1. Not finding help here? Sign up for a free 30min tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Advice for self-studying physics?

  1. Mar 14, 2013 #1
    I'm currently a senior in high school and I've been interested in self studying physics. I've picked up self studying since last year and have been going through various undergraduate introductory textbooks such as Goldstein and Griffiths for mechanics and EM.

    What exactly is the best way to self study to attain full understanding of the material? I usually read through the chapter where I gain understanding of it then go back and go through the problems. Should I go through the chapter and re-derive everything as well as do every problem? Or is it enough to read through the chapter and look through the problems and just see if I know how to approach it?

    I feel that mathematics is part of the problem as well. For example, as I was reading through Griffiths, I had to re-read how they used separation of variables to solve partial differential equations for laplace's equation several times in order to be able to understand it and do problems. Should I study some math first? I've already taken a course in multivariate calculus and am currently taking differential equations.

    In addition, is there a general method to self study efficiently? I feel that when I stay up late to study, I don't really learn/understand things as fast as I do when I read it during the day. When I take classes in school, it seems that I pick up things faster than when I self-study. For example, I somehow just sleep through all my math and physics classes and just ace the tests and I actually understand everything. However, when I self study and actually want to learn what I'm reading, it seems harder for me to understand things fast and retain it or have the drive to do more problems.

    Is the doing more problems part the most important? Most of the time, I spend more time going carefully through the chapter and re-reading proofs and re-deriving things and then I just skim through the problems and move on if I can figure out the general method to do it. I sometimes just get lazy and don't bother to go through with solving all the long integrals and math. Or I just get impatient and want to move on to learn new things.

    What tips do you guys have?

    thanks.
     
  2. jcsd
  3. Mar 14, 2013 #2

    Simon Bridge

    User Avatar
    Science Advisor
    Homework Helper
    Gold Member
    2016 Award

    Don't "self study" alone.
    Look for correspondence and "distance learning" materials to help you.
     
  4. Mar 14, 2013 #3
    i can't really find any online things for the things I'm currently studying.

    What tips do you have on self-studying online based on the above?
     
  5. Mar 14, 2013 #4
    I'm in a very similar situation to yours, but just two years behind you, and a bit less advanced. I have found it to be harder to learn from self-study than from school. What I do now is study only during the day, because usually I understand everything better, I talk to my physics teacher about what books are good for self-study (and ask him if I'm ever confused about a concept), and I try doing lots of problems (and coming up with my own as well, which is very important, I think). I think learning everything thoroughly (like you do) is really important, but with doing lots of problems, everything sinks in a little more.
     
  6. Mar 14, 2013 #5
    thanks. Do you guys think I should familiarize myself with higher level mathematics first like linear algebra or complex analysis? Or just continue with those books? And maybe move on to quantum mechanics or something later on.

    Just curious, on average, how long does it usually take one to complete a textbook like griffiths?
     
  7. Mar 15, 2013 #6
    What is your background in physics?

    Why jump to such advanced books?

    Have you looked into Kleppner and Purcell for mechanics & e&m?

    Studying linear algebra simultaneously with your physics wouldn't be a bad idea. I would hold off on complex analysis if I were you.
     
  8. Mar 15, 2013 #7
    Those are exactly the books I'm using! They're both amazing, and I highly recommend them, Ishida (if you haven't read them yet).
     
  9. Mar 25, 2013 #8
    Well I feel like I am capable of understanding these texts since I've already learned basic undergraduate classical mechanics and E&M, as for math: multi&diff. I feel I just have to brush up on some math like tensor notations and a bit of linear algebra. Like after looking through some math, I can solve problem sets.

    so just read through it and do the problems? Is it not enough to just go through the text and do mit ocw problem sets? Or do I have to do the problems in the textbook as well. Some problem sets usually assign questions from the textbook anyways.
    Do I have to work through the entire problem? Like I stated before, I sometimes jsut look at the problem adn figure out how to do it lol.
    And also, is there a general method to learning. Like staying up at night and studying is that usually not as efficient for learning?

    If I were to self study rigorously and to actually understand it completely, would I have to re-read everything carefully that I read through before when I didn't do problems? Or is it enough to just skim through it, re-derive things, work out examples, and do problem sets on mit ocw?
     
  10. Mar 31, 2013 #9
    I found that it's helpful to like write out a summary of what I've just read as like notes for someone else and I get a chance to rederive stuff and solve examples all over again. Is this a good idea?
     
  11. Mar 31, 2013 #10
    That actually sounds really good. I'm gonna start doing that as well!
     
  12. Apr 5, 2013 #11
    Can anyone give a list of the best/most comprehensive physics textbooks for each area of physics for someone just learning it? Would it be more useful/straightforward to just use the landau series?
    And is there a list of classic/popular math textbooks as well? For topics like analysis, differential geometry, topology, etc. Like the way there are more widely used textbooks in physics like griffiths, jackson, etc.

    thanks.
     
    Last edited: Apr 5, 2013
  13. Apr 5, 2013 #12
    I recommend some YouTube channels to: MinutePhysics and Varitasium are excellent.
     
  14. Apr 5, 2013 #13
    I would recommend studying for like 2 hours at a time and taking an hour or so break, lay down, rest your eyes, listen to some music, and just mull over everything you just went over. Then come back to it and review it a bit. Personally, I feel that when I re-read something after taking my break, it is more clear than reading it twice back to back. I vaguely recall some scientific reasoning to this, but I'm not confident enough in that memory to quote it. I think it has something to do with the way we solve problems and "store memories."

    I also recommend a similar method for when you get hung up on a problem. Step away from it for a bit, clear your mind, and then start again.

    If you're like me, and your mind is always wandering, I find it helpful to listen to familiar, or lyric-less music like film scores(there's actually a pandora station by that name is you are interested). It sort of cuts off everything but what I'm looking at and really helps me focus in on a problem. It may also be relevant to note that I have adhd, so this all may be completely useless to you. :)
     
  15. Apr 5, 2013 #14
    Don't forget sixtysymbols!
     
  16. Apr 6, 2013 #15
    Should I just go along with the problems in the textbook rather than problem sets like from mit ocw? Especially since they don't really have answers?
     
  17. Apr 7, 2013 #16
    A mixture of both would probably be ideal.
     
  18. Apr 8, 2013 #17
    Hp Ishida52134

    - I'm currently a senior in high school and I've been interested in self studying physics. I've picked up self studying since last year and have been going through various undergraduate introductory textbooks such as Goldstein and Griffiths for mechanics and EM.

    - What exactly is the best way to self study to attain full understanding of the material? Should I go through the chapter and re-derive everything as well as do every problem? Or is it enough to read through the chapter and look through the problems and just see if I know how to approach it?


    I think the best thing is to have a supplementary high school physics and a supplementary high school algebra text for reference as your very own firstly... I think a lot of students in the 60s 70s probably would self educate themselves more through life if some of the school textbooks were their 'own' to browse, if they really liked it and didnt have the time or maturity to 'really read it in school'.


    Goldstein i think would frustrate, boggle and intimidate people, and Griffiths any of his books will be useful one day... for sure. You're lucky to find a copy...

    i think reading *everything* in a text is a good but slow option.
    Reading a chapter or section many times...
    totally understanding the examples inside and out...

    and i'm of the school that 99% of students don't really do what the textbook author intended, doing all of the problems... You got a rare opportunity to really see *why* there are so many problems with your textbook, and i think it's almost like throwing half the textbook away, when teachers only toss 4 problems a week at people, and dont really 'educate' them on how to read, how to do the problems, how to self-study, and just how much time it takes, and what efforts are needed.

    I think all it takes is more time, just a few more hours. maybe some think 90 minutes a week is good, but sometimes real understanding can happen with 6-8 or maybe 15 hours working on a chapter.

    -----

    I would rather master 3 chapters of a textbook inside out than zoom through 67% of the textbook with a fair understanding of the 'whole book'...

    i think as a ball park 150 hours for one semester/half the textbook and
    300 hours for the whole book.

    -----

    - as I was reading through Griffiths, I had to re-read how they used separation of variables to solve partial differential equations for laplace's equation several times in order to be able to understand it and do problems. Should I study some math first? I've already taken a course in multivariate calculus and am currently taking differential equations.

    thats pretty good for high school...

    getting your calculus 4/vector calculus and your diff equations i think is 99% of all you need for physics really.


    Honours people will add complex variables later on

    and mathematical physics types will take a year of analysis [rudin] as well as more classes in math that need a course or two in differential equations.



    - In addition, is there a general method to self study efficiently?
    - Is the doing more problems part the most important? Most of the time, I spend more time going carefully through the chapter and re-reading proofs and re-deriving things and
    - then I just skim through the problems and move on if I can figure out the general method to do it.
    - Or I just get impatient and want to move on to learn new things.

    Well that 'does' work for a rough understanding and might be good enough to pass a course for a few years, but if you wanna really get a 95% understanding of the book and the problems...

    it all boils down to time... which is gonna be 10-15 hrs a week for 30 weeks [maybe 24 weeks if you rush] for a 68%-99% confident mastery of a textbook cover to cover]

    a lot of the frustrations i had melt away with the extra hours reading it again and doing all the problems.

    some textbooks hide stuff in the problems...too

    I think that the more solid you make your ladder the easier it is so there is less struggle on the higher rungs.

    IN a nutshell, i think get what it takes to totally master
    - a vector calculus textbook
    - one book on analysis

    - totally master Halliday/Resnick or Wolfson/Pasachoff for first year physics with calculus

    - take a whole year not 12 weeks for intermediate mechanics [I still think Symon and Kleppner/Kolenkow are the best reads, and not the marion route]

    - my style would be books 1 2 3 4 5 of the Berkley Physics Series
    [Mechanics/EM/Waves/Quantum/Statistical Mechanics]

    most people today only praise purcell's book 2 in the series and ignore the rest. I felt it was a great series and if you could own them and tackle them, i thought you went over the hump and mastered 'all the hard stuff in physics'. The rest is detail.... [though people with EM I II and QM I II III in third and fourth year might disagree!]

    - griffiths - all three of his books - if you can read those and bite off a few chapters, you got it made. Griffith was said to be such a good textbook writer and lecturer, he could teach physics to hamsters [yet he chose a small college to teach at, and not a big name school]




    - Well I feel like I am capable of understanding these texts since I've already learned basic undergraduate classical mechanics and E&M, as for math: multi&diff. I feel I just have to brush up on some math like tensor notations and a bit of linear algebra.

    if you can tackle Kleppner/Symon for Int Mech after Halliday Resnick, that's pretty impressive for someone still in high school!

    and knowing partial diff and starting on diff equations sounds like you might get the math hurdle done quickly.

    Some schools actually flop differential equations in third year now, and not in second year physics, which sorta bothers me. I think the sooner you get some exposure the better.

    and i'm seeing that disturbing trend with intermediate mechanics, it's like they wanna compress intermediate mechanics into a marion and landau mechanics class as one difficult and steep hurdle which is a mistake. I think Symon and Kleppner should be 24 weeks, and then goldstein, but some wanna get right into Landau or other books way way too fast


    - Can anyone give a list of the best/most comprehensive physics textbooks for each area of physics for someone just learning it? Would it be more useful/straightforward to just use the landau series?

    Well what have you seen that you liked, and what level of books do you want?

    Landau, i've seen in 4th year physics, and grad school syllabuses, but you can see them in some second and third year classes too. [if you got a russian physics teacher, you might be dead - his background would be quite different and so would his approach to teaching it way sooner than most would]

    [I really scratch my head at some universities where year one mechanics is Halliday-Resnick and then there's no year two of [Marion] but it's just a year three class with no text required, landau as the recommended optional text]


    And is there a list of classic/popular math textbooks as well? For topics like analysis, differential geometry, topology, etc. Like the way there are more widely used textbooks in physics like griffiths, jackson, etc.

    The Vijay list of physics textbooks is probably the best one in existence...
    basically they toss any textbook that people don't object strongly to, when more than one person thinks it's a classic. So there's endless and endless recommendations there.
     
  19. Apr 25, 2013 #18
    thanks.
    so it's more important to understand all the concepts presented in a specific textbook rather than worrying about every little word in the textbook right? How would I know that some concepts are presented in a textbook that isn't in another textbook though? What's the difference between all the same level, same topic textbooks out there if the concepts are the same? For example, Jackson and Schwinger.

