Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Computational Phyiscs

  1. Nov 10, 2004 #1
    [SOLVED] Computational Phyiscs

    Can anyone tell me what computational physics is all about? I'm "into" computers in general and was planning on doing a computer science degree until my interest in physics turned me away from it. The idea of a career that includes both computers and physics would be a dream come true for me. Also, what part do computers play in other divisions of physics?

    Thanks in advance.
  2. jcsd
  3. Nov 10, 2004 #2

    Dr Transport

    User Avatar
    Science Advisor
    Gold Member

  4. Nov 11, 2004 #3
    Thanks a lot!
  5. Nov 11, 2004 #4
    I'm not sure if this helps, but you can stimulate physics and physical systems with computer programming (like in games, for example).
  6. Nov 19, 2004 #5
    Could anyone give me some examples of the kind of work done as a Computational Physicist? How does there knowledge of physics compare to that of a Theoretical Physicist?
  7. Nov 20, 2004 #6


    User Avatar
    Science Advisor
    Gold Member

    With respect to information I think the phrase "its good have as much as possible, if not all", applies. As much theoretical background as possible added with computational skills would be my recipe for success. I think the best guys are those who actually know in-depth what they are doing ... great surprise. I for one work a lot with finite element, Galerkin etc. methods for various differential eqs and various physical scales, from the quantum scale to "our" macroscopic scale. Lots of it has to do with model development and implementation (I basically do various kinds of materials research, so its material constitutive models of different sorts), followed by analysis efficiency, convergence etc. questions.
  8. Nov 20, 2004 #7

    Dr Transport

    User Avatar
    Science Advisor
    Gold Member

    There are two types of theoretical physicists, the type that come up with the grand theory, QED, Strings, etc.... then there are guys like me, we come in and clean up, do some pick and shovel work and extend it to other applications. Computational physicists fall into the later, I take the general theory then apply it in specific cases and model specific systems.

    My area of expertise is transport in anisotropic semiconductors, the general theory for transport has been known for 50 years, but the solutions to the specific material systems have not been solved numerically to date, only isotropic crystals like Si and Ge have been published. There is nothing analytic about this work, it must be done numerically, hence I consider myself a computational physicist.

    Another area where a computational guy gets alot of work is in fluids, Computational Fluid Dynamics is a big deal, they use it in Aerospace, the automotive industry, pretty much anywhere where a body moves at large speeds.
  9. Nov 26, 2004 #8
    The application of computational physics to neuroscience is of particular interest, mapping neural networks, physics of action potentials, etc.. which is developing very rapidly and is under high demand for research.. (just a thought!)
  10. Dec 9, 2004 #9
    I agree with Dr Transport, a lot of the computational physicists are the guys that "clean up", or verify that the guys in the theory department (you know, the guys locked in rooms with large blackboards and empty coffee cups strewn about) are coming up with the right equations. I however, am one of those guys locked in a room with a blackboard, and computational physics is a regularly occuring theme in my research as well. I'm in a particle physics theory group working with QCD Hamiltonian renormalization, and I often find the need to write a piece of code to do this or that... while I might not spend as much time on a cluster as some of the other research assistants in my department... I think that anybody will tell you that if you're going into theory, computational techniques are a must-have set of tools...

    My suggestion therefore is if you really want to do physics for the rest of your life, find a field that you really enjoy working in, then, assuming you still have a love for computing, apply that to your selected field.
  11. Dec 9, 2004 #10
    I'm an undergraduate physics major, and just from what i've seen computational skills are essential, no matter what field of physics you go into. Our departmentheavily recommends computer science courses for physics majors.

    This next quarter i'll be working on a 'special problems for undergrad' class, nothing spectacular, just modelling energy transport mechanisms in stellar interiors, but it will involve a lot of computer modelling because of the non-linearity of just about every equation, everything has to be solved numerically.

    Can't wait to wind up being one of the guys locked up in the room with the blackboard and coffee cups everywhere. Looking forward to that part. :biggrin:
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook