[18863] in s-news-athena
[S] Watcom Compiler, Fortran, and S-Plus
daemon@ATHENA.MIT.EDU (Michael Prager)
Thu Aug 5 09:14:08 1999
Message-Id: <4.2.0.58.19990804104909.00a43530@mail.bea.nmfs.gov>
Date: Thu, 05 Aug 1999 09:09:12 -0400
To: s-news@wubios.wustl.edu
From: Michael Prager <Mike.Prager@noaa.gov>
In-Reply-To: <199908040630.BAA09376@wubios.wustl.edu>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
A few words about Fortran. I am conscious this is not a Fortran group, so
I will keep them brief.
Fred Balch recently wrote:
>> I found, at the Sybase site, that version 11 of Watcom's FORTRAN is
FORTRAN 90. <<
I believe that statement to be incorrect: the formal product name is
"Watcom FORTRAN 77 11.0". Unless Watcom made a stealthy last-minute
upgrade, it is FORTRAN 77 with some capabilities of Fortran 90 added, but
not many of the most useful and powerful ones.
>> Most of the additional capabilities of FORTRAN [90/95] appear to be
with regard to the stand-alone capability of the language (intrinsic
arrays) and are not related to the interface. <<
For the record: Fortran 90 was a major revision, with large improvements in
the following areas: code modularity and reusability, specification and
control of precision, inclusion of many common extensions that facilitate
error detection and structured programming, introduction of intrinsic array
operations, changes to improve optimization of parallel processing,
operator overloading, introduction of pointers, dynamic memory management
(allocatable arrays), and better file handling. In addition, there are
finally standard date and time functions. Fortran 95 was a minor revision
of Fortran 90, mainly to further facilitate high-performance computing and
parallel processing. The name of the language from the 1990 standard
onwards is "Fortran," not "FORTRAN."
Like Fred, I wonder what compiler MathSoft will use for new versions of
S-PLUS for Windows and how MathSoft intends to maintain and improve the
Fortran/S-PLUS interface. I was disappointed to read (between the lines)
that the DLL interface is troublesome, as otherwise it would seem to offer
a compiler-independent route to integration.
--
Michael Prager <Mike.Prager@noaa.gov>
NOAA Center for Coastal Fisheries and Habitat Research
Beaufort, North Carolina 28516 USA
Standard Disclaimers: Opinions expressed here are personal
and are not otherwise represented. Any use of tradenames
in the above does not constitute a NOAA endorsement.
-----------------------------------------------------------------------
This message was distributed by s-news@wubios.wustl.edu. To unsubscribe
send e-mail to s-news-request@wubios.wustl.edu with the BODY of the
message: unsubscribe s-news
