Hi Jonathan,
Thanks for your helpful message. Before I talk about the technical stuff I
should apologize for the title of my original message "dmarfadd seems
completely unreliable". When I _started_ the email (and typed the subject) I
didn't yet realize that my output was a copy of the first file -- I thought
the computations were just hosed (unreliable). :)
Anyway, the issue we're currently struggling with here is how to generate
reasonable ARF's for point sources that fall in the chip gaps, where the EA
(for a single CCD) can fall from nominal to zero in ~35 pixels. I know the
"right" way to do it is extraordinarily complex -- we need something
reasonable now. My thought was that computing several ARFs across our
extraction region and averaging would be a reasonable 1st order estimation.
And, of course, even if one chooses to accept ARFs calculated at single points
on the sky, there are lots of such locations that have data from multiple
CCDs. The user is forced to run mkarf on each CCD involved, then sum them.
(By the way I don't see why mkarf was written to deal with only one CCD at a
time.)
And, of course, even if one can ignore chip gaps, if you have multiple
observations of the same field and choose to fit merged spectra (rather than
simultaneous fits of spectra from each observation), then one must sum ARFs.
At PSU these are a non-trivial issues. Orion & HDF fields have hundreds of
sources, and both are multi-obsid campaigns. The fraction of the merged
field covered by chip gaps is substantial, and for dim sources it makes sense
to work with merged spectra (requiring summed ARF's).
Thanks for the pointers to the swap page -- I will take a look at what others
are trying.
Regards,
Pat
-- ==================================================================== Patrick S. Broos, Systems Analyst/Programmer patb@astro.psu.eduDepartment of Astronomy and Astrophysics My office 814-863-7947 Penn State University 525 Davey Lab FAX 814-863-8686 University Park, PA 16802-6305 http://www.astro.psu.edu Group office 863-9550 ====================================================================
This archive was generated by hypermail 2b29 : Thu Dec 13 2012 - 01:00:03 EST