fixpix (again)
caldwell wrote on Apr 16, 2007
Before I get going, perhaps someone can quickly tell me whether
fixpix problems are solved in 2.13 ?
If not, here's the story. I'm running iraf 2.12.2a. My system is Fedora 6,
but the problem occurs on a solaris machine as well. The problem does
not occur in 2.12.2-export, which I still have running elsewhere.
Whether I run fixpix as a program, or within ccdproc, the mask files are
not read properly. The consequence is that interpolation across the bad pixels is not correct, almost unpredictable in fact, with different behaviours depending on the particular files or the size and shape of the bad pixel areas. Sometimes the replacement value for a large bad pixel area is
a single value, etc weirdnesses.
It does appear that the WCS info in the file and mask have to be the same,
and I'm quite willing to accept the possibility that my headers have
conflicting keywords, but as someone mentioned in an earlier post, it's
surprising that fixpix pays attention to such things. However, even files
with minimal headers have problems.
I have put three files in iraf.noao.edu/pub:
test.fits (unfixed)
testmask.pl (the bad pixels are marked with value=3)
test2.fits (fixed)
All I do is run :
fixpix test2 testmask.pl
As one example, look at the bad pixels around 2510, 2230.
thanks very much,
nc
fixpix problems are solved in 2.13 ?
If not, here's the story. I'm running iraf 2.12.2a. My system is Fedora 6,
but the problem occurs on a solaris machine as well. The problem does
not occur in 2.12.2-export, which I still have running elsewhere.
Whether I run fixpix as a program, or within ccdproc, the mask files are
not read properly. The consequence is that interpolation across the bad pixels is not correct, almost unpredictable in fact, with different behaviours depending on the particular files or the size and shape of the bad pixel areas. Sometimes the replacement value for a large bad pixel area is
a single value, etc weirdnesses.
It does appear that the WCS info in the file and mask have to be the same,
and I'm quite willing to accept the possibility that my headers have
conflicting keywords, but as someone mentioned in an earlier post, it's
surprising that fixpix pays attention to such things. However, even files
with minimal headers have problems.
I have put three files in iraf.noao.edu/pub:
test.fits (unfixed)
testmask.pl (the bad pixels are marked with value=3)
test2.fits (fixed)
All I do is run :
fixpix test2 testmask.pl
As one example, look at the bad pixels around 2510, 2230.
thanks very much,
nc
Mike Fitzpatrick wrote on Apr 16, 2007
Appears fixed in v2.13. I think what you're seeing is Bug #537 (http://iraf.net/article.php?story=7416&query=fixpix)
-Mike
-Mike
caldwell wrote on Apr 16, 2007
thanks, that did the trick.
nc
nc
Last post on Apr 16, 2007