[fitswcs] Detector distortion correction representations in FITS

[Mark Calabretta]

On Tue 2000/02/01 16:18:35 -0000, Peter Bunclark wrote
and copied to: fitswcs at NRAO.EDU

>We're not getting close to saying, the CD matrix is too simplistic and
>you should have a `CD' polynomial instead?

I think I can kill this idea off very quickly simply by asking you to write
down the general form of the polynomial for a FITS image with N axes.

>  Which takes me back to an
>earlier point, I'd rather use the geometric projection which most closely
>matched the optical system (ARC for Schmidt, TAN for prime-focus are the 
>two closest to my heart) and apply the lowest-possible-powered
>correction to that to fit the real world.

OK, supposing we were to add the same polynomial to ARC as for TAN and
create a new ZED code for a straight zenithal equidistant projection
(noting that you can't get a zenithal equidistant projection from AZP as
you can for a gnomonic projection).

However, before I go any further with that I'd like you to answer the
three questions I asked in my last email on this subject.

Cheers, Mark