| appendix | Version-3.0 | appendix |
APPENDIX
A. CTIO Calibration Files
B. Merging Projector flats with Dome flats
C. Using the UNIX Mailer on the SUNs
D. Linking Between Computers
E. Using a PC at CTIO
F. Using Suntool Windows
G. Using the Imtool Window
H. Using the EHISTORY and EPARAM commands
I. Using IMEDIT to Fix Pixels
J. Using the Satellite Link to the US
1. Remote Login
2. Copying Files
3. Sending Images Home
4. Collaborations
K. Using the Computers at CTIO, Paper by Lisa Wells
1. Introduction
1.1. Notation
2. The Workstations
2.1. Resources
3. How To...
3.1. Logging In
3.2. Logging Out
3.3. Transferring Pictures from the CCD Computer
3.4. Displaying Images
3.5. Printing Files and Plots and Other Things
3.6. Allocating the Tape Drive
3.7. Managing Disk Space
3.8. Using Sub-Directories
3.9. Using the Exabyte Devices
4. Doing More with SunView Windows
4.1. Pop-up Menus
4.2. Specific Tools
4.3. Accelerators
4.4. SunView Default Tool Files
5. Other UNIX Features
5.1. man
5.2 enscript
5.3. TEX and LATEX
5.4. mongo
5.5. visual (vi) screen editor and emacs
5.6. mail
5.7. rlogin, telnet, rcp and ftp
6. IRAF Examples
6.1. rfits and rcamera
6.2. wfits
6.3. ctio
6.4. ccdred
7. Documentation
8. Games
Appendix A: System Information
Appendix B: Default Initialization Files
Lisa Wells
Mario Hamuy
April 30, 1993
A number of CTIO-specific calibration files are kept in the directories ccddb and onedstds. In order to access these subdirectories you must load first the noao package. The examples incorporate them where necessary however, if you plan to do data reductions at a different institution, be aware that these files may not exist there.
The subdirectory ccddb$ctio/
This is home for the translation files for those CCDs which have been studied and also the bad pixel files. The possible files currently available are:
ccd.dat
ech.dat
cfccd.dat
csccd.dat
fpccd.dat
epi5.dat
epi5_badpix.dat
instruments.men
The subdirectory linelists$
This contains the general CTIO site files for wavelength calibrations.
README
cuar.dat lowskylines.dat
vacidhenear.dat argon.dat
henear.dat ohlines.dat
vacthorium.dat ctiohear.dat
idhenear.dat skylines.dat
xenon.dat ctiohenear.dat
krypton.dat thorium.dat
If you require tables of emission lines for HeAr, or HeNeAr, these are available in the manuals located in each computer room and at the telescopes but they cannot be duplicated for online use. The Thorium-Argon lamp tables are available only in the computer rooms on Tololo and downtown. These are separate sources:
1. A CCD Atlas of Comparison Spectra: Thorium-Argon Hallow Cathode 3180 to 9540 Angstroms, by Daryl Willmarth of the Instrument Support Group, KPNO, 1987.
2. A Pictorial Atlas of the Emission Line Spectrum of a Thorium-Argon Hollow Cathode Lamp, by F. H. Chaffee and J. R. Peters, FLWO, 1983.
The subdirectory onedstds$
This contains the general CTIO site files for extinction corrections, and flux calibrations. It also includes files for the 2dfrutti multihole spacings and neutral density filter transmission files for the 1-m and 4-m telescopes.
Extinction Coefficients:
ctioextinct.dat -- magnitudes/airmass vs wavelength
The subdirectory onedstds$ctio/ contains:
Multi-hole Spacings in seconds of arc(theoretical):
multi1m.dat -- 1.0 meter spectrograph multi4m.dat -- 4.0 meter RC spectrograph multi4mech.dat -- 4.0 meter echelle spectrograph
ND Filter Transmission Curves:
nd1m.100mag.dat -- 1.0 meter "1-mag" filter nd1m.125mag.dat -- 1.0 meter "1.25-mag" filter nd1m.250mag.dat -- 1.0 meter "2.5-mag" filter nd1m.500mag.dat -- 1.0 meter "5-mag" filter nd4m.l000mag.dat -- 4.0 meter "0-mag" lower filter nd4m.l250mag.dat -- 4.0 meter "2.5-mag" lower filter nd4m.rc500mag.dat -- 4.0 meter "5-mag" for Richey-Chretien Spectrograph nd4m.u000mag.dat -- 4.0 meter "0-mag" upper filter nd4m.u025mag.dat -- 4.0 meter "0.25-mag" upper filter nd4m.u075mag.dat -- 4.0 meter "0.75-mag" upper filter nd4m.u150mag.dat -- 4.0 meter "1.5-mag" upper filter nd4m.u225mag.dat -- 4.0 meter "2.25-mag" upper filter ndfilters.men
The subdirectory onedstds$bstdscal/ contains the Hayes flux standards:
Directory of the brighter KPNO IRS standards at 29 bandpasses, data from various sources transformed to the Hayes and Latham system, unpublished.
hr3454.dat hr4534.dat hr7001.dat hr7950.dat
hr3982.dat hr5191.dat hr718.dat hr8634.dat
hr4468.dat hr5511.dat hr7596.dat hr9087.dat
standards.men
The subdirectory onedstds$spec16bluecal/ contains the Hayes flux standards in the range 3300-7550A:
Directory with fluxes at 16A steps (suitable for high-dispersion spectroscopy) for the bright secondary standards, published in Hamuy et al., 1992, PASP, 104, 533.
hr1544.dat hr4963.dat hr7596.dat hr9087.dat
hr3454.dat hr5501.dat hr7950.dat standards.men
hr4468.dat hr718.dat hr8634.dat
The subdirectory onedstds$spec16redcal/ contains the Hayes flux standards in the range 6050-10300 A:
Directory with fluxes at 16A steps (suitable for high-dispersion spectroscopy) for the bright secondary standards. These values were obtained in the spectrophotometric system described by Hamuy et al, 1992, PASP, 104, 533, and are accurate to better than 2%, except in the range 8400-9850A where the accuracy drops to 5% due to the lack of flux points in the calibrating spectrophotometric system. Work is underway to illiminate this problem at which time these data will be superseded.
hr1544.dat hr4963.dat hr7596.dat hr9087.dat
hr3454.dat hr5501.dat hr7950.dat standards.men
hr4468.dat hr718.dat hr8634.dat
The subdirectory onedstds$ctiocal/ contains the Stone/Baldwin standards:
Directory with fluxes in the range 3200-10400 A for the southern tertiary standards as published by Baldwin & Stone, 1984, MNRAS, 206, 241 and Stone and Baldwin, 1983, MNRAS, 204, 347.
