IRAF and More - For Mac Using Astronomers and Astrophysicists
I’m not running IDL of any stripe on my Macbook, but it would seem there is a patch for IDL 6.4 that will fix compatibility issues you may encounter with Leopard. From what I can tell from this thread the problem sounds related to the X11 changes in Leopard.
Tags: IDL, Leopard — .So, I admit I only recently installed SM(supermongo), and only have used it slightly. Between IDL, grace(installable via Fink or there are binaries for MacOS X at the HPC site), and Plot, I just don’t see why any Mac user would make plots in SM. It requires gnuplot-esque scripts/programs, and just seems in general a pain and esoteric.
Perhaps it is powerful, perhaps for automated plot creation from certain kinds of data, it’s great, but I don’t understand its popularity. Has anyone used all the above tools and yet prefers SM? If so, please post a comment. I’m curious if I’m missing something.
For casual plotting, say from analysis that I do in IDL, I just do quick plots in IDL itself, and I’ve written a fair number of simple IDL routines that call plot or ploterr or other people’s plotting programs. This works for day to day plotting, and is easy to integrate into my other IDL programs. For almost all plots for publication, I have always used the aforementioned grace, which I installed with Fink.
If if I wasn’t already familiar with grace, I’d probably try this unfortunately under-named program, Plot. It seems pretty good, but since it’s very similar to Grace except with all the options in different places, so I’ve just stayed with the familiar Grace. But, I vastly prefer a graphical interface like this for tweaking than altering parameters via some script. I have also used gnuplot. It was a nightmare. I recreated in grace in one hour what it took days to get gnuplot to do, and do poorly. I abandoned gnuplot as a first year grad student.
Anyway, I know sm must have its fans, so if that’s you, let me know why you like it below. But, please note if you’ve used any other plotting program.
Update I’m baffled by the “something else entirely” winning unless it’s non-astronomers voting for Excel. Whatever you are voting for, could you post a comment and tell me? Also, I recently got a new Macbook, and installing Grace via Fink was a royal pain requiring a selfupdate via rsync and then a very long compile-fest because they don’t have a binary available. It’s hard to recommend with all that pain, but I maintain it’s much better than SM.
Update II Ploticus sounds vaguely familiar (see in the comments), but not so much python/matplotlib. These sound like IDL/SMish sort of plotting though. Personally, I still prefer a GUI - hence my like for xmgr Grace.
Update III I certainly get the desire to have a plot automagically made while one is doing data reduction, analysis, etc. This is one reason I like IDL. However, I prefer to be able to finely tune the appearance in a GUI for a plot designed for publication. I suppose it’s a learning curve thing. Granted, once you write one sm script to plot a spectra … you’re set for all other spectra. However, I’m not convinced for my own work that the time/effort spent learning how to write such a script is worth it, since I’m obviously rather comfortable with Grace. (It’s worth noting for a certain repetitive plots I have used IDL.)
.If you have any issues getting IDL to properly record/recognize the cursor position or mouse clicks, you may want to try this command in the terminal
defaults write com.apple.x11 wm_click_through -bool true
Frank Varosi here at UF gave me this tip, which he found at this french site. What this command does is it changes a window manager preference in Apple’s X11. This probably only effects the so-called “Quartz VM” which is the default window manager. If you switched to the enlightenment or some other window manager, my guess is this click-through preference has no effect, but I can’t say for sure.
Personally, I’ve yet to write IDL programs that look for the cursor location, but I figure this may be helpful to others.
Update As a commenter noted, you will have to restart X11 for this to do anything, since it’s changing a preference in Apple’s quartz-vm window manager.
Update II As commenter Josh Shiode noted, the Macosforge X11 (and possibly any X11 with Leopard) requires a slightly different command:
defaults write org.x.x11 wmclickthrough -bool true
If you’re using IDL on a Mac, and also have the IDL license server running on the same Mac, it’s apparently necessary to set HOSTNAME in /etc/hostconfig. In the old days of OS X this was set by default in hostconfig with a line
HOSTNAME=-AUTOMATIC-
but this is no longer the case in Tiger, and maybe Panther. Your hostname should be something like mycoolmac.myschool.edu. I would think using -AUTOMATIC- would work as well, since that should grab your hostname from your DNS.
Thanks to Jessica Lair for this info.
FYI, I’m now trying out MarsEdit for my posts, which I think I may like more than ecto.
Tags: IDL — .I discovered MPFIT for tweaking parameters and optimizing curve fits. I had never used curvefit (and when I tried, it didn’t go that well.) MPFIT takes a bit of time to get the hang of, and I had to tweak my curve-generating code to make it play nice with MPFIT…. and the initial parameters you guess have to be close or it won’t converge. But, all said and done, I’m pretty happy with it.
Tags: fitting, IDL, MPFIT — .There is a handy IDL syntax coloring plug-in for Textwrangler that I find quite useful. It colors built in IDL functions, among other things.
Installation is fairly easy (you put the file in TextWrangler’s language module folder and then assign .pro to IDL) and described on the download site.
.At long last, a .pkg installer for the PPC version of IRAF 2.12 BETA 2 is available on the download page. You can download it directly from this link
.