Physics 329: Introduction to Computational Physics: Course News

This document will be updated throughout the course

Friday, November 22, 9:30 am

Alan pointed out that a tracing statement in my original version of newt2.f was printing out the same quantity twice, i.e. the code was 
            write(0,1000) iter, x(1), x(2), 
     &                    log10(max(nrm2dx,1.0d-60)), 
     &                    log10(max(nrm2dx,1.0d-60))
and should be (and now is) 
            write(0,1000) iter, x(1), x(2), 
     &                    log10(max(nrm2dx,1.0d-60)), 
     &                    log10(max(nrm2res,1.0d-60))
Both newt2.f and the sample output file have been updated to reflect the change.

Friday, November 15, 10:15 pm

I've installed the xforms toolkit on einstein. xforms is a GUI toolkit based on Xlib for X Window Systems which is designed to expedite construction of X-applications. Some of you may find the package useful in the context of your term-projects although you will have to program in C to use it. Type man xforms on einstein for basic usage information or see the xforms home page for more extensive information. Also, there are many demos along with source code in /usr/tmp/install/xforms/xforms/DEMOS on einstein. Thanks to Kurt Muehlner for digging this up.

Monday, November 4, 7:00 pm

Note the following correction to Homework 6, Problem 2: The third equation should be 
        x = yz
rather than 
        z = 1 / 2
The online version of the homework has been updated to reflect the correction.

Sunday, November 3, 11:00 pm

The second test has been rescheduled yet again: it will be held on Monday, Nov 11. Test coverage will include linear systems, finite difference techniques and non-linear equations and systems.

Monday, Oct 28, 12:30 pm

Please note the following change to the Class Schedule. There will be no lecture on Wednesday, Nov 6. Consequently, there will be a double lecture on Friday, Nov 1. Cookies will be provided.

Monday, Oct 28, 10:00 pm

There is a correction to the Oct 23 notes concerning initialization of the finite-difference scheme we discussed for the wave equation. Get your copy here if you didn't get one in class.

Monday, Oct 21, 11:00 pm

Note that you should keep total memory requirements of Fortran programs on charon to below about 2 000 000 *words*: i.e. 2 000 000 real*8 numbers. In particular, beware something like 
   real*8    bfa(10000,10000)
That's what you get for thinkin' too big, Alan :-) Seriously, thanks for "finding" the problem and my apologies for the time it cost you.

Monday, September 30, 6:00 pm

I have e-mailed you a message re possible re-scheduling of the first test. Here's the vote: 
                  WANT         DON'T WANT     DON'T CARE

   Wed. Oct 2      1                8              1       
   Fri. Oct 4      9                0              1
the respondent list 
   Choptuik, San Martin, Lorenzen, Finkenstadt, Shchekin,  Abdulai,
   Chiang, Mikula, Burns, Muehlner
and the final result 
   Friday, Oct. 4

Monday, September 23

Please note the following minor amendments to Homework 3. The online version of the homework has been updated to reflect the changes.

(1) In Problem 3, I now suggest that you use a write statement like 

 write(*,*) step, int(x), int(y), d
instead of 
 write(*,*) step, x, y, d
and you no longer have to produce a plot of the square of the particle's position as a function of step number: just plot the position itself vs step number.

(2) In Problem 4, there are some remarks at the end of the description concerning how the plots of the chaos game should be prepared.

Saturday, September 14

There is now a link in Kurt Lorenzen's Course page for downloading X-windows software for PCs. Interested parties are encouraged to check it out.

Tuesday, September 10

Please note the following correction to Homework 2. As usual I have updated the online version of the homework to reflect the change.

In Problem 5, the question now reads 

where ldata is of type list(list) (list of lists)
instead of 
where ldata is of type list(list(numeric))

Friday, September 6

You can use Maple V (Release 3) on either the Linux machines or the SGI machines (einstein recommended, see below). On either system type 
 % xmaple
to get the GUI form (make sure your DISPLAY environment variable is properly set if you're running it on a remote machine) or 
 % maple
to get the command-line form. Important Note: If you are on one of the Linux consoles but running xmaple remotely from one of the SGIs (presumably einstein) you will probably find that neither the "BACKSPACE" nor "DELETE" work (i.e. function as a destructive backspace/delete key). Use "CONTROL-h" instead. I know this is awkward and I am currently trying to track down the problem. However, for familiarizing yourself with the GUI, it should suffice to run xmaple locally on the Linux machines where everything should operate properly.

Thursday, September 5

I will be sending mail messages to all class members periodically (I have already sent two). These messages will be sent to your account on einstein, so be sure to create a file 
   ~/.forward
on einstein which contains your forwarding address. See the Unix notes for an example. Until you set up this forwarding file, there will be mail for you on einstein. Again, see the notes for information on how to read the messages using Mail

Thursday, September 5

Although you have access to the following SGI machines: I suggest that you use einstein as much as possible (possibly exclusively). The other systems, particularly infeld, aren't as completely configured, especially with respect to man pages, compilers etc.

Wednesday, September 4

Please note the following corrections/revisions for Homework 1: