Physics 381C: Computational Physics (also CAM 381C, Spring 1997)

Schedule: Tue, Thu 12:30 - 2:00 PM --- RLM 7.104 --- Unique # 53020

Instructor: M.W. Choptuik
Office: RLM 9.208A --- Office Hours: Mon, Wed 10:00-11:30 and by appointment
Phone: 471-1541 --- E-mail: matt@infeld.ph.utexas.edu
Instructor's Home Page: http://wwwrel.ph.utexas.edu/Members/matt/welcome.html
Course Home Page: http://wwwrel.ph.utexas.edu/Members/matt/Teaching/97Spring/Phy381C/Syllabus.html

Syllabus
News (last updated, Wed May 14, 8:45pm)
Course Topics
Course Notes
Course-Related Software
Course Resources and Online References
Suggested Hardcopy References
Homework Schedule, Problem Sets and Keys
Student Pages
Term Project Presentation Schedule

Course Overview: This course will cover methods of computational physics with particular emphasis on continuum problems (PDE's). Topics to be discussed include: Unix as a scientific programming environment, symbolic computation using Maple, issues in scientific programming and analysis, solution of ODEs, solution of linear algebraic systems, fast Fourier transforms and finite difference methods for elliptic, parabolic and hyperbolic PDEs. Applications will be drawn from various areas of physics, in part dependent on the particular interests of the class. Access will be provided to the UT Austin High Performance Computing Facility as well as some of the high-end graphics workstations in the Center for Relativity. See below for a brief Syllabus and the Course Topics page for a more detailed outline of the course coverage.

Note: There will be a significant programming component to this course: the instructor's current language of choice is Fortran 77, (hereafter f77), and, at least in the Homework assignments, students will be required to program in f77. Those students who do not know f77 will be responsible for picking it up on their own, with help from the instructor where and when necessary. Students will also be responsible for developing facility with the basic components of the Unix programming environment required for computational work: this includes familiarity with a suitable text editor, preferably vi or emacs, and proficiency with a Web browser such as netscape. Finally, students will be responsible for learning sufficient LaTeX or TeX to prepare their term papers. See the Course Resources and Suggested References pages, for on-line and hardcopy sources, respectively, on these and other topics.

Text: Due to the significant diversity in topics to be covered, there is no required text for the course. The optional text, Numerical Recipes (2nd edition), by Press et al is particularly recommended for those of you who anticipate doing further numerical work. Note, however, that the full text of the book is available on-line. Also note that there are distinct Fortran 77 and C versions of the book: choose the one which you feel will suit you best. See the Suggested References web page for texts and other references pertinent to the course, and the Course Resources web page for a collection of on-line reference/instructional material.

Grades: Homework and Term Projects:

Your mark in this course will be determined on the basis of your performance on a number of homework assignments and a term project, with the following weighting: Final marks will be subject to small adjustments based on overall class performance.

Homework: See the syllabus below for scheduled homework due dates. Homework will be assigned at least a week before it is due: late homework may be accepted at the instructor's discretion. As the course progresses, the Homework Schedule web page will contain information concerning current and past assignments. The weight of each individual homework towards your final mark will be based on the complexity and difficulty of the assignment.

Term Projects and Presentation: Either individually or in consultation with the instructor, each student must choose a topic for a term paper in some area of computational physics. All topics must be approved by the instructor: a one page outline of the project is due March 6th. Please secure the approval of the instructor regarding the topic before submitting the outline. Even if the bulk of the project involves programming, the term paper per se must be prepared in the style of a technical paper or a scientific essay. Term papers must be prepared using LaTeX (or TeX) mathematical typesetting software. Suggested paper length is 20-30 pages, including figures, graphs and source code listings. During the last two weeks of class, all students will be required to make a short in-class presentation on their project. The length of the presentation will depend to some extent on the enrollment, but will probably not exceed 15 minutes. Note that, in some cases, term projects may still be in progress at the time of presentation, but that this is to be avoided if at all possible. Speaking order will be determined via random selection by the third week of class. The presentation is intended to give you speaking experience as well as to educate the rest of the class and the instructor---it will not count significantly in the assessment of a grade for the project. All write-ups are due May 1, 1997 and this deadline is to be considered VERY FIRM.

Computer Access:

All students will be provided with accounts on the Physics Dept. public Unix systems located in the Graduate Computer Lab, RLM 3.118. You will be also be given remote access to Center for Relativity Unix machines, and you will be generally required to use these machines for completing homework assignments. To the extent possible physical access to Relativity machines will also be provided. Finally, accounts will be provided on both the Cray J90 and T3E systems of the High Performance Computing Facility. If you have a home computer and a modem, you are encouraged to contact the Physics Dept. Computer Group for information regarding access to campus systems via home machines.

Other Help:

You should also feel free to contact me via e-mail (preferred) or phone if you have quick questions, or if you are having difficulty getting something to work. Perhaps most importantly, you should strive to develop the ability to make effective use of the available documentation for the software you are using (on-line help, man pages, Web resources, etc.). On-line help tends to be extensive these days (particularly for systems such as Maple) and a little time invested in learning how to extract the information you are looking for usually pays off.

Syllabus

Due Tuesday Thursday

January 14
Class Cancelled		 January 16
Unix

January 21
Unix		 January 23
Unix/Fortran
H1 January 28
Fortran		 January 30
Maple

February 4
Maple		 February 6
Maple

February 11
Maple/ODEs		 February 13
ODEs
H2 February 18
ODEs		 February 20
ODEs

February 25
ODEs		 February 27
Linear Systems
Project Outlines March 4
Linear Systems		 March 6
FD Methods

March 11
Spring Break		 March 13
Spring Break
H3 March 18
FD Methods for Elliptic PDEs		 March 20
FD Methods for Elliptic PDEs

March 25
FD Methods for Elliptic PDEs		 March 27
FD Methods for Elliptic PDEs

April 1
FD Methods for Elliptic PDEs		 April 3
FD Methods for Elliptic PDEs

April 8
FD Methods for Time-Dep. PDEs		 April 10
FD Methods for Time Dep. PDEs

April 15
FD Methods for Time-Dep. PDEs		 April 17
FD Methods for Time-Dep. PDEs
H4 April 22
Miscellaneous/Presentations		 April 24
Miscellaneous/Presentations

April 29
Presentations		 May 1
Presentations/Class Evaluation

Syllabus Notes: