calcrmf2 bug fix, new tool, gfortran

[Alexey Vikhlinin]

BUG FIX:

Scott came across a bug in calcrmf2, when compiled with gfortran and
used with par.files with two levels of nested @includes. The bug does
not affect your fit results, even if you run into the same situation.
Also, if your spectral plots were not looking scaled by a factor of 3
in energy, you were unaffected.


mac osx binaries: http://hea-www.harvard.edu/~alexey/CHAV/chav-macox.zip
source code: http://hea-www.harvard.edu/~alexey/CHAV/chav-src.tar.gz

However, for those compiling from source it might be easier simply to
change rmf/calcrmf_simresp.f near line 165 to read:

if (.not.PICHAN) then
call get_cl_par ('gainfile',filename)
if (defined(filename)) then
calcebounds = .true.
else
calcebounds = .false.
endif
else
calcebounds = .false.
endif

and recompile using your previous config.


NEW TOOL:

Also, this release features a new tool, asol2aoff which can be used to
convert CIAO's pcad*asol.fits into the corresponding aoff. This might
be useful if you find small offsets between your obsrvations. CIAO has
a tool to apply offsets to asol's, and asol2aoff can be then used to
create the shifted aoff1 file for working with CHAV. Usage: as in

asol2aoff -o \!gov.aoff1 -refaoff acisf03182_001N002_aoff1.fits.gz \
-asol1 pcadf146054564N002_asol1.fits.gz -asol2
pcadf154544832N003_asol1.fits.gz


GFORTRAN:

I've started migrating from fortran77 to fortran2003 for the new code.
asol2aoff is written in 2003 and needs a fresh version of gfortran to
compile. Please consider using gfortran for your installation, it will
be the officially supported compiler. If you want to keep using g77
for now, please comment out the asol2aoff entry in the Makefile in
utils/