National Aeronautics and Space Administration

********************************************************************************
                         RELEASE NOTES FOR HEASOFT 6.5.1
                               September 18, 2008
********************************************************************************
The HEASoft 6.5.1 release solely consists of an update to the Suzaku software
(version 10).  Release notes for HEASOFT 6.5 follow the Suzaku notes below.

********************************************************************************
                                    SUZAKU
********************************************************************************

Suzaku software Version 10 - Changes since HEASoft v6.5

---------------------
 HXD specific tools:
---------------------

 hxdgtigen
     fixed bug when non-reset counters.

 hxdmkwamspec
     support version 2 format of HXD WAM event files.

 hxdbsttime
     a) Implement new method to assign time using ae_hxd_bstidt.fits
        which provide the first time estimate for the WAM events.
     b) Fill TIMECORR keyword (T: FRZN_TMn is valid, F: invalid), and
        if time assignment is invalid, FRZN_TMn (n=0,1,2,3) are set to
        the value 0.00.
     c) Update ae_hxd_bstidt.fits, when time assignment is valid.
        A bug remains: segmentation falt on gcc-4.1.2 environment

     Known bugs :1) hxdwambstidt where values of specific columns show
     platform dependence differences. This affect only the HXD_WAM data 
     not the HXD_WELL data. 2) hxdbsttime on the i686-apple-darwin9.3.0 
     platform gives values different for the column BST_OVER_FLOW valid
     only for the HXD_WAM data.

---------------------
 XIS specific tools:
---------------------
 
 xiscoord
     now set RAWX/Y=-999 for illegal PPUX/Y values for DarkInit/Update files.
 
 xissim
       fix to wraparound GTI when exposure is longer than that specifiled by
       the 'gtifile' parameter.

 mkphlist
       fix to check whether required WCS keyowrds (EQUINOX, CDELT1, CDELT2,
       CRPIX1, CRPIX2, CRVAL1, CRVAL2) exists in the image specified by the
       'fits_image_file' parameter.
 



********************************************************************************
                          RELEASE NOTES FOR HEASOFT 6.5
                                 June 26, 2008
********************************************************************************

The HEASoft 6.5 release is primarily driven by the release of the
Swift (version 2.9) software, but also includes to several libraries
and packages which have been revised to mirror the current development
versions in use at the HEASARC.

As before, configuration-related changes thoughout the package means that
users who already have HEASoft v6.x.x installed will need to completely
replace their software installation; there is no clean way to simply
overlay this new release or to update an existing installation.

********************************************************************************
                                 CONFIGURATION
********************************************************************************

- Updates to allow compilation with g++ 4.3.x.

- Updated support for GNU compilers on SunOS.

- Builds on SunOS using non-GNU compilers now have Fortran optimization
  turned off by default, due to runtime problems when using the SunStudio
  Fortran compiler.  Users who wish to increase optimization for e.g. the
  Workshop Fortran compiler may configure HEASOFT using the '--with-fopt='
  option.

- Add '--build=' configure option for users who want to name the
  architecture-specfic installation directory something other than the
  default value (which is of the form cpu-vendor-os[-glibcversion]).

- WCSLIB (ATNF) updated to version 4.3.

- PGPLOT perl module updated to v2.20.

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

Version 3.09 - 12 June 2008

   - fixed bug in the calculator function, parse_data, that evaluates 
     expressions then selecting rows or modifying values in table columns.
     This bug only appeared in unusual circumstances
     where the calculated value has a null value (= TNULLn).  The bug
     could cause elements to not be flagged as having a null value, or
     in rare cases could cause valid elements to be flagged as null. This
     only appears to have affected 64-bit platforms (where size(long) = 8).

   - fixed typo in imcomp_decompress_tile: call to fffi2r8 should have  
     been to fffi4r8.

   - in the imcopy_copy_comp2img routine, moved the call to 
     fits_translate_keywords outside of the 'if' statement.  This could
     affect reading compressed images that did not have a EXTNAME keyword
     in the header.
     
   - fixed imcomp_compress_tile in imcompress.c to properly support
     writing unsigned integers, in place, to tile compressed images.

   - modified fits_read_compressed_img so that if the calling routine
     specifies nullval = 0, then it will not check for null-valued
     pixels in the compressed FITS image.  This mimics the same 
     behavior when reading normal uncompressed FITS images.

Version 3.08 - 15 April 2008 

   - fixed backwards compatibility issue when uncompressing a Rice
     compressed image that was created with previous versions of
     CFITSIO (this late fix was added on May 18).

   - small change to cfortran.h to add "extern" to the common block
     definition.  This was done for compatibility with the version
     of cfortran.h that is distributed by the debian project.

   - relaxed the requirement that a string valued keyword must have a
     closing quote character.  If the quote is missing, CFITSIO will silently
     append a quote at the end of the keyword record.  This change was made
     because otherwise it is very difficult to correct the keyword
     because CFITSIO would exit with an error before making the fix.

   - added a new BYTEPIX compression parameter when tile-compressing
     images with the Rice algorithm.

   - cached the NAXIS and NAXISn keyword values in the fitsio structure
     for efficiency, to eliminate duplicates reads of these keywords.

   - added variants of the Rice compression and uncompression routines to
     support short int images (inaddition to the routines that support int).

   - moved the definition of LONGLONG_MIN and LONGLONG_MAX from fitsio2.h
     to fitsio.h, to make it accessable to application programs.

   - make efficiency improvements to fitscore.c, to avoid needless searches
     through the entire header when reading the required keywords that must
     be near the beginning of the header.

   - made several improvements to getcol.c to optimize reading of compressed
     and uncompressed images.

   - changed the compression level in the gzip code from 6 to 1.  In most
     cases this will provide nearly the same amount of compression, but is
     significantly faster in some cases.

   - added new "helper routines' to imcompress.c to allow applications to
     specified the "quantize level" and Hcompress scaling and smoothing
     parameters

   - modified the extended filename syntax to support the "quantize level"
     and Hcompress scaling and smoothing parameters.  The parser in
     cfileio.c was extensively modified.

   - extensive changes to quantize.c:
       - replace the "nbits" parameter with "quantize level"
       - the quantize level is now relative to the RMS noise in the image
       - the HCOMPRESS scale factor is now relative to the RMS noise
       - added routines to calculate RMS noise in image
      (these changes require a change to the main file structure in fitsio.h)

   - initialize errno = 0 before the call to strtol in ffext, in case errno
     has previously been set by an unrelated error condition.

   - added the corresponding long name for the ffgkyjj routine to longnam.h.     
   - changed imcomp_copy_comp2img (in imcompress.c) to not require the
     presence of the EXTNAME keyword in the input compressed image header.

   - modified imcompress.c to only write the UNCOMPRESSED_DATA column
     in tile-compressed images if it is actually needed.  This eliminates
     the need to subsequently delete the column if it is not used
     (which is almost always the case).

   - found that it is necessay to seek to the EOF of a file after
     turncating the size of the file, to reestablish a definite
     current location in the file.  The required small changes to 3
     routines: file_truncate (to seek to EOF) and fftrun (to set io_pos)
     and the truncation routine in drvrmem.c.

   - improved the efficiency when compressing integer images with
     gzip.  Previously, the image was always represented using integer*4
     pixels, which were then compressed.  Now, if the range of pixel
     values can be represented with integer*2 pixels or integer*1 pixels,
     then that is used.  This change is backward compatible with any
     compressed images that used the previous method.

   - changed the default tiling pattern when using Hcompress from 
     large squares (200 to 600 pixels wide) to 16 rows of the image.
     This generally requires less memory, compresses faster, and is more
     consistent with the default row by row tiling when using the other
     compression methods.

   - modified imcomp_init_table in imcompress.c to enforce a restriction
     when using the Hcompress algorithm that the 1st 2 dimensions of sll
     image tiles must be at least 4 pixels long.  Hcompress becomes very
     inefficient for smaller dimensions, and does not work at all with
     1D images.

   - fixed bug in the Hcompress compression algorithm that could affect
     compression of I*4 images, using non-square compression tiles
     (in the encode64 routine).


********************************************************************************
                                   ATTITUDE
********************************************************************************

- aspcorr: Added rotate parameter which updates CROTA2.

- makefilter: This task will be retired in the next HEASOFT release
  in favor of the existing 'newmakefilter' task, which provides the
  same functionality and is more stable.

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

- genrsp: Added LINEAR option to resol_reln.

- rbnrmf: Fixed bug that was causing the NUMGRP keyword in the output file
  to refer to the number of groups in the input response, not the output
  response.

- (callib) Modified to correctly handle old RMFs which only contain the
  RMFVERSN keyword and are missing HDUVERS. RMFVERSN='1992a' is equivalent
  to HDUVERS='1.0.0'.

********************************************************************************
                              EXTRACTOR & XSELECT 
********************************************************************************

- extractor:

  Changed the output MJDREF to double precision.

  Fixed timing filter file input to allow a filename[extension] option.

  Fix to allow filenames to include ']' and '+'.

  Made the interpretation of the grade string a bit more lenient - it
  now allows spaces as well as commas between individual elements.

  Added code to write the REF* keywords to the WMAP. These keywords,
  invented I think by the XMM folks, store the original WCS keywords
  for sky coordinates.  In addition to the EXT* keywords calculated by
  extractor they enable the calculation of RA and DEC for detector pixels
  in the WMAP (usual caveats apply).

  Modified FITS region writing for consistency with the CXC-defined
  standard.  Mainly switched the EXTNAME to REGION and used EXTVER to
  distinguish multiple REGION extensions.

- xselect:

  Added code to enable the option of passing the region filter file to the
  response generation script. Response generation now works for Suzaku XIS.

  Now allows the xselect.mdb to contain a wild card at the end of names
  of submissions, instruments, and datamodes. The wild card is a * ie XIS*
  will match to any of XIS0, XIS1, XIS2, XIS3.

  Added new spectrum rebinning option using a script whose name is retrieved
  from the MDB using the keyword rebin script.

  Missed a case in XSL_5X5TO3X3 - if the files had already been run through
  extractor in the current session then they were not handled correctly.

- xselect.mdb: 

  Added entry to specify whether the response generation script needs to be
  passed the region file. This is required for Suzaku XIS.

  Used new wild card capability to simplify the SUZAKU lines. Also removed
  some tabs that had got in - only spaces should be used as separators.

  Much simplification using the wild card syntax.  Replaced Suzaku XIS grpstr
  with the new rebin script.

  Add extended/point source response matrix option for Chandra

  Modification for ROSAT to deal with a problem which has recently turned up.
  Now that I allow wildcards in instrument names using only the first four
  characters of the INSTRUME keyword is no longer necessary.

  Fixed missing colon in one SUZAKU entry.

- xsl_chandra_acis_makeresp: v2.00 which switches between mkarf and mkwarf
  for point or extended sources, respectively.  Uses criteria recommended by
  CXC to decide whether to use the mkacisrmf or mkrmf tools.

- xsl_suzaku_xis_makeresp / xsl_suzaku_xis_rebinspec: 
  Moved rebinning of the Suzaku XIS spectrum from the response creation
  script to its own script.  makeresp: Simplified version which has the
  speed option hardcoded.

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

- fimgextract: Fixed problem with how null values are handled in integer
  images.  With the previous code, if a pixel value happened to equal the
  value of the nulval parameter, then it would incorrectly be assumed to
  be a null pixel even if there was no BLANK keyword in the image header.

********************************************************************************
                                      FV
********************************************************************************

FV version 5.2:

- Bug fix: zoom/unzoom of plot/image cause labels and units of the graph
  to disappear. 

- Bug fix: file selection panel was not displayed when exiting fv/pow
  application when FITS file is modified during the session. 

- Bug fix: when editing plot/image information, new title string was not
  updated in the graph. 

- Bug fix: fv/pow crashes during rendering of graph/image if FITS file
  includes inconsistent coordinates or unsupported coordinate system. 

- Bug fix: unnecessary X and Y labels displayed in POW window (no
  image/plot created yet) when xspec is invoked. 

- Enhancement: allow the selection of columns panel to remain visible
  during plotting. 

- Enhancement: allow renaming capability of created region files when
  running special analysis tools in Hera. 

********************************************************************************
                                    HEAGEN
********************************************************************************

- sprbnarf: New task which compresses a FITS ARF array in energy-space,
  intended to be used in parallel with the energy-binning option of rbnrmf.
  The default operating mode is to bin up by the factor given by the binfact
  parameter. If the hidden parameter binfile is set then an ascii file will
  be read to get the binning factor(s).

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

- ftdelrow: Add option to use an input list of rows ('@' syntax).

- ftpixcalc: Call HDfile_system_check with mode "e" instead of 0.  Was
  previously exiting with an error (for either value of clobber) when the
  output file existed.

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

Changes in version 4.4.1:

- Fixed two bugs in the skymap command related to ASDC catalog queries. Queries
  on ASDC catalogs failed, and Declination was not treated properly in returned
  results from ASDC queries

- Added new option out_file to skymap to dump RA/Dec of field sources to file

- Fixed wrong behavior in psf/pileup where a single pixel would be excluded
  from the psf if the pileup flag was specified, even if there was no pileup.

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

- lcmath: Fixed lcmath's handling of dead time and vignetting corrections
  for both source and "background" light curves. Added new function
  lcmathgabvd to determine the already applied corrections.  Added warning
  for asynchronous input light curves, and proper warnings when one curve
  is corrected while the other is not and docor=no.

********************************************************************************
                                   XSPEC 11
********************************************************************************

Increased the initial number of groups in the output response and improved
the diagnostics for overrunning the array size allocated when creating the
output response. This fixes a seg fault that appears when the input file has
been binned up in energy-space.

Fix for a possible seg fault on x86_64 machines.

********************************************************************************
                                   XSPEC 12
********************************************************************************

HEASoft 6.5 includes Xspec v12.4.0ad.  See

  http://heasarc.gsfc.nasa.gov/docs/xanadu/xspec/issues/issues.html

for information about v12.4.0ad and to get patches beyond level 'ad'.

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

v2.1kn8:

- Changes to atomic data file to update fe uta data to gu et al. 2006 values.

- Repair to bug on xstinitable.c which allows large grids of models to be run.



================================================================================
                           MISSION-SPECIFIC SOFTWARE:
================================================================================

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

- addascaspec: Check for the presence of ARFs in the list file and if they
  are not given then don't run marfrmf.

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

- gevt2rdf / img2us: Change variable-width Fortran READ formats to fixed-
  width formats to get around "Nonnegative width required in format string"
  compilation errors coming from gfortran v4.3.0.  Here we're assuming (based
  on examination of German data files) that the values of POINT_LAT and
  POINT_LONG will always start with ##H##M (and that POINT_LAT ends with ##S),
  and that the "ROR_NUM" to be extracted from OBS_ID will be the (last) 6
  digits of the OBS_ID string (6 digits that follow 2 letters).

********************************************************************************
                                    SUZAKU
********************************************************************************

Suzaku software Version 9 - Changes since HEASoft v6.4.1

---------------------
 XIS specific tools:
---------------------

xispi :
  - Eliminate flood of messages, e.g. "reading PARALLEL_CTI_xxx at **-th row" 
    when reading PARALLEL_CTI row in the makepi CALDB file and to access
    the makepi CALDB file much faster.

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

Swift software Version 2.9(Bld23) - Changes since HEASoft v6.4.1

--------------------
General  tools
-------------------
swiftxform :
        - specifying ra=-1 causes ra,dec,roll to be taken from {RA,DEC,PA}_PNT
-------------------
BAT Specific tools 
-------------------

batsurvey [NEW TOOL] :
        -  This task performs processing on BAT Survey DPH data.
        -  The sub-tasks batsurvey-aspect, batsurvey-detmask, batsurvey-erebin,
           and batsurvey-gti perform various temporal and spatial filtering
           operations.

batbinevt :
        - New binning method 'highsnr' which enforces a minimum signal to
          noise ratio (SNR), compared to 'snr' which enforces a maximum SNR.
        - Allow users to specify energybins=INFILE for event files, if the
          input file has discrete energy bins.
        - Add descriptive TTYPEn column comments for output files.

batcelldetect :
        - Add descriptive TTYPEn column comments for output files.

batdrmgen :
        - Change some WARNING messages to NOTEs

bathotpix :
        - Add descriptive TTYPEn column comments for output files.

batmasktaglc :
        - Add descriptive TTYPEn column comments for output files.

batmaskwtevt :
        - Add descriptive TTYPEn column comments for output files.

batmaskwtimg :
        - INTERFACE CHANGE: (version 1.19): now the 'infile' parameter is
          required for operation; the reason is that after the Aug-Sep 2007
          gyro problems, users can no longer rely on the default time of
          'now' to pull out the right alignment files from CALDB.
--------------------
UVOT Specific tools
--------------------

uvotapplywcs [NEW TOOL] :
        - Performs coordinate conversions

uvotgraspcorr [NEW TOOL]:
        - Aspect corrects grism images

uvotinteg [NEW TOOL]:
        - Determines centroid, area, sum for region of image

uvotunicorr [NEW TOOL] :
        - Aspect corrects image based on user defined matches

uvotdetect :
	- support operating on RAW and DET images
	- added regfile parameter for detected sources region file
	- added sexargs parameter to simplify user control of SExtractor
	- updated PHOT_APERTURES to 3 and 5 arcsec

uvotexpmap :
	- when aspect following information is missing for an IMAGE mode
	  exposure, create it based on attitude file
	- added masktrim parameter to allow user to control trimming of masks

uvotflatfield :
	- report warning if flat field is unity everywhere

uvotimage :
	- correct times in primary HDU

uvotimgrism :
	- by default, determine extraction angle from WHEELPOS
	- useful support for specifying source position ra and dec, provided
	  the image has been aspect corrected by uvotgraspcorr

uvotimsum :
	- by default, constrain FRAMTIME of inputs that are summed
	- write exposure weighted mean FRAMTIME to output
	- by default, exclude SETTLING mode images from sum

uvotmaghist :
	- allow user to specify uvotsource parameters (calibration files,
	  fwhmsig, centroid)
        - the UVOT team recommends using fwhmsig 15%

uvotproduct :
	- allow user to specify uvotsource parameters (calibration files,
	  fwhmsig, centroid)
	- added parameter timezero to control reference time
	- support automatically determining background region(s)
	- do not use SETTLING image for finding chart
        - the UVOT team recommends using fwhmsig 15%

uvotscreen :
	- propagate updated WINDOW extension to output

uvotskycorr :
	- parameter options=FORCE causes existing corrections to be overwritten
	- user can specify max.rate=value [counts/s] in the starid parameter
	  to control the brightest detections that will be used for matching

uvotsource :
	- fixed problem with processing non-circle source regions
	- report most corrected rate (currently LSS_RATE)
	- updated label for corrected rates
	- added support for determining centroid of source region
        - the UVOT team recommends using fwhmsig 15%

--------------------
XRT specific tools
--------------------

xrtlccorr [NEW TOOL] :
        - Generates  a  rate  correction file and a corrected light curve
          (optional)  accounting  for telescope vignetting and Point Spread
          Function  (PSF)  corrections  due to the geometry of the light
          curve  extraction  region. Photon losses due to bad pixels/columns
          falling within the extraction region are also taken into account.
          The XRT Photon Counting (PC) and Windowed Timing (WT) operational
          modes are supported.

xrtcalcpi :
        - Support new PHA to PI conversion and new CALDB file format in order to
          account for charge traps.

xrtpcbias :
        - New input parameter 'biascorr' to apply/or not bias correction.

xrtpccorr :
        - Not applies PHAS correction for telemetred bad pixels.

xrtcentroid : 
        - New input parameters 'dateobs' and 'timeobs'.

xrtexpomap :
        - Delete the sky coordinates instrument map and save the one in raw
          coordinates when 'cleanup' is set to "yes" (default).

 xrtproducts :
        - Changed WT default region for products extraction from BOX to CIRCLE
        - Removed 'roll', 'width' and 'height' input parameters.

xrtfilter :
        - Replaced 'makefilter' task call with 'newmakefilter' call
        - Removed 'tprec' input parameter.

xrtpipeline :
        - Handle Mac OS X AppleDouble file format
        - Removed 'tprec' input parameter
        - Changes to support Build 23.

********************************************************************************
                                      XTE
********************************************************************************

- asmappend / asmsrcupdate: Modified to accomodate new ASM file formats
  from MIT.

- pcabackest: Improved clarity of error messages and allowed for longer
  file pathnames.

- sbmerge: Major revision, v2.0. Fixed bugs related to double counting of
  events, TDDES2 keyword, and performance issues.

Web page maintained by: Bryan K. Irby