| What's New in IDL 8.4 [message #89610] |
Fri, 31 October 2014 14:03  |
David Fanning
Messages: 11724
Registered: August 2001
|
Senior Member |
|
|
Folks,
I just watched Beau Legeer's new screencast on what is new with IDL 8.4.
https://www.youtube.com/watch?v=_umxrSAEWis
All I can say is, "Wow!" I had no idea so many people were clamoring for
Big Integers (you know, 2^400, printed on your display), and I plan to
use the Prime and NextPrime static functions almost daily.
It sure would be fun to be a fly on the wall in the meeting in which
customer requests were prioritized for the next version of IDL. I have a
feeling I would be absolutely shocked at how out of touch I am with the
tools programmers need to do their jobs.
Cheers,
David
--
David Fanning, Ph.D.
Fanning Software Consulting, Inc.
Coyote's Guide to IDL Programming: http://www.idlcoyote.com/
Sepore ma de ni thue. ("Perhaps thou speakest truth.")
|
|
|
|
| Re: What's New in IDL 8.4 [message #89611 is a reply to message #89610] |
Fri, 31 October 2014 15:04  |
Lajos Foldy
Messages: 176
Registered: December 2011
|
Senior Member |
|
|
On Friday, October 31, 2014 10:03:43 PM UTC+1, David Fanning wrote:
> Folks,
>
> I just watched Beau Legeer's new screencast on what is new with IDL 8.4.
>
> https://www.youtube.com/watch?v=_umxrSAEWis
>
> All I can say is, "Wow!" I had no idea so many people were clamoring for
> Big Integers (you know, 2^400, printed on your display), and I plan to
> use the Prime and NextPrime static functions almost daily.
>
> It sure would be fun to be a fly on the wall in the meeting in which
> customer requests were prioritized for the next version of IDL. I have a
> feeling I would be absolutely shocked at how out of touch I am with the
> tools programmers need to do their jobs.
>
> Cheers,
>
> David
> --
> David Fanning, Ph.D.
> Fanning Software Consulting, Inc.
> Coyote's Guide to IDL Programming: http://www.idlcoyote.com/
> Sepore ma de ni thue. ("Perhaps thou speakest truth.")
There are priorities. The parallel processing API for IDL 8.4 was thrown out. BigInt is much more important.
https://groups.google.com/d/msg/comp.lang.idl-pvwave/bEr8Bh5 iLrI/60o6nJRNGdkJ
regards,
Lajos
|
|
|
|
|
|
|
|
| Re: What's New in IDL 8.4 [message #89614 is a reply to message #89613] |
Fri, 31 October 2014 16:08 |
David Fanning
Messages: 11724
Registered: August 2001
|
Senior Member |
|
|
fawltylanguage@gmail.com writes:
> There must be a number theory mathematician somewhere on the globe who uses IDL and posted a feature request. Python has bignum, too, so it was accepted.
Yeah, with a LOUD voice, apparently. :-)
Some of this stuff seems useful. The WatchFolder thing might be useful
(at least I've heard people ask for this kind of thing), and some of the
static functions might be handy, although God only knows when I'll need
to know the next prime of a BigNumber. It really does seem to me like
the kind of thing that might come up once in a lifetime, if then.
The code generator thing must be a marketing department's wet dream.
Take all the people who can't for the life of them figure out how to
make a plot in IDL (and, unfortunately, there are more and more of these
people every day) and get someone to generate lousy code for them that
somehow manages to work. Job done! Another successful IDL programmer.
But, really, who is asking for some of this stuff? Got to be the
marketing guys so they can tick off a couple of boxes in the Python vs
IDL matrix. It seems to me, with limited time and resources, you might
want to spend time working on stuff that actually matters to people and
how they do their jobs. Nobody is asking me, of course, but I could make
a few suggestions if anyone was interested. :-)
Cheers,
David
--
David Fanning, Ph.D.
Fanning Software Consulting, Inc.
Coyote's Guide to IDL Programming: http://www.idlcoyote.com/
Sepore ma de ni thue. ("Perhaps thou speakest truth.")
|
|
|
|
| Re: What's New in IDL 8.4 [message #89616 is a reply to message #89614] |
Sat, 01 November 2014 04:11 |
markb77
Messages: 217
Registered: July 2006
|
Senior Member |
|
|
On Saturday, November 1, 2014 12:08:56 AM UTC+1, David Fanning wrote:
> But, really, who is asking for some of this stuff? Got to be the
> marketing guys
hi David,
I can see why they're trying to update the language syntax to make IDL more like Python and less like FORTRAN... you need to keep the syntax current if the next generation of programmers are going to use IDL.
Same goes for function graphics.. while the current NG system has some glitches, you can understand why it's needed for the same reasons as above.
On the other hand, few USEFUL features have been added to recent releases for people who already know how to code in IDL. I appreciate the nice syntax of object programming, and the HASH is great for everyday use. However, the things desperately needed in my view are parallel / multi-threaded processing which is VM compatible (whether by an improved VM or whatever) and an updated Widget toolkit.
Will Exelis please stand up and answer when will we see some additions to parallel processing and widgets? I am suffering over here with one core of my CPU running 100% for 12 hours and the other 11 cores basically idle!!!
Mark
|
|
|
|
| Re: What's New in IDL 8.4 [message #89621 is a reply to message #89616] |
Sun, 02 November 2014 07:34 |
Karl[1]
Messages: 79
Registered: October 2005
|
Member |
|
|
On Saturday, November 1, 2014 5:11:06 AM UTC-6, superchromix wrote:
> On Saturday, November 1, 2014 12:08:56 AM UTC+1, David Fanning wrote:
>
>> But, really, who is asking for some of this stuff? Got to be the
>> marketing guys
>
> hi David,
>
> I can see why they're trying to update the language syntax to make IDL more like Python and less like FORTRAN... you need to keep the syntax current if the next generation of programmers are going to use IDL.
>
> Same goes for function graphics.. while the current NG system has some glitches, you can understand why it's needed for the same reasons as above.
>
> On the other hand, few USEFUL features have been added to recent releases for people who already know how to code in IDL. I appreciate the nice syntax of object programming, and the HASH is great for everyday use. However, the things desperately needed in my view are parallel / multi-threaded processing which is VM compatible (whether by an improved VM or whatever) and an updated Widget toolkit.
>
> Will Exelis please stand up and answer when will we see some additions to parallel processing and widgets? I am suffering over here with one core of my CPU running 100% for 12 hours and the other 11 cores basically idle!!!
>
> Mark
BigInts were probably added to support crypto operations. With a large integer library available, it isn't too hard to implement some of the tried-and-true crypto primitives, which could then be used to digitally sign data, validate digital signatures, perform hashing, and perhaps perform encryption and decryption. It isn't hard to understand the demand for this in today's computing environment.
I was going to point out that it would not be too hard to pull the large integer functions out of a library like cryptopp, which has very liberal licensing, and put them in an IDL DLM. However, the BigInt functionality in IDL looks like it is integrated better than that.
Karl
|
|
|
|
| Re: What's New in IDL 8.4 [message #89622 is a reply to message #89621] |
Sun, 02 November 2014 08:04 |
David Fanning
Messages: 11724
Registered: August 2001
|
Senior Member |
|
|
Karl writes:
> BigInts were probably added to support crypto operations. With a large integer library available, it isn't too hard to implement some of the tried-and-true crypto primitives, which could then be used to digitally sign data, validate digital signatures, perform hashing, and perhaps perform encryption and decryption. It isn't hard to understand the demand for this in today's computing environment.
OK, I guess that makes sense. For ENVI, then.
Cheers,
David
--
David Fanning, Ph.D.
Fanning Software Consulting, Inc.
Coyote's Guide to IDL Programming: http://www.idlcoyote.com/
Sepore ma de ni thue. ("Perhaps thou speakest truth.")
|
|
|
|
|
|
|
|
| Re: What's New in IDL 8.4 [message #89664 is a reply to message #89628] |
Sat, 08 November 2014 08:07 |
chris_torrence@NOSPAM
Messages: 528
Registered: March 2007
|
Senior Member |
|
|
On Sunday, November 2, 2014 11:44:51 AM UTC-7, David Fanning wrote:
> Fabien writes:
>
>> Is there something like a "feature request" for IDL? How do Exelis
>> actually decide on language evolutions, etc?
>
> [Comment deleted...]
>
> Cheers,
>
> David
> --
> David Fanning, Ph.D.
> Fanning Software Consulting, Inc.
> Coyote's Guide to IDL Programming: http://www.idlcoyote.com/
> Sepore ma de ni thue. ("Perhaps thou speakest truth.")
Ah, this is why I love this newsgroup.
I've been out for a couple of weeks, sorry I missed all the excitement.
Here's *an* answer:
BigInteger: nights & weekends for three weeks as a fun project
Parallel Processing: 6-9 months for multiple developers
New Widgets: see Parallel Processing
Cheers,
Chris
|
|
|
|
|
|
| Re: What's New in IDL 8.4 [message #89674 is a reply to message #89610] |
Tue, 11 November 2014 20:01 |
andrewcool777
Messages: 27
Registered: November 2012
|
Junior Member |
|
|
On Saturday, 1 November 2014 07:33:43 UTC+10:30, David Fanning wrote:
> Folks,
>
> I just watched Beau Legeer's new screencast on what is new with IDL 8.4.
>
> https://www.youtube.com/watch?v=_umxrSAEWis
>
> All I can say is, "Wow!" I had no idea so many people were clamoring for
> Big Integers (you know, 2^400, printed on your display), and I plan to
> use the Prime and NextPrime static functions almost daily.
>
> It sure would be fun to be a fly on the wall in the meeting in which
> customer requests were prioritized for the next version of IDL. I have a
> feeling I would be absolutely shocked at how out of touch I am with the
> tools programmers need to do their jobs.
>
> Cheers,
>
> David
> --
> David Fanning, Ph.D.
> Fanning Software Consulting, Inc.
> Coyote's Guide to IDL Programming: http://www.idlcoyote.com/
> Sepore ma de ni thue. ("Perhaps thou speakest truth.")
And I'm turning blue in the face waiting for READ_PNG to return metadata.
And maybe READ_JPEG to return EXIF data.
A one week task for one programmer, maybe...?
Andrew
|
|
|
|
| Re: What's New in IDL 8.4 [message #89720 is a reply to message #89610] |
Wed, 19 November 2014 21:51 |
SHENYUE JIA
Messages: 2
Registered: November 2014
|
Junior Member |
|
|
On Friday, October 31, 2014 2:03:43 PM UTC-7, David Fanning wrote:
> Folks,
>
> I just watched Beau Legeer's new screencast on what is new with IDL 8.4.
>
> https://www.youtube.com/watch?v=_umxrSAEWis
>
> All I can say is, "Wow!" I had no idea so many people were clamoring for
> Big Integers (you know, 2^400, printed on your display), and I plan to
> use the Prime and NextPrime static functions almost daily.
>
> It sure would be fun to be a fly on the wall in the meeting in which
> customer requests were prioritized for the next version of IDL. I have a
> feeling I would be absolutely shocked at how out of touch I am with the
> tools programmers need to do their jobs.
>
> Cheers,
>
> David
> --
> David Fanning, Ph.D.
> Fanning Software Consulting, Inc.
> Coyote's Guide to IDL Programming: http://www.idlcoyote.com/
> Sepore ma de ni thue. ("Perhaps thou speakest truth.")
There is a different way of running pros in a parallel mode not using their API. Try to loop and use SPAWN and throw the command that can execute a script to the terminal. If your CPU has multiple cores, it can run the program. It worked very well for under OS X.
|
|
|
|
| Re: What's New in IDL 8.4 [message #91626 is a reply to message #89664] |
Thu, 06 August 2015 15:54 |
pfp
Messages: 12
Registered: July 2009
|
Junior Member |
|
|
I just recently found BigInteger has a very handy feature: writing numbers in different bases. I had to convert numbers to base 36, and the first result I found when I searched for and IDL function to do this was BigIntegers's tostring method (http://www.exelisvis.com/docs/BIGINTEGER.html#ToString).
There is also a function for base conversion in the JHUAPL library (basecon).
I happened to need base-36 numbers to work with orbit IDs of Hubble Space Telescope observations, which are encoded in base-36 in STScI's datasets and observation IDs.
On Saturday, November 8, 2014 at 9:07:13 AM UTC-7, Chris Torrence wrote:
> Here's *an* answer:
>
> BigInteger: nights & weekends for three weeks as a fun project
> Parallel Processing: 6-9 months for multiple developers
> New Widgets: see Parallel Processing
>
> Cheers,
> Chris
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
