SPEX COMMENTS I have found some problems recently including the scattering component in the drm. I don't think that this makes an important contribution to the drm for almost all cases of interest, so for the time being I will set the default scatter file to a blank, and advise strongly against setting it to any real scatter matrix. Also, _1file and _2file may now be left blank. As long as the det_id and flare are set correctly, the call to time_history will find the correct files. Setting the files manually or through find will still work. When you first enter spex, a set of comments will be printed describing recent changes. A command, comments, will also be available to again print them to the screen. All of the SPEX code has been placed in the SDAC account so these procedures will be installed with the normal system wide idl_startup. Those of you who have been running an IDL_startup including some of Richard's directories should no longer do so, since those directories will be used to test code. 3-feb-1994 You may use the build_drm command in the DRM menu of SPEX. This will interpolate on a set of matrices spaced from a cosine of 0-1 for both SHERS and CONT values for data_type. These should work just as well as the drm's obtained through the read command, and will allow any data set to be used as soon as it is in a directory readable from SDAC. 13-Feb-94 SPEX users: The problem with the LAD drm's has been fixed. I suggest using the build command. For CONT data, be sure to start at energies above 27 keV using the erange parameter. New BATSE DATATYPE: High channel resolution data (128 channels) is available for the BATSE LADs in the burst mode. It is the HER and HERB data specified by setting data_type to BATSE HERS. The drm build command works for this data type. ============================================================================== Time Interval Selection: The time intervals for accumulation may be directly input now by setting the values in the array XSELECT to the start and end values for the time for each of N intervals in a 2xN double precision array referenced to the same base time as the array UT which specifies the start and end time of each data interval. Generally, this base is set to the start of the day. This feature facilitates keeping the same approximate time intervals when switching data types for the same event. ============================================================================== T_hist_mode This is both a new mode and a new parameter. Normally it is set to FLUX and time history plots (including GRAPH) are displayed per second per keV and per square cm. If it is set to COUNTS then the data are displayed as counts per second. ============================================================================== Scale_bands A new parameter with four values used as multiplicative scalers for the four bands specified by energy_bands. ============================================================================== Model Fitting The model fitting has been made more robust and unbiased by implementing weighting during fitting based on uncertainties calculated from the input model. This is considered the preferred technique, but I hadn't implemented it until now. Also, you may have noticed some short intervals with fits that did not agree with the data. This was due to some time energy bins with no counts within the energy fitting range. Using the model uncertainties has eliminated this problem. ============================================================================== The creation of Postscript files for plotting in SPEX has been included in the command list. The command CREATE_PS issued at the SPEX prompt will cause the last plot command to be reissued with the output directed to a Postscript file named for the command, i.e. graph.ps, time_history.ps, background.ps, bspectrum.ps, count_spectrum.ps, photon_spectrum.ps, fitting.ps. Overlay commands are included automatically if they were issued, i.e. the commands GRAPH, DISPLAY, and CREATE_PS will cause a time history in the selected energy bands and the selected intervals to be plotted in the graphics screen and to a Postscript file. Also, a new modifier has been added to the FITTING command. Call FITTING,PLOT to redo the plots without passing through the fitting loops again or FITTING,FIT to refit without any plot commands being issued. After calling CREATE_PS Remote users can then copy, mail, or ftp the desired *.ps files to their home machines. Only users at Goddard should issue the command HARD which causes the last created Postscript file to be sent to the printer here. ============================================================================== 9-March-1994 There was a problem with creating the time arrays for these data types: sherb, sher, her, herb The time arrays had been too low by about 2.048 s. The error has been fixed. If you performed any analysis that was sensitive to this level of time difference it should be redone. The timing of the CONT data within SPEX remains unchanged and consistent with the PI team software. ============================================================================== 12-March-1994 Parameter syntax update, Blank field syntax is now allowed. To change the second field in DATA_TYPE BATSE SHERS DATA,,CONT results in DATA_TYPE BATSE CONT To clear a string field (anything non-numeric), use a '' for that field and that value is set to a null string. ============================================================================== ONLINE? Wondering how to query the SPEX database in a simple way to figure out which flares are there? Answer, use the new command ONLINE? It searches the appropriate BATSE database for SHERB files and reports the corresponding flare number. It takes a few seconds to get this info, but successive queries in the same SPEX sessions just print out the already acquired list. ============================================================================== PREVIEW Another new feature is the PREVIEW command. Call this with the FLARE, DATA_TYPE, and DET_ID set and the program will report useful information for detector selection without having to wait for the data to be read. SPEX> PRE Command = preview These are the detector cosines for BATSE SHERS 0.970207 0.765352 0.315168 0.110312 -0.110312 -0.315168 -0.765352 -0.970207 These are the BATSE detectors stored in the SHERB and HERB files: 0 1 7 2 The lower edge for their pulse-height data is about 15.2772 11.1016 212.789 101.648 The detector cosines for these detectors are: 0.970207 0.765352 -0.970207 0.315168 Found this file: HXRBS::SYS$USER:[RICHARD.BATSE.SEQ_OBS.DB_00468]SHER_COR_IBDB_7_468.FITS;1 ============================================================================== BACKGROUND,CLEAR Don't like your current background subtraction and you want to see the lightcurves prior to background subtraction, call the command BACKGROUND,CLEAR That is the background command with a parameter, or for you minimalists bac,c is also the same command. ============================================================================== T_HIST_MODE This command may be called with or without one of two command parameters, T_hist,c - for time history plots in units of counts/s T_hist,f - for time history plots in units of counts/s/keV/cm2 Without a parameter it will query you for a selection. ============================================================================== FITTING The FITTING command can take one of two optional parameters, FIT, or PLOT. Normally, this command causes fits and plots between intervals IFIRST to ILAST but by calling FITTING,f it will only do fitting w/o plotting. or FITTING,P it will only plot the previous fits. ============================================================================== ATTENTION: Soon the look of SPEX will change. Parameters will be grouped functionally and some parameters may require a command to be displayed, such as parameters which control the look of the graphics, some of which I haven't discussed or included explicitly in the menu yet while they're under development. Kim Tolbert is working on this part of SPEX, and if you have any ideas you can send her mail, kim@sdac.gsfc.nasa.gov, or call her, 301-286-3965. Also, we are in the process of acquiring CONT data for all flares in our BATSE archive and we will soon make the call to that data transparent to the user, requiring only the flare number. I will include an explicit command to enter the BATSE MENU shortly making it possible to throw up light curves and do parametric data base searches. If you haven't used it before or haven't used it in a while, be sure to try it. ============================================================================== WIDGETS ARE HERE! 28-Aug-1994 A Graphical User Interface is available for SPEX. Type WSPEX at the IDL prompt. The interface won't perform well at a network site without really good communications to Goddard Space Flight Center, but try it anyway. The HELP buttons aren't yet active but many other aspects are. The data selection controlled under the SPECIAL buttons is changing, but the TIME_HISTORY and FITTING controls seem to be stable. New datatypes have been integrated into SPEX: From Yohkoh the HXT and the WBS instruments, SXS, HXS, and GRS. However, the SXS is still not understood and should only be used with that understanding. SMM HXRBS has been added with all of the flare files soon to be on-line in the FITS format expected by the default readers within SPEX. Hey!!! There is lots of good stuff here in COMMENTS. Enter the command COMMENTS