bd25.dat f56.dat l1020.dat l4816.dat
l9491.dat bd8.dat f98.dat l1788.dat
l6248.dat l97030.dat cd32.dat g9937.dat
l2415.dat l7379.dat lds235.dat eg21.dat
h600.dat l2511.dat l74546.dat lds749.dat
eg274.dat hz15.dat l3218.dat l7987.dat
ros627.dat f110.dat hz2.dat l377.dat
l8702.dat standards.men f15.dat hz4.dat
l3864.dat l9239.dat w1346.dat f25.dat
kopf27.dat l4364.dat l93080.dat w485a.dat
The subdirectory onedstds$ctionewcal/ contains the Stone/Baldwin standards:
Fluxes at 50A steps in the range 3300-7550 A for the tertiary standards of Baldwin and Stone derived from the revised calibration of Hamuy et al., 1992, PASP, 104, 533. This new calibration represents a significant improvement over the original and the fluxes in this directory are in general to be preferred over those in ctiocal.
cd32.dat f56.dat l2415.dat l4364.dat
l7987.dat eg21.dat h600.dat l3218.dat
l4816.dat l9239.dat eg274.dat l1020.dat
l377.dat l6248.dat l9491.dat f110.dat
l1788.dat l3864.dat l7379.dat standards.men
The subdirectory onedstds$spec50cal/ contains:
The KPNO spectrophotometric standards at 50 A intervals. The data are from (1) Table V, Spectrophotometric Standards, Massey et al., 1988, ApJ 328, p. 315 and (2) Table 3, The Kitt Peak Spectrophotometric Standards: Extension to 1 micron, Massey and Gronwall, 1990, ApJ 358, p. 344.
bd284211.dat feige67.dat hz14.dat pg0846249.dat
standards.men cygob2no9.dat g191b2b.dat hz44.dat
pg0934554.dat wolf1346.dat eg81.dat gd140.dat
pg0205134.dat pg0939262.dat feige110.dat hd192281.dat
pg0216032.dat pg1121145.dat feige34.dat hd217086.dat
pg0310149.dat pg1545035.dat feige66.dat hilt600.dat
pg0823546.dat pg1708602.dat
The subdirectory onedstds$iidscal/ contains:
The KPNO IIDS standards at 29 bandpasses, data from various sources transformed to the Hayes and Latham system, unpublished.
40erib.dat feige56.dat he3.dat kopff27.dat
ross627.dat bd253941.dat feige92.dat hiltner102.dat
l13633.dat ross640.dat bd284211.dat feige98.dat
hiltner600.dat l151234b.dat sa29130.dat bd332642.dat
g191b2b.dat hz14.dat l74546a.dat standards.men
bd404032.dat g4718.dat hz15.dat l8702.dat
ton573.dat bd82015.dat g9937.dat hz2.dat
l93080.dat wolf1346.dat feige110.dat gd140.dat
hz29.dat l97030.dat wolf485a.dat feige15.dat
gd190.dat hz4.dat lb1240.dat feige24.dat
grw705824.dat hz43.dat lb227.dat feige25.dat
grw708247.dat hz44.dat lds235b.dat feige34.dat
grw738031.dat hz7.dat lds749b.dat
The subdirectory onedstds$irscal/ contains:
The KPNO IRS standards at 78 bandpasses, data from various sources transformed to the Hayes and Latham system, unpublished (note that in this directory the brighter standards have no values - the `bstdscal' directory must be used for these standards at this time).
bd082015.dat feige110.dat g191b2b.dat hd217086.dat
hiltner102.dat bd174708.dat feige15.dat hd109995.dat
hd2857.dat hiltner600.dat bd253941.dat feige25.dat
hd117880.dat hd60778.dat hr7001.dat bd262606.dat
feige34.dat hd161817.dat hd74721.dat hz44.dat
bd284211.dat feige56.dat hd17520.dat hd84937.dat
kopff27.dat bd332642.dat feige92.dat hd192281.dat
hd86986.dat standards.men bd404032.dat feige98.dat
hd19445.dat he3.dat wolf1346.dat
The subdirectory onedstds$redcal/ contains:
Standard stars with flux data beyond 8370A. These stars are from the IRS or the IIDS directory but have data extending as far out into the red as the literature permits. Data from various sources.
40erib.dat gd140.dat he3.dat l93080.dat
standards.men bd174708.dat gd190.dat hz29.dat
l97030.dat wolf1346.dat bd262606.dat grw705824.dat
hz43.dat lds235b.dat wolf485a.dat feige24.dat
grw708247.dat hz44.dat lds749b.dat g191b2b.dat
grw738031.dat l13633.dat ross627.dat g4718.dat
hd19445.dat l151234b.dat ross640.dat g9937.dat
hd84937.dat l74546a.dat sa29130.dat
The subdirectory onedstds$spechayescal/ contains:
The KPNO spectrophotometric standards at the Hayes flux points, Table IV, Spectrophotometric Standards, Massey et al., 1988, ApJ 328, p. 315.
bd284211.dat feige67.dat hz14.dat pg0846249.dat
standards.men cygob2no9.dat g191b2b.dat hz44.dat
pg0934554.dat wolf1346.dat eg81.dat gd140.dat
pg0205134.dat pg0939262.dat feige110.dat hd192281.dat
pg0216032.dat pg1121145.dat feige34.dat hd217086.dat
pg0310149.dat pg1545035.dat feige66.dat hiltner600.dat
pg0823546.dat pg1708602.dat
Any other files that might be needed for data reduction can be obtained from Mario Hamuy (ext. 210) downtown, or Mauricio Navarette or Nelson Saavedra on Tololo (ext. 422).
It is often the case when you observe with CSCCD that the flats obtained from the white spot do not provide a sufficient illumination of the CCD throughout the spectral range. This is generally the case when you observe far in the ultraviolet (3100-4000A) or in the far red (1-1.1u). To get around this problem, the observer is forced to take a series of flats with the (projector) lamp located inside the spectrograph which provides a bright source of light. Given the brightness of the projector lamp, it becomes necessary to add in front of the light source a combination of filters in order to avoid the saturation of the CCD where the response of the detector is higher. Depending on your specific setup (grating, CCD, telescope) you have to choose the combination of filters that provides illumination of the CCD where the white-spot doesn't. Sometimes more that one combination of filters is required.
The goal of this appendix is to show you how to merge your dome flat with your projector flat(s) in order to produce a final image intended to remove the sensitivity variations of the CCD from your program frames. We assume in this appendix that your projector-flat and dome-flat have been properly processed through the point of overscan-subtraction, trimming and bias-subtraction. We assume as well that scattered light was subtracted from your flats. All these procedures are explained in the CSCCD manual.
The example of this section is relatively simple. We used a low dispersion grating on the 1.5-m telescope with a GEC CCD, with a resulting wavelength coverage of 3000-7700A. In this setup the white spot flat (hereafter dome-flat) provides sufficient illumination of the detector redward of 4000A. For the shortest wavelengths instead, we took a series of flats with the projector lamp together with the CuSO4 and Corning 9863 filters (hereafter projector-flat). This combination of filters provided enough illumination in the UV end and a sharp cutoff around 4500A.
Figure 1 shows the dome-flat with the evident lack of UV illumination at the bottom. On the other hand, Figure 2 shows the projector-flat with a strong illumination at the UV end. You can compare the two flats with the implot task.
cl> implot pflat
:c 180 (plot one of the central columns)
:i dflat
o (overplot command)
:c 180 (overplot the same column of dflat)
q (quit the task)
Figure 3a shows one of the central columns of the dome flat overplotted with the same column of the projector flat. It is evident that over the first 200 lines the projector flat predominates over the dome-flat. Beyond that line the dome-flat plays the most important role. Write down the intersection line of the two curves (line 200 in this example). Now, plot this specific line of the projector-flat and overplot the same line of the dome-flat.
cl> implot pflat
:l 200 (plot the desired line)
:i dflat
o (overplot command)
:l 200 (overplot the same line of dflat)
q (quit the task)
In this example (Figure 3b) a strong difference in the illumination of the slit is revealed from the two images. Later, we will remove the slit profile of the projector-flat before merging it with the dome-flat. At the moment you must determine from this plot the range of useful columns to be used in the merging procedure (columns 135-270 in this example). When making this decision you must include in your working sample only those columns which are completely contained in the decker limits; exclude any columns which are partially illuminated.
Basically, the merging procedure consists in removing the spectral signature of the white-spot and projector light (STEP 1), as well as removing the slit profile of the projector flat (STEP 2). Once these goals are achieved you will be able to get a final image with a smooth transition between the dome-flat and the projector-flat (STEP 3).
STEP 1: Normalizing the flats along the dispersion
We must start by removing the spectral signature of the white spot and the projector light. First, you must decide upon the area to be normalized in each image. For the dome-flat we will carry out the normalization between lines 190-572 and columns 135-270. For the projector-flat the normalization will be done between lines 1-210 and the same columns used for the dome-flat . Note that we have purposedly left an overlap between both images around the intersection line . That area is reserved in order to remove any difference in the slit illumination of both images. Figure 4 summarizes the five areas in which we have divided our frame.
Area 1: section [135:270,190:572] corresponds to the useful area of the dome-flat
Area 2: section [135:270,1:210] corresponds to the useful area of the projector-flat
Area 3: section [135:270,190:210] corresponds to the overlap area between both flats
Areas 4 and 5: sections [1:134,1:572] and [271:378,1:572] correspond to the columns that are excluded from the merging procedure
Load the packages twodspec and longslit and edit the parameters for the task response.
cl> twodspec
tw> longslit
lo> epar response (check the parameters given below)
\fIresponse task\fR
calibration = Longslit calibration images
normalizatio = Normalization spectrum images
response = Response function images
(interactive = yes) Fit normalization spectrum interactively?
(threshold = INDEF) Response threshold
(sample = "*") Sample of points to use in fit
(naverage = 1) Number of points in sample averaging
(function = "spline3") Fitting function
(order = 5) Order of fitting function
(low_reject = 3.) Low rejection in sigma of fit
(high_reject = 3.) High rejection in sigma of fit
(niterate = 1) Number of rejection iterations
(grow = 0.) Rejection growing radius
(graphics = "stdgraph") Graphics output device
(cursor = "") Graphics cursor input
(mode = "ql")
Now use the task response to normalize the dome-flat.
lo> response dflat[135:270,190:572] dflat[135:270,190:572] ndflat
response presents you with a plot of the averaged flat field columns [135-270] with a function fit through it (see Fig. 5a). If the curve does not fit the gross shape of the flat field, try different orders (type :order 10) Usually an order around 10 gives a satisfactory fit. Since this is a spline fit on equal intervals, a small change in the order (i.e. number of intervals) can have a relatively large effect. You may reperform the fit with the f key. A useful command is the k key which will present you with the ratio between the sample and the fit (see Fig. 5b). From this plot you can better judge the quality of the fit. You should end up with an order such that the ratio of the sample and the fit is one on the average, plus minus 5% due to the pixel-to-pixel sensitivity variations of the chip. You may type h to go back to the previous plot of the sample and the fit on top of it. When satisfied, exit with q. The output of the task is an image named ndflat obtained from dividing each of the specified columns of dflat by the fitted function. All of the pixels that lie outside the normalization area are set to one by response.
You may display or implot this image (see Figure 6).
lo> display ndflat 1 zs- zr- z1=0.95 z2=1.05
lo> implot ndflat
Now, you have to normalize the projector flat over area 2.
lo> response pflat[135:270,1:210] pflat[135:270,1:210] temp1
Figure 7a shows the spectral shape of the projector-lamp plotted together with a cubic-spline fit. Figure 7b shows the ratio between the fit and the sample which you can get with the k key. In this example we have removed from the sample a few lines that reveal a lower sensitivity of the CCD. You may do this with the help of the vertical cursor and pairs of s to define as many samples as you wish. In this case we fit a cubic-spline of order 20 to remove the pronounced spectral signature introduced by the color filters. The output image is a temporary image called temp1 which you may display it or plot it.
lo> display temp1 1
lo> implot temp1
STEP 2: Removing the illumination profile along the slit
So far, we have removed the spectral signature of both flats along the dispersion axis. On the other hand, the illumination along the slit has not been modified. In Figure 3b it is evident that both flats have extremely different illuminations, which has to be modified before attempting to merge them. In the next paragraphs we will remove the illumination of the projector-flat along the slit, with respect to the dome-flat . For this purpose it is crucial to have a few lines of overlap between the normalized images. In this example the overlap area ranges between lines 190-210.
Start by dividing the normalized projector-flat by the normalized dome-flat.
lo> imarith temp1 / ndflat ratio
This operation removes all of the pixel-to-pixel variations in the overlap area of the projector-flat. We use now the task illumination to fit the relative illumination left along the slit in the image ratio. Do an epar on the illumination task according to the parameters given below.
lo> epar illumination (check the parameters given below)
\fIillumination task\fR
images = Longslit calibration images
illumination = Illumination function images
(interactive = yes) Interactive illumination fitting?
(bins = "") Dispersion bins
(nbins = 1) Number of dispersion bins when bins = ""
(sample = "*") Sample of points to use in fit
(naverage = 1) Number of points in sample averaging
(function = "spline3") Fitting function
(order = 1) Order of fitting function
(low_reject = 3.) Low rejection in sigma of fit
(high_reject = 3.) High rejection in sigma of fit
(niterate = 1) Number of rejection iterations
(grow = 0.) Rejection growing radius
(interpolator = "poly3") Interpolation type
(graphics = "stdgraph") Graphics output device
(cursor = "") Graphics cursor input
(mode = "ql")
Now execute the task illumination on the image ratio.
lo> illumination ratio[135:270,*] illum
The parameters (135:270) must be the same employed in the response task. They correspond to the range of columns to be used in the determination of the illumination profile along the slit. These task is divided in two sections. In the first section, illumination presents you with the average of columns 135-270 (see Figure 8a). The horizontal bar represents the lines that will be used in the next section to determine the illumination of the slit. Now, you must restrict this sample only to the overlap area . To perform this operation, start by getting rid of the current sample (type i) and then redefine the new sample (type :bin 190:210) to match the overlap lines. A small bar should show up now at the bottom of the plot with the new sample. Once you perform this operation type q to quit this section.
In the second section you will be presented with a plot which corresponds to the average of the lines 190-210 (Figure 8b). This is the illumination profile of the projector-flat relative to the dome-flat. Here, you must fit a cubic-spline fit to this sample. To better judge the quality of the fit use the k key to look at the ratio between the sample and the fit (see Figure 8c). If you desire to increase the order of the fit type for instance :order 10 and reperform the fit (type f). Use an order that yields residuals no larger than 1%. When satisfied type q to quit this task. The result of illumination is called illum and contains the fit of the slit profile properly spread into a 2-dimensional image. You can examine the output image (Figure 9).
lo> display illum 1
Now, you have to divide temp1 by illum in order to remove the illumination profile along the slit from the projector-flat.
lo> imarith temp1 / illum temp2
The result of the previous operation is a temporary image called temp2. As a consequence of dividing the projector-flat by the slit profile, the resulting image has to be renormalized along the dispersion in order to match the same level of counts of the dome-flat. We use response to perform this operation over the same section used before for the projector-flat.
lo> response temp2[135:270,1:210] temp2[135:270,1:210] npflat
You will be presented with the average of the columns 135-270 with a fit on top of it (Figure 10a). Since this image was already normalized, the number of counts should be close to one, except for a small scaling factor introduced during the illumination correction. Choose a low-order fit in order to fit your sample. In this example a cubic-spline order 1 works well. Try k to look at the ratio between the sample and the fit (Figure 10b) and quit the task with q. Remember that everything outside the normalizing section is set to one by this task. You may display your output image called npflat (see Figure 11).
lo> display npflat 1 zs- zr- z1=0.95 z2=1.05
You should be ready now to merge ndflat and npflat. To make sure that the final image is not going to show discontinuities at the overlap region, we suggest to implot 'ndflat' together with npflat.
lo> implot ndflat
:l 190 210 (plot the overlap lines)
:i npflat
o (overplot command)
:l 190 210 (overplot the same lines)
q (quit)
If everything goes well you should get an almost perfect match between both images. If that is the case we will proceed to the merging itself.
STEP 3: Merging the flats
In this section we will sum both flats. Before doing this operation we will perform some pixel replacement with the imreplace task in the proto package. We have to start by replacing with 0 the pixels that were set to 1 by response. The columns must be the working columns originally chosen to perform the normalization, i.e., 135-270. For the projector-flat the lines to be replaced are the ones excluded from the normalization, i.e., 211-572. Analogously, in the dome-flat we have to replace columns 1-189.
lo> proto
pr> imreplace npflat[135:270,211:572] 0.
pr> imreplace ndflat[135:270,1:189] 0.
Now, you may sum both flats.
pr> imarith ndflat + npflat temp3
The output of the previous operation is a temporary image called temp3 which is not your final flat yet. This image has a step at the overlap area which has to be removed. In order to remove this step we are going to make a mask image which we will then multiply by temp3. Let's start by producing the mask by dividing any of your images by itself.
pr> imarith temp3 / temp3 mask
Let's check that mask is really filled with 1 everywhere.
pr> imstat mask
Let's now fill the overlap area 3 with 0.5.
pr> imrep mask[135:270,190:210] 0.5
Also we have to replace the pixels in areas 4 and 5 with 0.5.
pr> imrep mask[1:134,1:572] 0.5
pr> imrep mask[271:378,1:572] 0.5
Finally, let's multiply temp3 by mask to get the final flat.
pr> imarith temp3 * mask flat
You may display now your final flat (see Figure 12).
pr> display flat 1 zs- zr- z1=0.95 z2=1.05
This image should have areas 4 and 5 set to one. The area between columns 135-270 contain the pixel-to-pixel variations. The pixel values here are expected to be between 0.9 and 1.1.
Mario Hamuy - 30 April 1993
MAIL AND ITS OPTIONS
There are many commands that could be useful when using the UNIX mailer. This is to give a list of the commands available but does not include all the possibilities. If you would like more information, type !man mail. When sending a file and you would like to specify a subject, use the following:
mail -s "type subject here" address < filename
This will send the file "filename" to "address" with the specified subject. If you send a message larger than 20,000 bytes, it will be rebounded. To send a large file see Jim Hughes. Also, if the message is a binary file, it may be necessary to encode it before sending it, otherwise, bits will be lost along the way. If you receive mail, just typing mail will enter the mailer and give a listing of the message headers addressed to you currently in the system mailbox. If there are no messages waiting for you, you will get the message back, "No mail for v4". If you wish to suppress the message headers from being printed, use the option "-N". If you are in the middle of something else and just wish to see who has sent you mail, use the "-H" option. This will simply give the header summary only for the messages which are waiting for you. If you have logged on and missed the message telling you about mail, use the "-e" option to check for the presence of mail. Finally, if you wish to reread a message from your mbox file or another file, use the "-f filename" option. If the filename is not given, the default for this is the standard mbox file created when quitting mail (see below). The command mode is entered when a "&" prompt appears. Then you may use any of the interactive commands, or type help for a list (see the third section).
-N Suppress the header summary upon entering mail
-H List the header only of the unread mail
-e Check to see if there is mail
-f filename Read the messages in the file specified
TILDE COMMANDS WHILE WRITING A LETTER
When sending a message to someone, an example is shown below, you will then be prompted for the subject, just type it in and hit return. Now just type your message into the buffer that has been created for your message. When you are finished, just type a . on a blank line and type return to quit and send the letter. The mailer will echo "EOF" to specify the end of the file. There is a tilde command that works the same (see below).
v2> mail marsden@cfa.bitnet
Subject: A NEW SUPERNOVA!!!
We have found a new supernova in an anonymous galaxy located
at RA 05 23 45, Dec -69 46.3 (1950), at a visual magnitude of
9.31. You can't miss it!
Dr Magellan
.
EOF
In composition mode, you may use any of the many tilde commands. They must be used by typing the command on a blank line, and hitting return. They will not be echoed in your message. Any of the following commands may be used.
~? Prints a summary of the tilde commands
~b users Adds the users specified to the 'Bcc:'
(Blind carbon copy) list separated by
commas
~c users Adds the users specified to the 'Cc:'
(Carbon copy) list separated by commas
~d Read the dead.letter file into the buffer
~e Edit the message buffer using 'ed' editor
~f [messagelist] Forward the message specified, will be
prompted for the 'To:' list
~h Show the 'To:', 'Subject:', Cc:', 'Bcc:'
lists and add to or change them using
backspace and typing the new value
~m [messagelist] Read the message specified into the buffer
~p Print the message being composed in the
buffer as it is now
~q,~Q,^c Quit, and save the message in the home
directory under the name dead.letter
~r filename Read the file into the message buffer
~s subject Set the 'Subject:' to subject, or use
this to change the original
~t users Add users to the 'To:' list
~v Edit the message using the 'vi' editor,
using ":wq" to exit the editor will bring
you back to the current mode
~w filename Write the message into the file specified
~x Quit, do not save the file
~. End of input, send message
~! command Run a shell (UNIX) command
~| command Pipe the message through the command
~: command Execute a regular mail command, type
'~:help' for a list of the commands
The "users" is a list of one or many addresses. The "messagelist" is explained in the last section of this help page, but in brief, specifies the message number to be used. The "filename" is the file to be used or created. "Subject" is the new subject to be associated with this mail message. The UNIX or mailer commands are used in place of "command". In all cases, the mailer will return to the message input mode unless you use an exit command.
USING THE MAIL COMMANDS
When entering the mailer, the prompt becomes an "&" signifying the command mode. At this prompt you may type any number of commands. To read a mail message, just type the number of the message you wish to read. If you read a message and want to reply, just type r. You may specify the message number if you are not reading the file to which you want to reply. You can print the specified message by using the p command.
help or ? Prints a summary of the commands
<return> Print the next message
cd directory Change directory to one specified
copy messagelist filename Copy messages to file but don't
mark as saved
delete messagelist Delete the messages specified
edit messagelist Edit the message buffer using "ed"
editor
exit Leave the mailer, don't move
messages to 'mbox'
file filename Quit the current mail file and
read in the named file, will give
number of files within filename
from messagelist Print the header 'from:' the
message specified
headers Print out headers of the messages
being accessed
list Lists all commands available
without description
mail user-list Mail a message to the specified
user, will go to 'composition'
mode
next Go to and print the next message
in the list
print messagelist Print the message specified
preserve messagelist Put the messages specified back
in the system mailbox
quit Quit, and save the messages in
the home directory
under the name mbox
reply messagelist Reply to sender of message only
Reply messagelist Reply to sender and all
recipients of message
save messagelist filename Append the message to the file
specified
size messagelist Print the size in characters of
files specified
type messagelist Type messages specified, same as
print
top messagelist Show the top lines of the
message specified
undelete messagelist Undelete the message(s) specified
visual messagelist Edit the message using the "vi"
editor, using ':wq' to exit the
editor will bring you back to the
current mode
write messagelist filename
Append the message to the file
specified
xit Quit, do not change the system
mailbox
z [-] Display next [previous] page of
headers
! Shell (UNIX) escape
= Print the current message number
The user-list is the valid address(es) for the sending of the mail message using commas to separate multiple addresses. The file specified a file to be created in your directory. The messagelist given in these commands can be specified with any of the following;
. The current message
n Message number n
^ The first undeleted message
$ The last message
+ The next undeleted message
- The previous undeleted message
* All messages
n-m An inclusive range of message numbers
user All messages from user
/string All messages with 'string' in the subject line
(case ignored)
:c All messages of type 'c', where 'c' is one of:
d deleted messages
n new messages
o old messages
r read messages
u unread messages
One known bug, reply does not always generate the correct return address, so you may need to specifically input the correct address. An E-mail Directory is available to check for the correct address. See Mario Hamuy, or Jim Hughes for more information.
All the computers used at CTIO are linked together. These commands are all aliased as tasks in IRAF also so you need not be in the UNIX environment. If you are logged onto one machine you can easily login to any other. Telnet connects all the computers used at CTIO. If you want to login to the "VAX1" machine located downtown, use telnet to connect to it from another machine, ie:
mach% telnet ctiov1
You must then login in the usual manner by typing in your username and password when prompted. There are two VAXs which may be accessed this way, ctiov1 and ctiov2. This will also work if you are logged in and using IRAF. So if you are logged onto a VAX and would like to connect to a SUN, just use telnet. You must then be sure to turn off the caps lock when returning to a SUN, or the UNIX systems will be thrown into a strange mode. This can be gotten out of by logging out and then logging back in using lower case. The stateside computers are accessible using telnet also. The machine internet number can be used in place of "ctiov1".
mach% telnet 139.229.2.5
We ask that you not send images or big files over the link to the US unless you know what you are doing. Big files will tie up the network so be courteous. All the SUNs are linked via rlogin. NOTE THAT rlogin IS BETTER THAN telnet IF YOU ARE ONLY USING A REMOTE LOGIN TO ANOTHER SUN . If you are on one SUN and would like to login to get onto "SUN1" ie, ctios1 in La Serena then all you need do is type:
mach% rlogin ctios1
This will log you onto the "SUN1" machine in La Serena from the machine you are currently logged onto. The possible choices are:
\fBMountain\fR \fBLa Serena\fR
ctio4m ctios1
ctio1m ctios2
ctio60 ctios3
ctio36 ctios4
Rlogin is usable in IRAF, however if you are logged onto the VAX you can only get to the other machines by using telnet only. If you have any problems, contact Mario Hamuy, or Jim Hughes.
There are many PCs being used at CTIO to log onto the mainframe computers. This is done by running an emulator program on the pc which makes it more or less, simulate a vt100+retrographics terminal. If the pc is off, turning it on will boot the disk. If the pc is on but nothing is setup, ie; the prompt looks like "A>", then you can run the emulator by typing vt or em. Most of the PCs are equipped with hard disks. For those that are not, check to see that the correct floppy is installed. If this doesn't work, boot the disk again by typing Alt/Shift/Delete. Press all three keys at the same time. Then type vt again. Once the PC runs the emulator, type return several times. On the mountain each PC is connected directly to a specific SUN computer; you should get the normal login prompt for that computer. In La Serena the PCs are connected to a "micom" terminal switcher, which allows them to be used to log on to any of the La Serena computers. You will then be prompted for the number of the computer you wish to log onto, so type the number and hit return. You may now log onto the computer at the prompt.
If you are logged in and using IRAF from a PC, it is a good idea to set the terminal to the proper type. This helps when you are using the help facility for example by setting the screen size to the proper format. Do this by typing;
cl> stty pc640
This will properly set the terminal type.
There is a difference between text and graphics screens and there are two commands which link the two. In the vt100 screen you press Alt g to get into graphics mode. To get back to the text screen, Alt v. In graphics mode, for terminals which have only one arrow keypad, the cursor movement is in small steps, but using shift and the arrow gives larger steps. For terminals which have two arrow keypads, use the cursor pad key. The numlock key will light up as well. This uses the far right keypad for small increments and the left one for large increments in the cursor movement.
When you are finished first log off the mainframe computer in the normal way. In La Serena you must then type Alt/Shift b (type all three at once) to break the connection between the micom and the host computer. Finally, you can type Alt x to exit from the emulator program. The Ctrl/Shift key toggles back and forth between using the terminal emulator to talk to the mainframe and talking directly to the PC. You can safely switch back and forth without logging out, and can even leave a program running on the sun while you play chess on the pc.
To use the suntools windows, you must be logged in at a console. The windows are activated by a call of the suntools. This will usually be done automatically when you log in to the console to use iraf. If they do not come up, type "suntools". The screen will change to a grey background with windows. If the windows are not created properly, it means that there is not enough block memory for the window to use. This will result in an error message appearing somewhere on the console. This means that there are too many people using the computer, so try again later, or see Mario Hamuy.
Now that the windows are up, the mouse can be used to manipulate the windows. If you are using one of the windows, you must move the mouse cursor into that window. If the window you wish to use is covered by another window then move the cursor to the border of the window you want, and press the left button on the mouse. Your window will come to the front. If you wish to move the windows, move the cursor to the border of the window you wish to move. Now hold the middle button and move the window. A frame will appear showing the outline of the window. Releasing the button will set the position and will move the window to its new place.
If the console gets too cluttered and you wish to get rid of a few windows there are two ways of doing it. For windows you plan to use again it is easiest to just close the window. Take the cursor to the border and hold the right button. Move the cursor to frame and follow the arrow off the right edge. A new menu will appear so move to close and release the button. The window will change to a small rectangle. To bring back the window just move the cursor into the rectangle and press the left button. If you no longer need the window you can just quit out of it by moving the cursor to quit and releasing.
If you get error messages and the window borders are messed up, move the cursor to the grey region behind the windows and hold the right button. Move the cursor to redisplay all and release. To bring up more windows, again move to the grey background and hold the right button. Move the cursor to imtool or gterm for the type of window you wish and release.
When you are done for the day and wish to log out, move the cursor to the grey background area anywhere and hold the right button. DO NOT TRY TO LOGOUT FROM THE CONSOLE WINDOW OR ANY OF THE WINDOWS . Move the cursor to exit suntools and release. You will return to the blank screen and be logged off the computer. If you are not logged out immediately then type logout. This will log you off the computer. Note if you have IRAF background jobs running when you "exit suntools" they will continue to run until completed.
The imtool window is used to display images. This task can do many things. The "tv" package contains all the tasks which manipulate the imtool window. First load the tv package by typing:
cl> noao
no> tv
tv>
The tv> prompt means that the tv package has been loaded. Now display one of your images using display:
tv> display obj003 1
The "1" at the end of the command is the frame number in which you would like your image to be displayed. The image specified is now visible in the imtool window. Next press the F6 key. You will see coordinates appear at the lower right of the screen. Now when you move the cursor, the coordinates keep track of where you are. It is a good idea to check the coordinates by moving the cursor to the lower left corner of the image to look for the origin (1,1). If one of the corners doesn't give the origin coordinates then the imtool is not set up properly and you should ask Mario Hamuy for assistance.
If you wish to create a file with object positions from this image, the frame number at the top of the screen is important since you will begin writing the coordinates of your objects to a file in the uparm subdirectory by that title, i.e., frame.1.1. Now move the cursor to your object and press the left button on the mouse. A number will appear up and to the right of your object to indicate your choice and the coordinates are written to the file called frame.1.1 in your uparm directory. The cursor position is important since this does not use a centering algorithm.
If you would like a color display, move the cursor to the border of the window, press and hold the right button down. A menu will appear with many choices. Move the cursor to "Setup" and release the button. A smaller window opens with the setup parameters. Move the cursor into the window and point it at "Mono". Pressing the left button on the mouse will cycle through the display modes. If you would like more information about using the mouse and Imtool window, see appendix K.
There are several ways to recall or change a previously typed command in IRAF. The "eparam" and "ehistory" commands have two such modes; an immediate execution and editing. A summary of these shortcuts and their function follows. The "^" is for immediate execution and "ehistory" edits the command line recalled.
cl> ^ Rerun the previous command
cl> ^nite1^nite2^ Rerun the previous command but
substituting 'nite2' for the first
appearance of 'nite1'
cl> ^nite1^nite2^g Same as above but substituting
all occurrences of 'nite1' with
'nite2'
cl> ^rfits Rerun the last 'rfits' command
cl> ^?FLAT Rerun the last command containing
'FLAT' anywhere
cl> history List the last few commands in the
history buffer
cl> history 50 List the last 50 commands in the
history buffer
cl> ^37 Rerun command number 37 from the
history list
cl> ^-2 Rerun the command before the
previous one
cl> task ^2 ^1 Run the 'task' with the first
and second arguments of the
previous command reversed
cl> task ^$ Run 'task' with the last argument
of the previous command
cl> task ^* Run 'task' with all the arguments
of the previous command
Now the "^" command may behave like the editing command if the cl parameter "ehinit" is set to "verify". If you wish to change this, edit the parameters of "cl". The UNIX "ehistory" is much like the command recall in VMS. Typing an e will present the last command in the history list. To scroll up through this list just type the up arrow. Now to change something on the line, move the cursor using the left arrow and delete and type in the change (this deletes to the left). If a task name is specified after the "ehistory" command it recalls the last command using that task. For example, typing e display will recall the last display command used with its previous parameters. You may now edit this command. Typing return, (no matter where the cursor is) will run the task.
The task "imedit" is found in the `tv` (images) package. This task gives statistics on a region of the image, or edits pixels by background surface fit, interpolation from background, replacing one region by another, or replacement by a constant value. The image is automatically displayed if the parameter "display" is set to "yes", and will be redisplayed after each alteration if "autodisplay" is set to "yes". This may be slow and so the option to redisplay exists in the interactive mode so that several changes may be performed and the image redisplayed. The task uses the current values of the display task in the tv package.
If you would like the surface plot after each editing command, set "autosurface" to "yes", the default is "no". The "aperture" of the region can be designated by a circle or rectangle with a size governed by the "radius" parameter. If you want to fit the background, the parameters "xorder" and "yorder" define the number of terms in the polynomial fit. The maximum or minimum values of the region being fixed may not be centered in the cursor, so the parameter "search" is the radius over which a search is performed to find the pixel with the most extreme value. Caution must be used here however, the most extreme value may not occur in the center of the region you are fixing. The parameter "value" is the number for constant substitution. The horizontal and vertical viewing angles for the surface plots are set by "angh" and "angv". It is possible to run this task in fixpix style, see the examples at the end of the help page.
All of the parameters above are in the setup file for this task, and may be changed while running the task interactively by typing ":" and the parameter name followed by the new value, for example :search 6. Typing just the ":" and the parameter name will give the current value for that parameter. To edit the task parameters from the interactive mode, just type ":epar". Exit this menu as any other using "^z". Running this task in interactive mode, a circular cursor appears in the imtool window. This is controlled by the mouse and is used to mark the regions to be edited. There are many other commands that can be used interactively.
While running this task interactively, the radius is changed using the + and - keys. The t key toggles between a search for the maximum or minimum valued pixel within the "search" radius. The "space bar" gives statistics on the region defined by the task parameters around the position of the cursor. The "p" key gives a box of pixel values and statistics. "u" undoes the most recent change performed. If you have made changes that you want to redo, "i" initializes the image without saving the changes so you may begin again. To display the image after performing a few operations (if autodisplay is off), use "r". For the surface graph, type "g". To quit and save the image, type "q". If you screw up, "Q" exits without saving the changes.
The editing options are broken into pairs for the rectangle or aperture (square or circle defined in the task parameters) region. The rectangular region is defined interactively using two cursor positions, a "again:" will appear in the gterm window waiting for the second position to be marked. The "a" and "b" commands perform a background fit to edit the rectangle or aperture region, respectively. "d" and "e" are the constant value substitution option. Typing "j" or "k" replaces the defined region by input data. Line or column interpolation are done using a rectangular region with the "c" and "l" keys. The "f" key will interpolate across the narrowest dimension of the region marked. The "u" key will undo the last operation performed on the image and will redisplay if "autodisplay" is "yes". To get the interactive help, type "?" while running the task. You must type a "space bar" to get the next page and "q" to get out of the help mode. The mouse buttons work in interactive mode to center and zoom (middle) and change the transfer function (right).
CTIO in conjunction with NASA, now has a satellite link to the US which is open to all users, for remote logins, copying files to or from home, and real time communications with collaborators in the US. In 3-5 years, remote observing may be possible. We encourage you to use the link and if you do, please report it in your visitor suggestions form before you leave CTIO. This document explains how to log onto your computer at home, how to copy files in either direction, and instructions for copying images to your home institution. It is also possible to have collaborators in the US checking images as they are taken.
1. Remote Login
You may log onto machines at your home institution using the machine name or the internet number assigned to the machine. If you first try the name and cannot connect, use "nslookup machinename.path", to get the internet number for your machine. The UNIX command to remotely log onto other machines is "telnet". The following example shows the possible forms:
cl> !nslookup bootes.unm.edu
Server: ctio.edu
Address: 139.229.1.2
Non-authoritative answer:
Name: bootes.unm.edu
Address: 129.24.8.2
cl> telnet bootes.unm.edu
Trying 129.24.8.2 ...
Connected to bootes.unm.edu.
Escape character is '^]'.
+---------------------------------------------------+
| University of New Mexico |
| Computer and Information Resources and Technology |
+---------------------------------------------------+
The BOOTES Local Area Vax Cluster
Local Node UNMB -- VAX 6000-320 Running VAX/VMS V5.4-2
Username:
You may now log onto your home machine, provided you have an account. Using the address instead of the name, "telnet 129.24.8.2", would give the same result. Typing in your username will result in a prompt for the password. You may now read mail, run programs, and edit files on your home machine. Logging off your home machine will kill the connection.
2. Copying Files
The UNIX command "ftp" is used to copy files between machines. If you have a file on your home computer that you wish to copy down to your visitor account, for example, your loginuser.cl file, do the following:
cl> !ftp 129.24.8.2
Connected to 129.24.8.2.
220 BOOTES.UNM.EDU MultiNet FTP Server Process 3.0(12) at Thu
10-Oct-91 3:01PM-MDT
Name (129.24.8.2:username): iraf
331 User name (iraf) ok. Password, please.
Password:
230 User IRAF logged into CPD:[USER.IRAF] at Thu 10-Oct-91 17:02,
job 5c6b.
ftp>
Since "ftp" is not normally defined as a foreign task in IRAF, you must use an exclamation point in front of the command. Now you must go the the directory where your file resides and use "get" to pull it from your home directory to the CTIO computer. At the same time we will send a file home called backup.
ftp> cd cpd:[user.iraf.workspace]
250 Connected to CDP:[USER.IRAF.WORKSPACE].
ftp> dir
200 Port 6.65 at Host 129.24.8.2 accepted.
150 List started.
CDP:[USER.IRAF.WORKSPACE]
LOGINUSER.CL;1 2034 2-AUG-1991 08:07 [AP$,IRAF] (RWED,RWED,RE,)
FIRST_TRY.FITS;1 2020 2-AUG-1991 08:08 [AP$,IRAF] (RWED,RWED,RE,)
VARS.DST;1 33 2-AUG-1991 08:06 [AP$,IRAF] (RWED,RWED,RE,)
Total of 4087 blocks in 3 files.
