| IDL 6.0 [message #34528] |
Thu, 27 March 2003 10:04  |
jlhall
Messages: 1
Registered: March 2003
|
Junior Member |
|
|
Does anyone have any substantial information about IDL 6.0? When is the
planned release? What will the major differences be betwwen 6.0 and the
current version?
All I could find was this page in Chinese... which is still relatively
mysterious after babelfish translation:
http://astron.berkeley.edu/~jhall/export/IDL_6.0.pdf
------------------------------------------------------------ ------
Joseph Lorenzo Hall jlhall@sonic.net
|
|
|
|
| Re: IDL 6.0 [message #34589 is a reply to message #34528] |
Mon, 31 March 2003 08:23  |
JD Smith
Messages: 850
Registered: December 1999
|
Senior Member |
|
|
On Mon, 31 Mar 2003 08:50:45 -0700, Michael A. Miller wrote:
>>>> >> "JD" == JD Smith <jdsmith@as.arizona.edu> writes:
>
>> The problem is the IDL startup time is long enough that
>> interactive use is stressed much more than batch use. I.e. the
>> standard response would be that you should string together your
>> bits of code in an *IDL* script/batch-file/$MAIN$-level-routine.
>
> IDL starts in about 3 seconds or less on the machine I just timed it on
> (2 GHz pentiums). The jobs we're regularly running take from 5-500
> minutes, so start up time is negligible.
>
Yes, but that didn't use to be the case. It might be nice if IDL had a
non-command-line mode which skipped all the interface startup and went
straight into the interpreter. The basic issue is that IDL doesn't see
itself as an equal tool among many, but as *the* tool, to which can be
tacked on bits and pieces of code from other languages as necessary.
Writing an IDL script which drives a perl script is trivial. Writing a
perl script which drives an IDL script is less so. For good or for bad,
IDL's original authors were probably disillusioned with the "string
together various disparate pieces" method of problem solving, and wrote
IDL to be a "one-stop-shopping" data language.
JD
|
|
|
|
| Re: IDL 6.0 [message #34591 is a reply to message #34528] |
Mon, 31 March 2003 07:50 |
mmiller3
Messages: 81
Registered: January 2002
|
Member |
|
|
>>>> > "JD" == JD Smith <jdsmith@as.arizona.edu> writes:
> The problem is the IDL startup time is long enough that
> interactive use is stressed much more than batch use.
> I.e. the standard response would be that you should string
> together your bits of code in an *IDL*
> script/batch-file/$MAIN$-level-routine.
IDL starts in about 3 seconds or less on the machine I just timed
it on (2 GHz pentiums). The jobs we're regularly running take
from 5-500 minutes, so start up time is negligible.
Mike
|
|
|
|
| Re: IDL 6.0 [message #34596 is a reply to message #34528] |
Sun, 30 March 2003 13:54 |
wmconnolley
Messages: 106
Registered: November 2000
|
Senior Member |
|
|
David Fanning <david@dfanning.com> wrote:
> Yes, well, creating a filled contour on a map projection
> *always* requires CELL_FILL.
And is therefore slow. This is the point I was trying to make: I'd
like improved support for basics (such as this) rather than... well,
you know, widgets and stuff :-)
-W.
--
William M Connolley | wmc@bas.ac.uk | http://www.nerc-bas.ac.uk/icd/wmc/
Climate Modeller, British Antarctic Survey | Disclaimer: I speak for myself
I'm a .signature virus! copy me into your .signature file & help me spread!
|
|
|
|
| Re: IDL 6.0 [message #34598 is a reply to message #34528] |
Sat, 29 March 2003 17:08 |
David Fanning
Messages: 11724
Registered: August 2001
|
Senior Member |
|
|
William M Connolley (wmc@bas.ac.uk) writes:
> Hmmm, disregarding my spelling, always a good idea, trying looking at:
>
> http://www.antarctica.ac.uk/met/wmc/a.png
>
> the LHS plot is without /cellfill. The RHS is with /cell_fill.
> The RHS is correct. The LHS isn't.
Yes, well, creating a filled contour on a map projection
*always* requires CELL_FILL. I think I even found a warning
about this in the IDL documentation while I was looking for
something else. (At least I remember thinking, 'Humm, that's
interesting. I wonder what the hell it is doing *here*!"
In any case, the point is fairly dramatically made in this
article on my web page:
http://www.dfanning.com/color_tips/fill_colors.html
Cheers,
David
--
David W. Fanning, Ph.D.
Fanning Software Consulting, Inc.
Phone: 970-221-0438, E-mail: david@dfanning.com
Coyote's Guide to IDL Programming: http://www.dfanning.com/
Toll-Free IDL Book Orders: 1-888-461-0155
|
|
|
|
|
|
|
|
| Re: IDL 6.0 [message #34603 is a reply to message #34528] |
Sat, 29 March 2003 06:14 |
wmconnolley
Messages: 106
Registered: November 2000
|
Senior Member |
|
|
MC <markchan@shaw.ca> wrote:
> Quick and easy data visualization. I do what I needed for my use. Not to
> have fancy interfaces for others most of the time.
Indeed.
> Hope future releases will continue to improve our needs
BTW, does the contouring routine still need /cell_fill to work prooperly?
-W.
|
|
|
|
| Re: IDL 6.0 [message #34604 is a reply to message #34528] |
Fri, 28 March 2003 16:38 |
MC
Messages: 50
Registered: September 1996
|
Member |
|
|
I am with you fellows.
Direct Graphics and non-object programming was what attracted me to IDL.
Quick and easy data visualization. I do what I needed for my use. Not to
have fancy interfaces for others most of the time.
Hope future releases will continue to improve our needs so that IDL will not
become just 'another addition' to a whole lot of programming softwares (many
of which are much more powerful than IDL) already available to professional
programmers.
MC
"Randall Skelton" <rhskelto@atm.ox.ac.uk> wrote in message
news:Pine.LNX.4.33.0303280853520.2390-100000@moriarty.atm.ox .ac.uk...
>
> On Fri, 28 Mar 2003, Michael W Asten wrote:
>
>> This viewpoint shows an interesting divergence between different classes
of
>> users. I am of the troglodyte school who do not use objects (do I hear
a Coyote
>> howling from the next hill?). But I do use guis to drive my programs,
and I dont
>> like programming guis - I prefer to use ready-made items; I use IDL to
get quick
>> results from my data, not to generate fancy code.
>
> I may have hastily overstated my request... I too make guis so I can
> quickly load and visualize data and I agree that more ready-made items
> will probably make this easier. Provided, of course, that I can afford
> the next suite of IDL programming books that describe how to actually
> use these new features ;)
>
> I suppose my concern is that IDL stops being an 'Interactive Data
> Language' and instead becomes a glorified visual-basic competitor. The
> thought of double-clicking on each IDL object to get an 'objectsheet'
> where I enter formulas such as '=A1+$B$2+(SUM(D12:14))^2' is horrifying.
>
> For the record, many people in my department don't use guis, objects,
> functions or even procedures. Yes, many of the people I work with write
> 'routines' that are strung together using .run commands. With the
> aforementioned suite of code they '.r read_modelout.pro' or '.r
> modelradforce.pro' and, after answering a seemingly endless stream of
> console queries, they return to the IDL prompt to '.r diagnose.pro',
> 'print, res1, xh, bb, yobs', and '.r normplot1.pro' to produce
> direct-graphics windows and postscript plots of the atmosphere. You are
> certainly correct that such discussions do highlight the various classes
> of users.
>
> Cheers,
> Randall
>
> NB: from <http://www.winternet.com/~mikelr/flame39.html> one learns:
>
> Troglodyte -- an aggressive Philistine who categorically disdains all
> new-age enthusiasms and politics, and has a particular animus against
> "homos", "tree huggers", and artsy types. His/her loud brandishings of
> "common sense" and "reason" are nothing more than conceits to justify
> his/her close mindedness. Troglodyte's fierce, but predictable attacks
> are easily parried by more nimble Warriors. Sometimes, just for fun,
> Weenie or Issues will taunt him/her into a towering rage.
>
|
|
|
|
| Re: IDL 6.0 [message #34605 is a reply to message #34528] |
Fri, 28 March 2003 08:59 |
JD Smith
Messages: 850
Registered: December 1999
|
Senior Member |
|
|
On Fri, 28 Mar 2003 08:51:37 -0700, Michael A. Miller wrote:
>>>> >> "Michael" == Michael A Miller <mmiller3@iupui.edu> writes:
>
>> That and access to command line arguments like every other
>> language on the planet would go a long way to streamlining my IDL
>> use.
>
> To clarify my point a bit - what I'm getting at is that much of what we
> do here with IDL is interactive, so the "I" in IDL is really great. Even
> more of what we do is along the lines of "take what we've developed
> interactively and repeat it over and over and over ad infinitum." The
> cumbersome methods that are needed to pass parameters to IDL codes from
> outside IDL has resulted in some really ugly kludges around here.
>
> One source of that seems to be that many of us new-old-schoolers
> (troglodytes that is :-) tend to want to string a collection of programs
> together in sequence with a script. To do that with IDL, every tool we
> develop with IDL needs to have an additional layer of code wrapped
> around it to handle the fact that IDL has no access to command line
> arguments. That's something that we (well, ok, I) got used to back in
> the 80's and it is hard to give up. Even though IDL has lots of
> wonderful features, that "I" for interactive makes some harder to get at
> than I'd like.
>
> Ok, I'll get off that soap box now ;-)
The problem is the IDL startup time is long enough that interactive use is
stressed much more than batch use. I.e. the standard response would be
that you should string together your bits of code in an *IDL*
script/batch-file/$MAIN$-level-routine.
If you have other types of code you need to interact with in IDL, call it
from there (as opposed to visa versa). This can be awkward if IDL is
really just a small component of a large process, but it's really the
most convenient way.
JD
|
|
|
|
| Re: IDL 6.0 [message #34609 is a reply to message #34528] |
Fri, 28 March 2003 07:51 |
mmiller3
Messages: 81
Registered: January 2002
|
Member |
|
|
>>>> > "Michael" == Michael A Miller <mmiller3@iupui.edu> writes:
> That and access to command line arguments like every other
> language on the planet would go a long way to streamlining
> my IDL use.
To clarify my point a bit - what I'm getting at is that much of
what we do here with IDL is interactive, so the "I" in IDL is
really great. Even more of what we do is along the lines of
"take what we've developed interactively and repeat it over and
over and over ad infinitum." The cumbersome methods that are
needed to pass parameters to IDL codes from outside IDL has
resulted in some really ugly kludges around here.
One source of that seems to be that many of us new-old-schoolers
(troglodytes that is :-) tend to want to string a collection of
programs together in sequence with a script. To do that with
IDL, every tool we develop with IDL needs to have an additional
layer of code wrapped around it to handle the fact that IDL has
no access to command line arguments. That's something that we
(well, ok, I) got used to back in the 80's and it is hard to give
up. Even though IDL has lots of wonderful features, that "I" for
interactive makes some harder to get at than I'd like.
Ok, I'll get off that soap box now ;-)
Mike
--
Michael A. Miller mmiller3@iupui.edu
Imaging Sciences, Department of Radiology, IU School of Medicine
|
|
|
|
| Re: IDL 6.0 [message #34610 is a reply to message #34528] |
Fri, 28 March 2003 07:28 |
mmiller3
Messages: 81
Registered: January 2002
|
Member |
|
|
>>>> > "David" == David Fanning <david@dfanning.com> writes:
> P.S. Let's just say I was going to say something sarcastic
> about Mike Miller's "Here, here", but decided against it on
> further reflection. :-(
Don't hold back on my account :-)
Mike
|
|
|
|
| Re: IDL 6.0 [message #34614 is a reply to message #34528] |
Fri, 28 March 2003 01:47 |
Randall Skelton
Messages: 169
Registered: October 2000
|
Senior Member |
|
|
On Fri, 28 Mar 2003, Michael W Asten wrote:
> This viewpoint shows an interesting divergence between different classes of
> users. I am of the troglodyte school who do not use objects (do I hear a Coyote
> howling from the next hill?). But I do use guis to drive my programs, and I dont
> like programming guis - I prefer to use ready-made items; I use IDL to get quick
> results from my data, not to generate fancy code.
I may have hastily overstated my request... I too make guis so I can
quickly load and visualize data and I agree that more ready-made items
will probably make this easier. Provided, of course, that I can afford
the next suite of IDL programming books that describe how to actually
use these new features ;)
I suppose my concern is that IDL stops being an 'Interactive Data
Language' and instead becomes a glorified visual-basic competitor. The
thought of double-clicking on each IDL object to get an 'objectsheet'
where I enter formulas such as '=A1+$B$2+(SUM(D12:14))^2' is horrifying.
For the record, many people in my department don't use guis, objects,
functions or even procedures. Yes, many of the people I work with write
'routines' that are strung together using .run commands. With the
aforementioned suite of code they '.r read_modelout.pro' or '.r
modelradforce.pro' and, after answering a seemingly endless stream of
console queries, they return to the IDL prompt to '.r diagnose.pro',
'print, res1, xh, bb, yobs', and '.r normplot1.pro' to produce
direct-graphics windows and postscript plots of the atmosphere. You are
certainly correct that such discussions do highlight the various classes
of users.
Cheers,
Randall
NB: from <http://www.winternet.com/~mikelr/flame39.html> one learns:
Troglodyte -- an aggressive Philistine who categorically disdains all
new-age enthusiasms and politics, and has a particular animus against
"homos", "tree huggers", and artsy types. His/her loud brandishings of
"common sense" and "reason" are nothing more than conceits to justify
his/her close mindedness. Troglodyte's fierce, but predictable attacks
are easily parried by more nimble Warriors. Sometimes, just for fun,
Weenie or Issues will taunt him/her into a towering rage.
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
