iraf.net

Dear Jeanette,
Hofstra will be getting it's INTERNET node up next month. They do not
have much free memory on their SI workstation. How much disk memory
do you need to load and run IRAF on the Silicon Graphics and how much
to analyze spectral data? My dept. will have to buy the memory.
Likewise, they have a VAX mainframe running under VMS. What memory
requirements are needed for this system and am I better off using the
Silicon Graphics workstation. Finally, when you have a prot for AUX,
what Mac machine would you suggest, the IIci, IIsi, or II fx and how
much slower is IRAF on a MACintosh than on either a Sun or Si.licon
Graphics workstation?Thanks, Don Lubowich

Dear Jeanette,
I forgot to ask two questions. Will IRAF work on SUN clones that have
SPARC compatibility running licensed SunOS UNIX and X Windows or
clones running One Desktop SCO UNIX since these machines are less
expensive that Sun workstations? Will IRAF run on the NeXT computers
under their UNIX or will one need a specific port for the NeXT
machine?
Thanks,
Don

Hy,I'm Manrico Strappolini from astronomical institute in Rome and respon
sible for the application of IRAF in this institute.
I wish to Know if exist a command to obtain an hardcopy of a graphic display
but working outside from a task.(for example the command :.snap is inside
the splot task).Thanks for your cooperation.
Manrico Strappolini

Hello Manrico, In reply to your recent question about IRAF hardcopy: > I wish to Know if exist a command to obtain an hardcopy of a graphic
> display but working outside from a task.(for example the command :.snap
> is inside the splot task).The command you mention, :.snap, is available whenever you are in interactive
graphics mode. You are in interactive graphics mode whenever the graphics
crosshair cursor appears on the screen. The SPLOT task is an interactive task,
so this command is available within the task environment. If you have a plot
on the screen that you wish to make a hardcopy of, you can get into interactive
cursor mode with the CL command =gcur and then execute :.snap. The
cl> =gcur
actually requests the value of the graphics cursor, so the crosshair appears
and you can execute any of the commands that appear when you type :.help,
including :.snap. If you want a hardcopy of a CONTOUR plot for example, you'd
draw the plot with the CONTOUR task and then enter =gcur and :.snap. (Note
that with CONTOUR, you can also specify device=stdplot on the command line,
although you won't see the plot on the screen before the hardcopy is made.)
You can get a help page on CURSORS for more on using the graphics cursor. To
save keystrokes, the equal sign "=" can be used instead of :.snap to make the
hardcopy. If you're running SUN/IRAF, the Graphcopy Gterm Frame Menu Selection
can also be used to make a hardcopy of the contents of the Gterm graphics
window. This requires you have R_DISPOSE set to properly identify your
output device. In most cases, Graphcopy produces an inferior plot to what
is available from interactive cursor mode.Hope this helps; send any further questions to iraf@noao.edu.Suzanne Jacoby

HY !
I must analyse some spectrum.This spectrum have on the ordinate the flux
(y=flux) and on the x the wavelenght (x=wavalenght).But I need to have a
graphic where there is on the ordinate the logarithm of the flux (y=logflux)
and on the x the logarithm of the wavelenght (x=logwavelenght).
I wish to know if is possible to obtain this graphic and what is the task
or the procedure.
I have used the function mode inside the task SPLOT.This menu allows arithmetic
function to be applied to the spectrum,but this commands provides only the
logarithm of the flux but no of the wavelenght.How can I do ?
Thanks for your cooperation. Manrico Strappolini

Hello Manrico,There are several choices, none of which is ideal. This will be simple
to do in the next version of IRAF.One approach is to rebin the spectra to a logarithmic interval dispersion.
The task to do this is DISPCOR (which log+ and rebin+) or REBIN if
the spectra are onedspec format (i.e. a one dimensional image).
Be sure to use MSDISPCOR for multispec format; if you don't know what that
is then you don't have it. For the flux you can either do it
in SPLOT or use PROTO.IMFUNC.If your don't want to rebin the spectra then you can use GRAPH setting
logx=yes, logy=yes, lintran=yes, q1=(wavelength of first pixel),
q2=(wavelength of last pixel). However, in my quick test the problem is
that the minimum range in logx that can be plotted is a decade so for the
case 4000-5000 Angstroms I got a plot from 1000 to 10000 Angstrms.Ar more complex possibility is to dump the pixel values with
LISTPIX into a file and then to convert the values to log. Below is
the recipe: cl> listpix image > infile
cl> list="infile"
cl> while (fscan (list, x, y) != EOF) {
>>> z = 4110 + (x - 1) * 1.23
>>> print (log10(z), log10(y), >> "outfile")
}
cl> graph outfileIn the example you subsititutue your own values for the image, the infile,
and outfile, and the starting wavelength (W0) and wavelength per pixel
(WPC).Let us know if you need further help.Frank Valdes