next up previous
Next: About this document

COMPUTATIONAL PHYSICS

Nicholas J. Giordano

During the past several years I have worked to develop an undergraduate course on computational physics. The goal of this course has been to introduce students to some basic numerical techniques and then apply these techniques to a number of modern topics, that is, problems of current interest to physicists. I have long felt that the traditional introductory physics courses place too much emphasis on topics which have remained unchanged for a century or more. Very little effort is made to introduce the types of problems that today's physicists find interesting or to convey the excitement we feel in our work. A standard and perhaps somewhat justifiable excuse given for this shortcoming is that to treat such topics would require a level of mathematical expertise far beyond that of a typical student. It is here that computers can make a qualitative difference. Students with some experience in differential and integral calculus can readily grasp rather sophisticated computational techniques. These students can use computers as tools with which to attack and solve problems which they would not ordinarily encounter in the undergraduate curriculum. I have used this approach to try to convey the excitement of physics, with a variety of problems of current interest.

After I had formulated tentative plans for my course, I was faced with actually putting it together, and an early step in this process was to find a textbook. While there are many new books with the terms ``computers'' and ``physics'' in their titles, I was not able to find a text that was suitable for my needs. Most of the books in this area tend to focus heavily on numerical methods rather than physics. Since my goal was to teach a course on physics, rather than numerical methods, these books were not a good match. While there are a few books that emphasize the physics that can be done with numerical methods, they are either too advanced for use by undergraduates, or (more commonly) they fail to deal with the types of problems that can profit most from a numerical approach. In too many cases they tend to simply treat the standard problems, which are already dealt with in many traditional texts using analytic methods.

I was therefore forced to gather the material for my course from a wide variety of ``primary'' sources, as will become clear from the references at the end of each chapter. In many cases I started with papers from the recent physics literature and then ``distilled'' them to produce problems suitable for an undergraduate class. This book contains the results of my efforts. While it is necessary for this book to introduce a variety of numerical methods of interest to physicists, the overriding emphasis is on the physics that can be done with these methods. The majority of the problems described in this book cannot be solved with purely analytic techniques. A computational approach is required in most cases, and I have tried to use the computer to make the physics as clear and as interesting as possible.

How to use this book

When my course was in the design stages, one constraint imposed by local realities (i.e., space in our sequence of courses) was that it fit into the early part of the undergraduate curriculum. My course is taken mainly by sophomores whose only previous physics course is on introductory mechanics. Their mathematics backgrounds vary widely. It usually includes a total of two semesters of calculus, so one can assume only that they have been exposed to the basics of differentiation and integration. This background is easily sufficient for a wide variety of topics. My course currently includes projectile motion without and with realistic air resistance, planetary motion and the 3-body problem, chaotic motion of the pendulum, waves on a string, and some topics relating to fractal growth and stochastic systems. This material is covered in Chapters 1-4, 6 and 7, although these chapters contain many ``extra'' topics and examples, which time does not allow me to discuss in the course. In fact, this book contains far more material than could possibly be covered in a one semester course. This ``extra'' material is included to give the student (and instructor) some flexibility in the choice of topics and depth of coverage. I have used many of these extra problems for midterm or final projects. This book could also be used as a supplement to other courses. For example, a course in electricity and magnetism could make extensive use the material in Chapter 5, while an advanced mechanics course might find the examples on chaotic dynamics and waves on a realistic string useful. While a course aimed at advanced undergraduates could certainly make use of the material in the first few chapters, the emphasis would probably be on the examples from statistical physics and quantum mechanics in Chapters 8-10. Chapter 11 brings together ideas developed earlier in the book to deal with three interdisciplinary topics and could be of interest to students at virtually any level.

It is obvious that numerical methods play a major role in this book, but given my desire to place the primary emphasis on physics, there is simply not enough space to describe all of the subtleties associated with all of the numerical methods in full detail. Fortunately, there are several excellent books in this area (as mentioned in the references), so the student or instructor who desires an in-depth discussion can readily find one. However, this does not mean that I have cut corners with my treatment of numerical methods. I am perhaps a bit old fashioned, as I firmly believe that students should write their own original programs. This requires a careful treatment of the numerical methods, and this book contains ample explanation of the necessary algorithms. I believe that this book contains all the information that a student would need to write programs that perform all of the calculations that are described here.

While I am perhaps, as just admitted, old fashioned in my belief that students should write their own programs, I have relented to the requests of those who have refereed this book (and my publisher!), and provided many example programs. Some of these program listings are sprinkled throughout the book, in connection with discussions of the various problems and algorithms. In addition, Appendix 4 contains full listings of a number of representative programs; these listings are included on the diskette that accompanies the book and are also available though the Prentice Hall World Wide Web page. However, this book contains many example calculations, and most of these programs are not included. After all, some work must be left for the student.

The inclusion of source code leads naturally to the question of language. My course currently uses the True Basic language system, and all of the program listings are in this dialect of Basic. I fully realize that many readers will prefer some other language. My course has, at various times, used Pascal, C, and Fortran. However, if forced to choose one ``common denominator'' language, I feel that Basic is a good choice. Moreover, True Basic has available an excellent library of graphics routines, and can be used with a variety of machines. I have used it on IBM-compatibles, Macintoshes, Amigas, and Suns, and it is available for other machines as well. It is a well structured language (you will not find a single go to statement in this book), and in nearly all cases it can easily be translated, usually line-for-line, into the other languages mentioned above. In addition, it is quite readable as a ``pseudo-code'', so even students who prefer to work with another language should be able to take full advantage of the examples. In any event, I certainly understand that the choice of computer language can be a highly personal one, and I do not want to imply that any particular language is better than any other for the numerical work encountered in physics. My intent with this book has been to make the ideas underlying the numerical methods as clear as possible, so that students can use the language of their choice, be it C or Fortran or Mathematica or Maple.

This book contains a rather large number of example calculations. The pattern of my presentation is to show the results of specific computations as I introduce and discuss the physics. All of the example calculations in this book have been tested by the author using True Basic. The vast majority can be carried out in a reasonable amount of time (a few minutes or, usually, much less) with a Macintosh containing no more than a 68020 CPU. By the time this book reaches print, such machines will be hard to find, and most microcomputers will be faster by an order of magnitude or more. Hence, computational resources should not be a limitation in working out the problems in this book.

I owe many thanks to many people. The support of Arnold Tubis and the Department of Physics at Purdue, along with that of the National Science Foundation, made my course and hence this book possible. Many graduate students have helped me teach and develop the course; Miguel Castro, Chris Parks, Jan Spitz, Stuart Burnett, Todd Jacobs, and Dan Lawrence have taught (and corrected) me a great deal. Of course, the undergraduate students who have willingly submitted to the course have provided much useful feedback; there are too many to mention them all here, although Mike Pennington deserves a special thanks. Many colleagues have provided essential advice and encouragement, including Mark Haugen and Paul Muzikar. I am especially indebted to Paul and Todd for their detailed comments on the manuscript. The suggestions of the referees, Alejandro Garcia, Jan Tobochnik, and Rodney Varley (who were quite polite and constructive) have helped me a lot, as have the comments and advice of my editors, Ray Henderson and Alison Reeves. The final impetus to actually write this book was provided by the well-timed encouragement of Earl Prohofsky and Betsy Beasley. And finally, I thank my wife and children for helping in many ways, including numerous dinner-time discussions of the physics of everyday phenomena and the travels of Lucky Star.




next up previous
Next: About this document

Nick Giordano
Fri Sep 5 11:21:18 EST 1997