>>16646271
> Why would I need to know how do derive snell's law from maxwells equations from memory if I'm going into stat mech or anything other than high energy physics (not counting quantum optics, but still)?

I'm an EE, not a physicist, but people have the same arguments in our department about our qualifying exams (which often have quite rigorous representations of classic but niche proofs from probability, statistics and machine learning).

I think the point isn't that they expect you to derive it from memory, so much as they want you to be able to work through it yourself if you needed to later. If you stay in research you're going to inevitably run into variations on problems for which absolutely nobody in the literature has touched it. You should be able to figure out how to formalize the problem, work through a derivation of a modeling approach, and analytical evaluate relevant factors that will impact your model's performance.

Also, acoustic ray tracing is really cool but fuck Snell's law.

Back | Go to board

Board:

/sci/

Thread:

Purpose of Physics Predoctorals