Home > Archive > Fortran > April 2005 > MathCode F90: generating optimized Fortran 90 Code from Mathematica
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 |
MathCode F90: generating optimized Fortran 90 Code from Mathematica
|
|
| beliavsky@aol.com 2005-04-06, 12:21 pm |
| The Wolfram Research MATHwire recently announced the product MathCode
F90. Here are some details from
http://www.wolfram.com/products/app...ns/mathcodef90/ .
"The MathCode F90 code-generation system provides high performance,
connectivity, and easy-to-use matrix arithmetic for the Mathematica
developer. The core of the product is the translation of a subset of
Mathematica code to Fortran 90.
MathCode F90 requires an external Fortran 90 compiler to generate
executables from the Fortran 90 source code (Digital Visual Fortran or
Intel Visual Fortran). However, all interaction with the compiler is
handled by functions available in MathCode F90."
I wonder why the product does not work with every F90/95 compiler. The
U.S. commercial version costs $3800, with a 30% discount through June
30, 2005.
It may surprise a certain sci.math.num-analysis poster that Mathematica
and Fortran are complementary languages.
| |
| glen herrmannsfeldt 2005-04-06, 12:21 pm |
| beliavsky@aol.com wrote:
> The Wolfram Research MATHwire recently announced the product MathCode
> F90. Here are some details from
> http://www.wolfram.com/products/app...ns/mathcodef90/ .
> "The MathCode F90 code-generation system provides high performance,
> connectivity, and easy-to-use matrix arithmetic for the Mathematica
> developer. The core of the product is the translation of a subset of
> Mathematica code to Fortran 90.
> MathCode F90 requires an external Fortran 90 compiler to generate
> executables from the Fortran 90 source code (Digital Visual Fortran or
> Intel Visual Fortran). However, all interaction with the compiler is
> handled by functions available in MathCode F90."
> I wonder why the product does not work with every F90/95 compiler. The
> U.S. commercial version costs $3800, with a 30% discount through June
> 30, 2005.
The calling sequence would have to agree with what Mathematica expects.
They have had before systems which would dynamically link a user
supplied routine. If the DLL calling methods agree it should be
possible to make it work. It might also be that Mathematica can
directly call the compiler with appropriate options. A simple script
file should be able to fix that.
-- glen
| |
| Gerald F. Thomas 2005-04-06, 12:21 pm |
|
"glen herrmannsfeldt" <gah@ugcs.caltech.edu> wrote in message
news:LamdncUHqPgzI83fRVn-gg@comcast.com...
> beliavsky@aol.com wrote:
>
>
>
>
>
> The calling sequence would have to agree with what Mathematica expects.
> They have had before systems which would dynamically link a user
> supplied routine. If the DLL calling methods agree it should be
> possible to make it work. It might also be that Mathematica can
> directly call the compiler with appropriate options. A simple script
> file should be able to fix that.
>
> -- glen
>
You've got to be the biggest bull shitter since jesus christ, maybe more.
Isn't there a single post that you can spare CLF from your vacuous and
nauseous ad hominids? I guess not! Duboubtedly Wolfram has been waiting for
this, :-).
Get a fcuking life or perhaps you're post is ex --- [fill in the blank from
your mind, :-)] .
--
Ciao,
Gerry T.
--
As a native of the snow belt of upstate New York, I too claim the feisty
independence of the Northerner. I was raised, I like to say, breathing
cold, clear Canadian air. Camille Paglia.
|
|
|
|
|