[Cheeniandab]

3.14159/10

[ZX3URrBH]

Finserv coding isn't any nicer.

[grosqueneen]

Won't have to do much coding in my new job. Might have to read other people's code to ensure it's actually doing what they claim, but that's it.

Most of my work will be pen and paper apparently.

[Garry Hovard]

Won't have to do much coding in my new job. Might have to read other people's code to ensure it's actually doing what they claim, but that's it.

Most of my work will be pen and paper apparently.

The horror. I hope they are paying you over $80,000 a year to deal with such nonsense.

[searkibia]

I hated the last physics class I took in college. "Solve these differential equations." "Okay, should I use Java C++ or Fortran?" "By hand." "Jump up your butt!"

[Wdlglivi]

2nd year calculus? Who the hell takes longer than a year to learn calculus?

[soajerwaradaY]

Most of my work will be pen and paper apparently.

Newflash! We live in the digital age now - what is with this "pen and paper" ****?

[zoppereurvito]

"Calculus of functions in two or more dimensions. Includes solid analytic geometry, partial differentiation, multiple integration, and selected topics in vector calculus"

[fiettariaps]

I took a dumbed-down second year calculus course for computer science students, which covered some vector calculus and ghetto DEs. I had to do the full second-year multivariable and DE courses in 4th year as prerequisites for engineering, which was good.

SP

[Slchtjgb]

Urg. Up to there should be covered in 2.5 semesters, and no more.

Multi-variate calc is a joke. Can be taught in less than a week. Vector cal is more serious, and should also include Lagrange multipliers and perhaps some intro to calculus of variations.

That should cover 3 full semesters, followed by a semester of complex analysis. After that, there should be a full year of diffeq, starting with ODEs, integral transforms etc, followed by PDEs, integral equations, series sol'ns, Green's functions/fundamental solutions (this is assuming the student has already learned a decent amount of linear algebra)

After that, the only thing left is numerical methods. Finite differencing, Monte Carlo, the suitability and optimization of these techniques under various conditions.

That pretty much covers all the math an undergrad will ever need to know, and some of what a grad student might need. Specialized or abstract math classes might or might not be required depending on the area of research interest.

You know...I blame you for this. He wanted to be an engineer before he started reading your posts! Now I have to pay for how many years of grad school? Thanks

[GypeFeeshyTes]

You don't have to pay for grad school. They pay for him.

[swissloveone]

The diffeq year should include, of course, a thorough grounding in special functions. Introduce in the context of ODEs, flesh out in the context of PDEs.

[Jxmwzgpv]

You don't have to pay for grad school. They pay for him.

Awesome!!!! That is the best news I have heard in ages!

[Slchtjgb]

I have no idea about the summer thing. Be more descriptive. Is it a show-and-tell trip, or do they get their feet wet with research?

[WXQMQFIr]

Graduate schools base decisions on 3 factors:

GPA
Undergrad research (gives letters of recommendation)
Physics GRE

3.5/4.0 GPA, 1-2 summers of research (with good impressions from profs) and a 80+ percentile score in the GRE will get him into a top 10 school. 2/3 of those (assuming the missing element is not horrendously bad) will get him into top 20. Outside the top 20 things start to go rapidly downhill in terms of post-PhD career (either academic or nonacademic).

As an example, my undergrad GPA was ~3.3 (admittedly, in a more difficult than average degree for physicists), GRE was 91st percentile and I had decent to good undergrad research. Ended up at a school that's currently 10th-15th in my estimate.

[TriammaMade]

Graduate schools base decisions on 3 factors:

GPA
Undergrad research (gives letters of recommendation)
Physics GRE

3.5/4.0 GPA, 1-2 summers of research (with good impressions from profs) and a 80+ percentile score in the GRE will get him into a top 10 school. 2/3 of those (assuming the missing element is not horrendously bad) will get him into top 20. Outside the top 20 things start to go rapidly downhill in terms of post-PhD career (either academic or nonacademic).

As an example, my undergrad GPA was ~3.3 (admittedly, in a more difficult than average degree for physicists), GRE was 91st percentile and I had decent to good undergrad research. Ended up at a school that's currently 10th-15th in my estimate.

I guess I can see that (the 20 school thing). At UTK, there are 26,000 students and 83 of them are undergrad physics. Not really a whole lot of people persuing this I would imagine. Still....20 schools seems awful few...

[Liabmeasez]

Top 20 schools is ~350 physics PhDs per year (this is US only, but rest of world contributes slightly less than this to the top of talent pool). All physics PhDs total is ~1500 per year in the US

Every year in the US 150 faculty positions open up. 130 of those go to PhDs from top 20 schools. Top 10 might take 90 of those.

Anybody who has any ambitions of a faculty job basically needs to go to top 20, and in order to have a better than even chance, needs to go top 10 and preferably top 5.

Also, at top 20 schools something like 60% of the graduate students are non-US. But graduation rate is only 65% or so, so 350 PhDs means that there are ~300 slots open for US undergrads.

EDIT: fixed numbers

[Edifsdubs]

Cool. Mind paying off my credit card debt?

**** that noise. I got student loan debt out my perfectly-shaped ass.

[beonecenry]

[AdobeCreativeSuite]

Enjoy it will it lasts.