--------- GENERAL: --------- All the code was extensively rewritten to allow compatibility with linux. - The new code is not compatible with old source code. - It also runs on AIX, IRIX, etc. - On linux, it requires NAG f90 compiler, or you can get and use statically-linked binaries. - The linux version now uses the elf libraries. Added unix2unix (for byte swapping between unix systems). Removed old non-x-menu versions of pdedit, psedit. - use pdeditx and pseditx. Added unfold-type response function (.rsp) file handling to vms2unix, unix2unix and unix2vms. Changed logical/environment variable name GFONLINE_LOC to RADWARE_GFONLINE_LOC. This points to the directory containing the file gfonline.hlp. Be sure to change it in your .login / .radwarerc / LOGIN.COM file. There are several new logical names (VMS) or environment variables (unix) to control the behaviour of the programs. See also the new .radwarerc file. -- Cursor input routines (subroutine retic*): Now allow definition of RADWARE_CURSOR_BELL to turn off the bell. -- RADWARE_CURSOR_BELL = Y/y/T/t/1/undefined for bell on -- RADWARE_CURSOR_BELL = N/n/F/f/0 for bell off -- Works for all cursor routines, in gf2, escl8r, etc. For example, in VMS, type "define radware_cursor_bell n", or in unix, type "setenv RADWARE_CURSOR_BELL n" or "export RADWARE_CURSOR_BELL=n". -- In unix, you can now also set the environment variable RADWARE_OVERWRITE_FILE to specify whether you want to overwrite existing files without being asked. -- RADWARE_OVERWRITE_FILE = Y/y to always overwrite files -- RADWARE_OVERWRITE_FILE = N/n to never overwrite files -- RADWARE_OVERWRITE_FILE = anything else or undefined will cause you to be asked whether to overwrite, on a file-by-file basis (i.e. previous default). -- You can now also define RADWARE_AWAIT_RETURN to specify how you want the programs to behave when they are asking you questions that require only one input character, for example all those yes-no (Y/N) questions. -- RADWARE_AWAIT_RETURN = Y/y to require the return key to be pressed -- RADWARE_AWAIT_RETURN = anything else or undefined to accept a single keystroke (i.e. previous default). -- You can now also define RADWARE_XMG_SIZE to specify the desired initial geometry of the level scheme pixmap in xmgls, xmesc and xmlev. For example, setenv RADWARE_XMG_SIZE 400x450 will try to create an initial level scheme display (before zoom) of 400x450 pixels, rather than the default of 600x500. This lets you, for example, customize the aspect ratio of the level scheme display to suit your .gls file, thereby optimizing use of the display memory. -- You can now also define RADWARE_PROMPT to contain a text string to be appended to all questions asked by radware programs. Added automatic detection of byte ordering, and swapping if necessary, for the some common RadWare files (specifically, .spe, .eff, .cal and .tab files), in unix only. This means, for example, that you can FTP such files from a SUN to a linux box and not worry about having to run unix2unix, or share them on differently-byte-ordered systems. For FTP between VMS and unix, you will still need to run vms2unix or unix2vms. The hsicc routines (part of gls etc.) were modified to check the byte ordering of icc data files and automatically swap the bytes if necessary (for example on utrix/linux). The extra copy of the icc tables for dec/linux ordering was deleted. The position and sizes of all graphics windows are now saved at normal exit of the programs, and restored at next entry. In order to achieve this, files (Unix: .radware.xwg and .radware.xmg, VMS: radware.xwg and radware.xmg) are created and modified in your login directory. Added incub8r_lib fortran source code, a source library for creating your own incub8r replay programs. Added program sdgen to create gatelist files for SDB searches in levit8r. Added integer*4 matrix handling to escl8r and symmat. Specify file extension = .spn or .m4b for 4B / channel matrices. --------- GF2: --------- Added "autocalibrate" routine to gf2. This makes it very quick and easy to get energy and efficiency calibrations from source spectra. It is intended also to help in gain-matching data from large detector arrays. - the command is CA - it automatically writes a .sin file - you may wish to create a .sou file with some (doublets/contaminated) lines removed (see, for example, demo/eu_autocal.sou) Added "autobackground" routines to gf2, a procedure which tries to generate a background spectrum for matrix projections, for use in escl8r, subbgmat etc. - the command is BG Added new feature to the gf2 command OV: OV -1 now overlays with same vertical scale as last spectrum that was displayed. This is very useful for comparing spectra on the same counts scale. Changed gf2 to make initialisation routine look for files gfinit.dat and gfinit.cmd in your home/login directory if they are not found in the current default directory. Fixed major bugs in gf2x interface. Updated gf2.hlp and gfonline.hlp. --------- GLS, ESCL8R, LEVIT8R: --------- Removed old non-gls versions of escl8r, levit8r, and also the programs levels, lsedit and lseditx. - only support for .lvl files is now through gls_conv - escl8r_gls renamed to escl8r, levit8r_gls renamed to levit8r Escl8r and xmesc can now create and use 4kx4k-channel escl8r data files. The filename extension for such files is .e4k (rather than the standard .esc), and you need to specify the extension when you create or reopen the file. Added integer*4 matrix handling to escl8r and symmat. Specify file extension = .spn or .m4b for 4B / channel matrices. For portability across computer systems, added a new ASCII gls file format. This will allow you, for example, to e-mail a level scheme to a collaborator. - filename extension is .ags - can read and write .ags files instead of .gls files in gls, escl8r, etc., by specifying the filename extension as .ags - also added new options to gls_conv to convert .gls to/from .ags files gls, escl8r, levit8r, xmgls, xmesc and xmlev: shift-mouse_buttons can be used to pan, zoom and unzoom the level scheme display. This is not so useful for the Motif versions, since they already have zoom and scroll bars, but is a big help in the non-Motif versions. In the non-Motif programs, you must be in a cursor-input mode (for example, executing a command such as XG, MB...) for it to work. The gls programs now allow optional decimal place on level-energy and gamma-energy labels. - e.g. 123.4, (123.4) Added new command/menu option MM, "Move Mutiple Bands", to allow you to move a set of bands as a unit, while preserving their relative positions. All gls programs now have default .gls filenames for saving at exit. At the filename prompt, you must specify "q" to quit without saving. Escl8r and levit8r now create log files .esc_log and .lev_log, and record in them all calibration filenames, background-spectra filenames, peak shape and width parameter values, etc. If you forget which files or parameters you used, you can look back in the log file. Escl8r and levit8r now let you write out the non-background-subtracted gate as well as the subtracted gate, etc. These programs now keep 6 spectra for each gate, rather than the old 4. Fixed a problem with having .lev files messed up when levit8r crashes in unix (I hope). Level scheme hardcopy for expanded regions of the scheme now gives smaller .ps files and hence faster printing. This is achieved by writing to the .ps file only those levels, gammas and text labels that are close to the expanded region. The maximum number of fitted parameters in escl8r and levit8r used to be fixed at 500. There is now a parameter MAXFIT in the file LEVELS.PAR that determines the maximum allowed number of fitted parameters. Be a little cautious in increasing this beyond the default of 500. Added pointer-information display, to tell you the position of the pointer in the graphics windows, to xmgls, xmesc and xmlev. It also lists information on the nearest gamma and level.