murfinke Wrote:
-------------------------------------------------------
> I have just come out of this exam very despondent.
> I am sure I failed.
likewise, and i consider myself experienced with numerical methods; i've used numerical methods in my programming studies since early high school and specialise in monte carlo methods - which requires knowledge of which methods are best for which purposes etc.
this is not said to brag; i have taken a
personal interest in the subject for many years and found the exam to be unreasonably difficult (for no good reason - the course syllabus isn't easy of course but i mean difficult in the sense of requiring heroic feats of memorisation).
> The paper I wrote today did not resemble the past
> papers I worked off or the assignments I completed.
yes, it was a 180 degree turn from ALL the other exams. "didn't memorise the runge kutta 4th order formulae? too bad for you." 12% gone right off the bat, and is it so in reality that you'll have to memorise the formulae or not be able to use / understand them?
> It put me in a state that I even forgot
> the simple things I did know.
likewise, i was making the most embarassing mistakes for two pages on the last question where they wanted us to solve (not just set up) the system.
the worst part is that i could have EASILY given program code to solve it, on paper without any compiler assistance! the whole idea of this course is that we can solve "real world" problems which need more than just "sanitised" algebra with nice solutions and a little pocket calculator work - that i learnt to do this in this course helped me NOT ONE IOTA in the exam!
> I felt like it was set to trick me not test my
> knowledge or application of that knowledge.
it is this sentiment which i agree most with. that exam was nothing short of unfair and artificial - to not have the decency of giving a fat block of formulae with which we can work is just ridiculous. that implies that we should also have completely memorised:
1. modified euler and midpoint (i only remembered modified euler, not midpoint)
2. rk2 and all parameters
3. rk4 and all parameters
4. rkf and all parameters (!!)
5. adams and all parameters
6. adams-moulton and all parameters
7. milne's and all parameters
8. all the others...
what's that i hear the lecturers saying? remembering rk4 isn't so bad?
how were we to know we were expected to memorise only that, did you indicate it in any way so that we could avoid having to learn each and every fat block of formulae covered in the textbook? actually
exactly the opposite was indicated in the past papers - the formulae were always provided for the students to make sense of and use properly. is that suddenly not the proper way of doing it anymore?
i'm not a sissy who can't remember formulae when needed (and i do know most of them by heart), but at least indicate which are to be memorised. pointing to a textbook which is 50% formulae page-for-page is not very helpful, nor are past papers which indicate that we don't have to memorise these raw formulae at all!
> I would like to iterate that I did not feel like the
> lecturer enabled my learning of this subject or
> provided enough guidelines to prepare me for an
> exam like that.
yeah, and it's a crying shame because the year's assignments were great to solve on a computer.