226 Transfer completed.
295 bytes received in 0.86 seconds (0.34 Kbytes/s)
ftp> get loginuser.cl
200 Port 6.67 at Host 139.229.2.60 accepted.
150 ASCII retrieve of CDP:[USER.IRAF.WORKSPACE]LOGINUSER.CL;$ started.
226 Transfer completed. 16896 (8) bytes transferred.
local: LOGINUSER.CL remote: LOGINUSER.CL
16896 bytes received in 6.6 seconds (2.5 Kbytes/s)
ftp> send backup
200 Port 6.76 at Host 139.229.2.60 accepted.
150 ASCII Store of CPD:[USER.IRAF.WORKSPACE]BACKUP.;1 started.
226 Transfer completed. 620 (8) bytes transferred.
local: backup remote: backup
620 bytes sent in 0.0057 seconds (1.1e+02 Kbytes/s)
ftp> quit
221 QUIT command received. Goodbye.
cl>
The file will be copied to the directory from which you ran "ftp". There is a help page available in UNIX by typing "man ftp" within or outside of IRAF.
3. Sending Images Home
If you have an image that you would like to get to a collaborator quickly, then you must first write the image onto disk using wfits. You should use the proper format. For a fits format disk image you must first type "binary". This will set the mode for transfer, so then type the "send" command. This works that same for getting an image from home to compare with one you have just taken, use "get imagename".
cl> !ftp 129.24.8.2
Connected to 129.24.8.2.
220 BOOTES.UNM.EDU MultiNet FTP Server Process 3.0(12) at Thu
10-Oct-91 3:01PM-MDT
Name (129.24.8.2:username): iraf
331 User name (iraf) ok. Password, please.
Password:
230 User IRAF logged into CPD:[USER.IRAF] at Thu 10-Oct-91 17:02,
job 5c6b.
ftp> binary
200 Type I ok.
ftp> send image.fits
200 Port 6.76 at Host 139.229.2.60 accepted.
150 BINARY Store of CPD:[USER.IRAF.WORKSPACE]IMAGE.FITS;1 started.
226 Transfer completed. 620 (8) bytes transferred.
local: image.fits remote: image.fits
620 bytes sent in 0.0057 seconds (1.1e+02 Kbytes/s)
If you wish to copy all of your images at the end of the night to your home computer, there are several more steps required. You should find some space on a disk to write all your nights images to fits format. Use the UNIX "tar" utility to write them all to one tar file, and use "compress" on this file. The following example shows the steps to first get the images copied to the home machine and then, what to do when you get home to retrieve the images.
cl> cd home$tardir
cl> wfits *.imh image
File 1: may12.08.imh -> image001 sn1991t B Size = 509 x 506
xtype=short bitpix=16 blkfac=fixed scaling=none
2 Header 358 Data logical (2880 byte) records written
File 2: may12.09.imh -> image002 sn1991t V Size = 509 x 506
xtype=short bitpix=16 blkfac=fixed scaling=none
2 Header 358 Data logical (2880 byte) records written
File 3: may12.10.imh -> image003 sn1991t R Size = 509 x 506
xtype=short bitpix=16 blkfac=fixed scaling=none
2 Header 358 Data logical (2880 byte) records written
File 4: may12.11.imh -> image004 sn1991t I Size = 509 x 506
xtype=short bitpix=16 blkfac=fixed scaling=none
2 Header 358 Data logical (2880 byte) records written
File 5: may12.07.imh -> image005 sn1991t U Size = 509 x 506
xtype=short bitpix=16 blkfac=fixed scaling=none
2 Header 358 Data logical (2880 byte) records written
cl> !tar -cvf images.tar image*
cl> !compress images.tar
cl> !ftp 129.24.8.2
Connected to 129.24.8.2.
220 BOOTES.UNM.EDU MultiNet FTP Server Process 3.0(12) at Thu
10-Oct-91 3:01PM-MDT
Name (129.24.8.2:username): iraf
331 User name (iraf) ok. Password, please.
Password:
230 User IRAF logged into CPD:[USER.IRAF] at Thu 10-Oct-91 17:02,
job 5c6b.
ftp> binary
200 Type I ok.
ftp> send images.tar.Z
200 Port 6.76 at Host 139.229.2.60 accepted.
150 BINARY Store of CPD:[USER.IRAF.WORKSPACE]IMAGE.TAR.Z;1 started.
226 Transfer completed. 14432(8) bytes transferred.
local: image.tar.Z remote: image.tar.Z
620 bytes sent in 0.0057 seconds (1.1e+02 Kbytes/s)
ftp> quit
221 QUIT command received. Goodbye.
cl>
The file is in the same form as the file on the machine here. When you return home, first you must use the UNIX "uncompress images.tar.Z". This assumes that your home machine runs UNIX. Then use "tar xvf images.tar". And finally you must get the images out of fits storage format by using "rfits image* raw raw oldiraf+".
4. Collaborations
We have had astronomers log in remotely from the states to look over the shoulder of their collaborator at the telescope and do data reduction while the other observed. This requires that IRAF host files be set to the proper path, especially for displaying images remotely at the US terminal. If you would like to have your collaborator looking over your shoulder while observing, you must talk with Steve Heathcote, sheathcote@noao.edu, so that the link may be set up BEFORE your observing run. There is a UNIX command "talk" which enables two people to talk to one another in one of the windows while reductions are being done in another. To use "talk" you must specify the username of the person being contacted, if you are on the same machine, or specify the username@machinename for other machines. The online help for "talk" is available by typing "man talk".
1. Introduction
1.1. Notation
2. The Workstations
2.1. Resources
3. How To...
3.1. Logging In
3.2. Logging Out
3.3. Transferring Pictures from the CCD Computer
3.4. Displaying Images
3.5. Printing Files and Plots and Other Things
3.6. Allocating the Tape Drive
3.7. Managing Disk Space
3.8. Using Sub-Directories
3.9. Using the Exabyte Devices
4. Doing More with SunView Windows
4.1. Pop-up Menus
4.2. Specific Tools
4.3. Accelerators
4.4. SunView Default Tool Files
5. Other UNIX Features
5.1. man
5.2 enscript
5.3. TEX and LATEX
5.4. mongo
5.5. visual (vi) screen editor and emacs
5.6. mail
5.7. rlogin, telnet, rcp and ftp
6. IRAF Examples
6.1. rfits and rcamera
6.2. wfits
6.3. ctio
6.4. ccdred
7. Documentation
8. Games
Appendix A: System Information
Appendix B: Default Initialization Files