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

Simple Differential Problem

  1. Jul 2, 2005 #1
    This is a problem from a Calc 1 text book, I just can't figure out where they get dx from. The question is:
    Use differentials to approximate the value of the expresesion.

    the answer in the solution manual says:
    Let [tex]f(x)=\sqrt{x}, x=100, dx=-.6[/tex]

    then it solves the problem. But what I don't understand is where the heck is this dx quantity derived from?
  2. jcsd
  3. Jul 2, 2005 #2


    User Avatar
    Science Advisor

    [tex]f(x)=\sqrt{x}, x=100, dx=-.6[/tex]

    Your textbook is using the fact that df= f'(x)dx so that f(x+dx) is approximately
    f(x)+ df= f'(x)dx.
    Here, you want to evaluate f(99.4) and it easy to see that f(100)= 10 so take x= 100 and x+ dx= 99.4. What is dx?
  4. Jul 2, 2005 #3
    Wow, I get it. It's so simple. Dx is -0.6. This also makes it more clear why it is called a differential. Thanks Ivy! Hey Ivy, are these differentials the same thing that differential equations are based on?
    Last edited: Jul 2, 2005
  5. Jul 3, 2005 #4
    Though similar, they are not exactly the same thing.

    In terms of differentials df=f'(x)dx is true but taking dx= -0.6 is only an approximation. In the true sense a differential can be considered as an infinitesimally small change in the variable.

    You might be knowing the Taylor's series expansion of f(x+h)

    f(x+h) = f(x) + hf'(x) + (h^2)/2! f"(x) + (h^3)/3! f"'(x) + ...

    If h is very small relative to x, the first order approximation can be obtained as f(x+h) = f(x) + hf'(x), which is the same as the equation with differentials
    f(x+dx) = f(x) + f'(x)dx

    What I want to imply is that the dx in your question is an approximation but differentials in a differential equation do not signify any approximations.
  6. Jul 3, 2005 #5


    User Avatar
    Science Advisor

    Strictly speaking, differentials, like "dx", are "infinitesmals". When you write something like "dx= -0.6", it really is [tex]\Delta x[/tex], meaning a small change in x. One way of defining the derivative is [tex]lim_{\Delta x->0} \frac{\Delta y}{\Delta x}[/tex]. You can then use [tex]\Delta x[/tex] to approximate dx:
    [tex]\frac{\Delta y}{\Delta x}[/tex] is approximately [tex]\frac{dy}{dx}[/tex] so [tex]\Delta y[/tex] is approximately [tex]\(\frac{dy}{dx}\)\Delta x[/tex]. The smaller [tex]\Delta x[/tex] is, the better the approximation.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook