|
Version 6.0 of Scanpi (released in 2007) is a re-write of the classic software, Scanpi, version 5.0.
Some differences between the classic and new version exist. A full list of these
differences is noted below; some of the key differences are illustrated using
an example source.
- User parameter default
settings have been documented in the Scanpi User's Guide, where the classic default
settings are noted in red.
- The User Interface front page has been consolidated into one master page,
instead of a drop down menu to access documentation and parameter settings.
- Additional user input parameters have been added for more flexibility. The new parameter
"Show Plots of Intermediate Steps" gives access to all intermediate files/plots in a
summary page formated for ease of viewing. The new parameter "Results Offset (x) Display Range" allows the user to change the display range while keeping the fitting range separate; the classic version tied the background fitting range and the display range.
- Access to the workspace was allowed in the classic version, and is retained in the new version. However, the file structure and naming scheme have been changed; all the files have been organized and named using a well-defined naming convention. No binary files are in the workspace; all files are ASCII, jpg or gif. The workspace is unique for each source requested, unlike the classic version where all sources requested by the same user were placed into the one workspace, making it difficult to sort through the information. All files in the workspace are described in a
README.
- Scan Tracks are plotted on top of a background sky image in the new version. The user can chose to display the scan tracks using ISSA and/or DSS images as backdrops.
- UTC Date has been added to the headers of all the individual scans and results summary tables.
- Upload Table of multiple sources is now a single-step process, where the results for all the requested sourses are reported, with a success status and link to each result. The classic version used a multi-step process in order to run Scanpi on more than one source.
- Super Scanpi access has been removed from the new version of Scanpi. In order to coadd multiple sources, the user can download any/all results from the new Scanpi and coadd them using off-the-shelf software.
- Similarly to Super Scanpi, SCLEAN access has also been removed from the new version of Scanpi. Coaddition and fitting of multiple sources can be performed by the user using off-the-shelf software.
- Plotting is now done in color in order to better represent multiple item overlays on the same plot. Autoscaling on plots takes care of issues of truncated plots often seen in the classic version.
- Sigma was calculated using the "goodness" of the fit of the background. Sigma is now being calculated as the root mean square using the data within "Local Background Fitting Range", taking the "Source Exclusion Range" into account. Sigma calculations effect the value of 'snr' in the summary tables; the new and classic Scanpi results may vary for these two sets of values.
- Zero crossings for the new and classic Scanpi tend to vary; the crossings depend highly on the background fitting ranges and whether the source is extended or point-like. Dependent on the zero crossing results, is the value of fnu_z in the summary table. The new and classic Scanpi results may differ for the zero crossings and therefore fnu_z.
- Source Fit Plots and Data File access were available only for the coadd scans in the classic Scanpi; the new version has access to plot and data files for all individual scans as well as coadds.
- Classic Scanpi source fit plot for coadd scan #999 contained the background flux, while the rest of the coadds did not. In order to be consistent in the new Scanpi, all coadd scans are shown without background subtraction on the results summary pages. The user can also view background subtracted versions of the individual and coadd scans by selecting "Show Plots of Intermediate Steps".
- The scan direction may be reversed in the new version vs the classic.
The new code chooses the nominal scan direction so that the majority
of scans will be 'forward', the rest are 'reverse'. Classic Scanpi
used different logic, with the 'forward' scans controlled by whichever
one was closest to going due North.
- Source fitting uses a point source template to calculate the best fit with the lowest correlation coefficent.
There are three templates available for fitting - a forward direction template, a reverse direction template and
an average of those two. Classic Scanpi used either the forward or reverse template for individual scan source fitting,
depending on the direction of the scan. It used the average template for all coaddition source fitting.
The new Scanpi used the same template logic for the individual scan source fitting. However, for the
coadd source fitting, new Scanpi chooses the template according to a 2/3rds major of scans. For example,
if more than 2/3rds of the scans in the coadd were forward, then the forward template is used for
source fitting. If more than 2/3rds of the scans in the coadd are reverse, then the reverse template
is used for source fitting. If there is no 2/3rd majority of scans, then the average template is used.
This affects the "amp" and "corr_coef" results.
|