home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.lang.fortran
- Path: sparky!uunet!usc!rpi!batcomputer!msiadmin.cit.cornell.edu!bai
- From: bai@msiadmin.cit.cornell.edu (Dov Bai-MSI Visitor)
- Subject: Re: Robustness of MS Fortran 5.x & Vax extensions?
- Message-ID: <1992Sep5.121959.2311@tc.cornell.edu>
- Sender: news@tc.cornell.edu
- Nntp-Posting-Host: msiadmin.cit.cornell.edu
- Organization: /usr/local/lib/news/organization
- References: <Bu1CF5.HqL@news.cso.uiuc.edu>
- Date: Sat, 5 Sep 1992 12:19:59 GMT
- Lines: 86
-
- In article <Bu1CF5.HqL@news.cso.uiuc.edu> btbg1194@uxa.cso.uiuc.edu (Bradley T Banko) writes:
- >I am starting on a project to convert some several thousand line
- >RSX11/VMS Vax Fortran programs to run on a PC, and I am planning to
- >use MS Fortran 5.1. I would appreciate comments about MSF 5.x's
- >robustness in dealing with Vax code... including special types like
- >"BYTE" and out of order DATA/COMMON statements.
- >
- >Any comments are welcome. Thanks.
- >
- >Brad Banko
- >b-banko@uiuc.edu
- >
- >
- >--
- >Brad Banko; Dept of Physics; U of Illinois; b-banko@uiuc.edu, KB8CNE
- >=========================================================================
- > Tatiana Gutsu for President!
-
-
- Here is the list of MS-FORTRAN 5.0 VAX and IBM extensions. This is
- from my manual:
-
- Microsoft Fortran includes many (but not all) VAX extensions, which
- are listed below:
-
- * 31 character names
- * ACCESS selector 'APPEND' in the OPEN statements
- * allowing integer arrays to contain FORMAT statements
- * BLOCKSIZE and NML I/O keywords
- * Debug comment lines
- * DO statements without specified labels
- * DO WHILE statement
- * END DO statement
- * EQUIVALENCE of character and noncharacter items
- * EQUIVALENCE to a multi-dimensioned array with only one subscript
- * exponentiation of REAL and COMPLEX statements
- * IMPLICIT NONE
- * INCLUDE compiler directive
- * Initialization on the declaration line
- * In-line comments
- * Length specification within the FUNCTION statement
- * Length specifications within type declarations
- * Mixing of character and noncharacter items in COMMON statements
- * Noninteger alternate return values
- * Noninteger array subscripts
- * Numeric operands for .AND., .OR. etc.
- * Specified common-block variables in DATA statements outside a BLOCK
- DATA subroutine
- * STRUCTURE, UNION, MAP statements
- * UP to 99 continuation lines
- * Use of dollar sign ($) in names
- *.XOR. operator
-
-
- MS-Fortran includes all IBM SAA extensions which are listed below:
-
- * 31-character names
- * Bit-manipulation intrinsic
- * Case-insensitive source
- * COMMON allows character and noncharacted in same block
- * CONJG, HFIX and IMAG intrinsic functions
- * Data initialization in type statements
- * EQUIVALENCE allows association of character and noncharacter
- * Functions IOR, IAND, NOT, IEOR, ISHFT, BTEST, IBTEST, and IBCLR
- * IMPLICIT NONE
- * INCLUDE compiler directive
- * INTEGER*2, COMPLEX*16, and LOGICAL*1 data types
- * optional length specifications in INTEGER, REAL, COMPLEX, and LOGICAL
- type statements
- * Use of underscore (_) in names
- * Z edit descriptor
-
- ----------------------------- end list ------------------------
-
- I have been working with MS-FORTRAN 5.0 for few years and has
- been quite satisfied. With the HUGE memory model you can have
- arrays which are bigger than 64K. I have been working with it
- under OS/2 (it works under both MS-DOS and OS/2) on a large
- fluid dynamics code.
-
- Under MS-DOS you have of course the 640 limit, but under OS/2
- you can address at least 16MB ( depends on the version)
-
- The main problem with the compiler is that it is only a 16-bit
- and therefore is much slower than 32-bit compilers.
-
-