Skip to main content

Come analyze HEASARC, IRSA, and MAST data in the cloud! The Fornax Initiative is now welcoming all interested beta users.

******************************************************************************
                        RELEASE NOTES FOR HEASOFT 6.0
                               April 12, 2005
******************************************************************************

                                   SUMMARY

HEASOFT 6.0 is a software suite consisting of SWIFT 2.0, HEATOOLS 1.1,
FTOOLS 6.0, FV 4.2, XIMAGE 4.3, XRONOS 5.22, XSPEC (version 12.2.0 and
11.3.2), and XSTAR 2.1kn3.  This document contains notes about significant
changes made since the last major release of each.  All this software shares
common build, installation and initialization procedures.  Please see the
file HEASOFT-INSTALL.TXT for details.  Please see README.CYGWIN if you are
planning to install in the Cygwin environment.

New in HEASOFT 6.0 is the HEADAS build environment which replaces earlier
versions of configuration files and Makefiles as used in HEASOFT 5.3.1.
The build procedure remains basically the same however, i.e. modeled after
a typical GNU software distribution.

For each of the HEASOFT subpackages below are listed tools and/or features
which are new in version 6.0. More information on any of these packages
can be obtained using the fhelp utility.  Type "fhelp ftools" to obtain
a full listing of all the tools available, or "fhelp <sub-package-name>"
to list all of the tools in a particular subpackage, e.g. "fhelp swift".
Finally, type "fhelp <toolname>" for more detailed help on a particular tool.
FV, XIMAGE, and XSPEC all have online help available from within the program.
In addition, XIMAGE, XSPEC, and XSTAR have TeX formatted manuals.


******************************************************************************
                                  XPI / PIL
******************************************************************************

- In the HEADAS framework, all new tools use PIL (the ISDC Parameter
  Interface Library) instead of XPI (which is still used by the older
  FTOOLS).  New versions of all the parameter utilities (except for
  pquery), that is:

       pget plist pquery2 pset punlearn

  have been written using PIL, and are now the default versions you will
  find in your PATH after HEASOFT initialization.  These new versions are
  intended to reproduce the behavior of the xpi versions, but by virtue of
  using PIL there will be some differences.  If you find that you have a
  need for the older XPI utilities, you can still build them in the
  following way: after building HEASOFT, cd headas/ftools/xanlib/xpiutils
  and then "hmake xpitools" followed by "hmake xpitools-install".  The
  xpitools will be named with an "xpi" prefix, e.g. "xpipget".

******************************************************************************
                                    ASCA
******************************************************************************

- ascaarf: Fixed bug when accumulating the gis detector efficiency.
  Also, decreased run time on Linux.

- ascascreen: Fixed problem with output format with some versions of Perl.

- cleansis: Modified to work on event files from instruments other than the
  ASCA SIS.  The behaviour with ASCA SIS event files is unchanged. For other
  missions, cleansis gets the number of chips using the TLMIN/MAX of the column
  specified by the chipcol parameter and the image size from the TLMIN/MAX
  of the column specified by the rawxcol parameter.  Improved error handling
  if TLMIN/MAX keywords are missing for RAWX/Y or chipcolumns.

- mkgisbgd: Fixed problem which caused the background image in the primary to
  be corrupted when the GIS image bin size was 64x64 (which is different from
  the default 256x256).  The background spectral data in the first extension
  and other images in the second and third extensions were correct.

******************************************************************************
                                  CALTOOLS
******************************************************************************

- rsp2rmf: now available on Linux (now builds with g77/f90)

- genrsp: Corrected bug when reading from a file resolution information for
  a dispersive instrument.  Improved the diagnostic error if the user has
  given an invalid value for the res_reln parameter.

- caldbinfo: Modified to handle Swift/XRT namespace conflict with older
  XRT instruments. 

- cmprmf: Modified to update the value of the LO_THRES keyword in the
  output rmf file.

- rbnrmf: Fixed bug which generated a segmentation fault in the case where
  ebdfile=rmffile.

- udcif: Updated to allow users the option of including a file to be indexed
  in the CALDB even if it seems to be a "duplicate" of another entry in the
  caldb index file.  This is useful since the determination of a duplicate
  did not previously check for differences in the CBD block, which can be
  used to distinguish files.

******************************************************************************
                                   CFITSIO
******************************************************************************

CFITSIO 3.001:

Major changes have been made to the CFITSIO library in this release to fully 
support large FITS files which may have size parameters that are larger than 
2**31 = 2147483648, i.e., values that are larger than can be represented by 
a 32-bit signed integer.  CFITSIO now supports FITS files that have:

     - integer FITS keywords with absolute values > 2.1E09
     - FITS files with total sizes > 2.1 GB
     - FITS tables with more than 2.1E09  rows
     - FITS images with an axis length larger than 2**31 pixels

To support this change, the data type of the integer parameters in many of 
the CFITSIO subroutines has been changed from a 32-bit 'long' to a 64-bit 
'LONGLONG'.  Fortunately this change is transparent to existing software 
that only calls the public CFITSIO routines (as defined in the fitsio.h 
include file) because the C compiler and the cfortran.h macros called by 
Fortran programs will automatically perform the data type conversion between 
32-bit and 64-bit values as necessary.   Existing software should run 
exactly as before without any modification, but to fully take advantage of 
this new support for large FITS files some applications might need to be 
modified to use 64-bit integer local variables to store parameters such as 
the image axes sizes, or table row numbers.   There are a number of CFITSIO 
subroutines which pass the address to an integer parameter which could not 
be modified to support 64-bit integers without breaking existing software, 
so in these cases a new subroutine has been added to the CFITSIO library to 
specifically support 64-bit integer arguments.

This backward software compatibility of CFITSIO does NOT apply to any 
application programs that include the fitsio2.h file and then make calls to 
some of the privately defined routines inside the CFITSIO library.  Some of 
these internal routines are not compatible with previous versions of the 
library (especially those routines that now pass a 64-bit integer parameter 
by reference), so programmers that make use of fitsio2.h should carefully 
check which routines are called and make appropriate changes to the calling 
routine if necessary.  It is possible (and even likely) that some of these 
internal CFITSIO routines will continue to change in future releases, so 
programmers are strongly discouraged from using fitsio2.h or calling any of 
the undocumented internal routines in CFITSIO.

Other significant changes to CFITSIO in this release are:

  - New functions have been added to the lexical parser (as used by
    tasks such as fselect and fcalc), including MEDIAN, AVERAGE, STDDEV,
    ACCUM (accumulative sum), and ANGSEP (compute angular separation).
    Do 'fhelp calc_express' for more details.

  - Improved the routines that write 'DATE' keywords to rigorously
    verify that the input day, month, and year values are valid.

******************************************************************************
                                  FV / POW
******************************************************************************

FV Version 4.2:

New features/Bug fixes since V4.1.4 include:

    Add X axis range selection utility in POW.
    Add capablity to display 4D table and image (movie).
    Add capablity to create/display region on image.
    Add XPA entry points for X axis range selection and region display.
    Fix one dimension plot/image label problem.
    Update logic for display CAR projection.
    Rework background backup directory cleanup logics.
    Fix ploygon creation problem, allow right mouse button to drag the
      existing vertex, left mouse button for creation and move the region.
    Add color changes capability in Edit Region Panel.
    Change vizier/skyview/catalog file naming scheme.

******************************************************************************
                                   FIMAGE
******************************************************************************

- chimgtyp: modified to handle NANs in input image.

- fimgmerge: modified to handle null or NAN values in input image.

******************************************************************************
                                   FUTILS
******************************************************************************

- fpartab: fixed bug in writing values to logical ('L') columns

- fdiff: fixed problem with comparison of string and logical table columns.
  Fixed erroneous reporting of differences when two files have different
  numbers of extensions.  Fixed bug which caused the 'DATE' keyword to not
  be excluded from the file comparisons.

- fverify: fixed to eliminate bogus warnings when verifying floating point
  FITS images that have been tile-compressed.  Modified to accept '-' as the
  input file name, meaning read file from stdin.  Added compile option to
  force all error messages to go to stdout instead of stderr.  Modified to
  support testing for bytes at the end of LARGE files (>2.1 GB).  Modified to
  print out error if file doesn't exist or is not a FITS file, in the case
  where prstat=no.

******************************************************************************
                                   HEASARC
******************************************************************************

- Extractor:

  Filtering

  Modified the code to interpret filters attached to the event filename to allow
  multiple ranges (eg filename.evt[GRADE=0:0 2:2]).

  Added support for region files with sizes given in arcminutes ('), 
  degrees (d), or radians (r).

  Fixed bug in handling of DSVAL keywords that must be removed when filtering
  on the column set by the ccol parameter. 

  Cleverer handling of region extensions. Now looks in a region extension
  to see what coordinates are used. If the current filtering is with a
  different set of coordinates then a new region extension will be created
  (and the old ones copied). This fixes a bug which showed up with ASCA GIS
  data - the standard event files are filtered in detector space and then
  the user will often follow up by filtering in image space.

  Trapped NULL values of the TIME column and dealt with them correctly (ie
  rejected the event).

  Output Files

  Added a boolean parameter lctzero. The default is yes which provides
  the previous behaviour. Setting lctzero=no will cause the times in the 
  lightcurve to be written in spacecraft time units and the keyword
  TIMEZERO to be set to zero.

  Added WCSNAMEP keywords for image and WMAP.

  Removed setting of TIMEREF to 'local' in W_FBLC. This keyword is set to the
  correct (propagated) value in WSTDKY.

  Modified the way the EXPOSURE keyword is calculated when the ccol parameter
  is set if multiple GTI extensions contribute to the output. The old way was
  to calculate the exposure for each GTI extension and average. This does not
  work for the UVOT event files where the exposures must be summed. The new
  version combines individual ranges from all the selected GTI extensions and
  uses the merged list to calculate an exposure. This does the correct thing
  for the UVOT case. For missions with multiple CCDs if products are required
  for spatial regions covering more than one CCD the exposure may be slightly
  different. However, if the old and new ways give very different exposures
  then it is not correct to combine data across the CCDs anyway.

  Propagated the value of the RADECSYS keyword from the event file to the
  output files. Note that the spectrum file was always being written with
  RADECSYS='FK4' which was usually wrong but didn't matter.


- fovdsp (new in this release): display ISDC format reference catalog and
  INTEGRAL or Swift field of viewson either Galactic or Equatorial coordinates,
  aitoff or tangential projections. Also, FITS images may be displayed.

- gisxspec (new in this release): Do spectral analysis for gis data.

- makeregion (new in this release): Make a region file for an image with POWplot

- make2region (new in this release): Create two region files for an image
  through POWplot.

- pspcxspec (new in this release): Perform spectral analysis for PSPC data

- selectxrange (new in this release): Make an x range file for an event fits
  file with POWplot.

- spibkg_init: Relative source weighting parameters are now entered as
  floating point rather than as strings.  Some refinement to the background
  initialization algorithm was also made. In particular, some time-dependent
  background structures appearing on given detectors are detected and more
  accurately modeled.

******************************************************************************
                                   HEATOOLS
******************************************************************************

- The new package 'HEAtools' has been distributed in recent Swift software
  releases and as a beta version, but this marks its first inclusion in a
  major release of HEASOFT.

  The HEAtools package represents a "next generation" of FTOOLS, which currently
  reproduce the functionality of existing FTOOLS, and as such, the HEAtools
  package is not necessary if you already have FTOOLS.  However, having been
  written entirely in ANSI C for maximum portability, the streamlined HEAtools
  should also be more stable and provide faster results than their predecessors.

  For more information, see:

    http://heasarc.gsfc.nasa.gov/lheasoft/ftools/heatools/heatools.html

******************************************************************************
                                   INTEGRAL
******************************************************************************

New package containing:

- spibkgsubtrct: produces a background subtracted spectrum resulting
  from an initial XSPEC analysis of INTEGRAL/SPI data.

  ** NOTE: in order to compile spibkgsubtrct you must download the Xspec 12
           source code in addition to the INTEGRAL package.

- varmosaic: carry out mosaic of flux images with weighting of the
  accompanied variance images.  Primarily intended for INTEGRAL ISGRI
  and JEMX instruments, but may be used for other coded mask instruments.

******************************************************************************
                                    ROSAT
******************************************************************************

- pcexpmap: Fixed an array out of bounds condition which caused a segmentation
  fault under rare conditions.

******************************************************************************
                                    SWIFT
******************************************************************************

Swift software v 2.0:

 ------------------
 BAT specifc tools
 ------------------
 batbinevt -
   Correct name of POISSERR keyword in output spectra

    INTERFACE CHANGE: Allow the user to delete bins with zero counts.
    Before, the task always deleted these bins, in order to prevent
    battblocks from crashing. ('delzeroes' parameter)

    Some warning messages for unlikely user configs.

 batcelldetect -
    INTERFACE CHANGE: New parameter 'hduclasses' which allows
    filtering of input images by HDUCLASn keywords.

    Now able to read an image without an auxiliary tangent plane
    coordinate system.

    Input catalog can have galactic coordinates (if the image is
    also in galactic coordinates)
 
    Output images now have proper HDUCLASn keywords

 batclean -
    Can now enter custom background models via the 'bkgmodel' parameter.

    INTERFACE CHANGE: Can now rebalance the focal plane image before
    or after cleaning, with various rebalancing options ('balance' and
    'balancefirst' parameters).

    INTERFACE CHANGE: Can now ignore certain sources in the catalog
    via the 'ignore' parameter.

    Fixes and simplifications to the way the background model is applied.
 
    New CLEANLEV output keyword keeps track of the generation level of
    cleaning.

 batdrmgen -
    INTERFACE CHANGE: A new hidden parameter called "row" has been
    added that lets the user specify which row of a pha file a
    response matrix should be made for, in cases where there are
    multiple spectra in a pha file.

    The task will search for columns first, then keywords, with the
    following names: BAT_XOBJ, BAT_YOBJ, BAT_ZOBJ, PCODEFR, NGOODPIX,
    and MSKWTSQF. Before, it only searched for keywords.

    INTERFACE CHANGE: A correction function has been added to improve
    the response to the Crab spectrum. This is mediated by the
    "fudge" parameter (which until now has been ignored by batdrmgen).

    (Other changes are pretty transparent to the user, such as using
    finer binning to improve the response fidelity and changing the
    model of the passive material transmission to include the Ag and
    Au edges.)

 batcovert -
   INTERFACE CHANGE: Added cubic-residual ADU to energy conversion,
    with residuals computed by ADU or by corresponding DAC
    value. 'calmode' parameter options now include INDEF, LINEAR,
    QUADRATIC, CUBIC, DIRECTCUBIC, or FIXEDDAC. Except when using the
    LINEAR correction, residfile must be compatible with the calmode
    selection. FIXEDDAC is the recommended value for normal energy
    computation. Most users can enter INDEF, and use CALDB for the
    residfile. LINEAR is provided in order to reproduce the flight
    software energy calculation. The others are provided primarily for
    testing.
 
    A scaling bug that had a minor effect on results (because gains 
    and offsets have changed very little) was fixed. 
 
    A bug that prevented use of the rarely-used zeroit option was
    fixed.

 batrebin -
    Added cubic-residual ADU to energy correction, with residuals
    computed by ADU or by corresponding DAC value. The method is
    selected by selecting a residfile. Using the current CALDB file
    will apply the cubic-residual correction at fixed DAC values.

    Fixed an error that shifted all energy bins by approximately 0.1
    keV (0.5 ADU).

 batfftimage -
    INTERFACE CHANGE: Now can compute the theoretical variance and
    significance maps via the 'bkgvarmap' and 'signifmap' parameters.

    KNOWN BUG: output significance images contain bogus values in the
    "dead" areas. WORKAROUND for source detection: use 'pcodefile'
    option of batcelldetect.

    Output images now have proper HDUCLASn keywords
 
 batmasktaglc -
    INTERFACE CHANGE: A new hidden parameter has been added to
    "batmasktaglc." The parameter is "scale" which is a floating
    point value which is used to scale the input raw mask-tagged
    counts. This is to allow a correction to the input light curve
    due to any errors in how they are generated. The user must
    determine the correction factor.

 batmaskwtevt -
   Add additional error checking
 
 batupdatephakw -
   NEW with this release

 -------------------
 UVOT specific tools
 -------------------
 uvotdetect -
    TDISPn keywords modified to report 2 more digits RA,DEC

    Binning dependent SExtractor configuration files

    Option to display detected sources

 uvotimsum -
    Write RADECSYS and EQUINOX keywords to output image

    Fixed bug: was not propagating sufficient keywords to output
 
 uvotmag -
   TDISPn keywords modified to report 2 more digits RA,DEC
 
 uvotpict - 
   Modified graphical output

 uvotproduct-
   NEW with this release

 uvotscreen -
   Pass copyall=yes to extractor to retain WINDOW extension

   Write parameter history to output

 uvotsource - 
    NEW with this release

 uvotstarid-
    TDISPn keywords modified to report 2 more digits RA,DEC
 
    Write NULL MAG_DELTA when catalog magnitude is not available

    Separate parameters to limit position and rotation corrections 
 
    Index objects loaded from catalog

 uvottfc-
    Corrected postage stamp size indicator in output source table
 
 -------------------
 XRT specific ftools
 -------------------
 xrtcalcpi -
    Calculate PI dependent on the CCD temperature using a new CALDB 'gain'
    file format and the XRT science packet header File.

    New input parameter 'hdfile'.

 xrtevtrec - 
    Update to read the event and split thresholds from the science packet
    header which records the on-board settings as a function of temperature.
    This change is in support of the upcoming FSW. If the event and split
    thresholds parameters are negative, the event and split values are read
    from the header file otherwise xrtevtrec uses any input positive value
    as before.

    New input parameter 'hdfile'.

 xrtfilter -
    Add the columns 'TEN_ARCMIN, SETTLED, ACS_SAA, SAFEHOLD' to the output
    makefilter file.  These are derived from 'FLAGS' of the attitude file.
    This is to align the makefiler file generated by the xrtpipeline to the
    one stored in the archive.
 
 xrtflagpix - 
    Add the capability to flag event as 'bad' in the Photon Counting mode
    when the central pixel is below the event threshold. The default
    threshold is set to 80.

    New input parameter 'phas1thr', if set to 0 the check on the central
    pixel is not applied.

 xrthkproc-
    Add the capability to use for the source position the detector
    coordinates. By default the task is run with the detector coordinates
    of the center of the detector (300,300).
 
    New input parameters 'srcdetx','srcdety'.

 xrthotpix -
    Add the new input parameter 'usegoodevt' to include/exclude from the
    search of hot and flickering pixels events set 'bad' by 'xrtflagpix'
    when the central pixel is below threshold.

 xrtpcgrade -
    Update to read the split threshold from the science packet header
    which records the on-board settings as a function of temperature.
    This change is in support of the upcoming FSW. If the split threshold
    parameter is negative, the split threshold value is read from the
    header file otherwise xrtpcgrade uses any input positive value as before.
 
   New input parameter 'hdfile'.

 xrtpipeline -
    New input parameter 'obsmode' to process ALL obs_modes or only the one
    selected. This allows to specify different user input screening selection
    for the different observing modes. 'obsmode' values are (ALL,POINTING,
    SLEW, SETTLING).

    The "ROTIME" column is now maintained in the Lev 1a file for the Timing
    modes to allow the calculation of the LLD and SPLIT temperature dependent
    parameters. The "ROTIME" is now deleted in the Lev 2 files.

    The parameter 'pilow' is set to 30 in the xrtproducts call.
   
    New input parameter 'usesrcdethkproc' to allow users to process the header
    science packet file (xrthkproc task) using detector or sky coordinates.
   
    New input parameter 'evtfromarc' to allow to get input event files from
    the archive or from the users output directory. This is to allow users
    to apply only data screening or products generation (stage 2 and/or
    stage 3 of the pipeline) to the Calibrated Level 1 event files in the
    archive.
   
     New input parameter 'createmkffile' to enable/disable the generation of
     the Filter file ('.mfk' file).  If the 'createmkffile' is set to 'no'
     the pipeline uses the Filter file taken from the archive.

 xrtproducts-
    Changed image plot energy band to 0.5-10keV (PI=50-1000)

    Replaced xspec call with xspec11

    Added check on 'ra' and 'dec' input parameters if the input 'regionfile'
    is not set to 'default'.

 xrtscreen -
    Read a new format for the CALDB file where now the good values of the
    hk parameter are listed together with syntax to use in the screening
    expression.
 
    Add a new  input parameter 'obsmodescreen'. If the parameter is set to
    yes (default) add to the GTI expression, the parameters that specify the
    observing mode (i.e.: "SETTLED==0&&TEN_ARCMIN==0" for data taken
    during slew, "SETTLED==0&&TEN_ARCMIN==1" for data taken in
    10 arcmin from the source, "SETTLED==1" for data in pointing). These
    parameters are not added to the expression when generating GTI for
    the Image mode.
 
 xrttam -
    The extension ACS_DATA from the input attitude files is appended to the
    TAM corrected attitude output of xrttam.  This extension is used by
    xrtfilter.

 xrttimetag -
    Add the capability to use for the source position the detector coordinates
    to time tag the events. The default processing uses the source sky
    coordinates. 
 
   New input parameters 'srcdetx','srcdety', 'usesrcdet'.

 ------------------
 Swift General tool
 ------------------
 swiftxform -
    MJDREFI was mis-spelled causing a warning about missing MJDREF keywords

    Support attfile=CONST:*


 -------------------------------------------
 BAT known Issues on the analysis software
 -------------------------------------------

- batfftimage and batmaskwtimg: incorrect derived attitude

   Task: Imaging tasks (batfftimage and batmaskwtimg) 
   Version: All versions 
   What Builds: All builds 
   Problem: Attitude may be incorrect for observation with many snapshots 
   Status: Open 

   Two imaging tasks assume that the spacecraft attitude is fixed during
   an observation: batfftimage (to make sky images) and batmaskwtimg (to
   make mask weighting maps for flux extraction).  Both tasks take the
   attitude at the '''MIDPOINT''' of the observation start/stop times.

   If there are gaps in the observation, i.e. multiple snapshots, then it
   is possible, even likely, that the midpoint time will fall within a
   gap.  When this happens, the attitude may be erroneously interpolated.

   There are two workaround solutions,

    * analyze one snapshot at a time.  Or;

    * use the 'aspect' tool to generate a revised attitude file.  As of
      Swift 1.2, 'aspect' can create a new attitude file based on the
      median pointing direction during the observation good times
      only. You will also need to supply a good time interval extension
      to 'aspect', which should be available in the detector or sky
      images.

   In the future the BAT team will investigate how feasible it is to
    incorporate 'aspect'-like functionality into the BAT imaging tools.



 - Analysis: Earth and Moon Occultation

   Task: Flux extraction tasks (batmaskwtevt,batmaskwtimg,batfftimage) 
   Version: All versions 
   What Builds: All builds 
   Problem: Earth and moon may block parts of the BAT field of view 
   Status: Open 

   The BAT field of view is large, approximately 120 x 60 degrees fully
   coded.  The current spacecraft constraint excludes the sun with a 45
   deg constraint cone, and the earth limb and moon with 30 degree
   constraints each.

   Even so, the Earth and Moon may enter the BAT field of view.  This
   will most commonly occur at edges of the "long" BAT axis (i.e. large
   IMX in the image plane).  The effect will be to occult the flux of
   sources in that part of the sky.  Since the Moon and (primarily) the
   Earth move as a function of time, the blockage may have the effect of
   '''reducing,''' but not totally eliminating the source on-time.

   Example: in a 2000 second image, Sco X-1 might be blocked during the
   final 50 seconds.

   Users need to take special care regarding occultations, especially by
   the Earth.  Tools are in process to make this job easier.  These will
   involve image corrections for full images, and good time interval
   filters, for time selection of non-occulted data.



- Mask tagged light curve systematic flux errors

  Task: batmasktaglc 
  Version: All (bug is in BAT flight software) 
  What Builds: All 
  Problem: There are systematic flux errors in mask tagged light curves 
  Status: Open 

  Discussion: BAT mask tagged light curves are generated on board by the
  BAT flight software.  The on-board process involves generating a mask
  weight map via ray tracing (similar to the ground task batmaskwtimg).
  On the ground, the raw light curves require further processing before
  they are scientifically meaningful.

  The "mask tagging" process requires that the source position be known
  in advance, in instrument coordinates.  This transformation requires
  knowledge of the spacecraft attitude, the instrument-to-spacecraft
  alignment, and the source celestial coordinates.  A bug has been found
  in the BAT flight software which makes the incorrect transformation.
  The ray-traced position used for generating the mask weight map is
  several arcminutes off of the true position.

  The point spread function is thus sampled significantly off-peak; this
  irretrievably reduces the signal to noise of the mask weighted light
  curve fluxes.  Also, the flux itself is underreported compared to its
  true value.

  Ground software fixes are being investigated that would correct the
  flux to a truer value (although with larger error bars).  A long term
  solution to fix the flight software is also being investigated, but
  may not be feasible.
 

******************************************************************************
                                    TIME
******************************************************************************

- maketime: Implemented new prefr/postfr scheme: new filter files created by
  makefilter (or newmakefilter) need prefr=0.0 and postfr=1.0, and should have
  the PREFR & POSTFR keywords set.  If the PREFR/POSTFR keywords are not found,
  the standard value of 0.5 is used for each (for e.g. older missions).  The
  user may override either of these choices by entering prefr/postfr values
  on the command line.

******************************************************************************
                                   XIMAGE
******************************************************************************

XIMAGE VERSION 4.3

For full documentation on XIMAGE, refer to:

  http://heasarc.gsfc.nasa.gov/docs/xanadu/ximage/

   * Better handling of images with no sky coordinates
   * Improved GTI merging capability
   * New command uplimit, which calculates an upper limit
   * New options
      + ccorr
         -> symlwidth - Specify symbol's line width for guide source markers
      + read_image
         -> gtiext - Specify extension to read GTI from event file
      + wcs
         -> frameid - Switch between image coordinate frames

For complete change history of XIMAGE see:

  http://heasarc.gsfc.nasa.gov/docs/xanadu/ximage/ximrelease.html

******************************************************************************
                                   XRONOS
******************************************************************************

- earth2sun: Fixed uninitialized variable that resulted in incorrect results
  on Darwin (Mac OS X).

- rbf2fits: compiles with g77 & f90 now.

******************************************************************************
                                   XSELECT
******************************************************************************

Added offset switch option to extract curve. Offset=no now creates a lightcurve
whose times are given in spacecraft units. This differs from the default which
writes out times relative to the start bin.

Added filter column option which allows filtering on event attributes which
are included in columns in the event file. At the moment allows filters of
the form "NAME=val:val val:val..." because these are handled by the extractor
key filtering. Note that this filter can be used on Astro-E2 XRS data to
get events from subsets of pixels.

Replaced getting MJDREF from the MDB by reading MJDREF[F/I] and TIMESYS from
the event file. These are used when time filtering based on UTC or MJD ranges.
Now includes conversion from UTC to TT if TIMESYS='TT'.

Allowed "select expref" as an alias for "select chip".

Updates to MDB for Swift, Integral, and Astro-E2 XRS.

******************************************************************************
                                    XSPEC
******************************************************************************

*** Note that this release includes both xspec v12 (invoked with "xspec")
    and xspec v11.3.2 (invoked with "xspec11").  xspec v12 is not currently
    supported on the OSF/alpha or Cygwin platforms, so for those platforms
    only, invoking "xspec" will call xspec11.

**************
XSPEC v12.2.0:
**************

Xspec v12 is a major revision from v11. The core of the program is now
primarily written in double precision ANSI C++ instead of single
precision Fortran77 and its internal design, layout, and data
structures, have been reorganized into an object-oriented framework
using design patterns and generic programming techniques made
available with C++. Most of the xspec models library however has been
retained in its original Fortran77 code.
  
Some of the enhancements, in particular the use of multiple models
described below, require corresponding changes in command-line syntax.
However, with few exceptions the program remains fully
backward-compatible with xspec v11, and previously existing user
scripts are expected to run in v12 with little or no
modification. However, compatibility issues are possible with this
initial release, and we welcome any user reports to help us correct
these and other bugs.

Key modifications:

The implementation of user models has been rewritten to allow users to
write models not only calculated in single precision Fortran77, but
double precision Fortran77, C, and C++.  Further, XSPEC can now be used
as a development environment for user models by allowing recompilation
and plotting from the XSPEC command prompt.

A spectrum can be fit with the sum of separate models each having its
own response matrix. This feature is useful for analyzing data from
coded aperture masks.

A new internal dynamic expression implementation allows more complex
(multiply-nested) models, and also allows parameter links to be
polynomial functions of one or more parameters.
  
A ctrl-c breaking mechanism has been implemented to allow early exit
from the more time-intensive tasks such as fitting and error
calculations.  An "undo" command has also been added to return the 
program to the state before its most recent command was issued.

The CERN Minuit/migrad algorithm has been better integrated into the
code and its documentation is now directly accessible to the user
during XSPEC sessions.Type II (multi-spectrum) OGIP files are now
fully supported. Multiple ranges can be selected in the data command,
and support is present for Type II background and arf files.
Observation simulations (the fakeit command) now operate on Type II
inputs.

The online documentation scheme is now implemented using pdf files,
replacing the older VMS-style help system. The help scheme can be
configured to use external applications such as Adobe Acrobat or the
xpdf readers. Users can document their own local models and  
tcl-scripted procedures in pdf files and add them to the help system.

For a more complete list of modifications, enhancements, and
descriptions, please refer to the xspec12 manual. 


**************
XSPEC v11.3.2:
**************

This is mainly a bug-fix update of 11.3.1 with the addition of three
new models. We are continuing to distribute v11 to aide comparisons
with the new v12 however all future development work will occur on
v12. We will however attempt to supply fixes for any serious bugs that 
show up in v11.

Changes to commands

"tclout simpars" returns a list of simulated parameter values for the
model in use (based on the covariance matrix at the end of the last fit).

Changes to models

All the models that use the mekal code (eg mkcflow) can be made to use 
the APEC code by setting the switch parameter to 2.

The APEC versions of NEI models are modified by the APECTHERMAL and 
APECVELOCITY variables in the same way as the standard APEC models.

Models added

  compbb    -  Comptonization model of Poutanen and Svenson.

  ezdiskbb  -  Replacement for diskbb with a zero-torque inner boundary 
               condition (Zimmerman, Narayan & McClintock).

  kerrbb    -  Multi-temperature blackbody for thin accretion disk around a
               Kerr black hole (Li et al. 2005).

  zredden   -  Redshifted version of redden.

******************************************************************************
                                    XSTAR
******************************************************************************

Xstar Version 2.1kn3 (April 2005):

- Two bugs were found in version 2.1k in the implementation of the
  Voigt function when calculating line absorption and the calculation of
  line broadening:

  The Voigt function bug affected primarily lines with small damping
  parameters, and resulted in non-fatal numerical errors in the xstar output
  absorption spectrum (INFs). When xstar was called as part of xstar2xspec
  this resulted in fatal errors because the cfitsio routines which read the
  xstar output could not interpret the INFs. The line broadening bug resulted
  in too large absorption line depths when turbulent broadening was important.
  Neither of these bugs affected the temperature, ionization balance or
  emission spectrum. The bugs have been repaired in version 2.1kn3.
                                                                                
  Version 2.1kn3 also has an added feature, which is the addition of
  ion-by-ion  heating and cooling rates as extensions to the output file
  xout_abund1.fits.
                                                                                
  Also added is the capability to set the value of niter to a negative
  number, which allows the solution of charge conservation without solving
  thermal equilibrium.  As before, if niter=0 then neither charge conservation
  nor thermal equlibrium is calculated.

******************************************************************************
                                    MISC
******************************************************************************

- A final note about libraries - new in this release (relative to
  HEASOFT 5.3.1) are:

  atFunctions 2.3
  readline 4.3
  SLALIB version 2.4-13 (contained in xanlib)