How Does Krane Compare to Kleppner & Purcell in Depth and Difficulty?

[Thinker301]

I know that Kleppner and Purcell are deeper, but does HRK even come close? Also how far is HRK from HRW in depth of coverage and difficulty? Thanks!

 

[WannabeNewton]

HRK is noticeably more difficult than HRW, both in terms of problems and coverage of material. HRK isn't that much easier than Kleppner, in my opinion. Purcell is definitely leagues above HRK however.

 

[Thinker301]

Hmm Thanks! So A reasonable First Year Curriculum could be, Kleppner(or Morin) for CM, Purcell for EM, and HRK for everything else?

Also I have always wanted to ask you(I have been on the forums a lot and your name keeps popping up). If you had to chose one, Morin or Kleppner?

 

[WannabeNewton]

Hmm Thanks! So A reasonable First Year Curriculum could be, Kleppner(or Morin) for CM, Purcell for EM, and HRK for everything else?

Is this spread over two semesters or all in one semester? If the former then I think it's a reasonable endeavor to take on but if your school uses other books then it might be a lot of work going through the assigned texts on top of those. If your school has honors intro physics courses then all the better because you'll more than likely be dealing with some of those texts anyways. At my school it goes Kleppner then Purcell for the first two honors physics courses and I recall from a couple of years ago that RPI used HRK for their honors physics sequence.

If you had to chose one, Morin or Kleppner?

I think if I tried to go through Morin before Kleppner my self-esteem would have been thoroughly destroyed after attempting all of the problems in Morin :)

If you have time, on top of your assigned coursework, then go with Kleppner.

 

[Thinker301]

Alright so this is what I am getting. If I were to make a hierarchy of difficulty of books, this would be it.

Level 1 - HRW - Level
Level 2 - HRK
Level 3 - Kleppner
Level 4 - Morin(CM) and Purcell + Morin(EM)

Sound about right?

Also can you comment on my plan, here it is below.

Summer
HRK
Computational Calc(SV)(Larson)(I did take Calc(SV) before, but back then, I didn't really know how to learn and there are some issues with computing integrals)
MV Calc(Colley)

School
Honors Calc course (Uses Spivak then Advanced Calculus by Folland)
Honors Intro Physics(Uses Six Ideas that shaped physics)
Kleppner + Purcell
Feynman(With the new exercise book coming out)

Summer
Morin
Boas

Also what are your thoughts about books like Spivak for physicists?

I know it is a a lot of questions, but I have had them for a while and never really asked.

And seriously, thank you!

 

[WannabeNewton]

Sound about right?

Honestly I think HRK and Kleppner belong on the same level but apart from small things like that yeah I would agree.

Also can you comment on my plan, here it is below.

If you can stick to the plan then all I can say is, it looks pretty solid to me, at least the physics parts. I can't say much about the math stuff because I don't know much about math books. The hard part is sticking to the plan :)

Also what are your thoughts about books like Spivak for physicists?

I'm not much of a math person so I would say it's a waste of time unless you're interested in pure math for the sake of it and have the extra time to invest in a book like Spivak.

 

[Thinker301]

HRK sounds pretty awesome!

The sticking to it part is always a little hard lol :)

The questions are all flooding me. I think this is the last on that was bothering me. Wouldn't not doing all the problems make you lose out on some learning?

 

[WannabeNewton]

The questions are all flooding me.

Well keep em coming :)

Wouldn't not doing all the problems make you lose out on some learning?

Not in my experience, no. Usually it's safe to skip the "easy" problems and hit the "hard" problems with full force. This way you'll learn the most in the most efficient manner with regards to time and effort. The problem with this is of course that some problems seem ostensibly "easy" but after actually attempting them show their true difficulty. In graduate textbooks there tend to be few enough problems that one can actually do all of them to completion but in books like HRK, Purcell etc. there are just way too many problems for a person to finish with average motivation and available time. Thankfully in books like Purcell (at least in the newest edition) the problems are already labeled by difficulty so you can pick and choose from the hard problems.

 

[Thinker301]

Awesome!

And the that seems like a good solution, but I remember looking at Morin's CM and the one star problems seem difficult enough to be worth doing. Were there problems in Morin and Purcell that were easy enough to skip?

 

[verty]

HRK sounds pretty awesome!

The sticking to it part is always a little hard lol :)

The questions are all flooding me. I think this is the last on that was bothering me. Wouldn't not doing all the problems make you lose out on some learning?

I read a book once called "how to read a book", it sounds silly but it was about how to read a book in the best way to learn. It wasn't a particularly good book but it is nice sometimes to see questions like this worked out.

Like, how to stick to it? Imagine there was a book "how to stick to it", it would say stuff like set goals and checkpoints, keep a calendar with study hours, stuff like that. One of the things it would say is try to finish a chapter in one sitting, even if you go over it again the next day. That way you always feel like you learned something.

On the other hand, it is true what they say about calculating, if you can't calculate it, you don't really know it. So with regard to the problems, it seems like it is a good idea to do all the problems, but there just isn't enough time and usually it is too tiring. Good books usually have answers to odd-numbered questions or answers to the easy questions, so it can help to do some easy questions and then look at the answers, sometimes a better method will be mentioned.

I think knowing the best method to solve the easy questions is a good stepping stone to the more difficult questions. Otherwise it can be difficult because the difficult questions are much larger and are built up from those easier components. Then also, there are some questions that one needs insight to answer, and I think one can gain a lot of insight by mastering the easy problems, and I mean not just solving them but in the best way. One can see how they relate to each other and concepts are made much clearer.

