data.

As I tried to answer personally to Rick Armstrong this morning (but the
mesage was rejected because the e-mail address was wrong so I repeat it
here), that's not the only reason, particularly from an historical point
of view.

Astronomers use FITS because that exists from more than 20 years, back in
the magtape and mainframe era (including 16-bit, 32-bit, 36-bit, 60-bit
machines which explain the otherwise funny 2880 byte record length !) and
possibly pre-dates "all some other" formats.

Astronomers use FITS because they are interested in the exact measure
content : even in FITS images (at the beginning all FITS files were images
... hmm or random groups, but I never used random groups) each pixel is a
measure. Most other image formats are mainly interested in the look.

For instance some of them use lossy compression algorithms (JPEG I
believe), which could never be acceptable for an astronomer.

----------------------------------------------------------------------------
Lucio Chiappetti - IFCTR/CNR - via Bassini 15 - I-20133 Milano (Italy)
For more info : http://www.ifctr.mi.cnr.it/~lucio/personal.html
----------------------------------------------------------------------------