In Defence of Fortran
When I teach a course which requires extensive numerical computation, I expect my students to write their codes in Fortran. The reason for this is that the Fortran continues to be the nearly universal language of scientific computing. It seems to me worthwhile to force the students to learn something about this important scientific computing language. For those who think that the Fortran is going to disappear, let me remind you that the demise of Fortran has been predicted for many years. As we all know, Fortran has survived and flourished, while languages like PL/I and Pascal which were developed during the same period, never lived up to its initial expectations.
Fortran 90
Fortran 90 is a general-purpose programming langauge that provides superior facilities for dealing with numerical data, and it is far and away the best language for most applications that are dominated by mathematical, engineering, or scientific analysis. Of course, Fortran 90 supports a free-source form, as well as modern control-flow and data structures, but it offers much more. Fortran 90's array-handling capabilities are outstanding, it provides a rich collection of intrinsic procedures.
Also, Fortran 90 provides excellent mechanisms for partitioning large programs into smaller, more manageable pieces. Further more, it supports features that facilitates data hiding and data abstraction. Yet even with these dramatic enhancements to previous versions of Fortran, an exceptionally high degree of backward compatibility is maintained. However, the price that is paid for all this is that Fortran 90 is a sprawling, untidy language whose complexity is roughly comparable to that of Ada or C++.
An argument for Fortran 90 instead of C++
A word about F
F is designed to be a subset of Fortran 90 containing only modern features. Fortran has evolved over more than 40 years by constantly adding new, modern features; for compatibility with older versions, however, few obsolete features have ever been removed.
F retaining the modern features of Fortran - modules and data abstraction, for example - but discards those which are obsolete, or which are difficult to teach or use. Thus, F contains no features categorized as obsolescent in the Fortran 90 standard. Although every F program is also a Fortran program, F is more than just a subset. It is a safe and portable language that is ideal for programming style, which encourages module-oriented programming.
Ftnchek
Ftnchek (short for Fortran checker) is intended to assist users in the debugging of their Fortran programs. It is designed to detect certain errors in a Fortran program that a compiler usually does not. ftnchek is not primarily intended to detect syntax errors. Its purpose is to assist the user in finding semantic errors. Semantic errors are legal in the Fortran language but are wasteful or may cause incorrect operation. For example, variables which are never used may indicate some omission in the program; uninitialized variables contain garbage which may cause incorrect results to be calculated; and variables which are not declared may not have the intended type. ftnchek is not intended to catch all syntax errors. This is the function of the compiler. Prior to using ftnchek, the user should verify that the program compiles correctly.
Ftnchek is written in C. You must have a C compiler for the machine on which you wish to build it. This is a free ware and different versions are available for different platforms. The URL for ftnchek's home page is http://www.dsm.fordham.edu/~ftnchek. Source code, binary executables for some platforms, and other information are available there.
Numerical Recipes in Fortran Here are some of the Fortran routines that I have developed over a period of time. If you find any of these useful, click on the links below to see the source code and then download.
 |
Algebraic and Transcendental Equations
|
|
System of Linear Equations
|
|
System of Nonlinear Equations
|
|
Interpolation Routines
|
|
Regression Analysis and Curve Fitting
|
|
Numerical Integration
|
|
Numerical Solution of Ordinary Differential Equations
|
|
Eigenvalue Routines
|
|
Special Functions
|
|