And also, it is easy to skip a few easy questions without being too worried that you might miss something, because if you are mastering them, you'll quickly learn to see if the questions are more devious. And, more difficult questions are often more abstract and the concrete knowledge can help.

So I think that a lot of the trouble that people have with difficult questions is that they don't know everything they should about the easy questions.

But as WannabeNewton says, doing the difficult questions is important, that is where your knowledge is tested and partial solutions can be enlightening even if you can't find the answer.

But everybody is different and they learn differently so not everything has a simple answer.

 

[WannabeNewton]

Were there problems in Morin and Purcell that were easy enough to skip?

Yes but what's "easy" and what's "hard" is quite subjective. As verty noted in his excellent reply, the entire process of working through a textbook can be quite subjective.

 

[Thinker301]

So taking in both of your ideas, how does going through the book and doing only the one star problems, then going back and rereading and doing the two star problems, etc?

 

[micromass]

So taking in both of your ideas, how does going through the book and doing only the one star problems, then going back and rereading and doing the two star problems, etc?

I don't see the benefit of that over just doing the one and two star problems immediately.

 

[Thinker301]

Alright, so I go through twice, 1 and 2 the first time, 3 and 4 the second time?

 

[verty]

Alright, so I go through twice, 1 and 2 the first time, 3 and 4 the second time?

Sounds good.

 

[micromass]

Alright, so I go through twice, 1 and 2 the first time, 3 and 4 the second time?

I don't get why you want to go through twice. Just do all the exercises the first time.

 

[Thinker301]

So I watching this video done by this guy who worked through the MIT curriculum on his own. He had a lot of ideas about how one learns things. He was suggesting that if you go through it all multiple times you will have a better grasp on the material.

 

[guitarphysics]

Hi Thinker, my two cents:

I've self-studied a lot over the past couple of years or so, and I found out fairly quickly a nice way to study from a textbook. First, I look at the contents to get a general (*VERY* general) idea of the subject. Then I go chapter by chapter. But at each chapter, what I do is this: first skim the chapter (to sort of understand the motivation behind introducing some concept at the beginning, and just get a general picture of what you'll learn). Then actually read the chapter (thoroughly and actively, obviously). After that, I go for the exercises straight away. If I can do them all easily (rarely happens, ha!), I'll move on. If not, then I'll re-read parts of the chapter, and try some more.

I'd say that plan is pretty good. Kleppner is a book I really love; it's the first 'real' physics book I read, and it's really hard for someone starting out, but it gives you 'character' in physics. You'll learn that it's not easy but it's really satisfying to understand something so well that you can apply it to hard problems.

So you'll be finished with Kleppner, and feel like a physics wiz. After that book, you can solve anything, right? Nope! Purcell is much harder, in my opinion- but also a lot more fun and even more physics-y (I'd say Purcell is the author, besides Feynman, whom I've read with better physical reasoning than anyone else).

Anyway, that seems like a good plan. Stick to it and I'm sure you'll be fine. (That calendar thing helps- also if you like physics enough, it should fly by fairly quickly.)

Another thing- maybe after K&K, instead of Morin you should consider reading up on Lagrangian mechanics (maybe Morin talks about this, though. Not sure.). It's relevant to many parts of physics, and relatively accessible if you're equipped with Boas (you'll need to know about variational methods). Maybe try Taylor, which is a pretty gentle approach.

 

[ParticleGrl]

He was suggesting that if you go through it all multiple times you will have a better grasp on the material.

What I always found helpful was to save a few hard problems from earlier chapters and revisit them as I got farther in the book. i.e. when I'm at chapter 4, go back and do the saved chapter 1 problems WITHOUT referring to the book/notes,etc.

I would say forcing myself to work through problems without reference to notes or texts helped develop me as a physicist (and now as a statistician/machine learning specialist) much more than any other study skill.

 

[atyy]

I second guitarphysics's suggestion of reading up on Lagrangian and Hamiltonian mechanics after Kleppner and Kolenkow. Kleppner and Kolenkow is not conceptually "advanced", but it is conceptually very efficient on the basic stuff. It is possible to benefit from Kleppner and Kolenkow's exposition while doing simple problems from another book. Why should one do at least some of the hard problems from Kleppner and Kolenkow? Because doing them enables one to appreciate how many of these become easy once one has the Lagrangian formalism. For Lagrangian and Hamiltonian mechanics, one can try Landau and Lifshitz, and Fetter and Walecka. Susskind's Theoretical Minumum is also very good. Landau and Lifshitz have a reputation for being formidable, but there's plenty to be gained even by a casual reading to see how great physicists organize the material.

While Purcell has many great insights (eg. using relativity to relate electricity and magnetism, and the derivation of the Lamor formula), I found it hard to understand the total conceptual structure of electromagnetism from it, since he builds it from simple cases. So a complement might be a book that views the simple cases as various approximations to the entire structure of Maxwell's equations. Dugdale, and Haus and Melcher take this complementary approach to the material.

I haven't seen the final version of Spivak's physics book, but a draft was horrible.

For electromagnetism, one needs a couple of things from multivariable calculus beyond elementary calculus: the change of variables formula involving the Jacobian (and the concept that an area is involved), and the various versions of Stokes's theorem. It is interesting to contrast the presentation of Stokes's theorem in the Feynman lectures with that in Spivak's Calculus on Manifolds (both are excellent).