Appendix 5 - Quick-Look & On-line Reduction

  1. VAX or Sun ?
  2. Where's the Data?
  3. Using IRAF and FIGARO
  4. Using the FIGDISP display
  5. Quick Look with QIKLOOK
  6. Quick Look with IMEXAM (in FIGARO)


Introduction The Telescope & Optics The Detectors
The Imaging Cameras An Imaging Cookbook The Data you Take Away
Exposure times OFFSET_RUN files CCD Windows Data Catalogs
On-line Reduction Filters Flat-fields Blank Fields Orientation Shutters

Back Contents Next

Last Contents Next

1. VAX or Sun?

The data can be examined on-line with either the VAX or Sun workstations.  

The Sun can be used directly from its console, as well as any of the X-terminals in the control room. The VAX can be accessed by any of the X-terminals. (The two VT terminals can be used to access the VAX, but have no graphics capabilities.)

The Sun is somewhat faster than the VAX for actual CPU operations, but slower in the sense that the data must be read across the network from the VAX.


2. Where's the data?

The VAX data goes in files like

DISK$INST:[CCD_1.960237]37FEB0001.DST         (for RUNs)
DISK$DATA:[CCD_1.960237]A.DST                 (for DUMMYs)

which you can see from the Sun in files like

/vaxinst/ccd_1/960237/37feb0001.dst           (for RUNs)
/vaxdata/ccd_1/960237/a.dst                   (for DUMMYs)

However, you should be aware that these files are NFS-mounted from the VAX, so access to the files can be somewhat slow. To help get around this, the unix_server process should be started when you start up OBSERVER on the VAX. This will automatically copy the files created on the VAX across to the Suns. This means the files can be accessed on the Suns without using the network (after the initial transfer that is). The files will appear on the Sun about 20s after they are created on the VAX . Moreover, FITS and IRAF versions of the files will be automatically created. The files will appear in the following locations

/data/ssf/1/obsred/inst/yymmdd/ddmmm0001.[dst|sdf]
(for Figaro files)
/data/ssf/1/obsred/fits/yymmdd/ddmmm0001.fits
(for FITS files)
/data/ssf/1/obsred/iraf/yymmdd/ddmmm0001.{imh,pix}
(for IRAF files)

You will notice that below each of the inst, fits and iraf directories, we have a direct copy of the structure which is created on the VAX. Note also that theses files on the Sun are NOT archived (that is still done from the VAX disks), and can be freely deleted - in particular if you run out of space and are not using all three copies of the files (say you are only using the Figaro or IRAF files), you can delete the other two copies.

You should log into the OBSRED accounts on either machine to do on-line reduction (your support astronomer will give you the password). Note that you cannot write data into the directories where the archived data is stored. You must write files into other directories. So before you start looking at data you should change into a directory with some space. On the Sun these are /data/ssf/1/obsred and /data/ssf/2/obsred. On the VAX this is DISK$DATA:[OBSRED].

You can feel free to find anything you find in here at the start of your run. But be warned - so will the next observer, so you must back your reduced data up to tape before the afternoon of the next observer's run.


3. Using IRAF and FIGARO

On the Sun you can use either IRAF or FIGARO to look at your data.

To start FIGARO simply type on of the figaro commands

figaro42          for the latest beta version of Figaro4(4.2)
figaro4           for an older version of Figaro4 (4.0)
figaro3           for Figaro3

Then type

figdisp           to start an image display

Then to look at your data use

image /vaxinst/ccd_1/960237/37feb0001 au re opt=0 \\           or
imexam
 /vaxinst/ccd_1/960237/37feb0001

To use IRAF you may have to re-create a login.cl file in the directory where you want to look at the data. Then start IRAF with CL. To get the data into IRAF you currently have to use FIGARO to write a FITS file, then RFITS in IRAF to read the FITS file into IRAF format. At the UNIX prompt type

wdfits /vaxinst/ccd_1/960237/37feb0001 crap.fits \\

then within IRAF

rfits crap.fits 37feb0001

will create a file called 37feb0001 which IRAF can display. (Note that a symbolic link to the /vaxinst directory can save a lot of typing.)

On the VAX you can start FIGARO as follows

XON your_X-terminal's_name             (the name should be written on the terminal)
FIGARO
FIGDISP


4. Using the FIGDISP display

The FIGARO display device is an application called FIGDISP. It is started with the command `figdisp', though if you are using an X-terminal you must have the DISPLAY or DECW$DISPLAY variable set.

Use F5 once the display has started to get some help. The main features you will want to use are

Figdisp Commonly Use Functions
F2 Zoom in
F4 Zoom out
F3 Return to 1-1 zoom
F5 Help!
F7 Recenter the image in the center of the display
F9 Quit
Click Left Mouse Button Move that image location to the center of the display
Drag Middle Mouse Button Plot a cut through the image display
Drag Right Mouse Button Change display look up table
Click Right Mouse Button Return display look up table to default setting.

 

5. Quick look with QIKLOOK on the Suns

QIKLOOK is a graphical user interface to the IMEXAM routine. It allows you to perform quick look examination of imaging (and some spectroscopic) data, without actually knowing any of the details of how Figaro or IMEXAM really works (although having a small clue how Figdisp works will help).

QIKLOOK is launched from the obsred account on the Suns using the command

aatssf> qiklook

You can also launch it from the Reduction pull-down in the AAT menu system, which you can start from the obsred account using

aatssf> AAT

In eithert case, when qiklook start you'll get

  1. a small xterm window (on which some useful logging information will be output)
  2. a Figdisp window (if there is not already one running) , and
  3. the QIKLOOK panel, which looks like

The QIKLOOK panel itself is really quite easy to operate. The green section is the file chooser. You can use this to examine directories - either the remote mounted ones from the VAX, or the directories on the Sparc stations to which unix_server copies its files. The buttons at the bottom give you `accelerators' to get to the useful directories. You "select" a file for further action by simply clicking on its name.

Next, there are display buttons in the red area. Once you've selected a file, its name will appear next to the `Show' button. You can use this to display the image. You can use the `Show Latest' button to automatically pick and display the latest image. The button with the three dots `...' produces another pop-up window which allows you to set some useful default parameters (image vs spectral auto-scaling, whether to display saturated pixels, what colour table to use, ...). If you make a choice here it will apply next time you perform any QIKLOOK actions on the image.

Then on the right there's a series of action buttons in the blue area. These will display a new image (or use an already displayed image if you haven't selected a new image) and allow you to interrogate the image. Each set of actions has a `...', which allows you to define default parameters. For example, the `...' button next to the `Seeing & Photom.' button produces a pop-up like so

This allows you to change the image scale, expected gain and read-noise of you image. You can also define the radii QIKLOOK will use when it estimates seeing FWHM and photometry. You can use the menu lists to select useful pre-defined values for these parameters, or enter new values by hand in the appropriate boxes. Once you are happy, dismiss this pop-up and hit 'Seeing & Photom.'. QIKLOOK will spawn an special IMEXAM process, which will show you a point-source profile, seeing and photometry for wherever the cursor is in the Figdisp display window when you hit the space bar. When done measuring seeinf, hit 'Q' in the Figdisp window.

The other action buttons allow you to measure X-Y Profiles (as opposed to radial profiles), estimate best focus from a focus sequence image, and estimate telescope offsets using your image. In each case, there's some information about your image which QIKOOK needs to know, which can be set using the `...' button. In some cases (focus and offset), you'll always get a pop-up to fill in.

Figdisp

The image display used is the figdisp one used by Figaro 4.x. It is useful to know a bit about its functions.

Expert users

If you know anything about Figaro you'll quickly see the QIKLOOK is just a GUI (written in PerlTK) which operates as a front end to the IMEXAM task, and allows you to use some IMEXAM functions without needing to know, for example, that `P' stands for radial profile, and `I' stands for spectral profile, and `F' stands for focus, and `C' stands for change parameters. If you do know those things, then you may prefer to access all the IMEXAM functions by simply use the `IMEXAM' action button. You may still want to set useful parameters with the relevant `...' buttons however.

For even more expert Figaro users, it may help to know that IMEXAM grew out of the CPOS routine.

Last things

QIKLOOK is still under development. Comments, bug reports and suggestions of features you'd like to see will be gratefully received by cgt@aaoepp.aao.gov.au.


6. Quick Look with IMEXAM in FIGARO

IMEXAM is a local Figaro application. It is not part of standard Figaro (though it may become so). It has not yet been implemented on the VAXes.

The QIKLOOK system is based on a GUI front end to the IMEXAM routine in Figaro. If you don't like GUI's, or want to know a bit morre about IMEXAM and its functions (which is useful for using QIKLOOK) then read on.

The IMEXAM program should be the only FIGARO application you will need to run in order to look at your data while at the telescope. For those with a knowledge of FIGARO, it may help to know that it grew from a slightly bodged-about version of the CPOS program, until now it includes just about everything you'd want to do with data at the telescope. Among its functions are:

Because IMEXAM uses the FIGARO parameter system, which `remembers' the defaults you used last time you ran IMEXAM, the program is designed to require as user input only the name of the image to be displayed. Once you have set the defaults at the start of your run (image scale, whether its a direct or spectral image, magnitude zero-point, object and sky aperures etc), you should only really need to type somthing like

imexam /vaxinst/ccd_1/960228/28feb0010                       (Sun)

and if you create a logical link to the data directory

ln -s /vaxinst/ccd_1/960228/28feb0010 mydata

then you can shorten that to

imexam mydata/28feb0010

In order to see all the options for IMEXAM (which are normally not prompted for) use

imexam prompt

You can fix the value of a numeric parameter, by simply typing in a new value when prompted. Next time you run the program, the default value for that number will be what you used last time.

You can fix the value of a keyword (ie a parameter requirig a yes or no answer) by typing yes! or no! (if you just type yes, or no, only the current value is changed, the default value will remain the same).

You can always see what the current default value of a parameter is, because FIGARO always shows it to you (in square brackets) when it prompts you.

A typical useful set of defaults would be something like the following (for B band,  f/3.3 imaging with the TEK chip in FAST readout and typical (1-2") seeing) :-

imexam mydata/28feb0010 pr              
      pr is an acceptable abberviation for prompt
(APpend) Append points to existing file [NO] -
(AUto) Set Initial Display Limits automatically [YES] - yes!
(SPECTRAL) This is a spectral image [NO] - no!
(CRoss) -ve=Cross,0=Point,+ve=Box size [20] -
(CURsor) Interrogate image with cursor [YES] -
(QUEry) Prompt before quitting [NO] -
(SATCOLour) Saturation colour (Red, Blue, Green, None) [none] - g
(SATVALue) Threshold to flag as saturated on image. [32000] - 65000
(LUT) Colour table file (use NONE for greyscale) [heat] - grey
     You can use any FIGARO colour table. GREY and HEAT are the two most useful I find.
(IMERase) Erase screen before display [YES] -
(IMAGESCALE) Arcseconds/Pixel [0.3333333] - 0.391
(IMAGERAD) Radius of Photometry Aperture [15] -
(PSFtype) Type of PSF to fit (Moffat or Gaussian) [Moffat] -
(IMAGERA1) Inner radius of sky aperture [15] -
(IMAGERA2) Outer radius of sky aperture [20] -
(SPX) X Box size for spectral profile fits [15] -
(SPY) Y Box size for spectral profile fits [15] -

   
These last two determine the size of a box collapsed for a 1-D Gaussian - this is most useful for
       fitting a width to a spectral line.
(IMAGEBAD) Bad pixel rejection threshold [-150] -
(IMAGEPHotons) Number of photons per ADU [2] - 5.5
(IMAGEREAdns) Readout noise [10] - 7.2
(IMAGEMZero) Magnitude offset [24.6] - 24.6

If you are using QIKLOOK, all the above parameter stuff is deault with by the GUI. The only reason you'd need to come accross it is if you use the 'C' function while in one of the action tasks. Which you can do, if you want, of course.


Introduction The Telescope & Optics The Detectors
The Imaging Cameras An Imaging Cookbook The Data you Take Away
Exposure times OFFSET_RUN files CCD Windows Data Catalogs
On-line Reduction Filters Flat-fields Blank Fields Orientation Shutters

Back Contents Next

Last Contents Next

This Page maintained by : Chris Tinney (cgt@aaoepp.aao.gov.au)
This Page last updated:  7 Sep 1997, by Chris Tinney