Scientific Computing mid thgouhts
Scientific Computing is Hard! :(
I started programming at university, but the only experience were found in the two previous term-length modules on python, where the hardest things we had to do were error approximations via MonteCarlo Sims and solving ODEs with premade packages.Â
SciComp is a module about C, and thus it's the first module where we actually need to care about memory and have to spend a lot of care with the code.
First two assignments weren't terrible, about functions and pointers, or pointers to arrays which act as pointers and pointers to pointers to arrays etc, and although it was initially easy to get confused about * and & operators for dereferencing and initializing pointers / memory locations, they were okay.
But 3rd and 4th assignments are horrible, the 3rd assignment gave us a piece of botched code and a specification to solve an inhomo complex pde (or two inhomo coupled pdes) and just let us at it, almost all of what I typed I had to try and learn from online, there was so much I didn't feel I knew or even knew I had to know. And the 4th isn’t much better, this time we have a 2d state vector nonlin inhomo pde and have to solve it completely from scratch. We haven’t discussed stability or discretization for nonlinear situations at all and it’s definitely taking up most of my time (I should definintely be working on it rn but haha?)
I’m continuing with the module, but I hope I take enough modules by the end of the year such that my Year’s Grade doesn’t get overly screwed by my inevitable bad mark.
At least I do think I’m actually learning a lot, even if it is a trial by fire.
