home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!ogicse!uwm.edu!linac!convex!convex!joelw
- From: joelw@convex.com (Joel Williamson)
- Newsgroups: comp.lang.fortran
- Subject: Re: Welcome to HPFF
- Keywords: HPFF, High Performance Fortran
- Message-ID: <1992Aug14.181229.12520@news.eng.convex.com>
- Date: 14 Aug 92 18:12:29 GMT
- Article-I.D.: news.1992Aug14.181229.12520
- References: <Bsy43C.9BG@rice.edu> <1992Aug14.092952.11572@ibr.cs.tu-bs.de>
- Sender: usenet@news.eng.convex.com (news access account)
- Organization: Engineering, CONVEX Computer Corp., Richardson, Tx., USA
- Lines: 59
- Nntp-Posting-Host: mozart.convex.com
- X-Disclaimer: This message was written by a user at CONVEX Computer
- Corp. The opinions expressed are those of the user and
- not necessarily those of CONVEX.
-
- In article <1992Aug14.092952.11572@ibr.cs.tu-bs.de> i2041101@rzphy1.rz.tu-bs.de (Paulini) writes:
- >In article <Bsy43C.9BG@rice.edu>, chk@rice.edu (Charles Koelbel) writes:
- >|>
- >|> Welcome to the High Performance Fortran Forum!
- >|>
- >|>
- >|> WHAT IS HPFF?
- >|> =============
- >|>
- >|> ..... deleted ....
- >|>
- >|> Our intent is to develop extensions to Fortran which
- >|> provide support for high performance programming on a wide variety of
- >|> machines, including massively parallel SIMD and MIMD systems and
- >|> vector processors. Some of the topics considered will be
- >|>
- >|> ..... deleted ....
- >
- >Stop it! I'm programming now for several years scientific applications
- >in FORTRAN 77 because I must do it, due to libraries and already existing
- >programs in FORTRAN. Compared to other languages (even to cryptic C - Codes)
- >FORTRAN is not recommendable as a programming language at all. It is merely
- >a collection of erratic ideas and lacks a real conception. In my opinion it is
- >not useful to make any extensions or "improvements" to the language (this also
- >holds true for Fortran 90). A better way seems to me to design a comptetely
- >new language or to use an existing better (like C, Ada, Modula or whatever).
- >And then to decide no longer to support programming in FORTRAN but to use
- >existing FORTRAN codes in the new language in a portable way (e.g. by defining
- >an interface or a standard linking method or something similar).
- >
- >The situation now is that people who are used to a 'real' programming language
- >(mostly C) only write programs in FORTRAN if there is no other way ... like me.
- >If in some years most of the numeric software is available in C, all new
- >numeric programmers won't code in FORTRAN and many of the old will convert too.
- >
- >
- >--
- >#-----------------------------------------------------------------------------#
- ># Joachim Paulini #
- ># Tel (0531) 391-5190 #
- ># TU Braunschweig Fax (0531) 391-5833 #
- ># Institut f. Theoretische Physik e-mail i2041101@ws.rz.tu-bs.de #
- ># Mendelssohnstrasse 3 i2041101@dbstu1.bitnet #
- ># 3300 Braunschweig / Germany #
- >#-----------------------------------------------------------------------------#
-
- This debate has gone on for decades now, and the scientific marketplace
- continues to choose Fortran. There are literally billions of lines of
- existing Fortran code, and their owners have a strong predilection to
- continue to run them on modern machines. When the proponents of the
- better languages offer to 1) convert all existing code to the new
- language; 2) assure that equal or better performance is attained; and 3)
- re-educate (for free) all of the Fortran programmers so that they are
- proficient in the new language, then perhaps the marketplace will vote
- differently. All of the above may occur naturally in the next sevral
- decades, but until then it is a losing position to promote.
-
- Joel Williamson
- --
-