Документ взят из кэша поисковой машины. Адрес оригинального документа : http://xmm.vilspa.esa.es/sas/6.0.0/doc/rgsrmfgen/ChangeLog Дата изменения: Mon Mar 15 19:03:30 2004 Дата индексирования: Sat Dec 22 09:07:55 2007 Кодировка: Поисковые слова: arp 220

ChangeLog for rgsrmfgen
===============================

Version 1.5.3 - 2004-03-15 (CG)
-------------
+ parameter name update in docu (ebins >> rows)

Version 1.5.2 - 2004-02-06 (CG)
-------------
+ bogus with executable permission

Version 1.5.1 - 2004-02-05 (CG)
-------------
+ bug introduced in 1.5 (calib.cc - small angle scattering) fixed

Version 1.5 - 2003-06-27 (CG)
-----------
+ replacement from 1.4 now correct

Version 1.4 - 2003-06-25 (CG)
-----------
+ maximize replaced by STD max in calib.cc

Version 1.3 - 2003-04-22 (CG)
-----------
+ few changes (required by gcc3.2) to improve standard C++ compliance

Version 1.2 - 2002-08-19 (UL)
-----------
+ now DEPENDs on package sas
+ a few transparent changes to improve standard C++ compliance

Version 1.1 - 2002-06-28 (JDR)
-----------
- replaced parameter --whichmirrorpsf with --withmirrorpsf and --withangdist

Version 1.0 - 2002-06-25 (JDR)
-----------
- requires cal-3.142
- switched roles of mirror PSF and grating misalignment distributions in
the convolutions to allow the mirror PSF to become optional
- the user may supply a custom angular distribution to replace the mirror PSF
- added a matrix editing mode
- in HTR mode the background correction is silently disabled
- changed standard file name tag to RSPMAT for future compatibility with PPS
- progress messages are considerably less verbose
- task description not yet updated

Version 0.51 - 2002-05-09 (JDR)
------------
- requires param-2.0

Version 0.50.2 - 2002-02-12 (JDR)
--------------
- task description updated with correct default for the --ebins parameter

Version 0.50.1 - 2002-01-17 (JDR)
--------------
- supports arbitrary energy selection regions
- bug fix: does not rely on implicit initialization of global variables

Version 0.50 - 2002-01-16 (JDR)
------------
- changed default to --ebins=4000

Version 0.49 - 2001-10-10 (JDR)
------------
- changed parameter name --set to --evlist and --sourceset to --srclist

Version 0.48 - 2001-10-01 (JDR)
------------
- updated runtime default filenames

Version 0.47 - 2001-08-30 (JDR)
------------
- replaced 0.678 with 1/0.678 in empiricalCrossLA()

Version 0.46 - 2001-08-17 (JDR)
------------
- added parameter --fftdim to control the convolution size

Version 0.45.2 - 2001-08-01 (JDR)
--------------
- the small angle LSF is computed more carefully: the convolution size
now hovers between 512 and 1024; runtime is a bit slower

Version 0.45.1 - 2001-07-16 (JDR)
--------------
- DEPEND on rgslib and cal backed down to SAS 5.1 version, which is fine

Version 0.45 - 2001-07-13 (JDR)
------------
- code fixed for OSF platform compiler

Version 0.44 - 2001-07-11 (JDR)
------------
- hardwired lfactor value as 0.678 in empiricalCrossLA()
- rgsrmfgen: fixed large-angle loss function formula
- rgsmcrgen: removed temporary parameter --lfactor

Version 0.43 - 2001-06-29 (JDR)
------------
- rgsrmfgen: optimized the convolutions (about five times faster)
- rgsmcrgen: added temporary parameter --lfactor

Version 0.42.1 - 2001-06-11 (JDR)
--------------
- added check for null bowing distribution

Version 0.42 - 2001-06-08 (JDR)
------------
- rgsrmfgen: reorganized to further separate large and small angle treatment
- rgsrmfgen: chip 1 CCD redistribution function used for all chips
- added proprietary empiricalCrossLA (large angle scattering profile)

Version 0.41 - 2001-05-10 (JDR)
------------
- added CAL LSF defocus distribution
- replaced proprietary bin-size broadening distribution with CAL interface
- rgsmcrgen: separate tensors for small and large angle scattering
- rgsmcrgen: added proprietary large angle xdsp scattering distribution

Version 0.40 - 2001-04-20 (JDR)
------------
- added CAL bowing figure distribution

Version 0.39 - 2001-04-11 (JDR)
------------
- updated for rgssources-4.0

Version 0.38 - 2001-04-06 (JDR)
------------
- fixed bug introduced in version 0.36 (corrupted LSF)

Version 0.37 - 2001-03-23 (JDR)
------------
- replaced proprietary large-angle scattering distribution with CAL interface
- requires rgslib-1.10

Version 0.36 - 2001-03-09 (JDR)
------------
- replaced proprietary small-angle scattering distribution with CAL interface

Version 0.35 - 2001-02-07 (JDR)
------------
- updated for and requires rgslib-1.5
- replaced proprietary grating misalignment and mirror psf with CAL interface

Version 0.34 - 2001-01-10 (JDR)
------------
- updated for and requires rgslib-1.4
- replaced proprietary CCD quantum efficiencies with CAL interface

Version 0.33.1 - 2000-11-06 (JDR)
--------------
- CAL state set by hand to eliminate spurious warnings

Version 0.33 - 2000-11-03 (JDR)
------------
- requires rgslib-1.2

Version 0.32.1 - 2000-10-27 (JDR)
--------------
- updated GROUPS

Version 0.32 - 2000-09-25 (JDR)
------------
- MirrorDataServer --> EffectiveAreaDataServer (cal-3.81)

Version 0.31 - 2000-09-22 (JDR)
------------
- updated default parameters for compatibility with rgsproc-0.82
- parameters and errors now documented in task description

Version 0.30 - 2000-09-15 JDR
------------
- changed default rgs1fudge to 1.5
- fixed two bugs in the background correction logic
- updated hard-coded parameters of LSF broad scattering component
- removed internal calculation of FOV coordinates
- rgsmcrgen: adopted CAL sign convention for RGS_XDSP angles
- rgsmcrgen: workaround for theta<0 bug in CAL

Version 0.29 - 2000-08-25 JDR
------------
- fixed bug in rgs1fudge (it was being applied to RGS2)

Version 0.28 - 2000-08-25 JDR
------------
- compiles under gcc-2.95.2
- fixed channel space compatibility test
- fixed uninitialized variable in spatial mask filter
- added --rgs1fudge for extra broadening in RGS1 only
- separated LSF into broad and narrow components
- rgsmcrgen: fixed sign error in offaxis angle conversion

Version 0.27 - 2000-07-21 JDR
------------
- the background contribution is deducted (optionally)
- added line-broadening convolution (finite bin-width effect)
- exposure map is normalized for each node independently
- compatible with count rate spectrum as per rgsspectrum
- not compatible with count spectrum as per evselect

Version 0.26 - 2000-07-11 JDR
------------
- changed --emax default to 2.8
- now depends on dal-1.21

Version 0.25 - 2000-07-06 JDR
------------
- now depends on rgslib-1.0

Version 0.24.2 - 2000-06-15 JDR
--------------
- some hardcoded geometric constants replaced with CAL interface
- added dataset attributes: SOURCEID, RFLORDER, BETA_REF, BETA_WID
- fixed exposure map rebinning error

Version 0.24.1 - 2000-06-01 JDR
--------------
- hardcoded RGA distribution data updated
- replaced CAL mirror effectiveArea with intrinsicEffectiveArea
- replaced proprietary self-vignetting function with CAL interface

Version 0.24 - 2000-05-25 JDR
------------
- replaced proprietary cross-dispersion PSF with CAL interface
- reorganized for code-sharing with rgsmcrgen
- added rgsmcrgen: generates response tensors for monte carlo analysis

Version 0.23 - 2000-05-10 JDR
------------
- output channels now defined by spectrum rebinning
- external proprietary calibration database no longer used
- hardcoded CCD layer thicknesses updated

Version 0.22 - 2000-04-06 JDR
------------
- tied channel space definition to events table
- improved parameter interface

Version 0.21 - 2000-03-14 JDR
------------
- attempt to fix DEC platform build error: added string.h to prf.c includes

Version 0.20 - 2000-03-10 JDR
------------
- extraction region table names changed by rgsregion

Version 0.19 (JDR 2 February 2000)
----------------------------------
- replaced proprietary Henke data with CAL interface

Version 0.18 (JDR 1 February 2000)
----------------------------------
- added selectlib to package dependencies

Version 0.17 (JDR 28 January 2000)
----------------------------------
- class Convolver moved into slatec

Version 0.16 (JDR 24 January 2000)
----------------------------------
- bypass computation of combined exposure and property maps
- replaced Convolve3 with generalized Convolver

Version 0.15 (JDR 11 January 2000)
----------------------------------
- removed exposure map normalization
- moved ShapeScanner to rgslib
- eliminated all Numerical Recipes code
- replaced fitsio calls with DAL calls

Version 0.14 (JDR 23 December 1999)
-----------------------------------
- apply crude normalization to exposure map

Version 0.13 (JDR 22 December 1999)
-----------------------------------
- replaced sron_ccd_prf() with CAL Redist interface

Version 0.12 (JDR 21 December 1999)
-----------------------------------
- updated for compatibility with selectlib-4.23

Version 0.11 (JDR 17 December 1999)
-----------------------------------
- compatible with rgsregion-2.6 extension names and ra/dec units

Version 0.10 (JDR 16 December 1999)
-----------------------------------
- first version ready for SAS integration, but not yet tested
- still relies on proprietary calibration data

Version 0.9 (JDR 8 December 1999)
---------------------------------
- Added scan converter for efficient region processing.
- Conversion of prf interface in progress; task does nothing as yet.

Version 0.8 (JDR 30 November 1999)
----------------------------------
- Directory structure created by pkgmaker.
- Fresh start; only numutils subdirectory inherited from version 0.7.
- Conversion of prf interface in progress; task does nothing as yet.