| IDL 5.4 Stability [message #22359] |
Mon, 06 November 2000 00:00  |
davidf
Messages: 2866
Registered: September 1996
|
Senior Member |
|
|
Hi Folks,
Has anyone had any problems with IDL 5.4 stability? I've
managed to completely crash it three times in the past
couple of weeks. It's odd, because I don't recall
IDL 5.3.1 *ever* crashing on my Windows NT machine in
all the times I've used it.
Twice I was doing something stupid, I'm sure. But
once it crashed running one of my programs on my
web page. (This was just before a lecture, so I didn't
follow up and now I can't even remember what program
I was running. I switched to IDL 5.3.1 for the lecture.)
I'm just curious if others have experienced something
similar.
Cheers,
David
P.S. Please, I am *NOT* trying to start a flame war
here. I'm just beginning to think I may have installed
something incorrectly. At this point, I'm more inclined
to think this is my problem (or, more likely, a Windows
problem) rather than IDL's. :-)
--
David Fanning, Ph.D.
Fanning Software Consulting
Phone: 970-221-0438 E-Mail: davidf@dfanning.com
Coyote's Guide to IDL Programming: http://www.dfanning.com/
Toll-Free IDL Book Orders: 1-888-461-0155
|
|
|
|
| Re: IDL 5.4 Stability [message #22386 is a reply to message #22359] |
Fri, 10 November 2000 00:00  |
Liam E. Gumley
Messages: 378
Registered: January 2000
|
Senior Member |
|
|
Med Bennett wrote:
> Well, I'm still using 5.0.2, circa mid 1997. It's hard to know what the
> "best" version for one's particular feature needs and stability is. Maybe we
> should have a vote as to what the best version of IDL is, taking into account
> both features and stability. Since I haven't (I hate to admit it) gotten into
> object graphics yet, I don't consider that functionality to be important
> right now.
My experience with writing IMDISP
(http://cimss.ssec.wisc.edu/~gumley/imdisp.html) taught me it can be
frustrating when you try to support multiple versions of IDL (IMDISP
supports IDL 5.0-5.4). Little things keep creeping in which make it
harder to write version-independent code. For example, the SIZE function
keywords n_dimensions, dimensions, type, tname, and n_elements
introduced in IDL 5.2 make it much easier to determine array
information. Decomposed color handling is also much improved in IDL 5.2,
as is the PRINTER device support. Also, once I made the switch to
square-bracket array syntax, I could never go back to the pre-5.0
parentheses.
For my work-related programming, which is mostly done on UNIX boxes, I
use IDL 5.3 routinely and find it to be reliable, stable, and
worry-free. The transition from IDL 5.1 to 5.2 to 5.3 over the last few
years was painless AFAIK for all IDL users on these systems. I have
installed IDL 5.4 on a couple of NT PCs and it works fine, but we need
to decide how to handle GIF images before we can upgrade our UNIX
systems to 5.4.
Cheers,
Liam.
http://cimss.ssec.wisc.edu/~gumley
|
|
|
|
|
|
| Re: IDL 5.4 Stability [message #22388 is a reply to message #22359] |
Fri, 10 November 2000 00:00 |
Med Bennett
Messages: 109
Registered: April 1997
|
Senior Member |
|
|
Well, I'm still using 5.0.2, circa mid 1997. It's hard to know what the
"best" version for one's particular feature needs and stability is. Maybe we
should have a vote as to what the best version of IDL is, taking into account
both features and stability. Since I haven't (I hate to admit it) gotten into
object graphics yet, I don't consider that functionality to be important
right now.
LC's No-Spam Newsreading account wrote:
>>>> That's why I'm still ysing 4.0.1 at home.... :-)
>>> And it does all you want? You must be easy to please!
>> Indeed, it does.
>
> Me too. I'm still on 4.0 on a DU 3.2 system which has reasons to stay
> that why. I'd like a stable cheap language.
>
> I would be more agreable to pay cheaper licenses (for instance linked to
> hours of usage, and "rechargeable") than the current floating licenses,
> and a small upgrade/maintenance charge for the rare cases (one is
> approaching) when I have to change the hostid in the license server file
> because I write off an old workstation. I'm really not interested in
> getting CDs of "upgraded" versions which give me lots of GUI oriented
> stuff I do not need, or changes which can be annoying (like the [] stuff
> for arrays) or worsened performance (our license server is tied to 5.1
> because 5.3 does not support OPTIONS.DAT, and I would carefully avoid to
> go to 5.4 since it does not write GIFs anymore).
>
> --
> ------------------------------------------------------------ ----------
> nospam@ifctr.mi.cnr.it is a newsreading account used by more persons to
> avoid unwanted spam. Any mail returning to this address will be rejected.
> Users can disclose their e-mail address in the article if they wish so.
|
|
|
|
|
|
|
|
|
|
|
|
| Re: IDL 5.4 Stability [message #22461 is a reply to message #22359] |
Sun, 12 November 2000 13:23 |
Mark Hadfield
Messages: 783
Registered: May 1995
|
Senior Member |
|
|
"Lyn Doose" <ldoose@dakotacom.net> wrote in message
news:3a0f0185_1@corp.newsfeeds.com...
