How to Obtain the Argus Simulator

The Argus simulator runs best on an unsophisticated PC running POD (Plain Old Dos - not Windows) with a standard VGA monitor, at least 640 K of memory and a Microsoft compatible mouse. A math coprocessor is highly recommended. As is true with a lot of DOS programs, it sometimes doesn't get along well with TSR (Terminate and Stay Resident) programs.

It needs nearly all of the basic 640K of memory of a PC so if you have any problems, boot the PC up with nothing else running in the background and try using "memmaker". It usually runs pretty well within the DOS window on a Windows 95 or NT machine, but it is more reliable and the graphics are significantly faster and better under POD. If you have problems using the simulator under Windows NT or 95, try rebooting the system from a barebones MSDOS floppy disc. This will almost always work.

The following files will be generated when you extract the simulator:

1. sargus.exe-the simulator itself
2. haloibmv.dev-a DOS device driver needed to do the graphics
3. argus.par-the Argus parameter file
4. afield_a.txt-A sample "assignment" file.
5. afield_c.txt-A sample "coordinate" file
6. aformat.txt-a description of the format of the preceeding two kinds of files.

The first three of these files MUST be on the same disc and in the same directory for the simulator to run. WARNING: The simulator has undergone several revisions. BE SURE you get these three files from the same place to insure they are compatible.

Start the simulator with the command: sargus

where is the name of a hard disc drive with at least one mbyte of free space to use as a scratchpad. Usually, one uses either drive c or d on a PC.

Once you have the simulator working properly, it will come up with an observing menu. You only have to follow through the menu system to get some idea of how it works. There is a manual for the software on the Argus WWW page, called argus_soft.html. The program is quite intuitive to use so the manual is seldom necessary.

The simulator works almost instantaneously. Argus does not. To make the process of learning to use Argus more rapid, we decided not to introduce artificial delays into the simulator. True seek time with the Argus for a new field is 1-2 minutes. The periscope goes to a new position in about 10 seconds. You should be sure and keep these delays in mind when planning observing procedures.

Argus uses two kinds of files:

1. Object files, which give coordinate lists in R.A. and Dec.
2. Assignment Files, which assign positioners to objects.

The formats for these two types of files are described in aformat.txt

A sample pair of coordinate and assignment files is included: afield_c.txt and afield_ass.txt, which give the user a set of files which are known to work.

afield_c.txt has two groups of objects on it. Objects 1-201 are a grouping which fills the Argus field uniformly and randomly with targets. Object #1 is in the center.

Objects 202-214 form a "Southern Mini-Cross" in which the targets are spaced at 15 arcsec intervals. Center on object #202 and zoom in by 16:1 in order to experiment with what field positions will cause the positioners to collide and which will not.

Centering on object #136 at the edge of the field will show both the main field and the mini-cross at the same time.

Observers should try to produce and check object and assignment files before arriving at the telescope. All field configurations should be tested with the simulator software to discover if there are collisions. The assignment of targets to fibers is not something you will want to do while sitting at the console of our 4M telescope with a beautiful dark Chilean sky outside.

A few additional comments:

This is permanent "Beta-release" software. We have done the best we can, but it is not guaranteed to be error-free. Bug reports and comments are always welcome. The actual user interface is always subject to refinement, so that the version you will actually use at the telescope may be slightly different from this release. Any differences will be small and will not affect object and assignment file formats.

The user interface is a series of menus. An option can be selected with the arrow keys followed by a carriage return or by simply typing the first letter of the option, which is faster. ESC will almost always fix any problem or stop unwanted action and return to the previous menu.

An observer normally never uses anything but the "Observing" menu and its subsidiary modes. The observer should normally not need to use the "Diagnostics".

COLLISIONS:

When you experiment with the positioners, sometimes you may think that the software detects a collision when it is "obvious" that the position is possible. This is often because the positioners have rear ends on the other side of their pivots and can actually collide when the probe tips aren't even close. These ends are not visible on the graphics screen, but the software knows where they are.

The Argus simulator assumes the objects are at zenith when predicting collisions and does not include the effect of atmospheric refraction. When observing at high air masses, occasionally Argus will refuse to go to a field because it finds that a collision will occur even though the simulator said there would be no collisions. Should this happen, one must go to the interactive mode in the Argus program itself and do the assignment over again or disable the one of the offending fibers.

To experiment with using Argus, try the following recipe:

1. From DOS, execute SARGUS discdrive . You must give the name of an existing disc drive (c, d,...) or the program will not execute. A RAM drive is preferred, in order to make the graphics run faster but a hard disc will usually work fine; e.g. type SARGUS C.
2. This will give you the Observing Menu. It will first ask you if you want to reset (initialize) Argus. Under normal observing conditions, you usually say "no" here because the instrument once operational is in a calibrated and functioning state. If it has not been touched since the last person used it, it will have stored all of its operational parameters on disc and will be in a fully operating condition without initialization. If you say "yes", Argus will initialize itself automatically, but some hand calibrations are required to put it in the best possible operating condition so it is better not to initialize it unless there is reason to believe it has a problem. When using the simulator, it is best to answer "yes' the first time you use it. After that, it doesn't make any difference.
3. Select S (Star Operations)
4. Select S (Get Stars from File). It will say "Enter object's filename"
5. Give it the coordinate filename (e.g. AFIELD.COR)
6. Select F (Assignments from file). Here it will say "Enter filename"
7. Give it the assignment filename (e.g. AFIELD.ASS)
8. Select M (Move to assigned positions). It will now check for collisions and then move to the positions. If collisions are detected, it means your assignment files are bad. This should never happen if the files were generated via the Interactive Mode. Generally collisions are detected when one tries to use an assignment file which corresponds to the wrong object file.
9. Select K (Keypad Operations).
10. Experiment with the keypad. Note the following:
11. The bright box at the bottom of the field is the periscope.
12. The crosshatched area at the top is the guide tv pickoff mirror.

NOTES:

A. You can run the probes into each other in this mode. This is because in the keypad mode, no checking for collision is done in software. This improves interactive response time. In Argus itself, the hardware detects physical collisions and stops probe movement automatically, pulling the probes back harmlessly from the collision. In the simulator, no collision will be detected since there is no hardware. Thus the probes are apparently allowed to criss-cross each other. Obviously this will not happen when you are actually connected to Argus. Letting collisions be detected in hardware under real observing conditions is a nuisance and a waste of time. Try to get in the habit of thinking about what you are doing when you move them manually to try and make sure you do not run them into one another accidentally.

B. The motion of the probes will not be obvious on the screen unless you select a large step size. To practice, set it to 250, which is the maximum. Each step is 10 microns (.2 arcsec).

C. The arrow movements in the keypad mode are always in the N-S and E-W directions for all positioners, regardless of their location.

D. You can switch between moving the periscope and fibers with the arrow keys by selecting P or F, change fibers with page up/down, etc.

E. To exit, type to get back to the Observing mode and type T (terminate program). It will ask you if you want to reset (initialize) Argus. Generally when observing, you should reset it on leaving in order to leave it in a known state.

Once you have the hang of using Argus, if you have a mouse, try the Interactive Assignment mode. It will put you into the keypad display and ask you for an Object file and a reference object. It will then put the reference object in the center of the field and draw the rest of your objects in the field in their appropriate locations.

At this point, you can use the mouse to pick up objects and interactively assign them to positioners. Their locations will be shown on the screen and software will do collision detection for you in real time. It will generate an assignment field for you which you can store once you have generated it.

When stars have been assigned, they are removed from the field of view. However, once you leave the Interactive Mode, if you re-enter it, all the stars will return when you select the object file. You then must call up all assignment files and re-store them in order to remove the objects from view. The Interactive Assignment Mode is very easy to use. It is rather like a video game. Try it. You'll like it! Its even better if you have a 13 year old boy in the house. You may be able to get him to do all you assignment fields for you. Crowded fields are especially challenging.

Feel free to address any questions to:

Good luck!

Appendix - Known bugs or warnings:

In the keypad mode, in the E-W direction, probes on the screen move opposite to the sense of the keypad arrows. The screen is correct. This has to do with an optical/mechanical problem which was fixed in software by deliberately inverting the sense of the arrows. This will be corrected at some time in the future.

BEWARE of using extreme zooms of less than one for anything serious. It appears to work and could be useful for selecting centers of target fields, but the reliability of the mapping software is untested in this mode.