22.15  Essential Numerical Methods

You have heard your official final exam grade from Clare Egan. Some folks were asking about their score. Your score for the final exam is now available through the stellar site. Go to Membership (left menu), Gradebook (top menu), and it should be there. The A-B threshold was 70%. The B-C threshold was 56%. I believe you can see the distribution of scores through a click somewhere in the gradebook (view distribution). If you see some discrepancy, especially between the official final exam grade and the Stellar entry, please let me know. If there's any other problem that arises from gradebook settings also please let me know. It's not user-friendly. Grades for the course and for the final exam are not all the same. Nothing on the Stellar site addresses the exercises or course grades. The course grades are submitted to the registrar's office and I think they don't release them till later.

Dear 22.15 Participants,

The graded Exercise 7 is now available in your unique pset sharing folder.

Please don't hesitate to let Prof. Hutchinson or Mingda know if you have any questions.

Dear 22.15 participants,

Please find your graded exercise 6 in your unique pset sharing folder. 

An important comment from Prof. Hutchinson regarding to exercise:
 
Some have error at Nx = 10 greater than Nx=30. Such results are not consistent with a finite differences in the spatial direction that are second order accurate. Other people found values that had very small errors at Nx=10. The reason is that the solution is in fact a parabola -- second order -- so that a properly differenced second order accurate scheme (including the boundary conditions) gives theoretically exact accuracy in the spatial differences. The reason the higher Nx cases give larger error is that they are in fact NOT converged in time, because Jacobi iteration is so slow that the convergence criterion I prescribed is inadequate.

I hear that exercise 8, which is due Monday, has stretched the capacities of quite a few folks. On Monday before 9:30am, I will answer questions about it, but without exactly telling you the answers. On Wednesday at 9am, when you will have already submitted your solutions, if it seems useful I will work the problems on the board for interested parties.