> Based on a very limited number of uses, I agree that IDL 5.4 on
> Windows NT seems unstable. I haven't documented the circumstances,
> but I kind of resent being a beta tester.
I think that's a little unfair. There *was* a beta test programme for IDL
5.4. I don't know if it revealed any stability problems on Windows--I know I
didn't encounter any.
For what it's worth, I have found 5.4 (beta & final) more stable that 5.3.1
(this is on NT4 + SP5). Version 5.3.1 used to crash from time to time,
usually after an attempt to reset an IDL session, but to the best of my
recollection version 5.4 hasn't. I haven't bothered to mention this
previously in this thread, because when you say "this application crashes on
my machine" it's not very helpful for others to say "it doesn't on mine".
But I the general opinion that 5.4 is less stable than 5.3.1 is not yet
proven, and I think the assertion that RSI is guilty of not testing its
products is unfair.
Reading over that last sentence I am reminded of the time when a version of
IDL came out with its netCDF support (or was it CDF?) completely broken on
Windows. Now *that* was an example of inadequate testing. The email I wrote
to RSI sizzled with barely concealed fury. Still, they sent me a T shirt to
calm me down, so I guess it wasn't all bad.
---
Mark Hadfield
m.hadfield@niwa.cri.nz http://katipo.niwa.cri.nz/~hadfield/
National Institute for Water and Atmospheric Research
PO Box 14-901, Wellington, New Zealand
|
|
|
|
| Re: IDL 5.4 Stability [message #22464 is a reply to message #22461] |
Sun, 12 November 2000 00:00 |
noymer
Messages: 65
Registered: June 1999
|
Member |
|
|
I have not used 5.4, but don't forget what
a major pain it must be to come out with software
that runs on Win, Mac, VMS, and N flavors of Unix.
To quote Danny Thorpe of Borland (you remember Turbo
Pascal, don't you?):
"Proper testing of Win32 applications today
should include testing on the following distinct
platforms: 'virgin' Windows 95, Windows 95 OSR2,
Windows 95 with IE4, Windows 95 with IE5,
Windows 95 with DCOM, Windows 98, Windows 98 Second Edition,
Windows 98 with IE5, Windows NT 4.0 SP3,
Windows NT 4.0 with IE5, and Windows 2000"
And he forgot Windows ME...
I don't know if the Win/IE interactions are relevant
to IDL, but it's still a lot of work.
Platform-specific instability is less of an
indictment than overall instability.
-- Andrew
Sent via Deja.com http://www.deja.com/
Before you buy.
|
|
|
|
|
|
|
|
| Re: IDL 5.4 Stability [message #22493 is a reply to message #22461] |
Sat, 18 November 2000 00:00 |
Lyn Doose
Messages: 4
Registered: November 2000
|
Junior Member |
|
|
There certainly are real problems with IDL 5.4 under NT. I have no idea how
wide spread they are; in my case I'm not doing anything exotic. I am
currently working with RSI to resolve a problem with code that runs fine
under 5.2, throws a Dr. Watson every time under 5.4, but then runs fine if a
breakpoint is inserted. I'll let you know the outcome.
Lyn Doose
-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----== Over 80,000 Newsgroups - 16 Different Servers! =-----
|
|
|
|
|
|
| Re: IDL 5.4 Stability [message #22522 is a reply to message #22461] |
Thu, 16 November 2000 00:00 |
Lyn Doose
Messages: 4
Registered: November 2000
|
Junior Member |
|
|
>> Based on a very limited number of uses, I agree that IDL 5.4 on
>> Windows NT seems unstable. I haven't documented the circumstances,
>> but I kind of resent being a beta tester.
>
> I think that's a little unfair. There *was* a beta test programme for IDL
> 5.4. I don't know if it revealed any stability problems on Windows--I know
I
> didn't encounter any.
>
> For what it's worth, I have found 5.4 (beta & final) more stable that
5.3.1
> (this is on NT4 + SP5). Version 5.3.1 used to crash from time to time,
> usually after an attempt to reset an IDL session, but to the best of my
> recollection version 5.4 hasn't. I haven't bothered to mention this
> previously in this thread, because when you say "this application crashes
on
> my machine" it's not very helpful for others to say "it doesn't on mine".
> But I the general opinion that 5.4 is less stable than 5.3.1 is not yet
> proven, and I think the assertion that RSI is guilty of not testing its
> products is unfair.
>
> Reading over that last sentence I am reminded of the time when a version
of
> IDL came out with its netCDF support (or was it CDF?) completely broken on
> Windows. Now *that* was an example of inadequate testing. The email I
wrote
> to RSI sizzled with barely concealed fury. Still, they sent me a T shirt
to
> calm me down, so I guess it wasn't all bad.
Gee, you want copies of my Dr. Watson logs. RSI doesn't. They want me to
boil the problem down to its "essential failing code" and send it to them.
Maybe after enough hours of work on this, I'd get a T shirt too. Wheee!
Lyn Doose
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
