| Re: Best platform for IDL 6.2? [message #46335 is a reply to message #46334] |
Wed, 16 November 2005 13:07   |
Michael Wallace
Messages: 409
Registered: December 2003
|
Senior Member |
|
|
> Pluses (Windows vs Linux) are:
>
> * The IDLDE environment is quite nice on Windows and woeful on
> Unix.
> * Graphics performance was somewhat better on Windows.
> (Originally I found IDL on Linux *very* slow, but I traced
> this to a setting like RETAIN. I posted about this on this
> newsgroup.) Last time I checked, IDL/Linux was still 30-40%
> slower for graphics than IDL/Windows on the same hardware.
> I know Karl Schulz has put some work into this, so this may
> not be true any more.
> * The IDL2AVI DLM is very nice--I use it all the time.
>
> Minuses (Windows vs Linux):
>
> * Can't integrate properly with (X)emacs and IDLWAVE. The
> problem is the lack of a console-mode IDL executable on
> Windows. It would not be difficult for RSI to produce such
> a thing but they choose not to. I do use Xemacs & IDLWAVE
> to edit files but then have to switch back to IDLDE to
> compile & run. This works better than you might think,
> but still...
> * Poorer memory handling. This has only become an issue for me
> in the last year or so, as dataset sizes have increased. It's
> still not a *serious* problem.
>
And one more thing I'll throw in to your list is that IDL under Windows
still doesn't have a command line version. That may not be important to
some of you, but because of integration we have with other languages, it
is very nice to be able to spawn a quick IDL process when you need it
whether it's in some automated nightly processing, responding to a
request for data on a web site, or other task where you're not sitting
in front of your computer.
Putting zealousness aside, a lot of people here will write their code on
Windows to take advantage of the IDE and then deploy it on Unix/Linux.
That setup seems to work well.
-Mike
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
