is self-studying advanced physics a silly idea?

nabil0

I don't solve all the problems at the end of every chapter,but most of the time I solve problems .I solve some problems at the end of the chapters . I try to derive formulas by myself rather than just follow the author's derivation of results . I try to apply the ideas I learn to other contexts . If you try to apply things you learn to simpler contexts you will gain more understanding of the methods used in physics ...
I will try to address your questions:
1-If you get stuck somewhere , you can find the answers you want and ask questions at physics.stackexchange and math.stackexchange websites .
2-You will benefit from having a computer-algebra system and graphing software . I found matlab and mathematica very helpful . Also,There are books that have a complete solution manuals (Griffiths electrodynamics and the quantum mechanics book by the same author for example) You must try to solve the problems without looking at the solutions and also find many ways to solve the same problem.I think that really understanding the problems , finding the correct tools needed to solve it and being able to figure out what the answer should look like is more important than arriving at the correct answers.
3-If you want to read research papers in string theory , you have to learn string theory from a textbook or lecture notes on the subject (Which may be out-of-date . after you finish the book you will be able to read review papers in the field that mostly interests you (e.g. D-branes )

nabil0

I didn't study any calculus based general physics textbook . I recommend Gregory's classical mechanics or Taylor's Classical mechanics. They are very easy to follow . Thermodynamics and statistical mechanics are extremely important but they are generally not needed in introductory QFT (in more advanced field theory topics ,you will need them).
You don't need to study particle physics before field theory though it would help if you know elementary stuff about particle physics ..
For E&M the level of Griffiths is enough . You have to learn special relativity before GR .It's covered nicely in Griffiths and landau Classical field theory
I'd like to add that Differential geometry is extremely important in theoretical physics and I recommend that you learn it.

nabil0

I want to add that you can know a lot by just reading a textbook carefully without working through the book using a pen and papers. I learned a lot in QFT this way . In a couple of weeks ,you can read most of srednicki's QFT (without really understanding a lot of things) but you will know things like how dirac and Maxwell's fields are quantized , how to calculate things and the conceptual basis of renormalization . Actual understanding comes later when you try to derive things by yourself.

WannabeNewton

Don't rush things is all I can say. If you are in a hurry to get to the subject of your choice then not only will you not properly comprehend that subject at the apex of your cursory climb but also you will not have fully and elegantly grasped the subjects preceding it. Take your time as learning physics isn't a marathon (unless of course someone is quite literally holding a gun to your head =D).

A. Bahat

An important part of learning e.g. advanced physics (the same applies to math, with which I have personal experience, and I imagine to many other fields as well) is getting confident in your solutions. Naturally, sometimes you will be unsure, especially at the beginning, but eventually you've got to wean yourself off of being given answers. After all, people doing research have no answer book.

Again, I agree that it is generally pedagogically preferable for autodidacts to have some sort of hints or solutions, at least when getting started, but I think most elementary books do provide something of the sort. By the time you get more to more advanced stuff, I don't think this will be a problem.

redoopi

Thank all of you for your valuable comments!