[Aavso-photometry] In defense of the ST-9 & CCDchoiceConsiderations

[arne]

Greg Crawford wrote:
> A couple of comments in the hope that I will learn from replies:
> 
> 1) I am not sure why this thread is picking on the ST9. It is just one of a
> number of CCDs with larger pixel sizes. Ultimately, pixel scale will depend on
> other factors such as focal length, so that the larger pixels of the ST9 and
> other larger pixel cameras may well give an appropriate pixel scale with some
> systems. Furthermore, fewer pixels leads to faster download for fast-changing
> targets. At the end of the day, every decision is a compromise, including
> economic factors. Some factors beyond control, such as the exchange rate between
> US and other countries, may make an ST9 the only economic choice and other
> factors, such as local seeing, may make it the only practical choice; in both
> cases achieving satisfactory results.
> 
This is correct; the ST9 was chosen in this dialog because the original
poster give the ST9 and the ST8 as his choices.  I do not think most of
the postings were "picking" on the ST9.  There are nearly an infinite
choice of CCD detectors and cameras, so there is rarely a "perfect" camera
for all situations.

> 2) I have always been puzzled as to why FWHM is chosen as the point at which
> sampling is measured. After all, the star profile usually continues to broaden
> below half maximum, even though this may not be apparent from the contrast
> regime of an image. If a star profile only covers 2 pixels at half maximum, may
> it not cover 2.3 or even 3 pixels at one quarter maximum, so that a larger
> measuring aperture will include sufficient sampling? 
> 
Again, everything is a compromise, but some guidelines have to be
set somewhere.  It is not "take one more step and you will fall
off of the cliff."  If you look at figure 9.6 of our Astronomical
Photometry book, you will see that roughly 30percent of the total
flux is contained within the fwhm, so any error in measuring inside
of one fwhm will affect your photometry at a significant level.  Analytical
functions, such as Gaussians for psf fitting, require proper sampling
of the profile in order to fit properly.  fwhm is a convenient
and easily defined parameter to use when specifying pixel size.
Arne