Home > Archive > Fortran > May 2004 > Re: article on Intel compilers and tools
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
Re: article on Intel compilers and tools
|
|
| Steve Lionel 2004-05-12, 9:08 pm |
| On 6 May 2004 14:56:13 -0700, gary.l.scott@lmco.com (Gary Scott) wrote:
>Still, the title including the term "stealth source" is telling.
>Being an expert in such matters of course :-), I recommend a minor
>advertising/marketing makeover so that it is less stealthy (stealth is
>a good thing for my industry, but probably not for commercial
>products).
Again, this shows that either the article was written more than a year ago or
that the author has had their eyes closed for a long time. We have been
aggressively advertising the Intel compilers and other development tools in
catalogs (for example, Programmer's Paradise), magazines (Dr. Dobb's, MSDN)
and web sites (Slashdot, OSDN, Google AdWords (very effective!)) Resellers
outside the US have also been marketing the Intel compilers extensively. I
run across the ads frequently in my reading and web browsing.
We also promote at symposiums such as Linux Expo, Supercomputing and more. In
fact, I'm working on a Fortran compiler demo for Supercomputing right now. ( I
will almost certainly be attending this year's SC show in Pittsburgh. There's
a smaller chance I'll be at the International Supercomputing show in Germany
next month.)
No, you won't see compiler engineers dancing in bunny suits on TV, but there's
a lot of promotion going on and it is getting results.
Something must be working - the number of posts I've seen here about Intel
compilers has increased dramatically over the past year. We've also seen a
significant rise in traffic in our own user forums. This tells me that many
more people are using our compilers than in the past.
The article's comment on "stealth" would indeed have been accurate two years
ago. I recall watching in 2001 as a focus group was quizzed about software
development tools, initially without mention of a vendor. When the
facilitator revealed that they wanted to ask about Intel software development
tools, only one of the participants had ever used one (VTune) and one guy
steadfastly refused to believe that Intel sold software at all! It was very
enlightening - and depressing - and triggered some major changes.
Steve Lionel
Software Products Division
Intel Corporation
Nashua, NH
User communities for Intel Software Development Products
http://softwareforums.intel.com/
Intel Fortran Support
http://developer.intel.com/software/products/support/
| |
| Gary L. Scott 2004-05-12, 9:08 pm |
| Steve Lionel wrote:
>
> On 6 May 2004 14:56:13 -0700, gary.l.scott@lmco.com (Gary Scott) wrote:
>
>
> Again, this shows that either the article was written more than a year ago or
> that the author has had their eyes closed for a long time. We have been
> aggressively advertising the Intel compilers and other development tools in
> catalogs (for example, Programmer's Paradise), magazines (Dr. Dobb's, MSDN)
> and web sites (Slashdot, OSDN, Google AdWords (very effective!)) Resellers
> outside the US have also been marketing the Intel compilers extensively. I
> run across the ads frequently in my reading and web browsing.
>
> We also promote at symposiums such as Linux Expo, Supercomputing and more. In
> fact, I'm working on a Fortran compiler demo for Supercomputing right now. ( I
> will almost certainly be attending this year's SC show in Pittsburgh. There's
> a smaller chance I'll be at the International Supercomputing show in Germany
> next month.)
>
> No, you won't see compiler engineers dancing in bunny suits on TV, but there's
> a lot of promotion going on and it is getting results.
I like butterflies! I thought though that a "dancing dinosaur" might be
nice (re: article title "Dinosaurs Can Dance" (or something similar)
when DVF came out).
>
> Something must be working - the number of posts I've seen here about Intel
> compilers has increased dramatically over the past year. We've also seen a
> significant rise in traffic in our own user forums. This tells me that many
> more people are using our compilers than in the past.
Wonderful. Admittedly, my focus has been in other areas of late and I
wasn't paying as much attention.
--
Gary Scott
mailto:garyscott@ev1.net
Fortran Library: http://www.fortranlib.com
Support the Original G95 Project: http://www.g95.org
-OR-
Support the GNU GFortran Project: http://gcc.gnu.org/fortran/index.html
Why are there two? God only knows.
Democracy is two wolves and a sheep, voting on what to eat for dinner...
Liberty is a well armed sheep contesting the vote. - Thomas Jefferson
| |
| beliavsky@aol.com 2004-05-12, 9:08 pm |
| My unsolicited marketing advice to Steve Lionel and other Fortran 95
compiler vendors is to carefully study the Matlab product and
marketing. They are extremely successful in selling both the base
Matlab product and assorted toolboxes for a few $K each. Figure out
what makes people willing to pay so much for what is essentially
interpreted Fortran 95, with nice graphics and hooks to spreadsheets
and databases.
Intel Visual Fortran with IMSL 5 is not cheap at $1200, but it is
cheaper than Matlab alone, not to mention the additional cost of a
Matlab compiler and all the toolboxes you would need to buy to get the
functionality of IMSL and the ability to distribute executables. Tote
up the costs and compare the speed of the resulting executables -- I
have read that compiled Matlab code often runs more slowly than
interpreted Matlab.
(A Matlab salesmen was demonstrating the array functionality of Matlab
and claimed that "you can't do this in C or Fortran". He was startled
when I corrected him about Fortran 90/95.)
Also look at S-Plus. As the data sets people use keep growing, S-Plus
users should be willing to consider faster alternatives, but most do
not have the time, inclination, or ability to code the algorithms
themselves in Fortran. IMSL should continue to add statistical
capabilities to its library. The Fortran 95 interfaces in IMSL 5 and
the new statistical routines (described at
http://www.timberlake.co.uk/software/vni/imslmore.html for example)
are what made me shell out $800 of my own money to upgrade CVF Pro to
IVF Pro.
| |
| Gerry Thomas 2004-05-12, 9:08 pm |
|
<beliavsky@aol.com> wrote in message
news:3064b51d.0405071607.67610a28@posting.google.com...
> My unsolicited marketing advice to Steve Lionel and other Fortran 95
> compiler vendors is to carefully study the Matlab product and
> marketing. They are extremely successful in selling both the base
> Matlab product and assorted toolboxes for a few $K each.
From the latest NA digest:
"In the presence of Her Majesty, Queen Margrethe of Denmark,
Cleve Moler has been appointed Honorary Doctor (doctor technices,
honoris causa) at the Technical University of Denmark.
The degree was given at a ceremony on April 30.
It is motivated by Cleve's outstanding contributions in
research and software development, and especially for his
ground-breaking work with the software environment Matlab, which
has revolutionized the art of numerical computing, for the benefit
of scientists and engineers in a huge number of subjects throughout
the world."
> Figure out
> what makes people willing to pay so much for what is essentially
> interpreted Fortran 95, with nice graphics and hooks to spreadsheets
> and databases.
>
Matlab is not interpreted Fortran, Fortran 95 or otherwise.
Current Matlab is ensconsed in C/C++, whatever its humble origins and has
arrived: how's that jingle go?, 'movin' on up to the East Side'.
> Intel Visual Fortran with IMSL 5 is not cheap at $1200, but it is
> cheaper than Matlab alone, not to mention the additional cost of a
> Matlab compiler and all the toolboxes you would need to buy to get the
> functionality of IMSL and the ability to distribute executables. Tote
> up the costs and compare the speed of the resulting executables -- I
> have read that compiled Matlab code often runs more slowly than
> interpreted Matlab.
>
>
> Also look at S-Plus. As the data sets people use keep growing, S-Plus
> users should be willing to consider faster alternatives, but most do
> not have the time, inclination, or ability to code the algorithms
> themselves in Fortran. IMSL should continue to add statistical
> capabilities to its library. The Fortran 95 interfaces in IMSL 5 and
> the new statistical routines (described at
> http://www.timberlake.co.uk/software/vni/imslmore.html for example)
> are what made me shell out $800 of my own money to upgrade CVF Pro to
> IVF Pro.
The Fortran 95 interfaces in IMSL 5 is primarily of benefit to VNI; it does
squat for me. I see no evidence that VNI made any effort to advance Fortran
usage in any way through IMSL 2 (Microsoft FPS 4) - 4 (CVF 6)other than to
maintain their hegemony among a naive and undiscerning user base. That's
smart.
--
E&OE
Ciao,
Gerry T.
______
"A cynic is one who knows the price of everything and the value of
nothing." -- Oscar Wilde.
| |
| Gerry Thomas 2004-05-12, 9:08 pm |
|
"Steve Lionel" <Steve.Lionel@intel.com> wrote in message
news:soan90hr1kobmumc2fa3lpju70qq0mu6md@
4ax.com...
>
> Again, this shows that either the article was written more than a year
ago or
> that the author has had their eyes closed for a long time. We have been
> aggressively advertising the Intel compilers and other development tools
in
> catalogs (for example, Programmer's Paradise), magazines (Dr. Dobb's,
MSDN)
> and web sites (Slashdot, OSDN, Google AdWords (very effective!))
Resellers
> outside the US have also been marketing the Intel compilers extensively.
I
> run across the ads frequently in my reading and web browsing.
>
No slight on Intel intended (well only a little!), but throughout the 'Two
Years in the Making ..." of IVF 8 its Canadian Programmer's Paradise
catalog entry display (the ads copy was fine) was for Intel C++ Compiler.
It's OK now. HP seems to have discontinued promoting CVF in the same
catalog.
--
E&OE
Ciao,
Gerry T.
| |
| Gerry Thomas 2004-05-12, 9:09 pm |
|
"Steve Lionel" <Steve.Lionel@intel.com> wrote in message
news:soan90hr1kobmumc2fa3lpju70qq0mu6md@
4ax.com...
>
> Something must be working - the number of posts I've seen here about
Intel
> compilers has increased dramatically over the past year.
>
Perhaps it's high time to rein in IFC-specific (sic Linux) posts to their
rightful paddock, the Intel forum(s), as so often has been customary for
CVF/IVF-related (sic Windows) posts.
Not only does Intel generously provide IFC for the qualified, but free
support is also thrown in, so what more do they want?. c.l.f. isn't the
place to air IFC frustrations: Intel is way and by far the most efficient
route for such support in contrast to that provided by the usual jackals
herein.
--
E&OE
Ciao,
Gerry T.
______
"We are all in this together son!" -- Harry Tuttle
| |
| Eric K. 2004-05-12, 9:09 pm |
| beliavsky@aol.com wrote in message news:<3064b51d.0405071607.67610a28@posting.google.com>...
> My unsolicited marketing advice to Steve Lionel and other Fortran 95
> compiler vendors is to carefully study the Matlab product and
> marketing. They are extremely successful in selling both the base
> Matlab product and assorted toolboxes for a few $K each. Figure out
> what makes people willing to pay so much for what is essentially
> interpreted Fortran 95, with nice graphics and hooks to spreadsheets
> and databases.
Both Matlab and Intel Fortran are fine products, but they address different
needs. Matlab is especially good for throwaway programs, i.e., cases
where people need to get something running with minimal programmer time.
Many undergraduate engineering programs have standardized on Matlab, because
the Mathworks offers attractively-priced site licenses to universities, and
these students naturally prefer to continue with Matlab when they go into
industry.
--Eric
| |
| Steve Lionel 2004-05-12, 9:09 pm |
| On Sun, 9 May 2004 04:05:57 -0400, "Gerry Thomas" <gfthomas@sympatico.ca>
wrote:
>Perhaps it's high time to rein in IFC-specific (sic Linux) posts to their
>rightful paddock, the Intel forum(s), as so often has been customary for
>CVF/IVF-related (sic Windows) posts.
>
>Not only does Intel generously provide IFC for the qualified, but free
>support is also thrown in, so what more do they want?. c.l.f. isn't the
>place to air IFC frustrations: Intel is way and by far the most efficient
>route for such support in contrast to that provided by the usual jackals
>herein.
I agree, but I don't know how to get people to stop posting such things here.
I do what I can to promote our user forum (it's mentioned in my sig.) Once
people find the forum, they are fine, but the newsgroups seem to be the first
choice of many looking for vendor-specific support.
Steve Lionel
Software Products Division
Intel Corporation
Nashua, NH
User communities for Intel Software Development Products
http://softwareforums.intel.com/
Intel Fortran Support
http://developer.intel.com/software/products/support/
| |
| Gary L. Scott 2004-05-12, 9:09 pm |
| Steve Lionel wrote:
>
> On Sun, 9 May 2004 04:05:57 -0400, "Gerry Thomas" <gfthomas@sympatico.ca>
> wrote:
>
>
> I agree, but I don't know how to get people to stop posting such things here.
> I do what I can to promote our user forum (it's mentioned in my sig.) Once
> people find the forum, they are fine, but the newsgroups seem to be the first
> choice of many looking for vendor-specific support.
Part of it may be that news group facilities are simpler to use. I find
most forum software somewhat annoying to use and so avoid it when I
can. I dislike mailbase style even more because of the "unsolicited"
emails.
>
> Steve Lionel
> Software Products Division
> Intel Corporation
> Nashua, NH
>
> User communities for Intel Software Development Products
> http://softwareforums.intel.com/
> Intel Fortran Support
> http://developer.intel.com/software/products/support/
--
Gary Scott
mailto:garyscott@ev1.net
Fortran Library: http://www.fortranlib.com
Support the Original G95 Project: http://www.g95.org
-OR-
Support the GNU GFortran Project: http://gcc.gnu.org/fortran/index.html
Why are there two? God only knows.
Democracy is two wolves and a sheep, voting on what to eat for dinner...
Liberty is a well armed sheep contesting the vote. - Thomas Jefferson
| |
| Christian 2004-05-12, 9:09 pm |
| Eric,
You might want to check out Intel's offerings for education at
http://www.intel.com/software/produ...l/academic.htm. Some
resellers sell our tools at $30 to students and classroom licenses for
educational institutions go as low as $15 per license depending the
quantity.
Christian
Intel Corp.
ejko123@yahoo.com (Eric K.) wrote in message news:<4ccbcc1c.0405092215.339ff8f5@posting.google.com>...
> beliavsky@aol.com wrote in message news:<3064b51d.0405071607.67610a28@posting.google.com>...
>
> Both Matlab and Intel Fortran are fine products, but they address different
> needs. Matlab is especially good for throwaway programs, i.e., cases
> where people need to get something running with minimal programmer time.
> Many undergraduate engineering programs have standardized on Matlab, because
> the Mathworks offers attractively-priced site licenses to universities, and
> these students naturally prefer to continue with Matlab when they go into
> industry.
>
> --Eric
| |
| Gerry Thomas 2004-05-12, 9:09 pm |
|
"Eric K." <ejko123@yahoo.com> wrote in message
news:4ccbcc1c.0405092215.339ff8f5@posting.google.com...
> beliavsky@aol.com wrote in message
news:<3064b51d.0405071607.67610a28@posting.google.com>...
>
> Both Matlab and Intel Fortran are fine products, but they address
different
> needs. Matlab is especially good for throwaway programs, i.e., cases
> where people need to get something running with minimal programmer time.
> Many undergraduate engineering programs have standardized on Matlab,
because
> the Mathworks offers attractively-priced site licenses to universities,
and
> these students naturally prefer to continue with Matlab when they go into
> industry.
>
You don't necessarily have to throw away your Matlab projects: they can be
compiled to C/C++ standalones (exe's, DLL's, and COM's) but you'll have to
have the costly tools to do so. Also, for heretics, there's
https://sourceforge.net/projects/f2matlab/.
--
E&OE
Ciao,
Gerry T.
______
"Facts are meaningless. You could use facts to prove anything that's even
remotely true." -- Homer Simpson.
| |
| Dr Chaos 2004-05-12, 9:09 pm |
| On Mon, 10 May 2004 10:11:40 -0400, Steve Lionel <Steve.Lionel@intel.com> wrote:
> On Sun, 9 May 2004 04:05:57 -0400, "Gerry Thomas" <gfthomas@sympatico.ca>
> wrote:
>
>
> I agree, but I don't know how to get people to stop posting such things here.
> I do what I can to promote our user forum (it's mentioned in my sig.) Once
> people find the forum, they are fine, but the newsgroups seem to be the first
> choice of many looking for vendor-specific support.
One of the main reasons to post here (comp.lang.fortran) is to read a
wider variety of highly educated opinions. And a wider variety of
vigorously asserted opinions. :)
| |
| Paul Van Delst 2004-05-12, 9:09 pm |
| Dr Chaos wrote:
> On Mon, 10 May 2004 10:11:40 -0400, Steve Lionel <Steve.Lionel@intel.com> wrote:
>
>
>
> One of the main reasons to post here (comp.lang.fortran) is to read a
> wider variety of highly educated opinions. And a wider variety of
> vigorously asserted opinions. :)
I think the problems (or successes) people have with various compilers is important
information for prospective buyers. I don't have the time (or the inclination for that
matter) to test a bunch of different compilers to see how easy they are to install, or
whether they'll install at all, etc etc so I appreciate others filling me in on their
efforts. Both the highly educated and vigorously asserted ones. :o)
cheers,
paulv
| |
| Dick Hendrickson 2004-05-12, 9:09 pm |
|
Dr Chaos wrote:
> On Mon, 10 May 2004 10:11:40 -0400, Steve Lionel <Steve.Lionel@intel.com> wrote:
>
>
>
> One of the main reasons to post here (comp.lang.fortran) is to read a
> wider variety of highly educated opinions. And a wider variety of
> vigorously asserted opinions. :)
Gee, aren't those normally the same?
Dick Hendrickson
| |
| Richard Maine 2004-05-12, 9:09 pm |
| Dick Hendrickson <dick.hendrickson@att.net> writes:
> Dr Chaos wrote:
>
> Gee, aren't those normally the same?
I tend to notice an inverse correlation. The more careful someone
is to qualify their posts as being their own evaluation instead of
divine truth, the more likely I am to trust that their evaluation
as accurate. But then, that's just a general reflection of
professional technical writing style.
(Yes, I saw the smiley...and I also understood the implied one in
Dick's reply..)
--
Richard Maine | Good judgment comes from experience;
email: my first.last at org.domain | experience comes from bad judgment.
org: nasa, domain: gov | -- Mark Twain
|
|
|
|
|