    And also, is it necessary for me to master everything in halliday/resnick first of all? Or can I just pick up individual topics with individual undergrad level books in topics such as thermodynamics, waves, optics, etc. So far, my high school only uses the mechanics and em sections for physics C. And I don't think colleges really use the other sections of that book for topics like waves/optics/quantum mechanics.
     
  20. Apr 26, 2013 #19
    I hope that, being in a similar situation to you, I can provide a bit of help.

    If not immediately obvious, a method that has worked for me is to complement textbooks with other sources (wikipedia, hyperphysics, MIT OCW, Berkeley, online notes (more) and some introductory papers I find particularly helpful, most universities will have quite a few lecture notes online: if you hate the current textbook, chances are there's a better one online for free), as reading from a limited number of sources, some subtleties that would be obvious if I were actually sitting the course at university can elude me when teaching myself. (This website seems good for homework problems, and Stackexchange (math and physics) is a useful way to get answers from experts concerning problems with the content (not homework), and often find troves of information, interesting papers or new areas of mathematics).

    The majority of my time is spent learning mathematics, which I find more challenging, as some physics at low-undergraduate level I find a little trivial (as opposed to, say, Landau Volume 1), and according to previous experience that learning physics is much easier if you don't have to worry about the maths, and can focus on the core issues. I also quite like reading historical papers or overviews for wider knowledge, but that's for fun and hardly essential. Also, things like number theory are really interesting for their own sake.

    However, I am cautious of turning into the mathematical physicist in Feynman's lectures that understands the mathematical structures and not the underlying physics.


    Like you, I also have a habit of flitting to the more interesting topics away from the stagnant ones (most often deterred by a roadblock hit in an exercise: ask online instead of switching topics because of refusal to go on without completion). Occasionally, this lends better intuition (with more time to mull over things or allowing me to forget, highlighting rote-learned subtopics for regrokking), but is slow. Perhaps if I pressed forward with one at a time, more net progress would be made, but I wouldn't enjoy learning as much.

    Always do at least all the non-trivial problems (I skip over those I know that I know the solution's method)! Especially in mathematics, the exercises are extensions of the chapter: some things are missed out of the main text with the assumption that you'll rediscover it yourself. Olympiad problems and a plethora of similar websites (Brilliant, Komal for instance) are useful also, but be sure to explore around the problems if they're interesting.
    I wouldn't solely re-derive formulae immediately after doing them, and would certainly search for (or come up with) alternate derivations, else it's all mechanical and no bellyfeel. I tend to re-derive it (in paper, and later in my head) when doing every problem involving the formula until I feel I know it inside out.

    From what I've gathered, real (university) physics has a much higher information density (per page or unit study time) than high-school physics, so will of course take longer to self-study.
     
    Last edited: Apr 26, 2013
  21. Apr 26, 2013 #20
    thanks. so basically, I have to spend time on thinking and applying what I've learned and do problems more so than just reading it right?
    what do u think about these questions?

    1) Is it more important to understand all the concepts presented in a specific textbook rather than worrying about every little word in the textbook right? How would I know that some concepts are presented in a textbook that isn't in another textbook though?

    2) When would I know when to read/look at lecture notes besides just reading the textbook and which combination of textbooks to use on specific areas within in one topic? And which problems to do since there are different problems in every textbook or lecture notes?

    3) What's the difference between all the same level, same topic textbooks out there if the concepts are the same? For example, Jackson and Schwinger, where both are graduate level EM books.

    4) And also, is it necessary for me to master everything in halliday/resnick first of all? Or can I just pick up individual topics with individual undergrad level books in topics such as thermodynamics, waves, optics, etc. So far, my high school only uses the mechanics and em sections for physics C. And I don't think colleges really use the other sections of that book for topics like waves/optics/quantum mechanics.

    5) Are Feynman's lectures as useful as a regular textbook? Or just a fun aside reading?

    6) Does this advice apply to studying math as well?

    thanks a lot.
     
    Last edited: Apr 26, 2013
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: Advice for self-studying physics?
  1. Advice on Self-Study (Replies: 15)

Loading...