Preface
Seventeen years have past since Wiley first published Landau and Páez’s Computational Physics and twelve years since Cristian Bordeianu joined the collaboration for the second edition. This third edition adheres to the original philosophy that the best way to learn computational physics (CP) is by working on a wide range of projects using the text and the computer as partners. Most projects are still constructed using a computational, scientific problem-solving paradigm:
Our guiding hypothesis remains that CP is a computational science, which means that to understand CP you need to understand some physics, some applied mathematics, and some computer science. What is different in this edition is the choice of Python for sample codes and an increase in the number of topics covered. We now have a survey of CP which is more than enough for a full-year’s course.
The use of Python is more than just a change of language, it is taking advantage of the Python ecosystem of base language plus multiple, specialized libraries to provide all computational needs. In addition, we find Python to be the easiest and most accessible language for beginners, while still being excellent for the type of interactive and exploratory computations now popular in scientific research. Furthermore, Python supplemented by the Visual package (VPython) has gained traction in lower division physics teaching, and this may serve as an excellent segue to a Python-based CP course. Nevertheless, the important aspects of computational modeling and thinking transcends any particular computer language, and so having a Python alternative to our previous use of Fortran, C and Java may help promote this view (codes in all languages are available).
As before, we advocate for the use of a compiled or interpreted programming language when learning CP, in contrast to a higher level problem-solving environment like Mathematica or Maple, which we use in daily work. This follows from our experiences that if you want to understand how to compute scientifically, then you must look inside a program’s black box and get your hands dirty. Otherwise, the algorithms, logic, and the validity of solutions cannot be ascertained, and that is not a good physics. Not surprisingly, we believe all physicists should know how to read programs how to write them as well.
Notwithstanding our beliefs about programming, we appreciate how time-consuming and frustrating debugging programs often is, and especially for beginners. Accordingly, rather than make the learner write all codes from scratch, we have placed a large number of codes within the text and often ask the learner only to run, modify, and extend them. This not only leaves time for exploration and analysis, but also provides experience in the modern work environment in which one must incorporate new developments into the preexisting codes of others. Be that as it may, for this edition we have added problems in which the relevant codes are not in the text (but are available to instructors). This should permit an instructor to decide on the balance of new and second-hand codes with which their students should work.
In addition to the paper version of the text, there is also an eBook of it that incorporates many of the multimodal enhancements possible with modern technologies: video lecture modules, active simulations, editable codes, animations, and sounds. The eBook is available as a Web (HTML5) document appropriate for both PCs or mobile devices. The lecture modules, which can be viewed separately from the eBook, cover most of the topics in the text, are listed in Appendix B, and are available online. They may provide avenues for alternative understanding the text (either as a preview or a review), for an online course, or for a blended course that replaces some lecture time with lab time. This latter approach, which we recommend, provides time for the instructor to assist students more personally with their projects and their learning issues. The studio-produced lectures are truly “modules," with active slides, a dynamic table of context, excellent sound (except maybe for a Bronx accent), and with occasional demonstrations replacing the talking head.
The introductory chapter includes tables listing all of the problems and exercises in the text, their locations in the text, as well as the physics courses in which these problems may be used as computational examples. Although we think it better to have entire courses in CP rather than just examples in the traditional courses, the inclusion of examples may serve as a valuable first step towards modernization.
The entire book has been reedited to improve clarity and useability. New materials have also been added, and this has led to additional and reorganized chapters. Specific additions not found in the second edition include: descriptions of the Python language and its packages, demonstrations of several visualization packages, discussions of algebraic tools, an example on protein folding, a derivation of the Gaussian quadrature rule, searching to obtain the temperature dependence of magnetization, chaotic weather patterns, planetary motion, matrix computing with Numerical Python, expanded and updated discussion of parallel computing including scalability and domain composition, optimized matrix computing with NumPy, GPU computing, CUDA programming, principal components analysis, digital filtering, the fast Fourier transform (FFT), an entire chapter on wavelet analysis and data compression, a variety of predator–prey models, signals of chaos, nonlinear behavior of double pendulum, cellular automata, Perlin noise, ray tracing, Wang–Landau sampling for thermodynamic simulations, finite element (in addition to difference) solutions of 1D and 2D PDEs, waves on a catenary, finite-difference-time-domain solutions for E&M waves, advection and Shock waves in fluids, and a new chapter on fluid dynamics. We hope you enjoy it all!
Redmond, Oregon, June 2014