1. Limited time only! Sign up for a free 30min personal 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!

Taking Derivatives in Physics

  1. Sep 29, 2012 #1
    When I take derivatives in math I think of it as the amount of infinitesimals that change in one variable with respect to another, when the latter changes by one infinitesimal. But in physics those variables have real life meanings, so when I take the derivative of position with respect to time I feel like I am asking myself how many infinitesimals of time change in distance. Which is kind of weird.

    Is that because all infinitesimals are the same? Like: dx=dt=dy=dz? And even if that's not the explanation for this particular question; are they all the same? Are the infinitesimals of all variables basically 1/∞?

    Anyway; or is it that some mathematical equation just happens to explain a physical occurrence? Like, we just assign real meanings to abstract variables and the derivatives are actually speaking of the variables, and only as a consequence about the physical meaning we assign to them?
  2. jcsd
  3. Sep 29, 2012 #2


    User Avatar
    Science Advisor

    Derivative in physics means the same thing as derivative in math. It's the rate of change of something with respect to something else. Are you familiar with the definition of a limit?
    "d" doesn't mean anything by itself. There must be at least 2 "d"s in an equation for it to mean anything, because "d" is really a device for comparing how two variables are changing with respect to each other.
    dx = 2dy means that anytime you change y by a small amount, x changes by twice that amount. Infinitesimals like dx and dy are not the same. It only makes sense to equate them with other infinitesimals (or expressions containing infinitesimals) because they aren't real numbers.
  4. Sep 29, 2012 #3
    Same for physics:


    [see the discussion there on differentiation]

    no; dx = dt only when dx/dt = 1.....

    we can assign real physical meaning to variables and their derivatives....if x is a distance
    measure, then dx/dt is the velocity [change in distance with respect to x]....and also
    dx/dt = dx/dy times dy/dt
  5. Sep 29, 2012 #4


    User Avatar
    Gold Member

    I suspect you are mixing up Delta x with dx since you speak of infinitesmals.

    An expression such as Delta y / Delta x = 2, would be more applicable to your inquiry about infinitesmals. Here, Delta y or Delta x has a certain size, value and can be made smaller and smaller until infinesimally small when the limit is reached whereDelta y or Delta x approach zero. For this expression Delta y will always be twice as large as Delta x. As the limit appraoches zero, you will recall, the slope of a graph at that certain point with which you are interested in will be the slope of a line tangent to the curve at that point.

    On the other hand when you are speaking of dy/dx you are really expressing the slope of the tangent line as it follows the curve of your initial expression of which you have taken the derivative. dy/dx is not the same as Delta y / Delta x, but rather the change in x versus the change in y , and that can be calculated for values of x for a function y = f(x) throughout the whole curve.

    For example, for a simple expression y = 2x, with derivative dy = 2 dx, or dy/dx = 2 , the slope, dy/dx, along the curve is always 2. and dx or dy on there own do not have a real value as mentioned in a previous post.

    Until you encounter more indepth calculus this brief explanation may suffice for your immediate purposes.
    Last edited: Sep 29, 2012
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook