If you're seeing this message, it means we're having trouble loading external resources on our website.

If you're behind a web filter, please make sure that the domains *.kastatic.org and *.kasandbox.org are unblocked.

Main content

Rewriting before integrating: challenge problem

In this example, we find the antiderivative an expression which is not so simple.  Created by Sal Khan.

Want to join the conversation?

  • leafers sapling style avatar for user S Talbot
    Correct me if I'm wrong, but when you apply the sum rule to break the antiderivative into smaller pieces, you would technically have an added constant for each piece. At when you add the constant, that constant would technically represent the sum of all smaller constants found when taking the general antiderivative of each of the smaller pieces, correct? It would be almost impossible to find a particular solution given an initial condition, since it we would be solving for multiple constant C's.
    (89 votes)
    Default Khan Academy avatar avatar for user
    • old spice man green style avatar for user Boschey
      c can be anything, as opposed to x, which refers to a single, unknown value. We add c to generalise the (INDEFINITE) integral, so c can be anything, and everything, it doesn't matter!!
      c = 0, 1, 2, 3, 4...(to infinity) and all the values in between, and all the negative numbers and and and...(you get the idea...basically all the numbers in the universe - can someone please let me know if this includes complex numbers...?)

      To conclude,, any number, plus any number is still, any number! (c + c = c)

      I emphasised INDEFINITE, because definite integrals refer to the area under the curve/line. (gross oversimplification!!) Clearly, changing c (or the height) would change the area. So c for definite integrals can only be one, specific, value, if any..
      (6 votes)
  • male robot johnny style avatar for user Sidhusan Devamanoharan
    why is it necessary to put that dx after every integral sign. Whats the purpose?
    (7 votes)
    Default Khan Academy avatar avatar for user
    • leafers ultimate style avatar for user Tim Lindsley
      I would disagree that the dx is mostly for show.

      You can think of integration simply as the sum of a lot of really skinny (infinitely skinny, but not zero) rectangles. Now, how do you get the area of a rectangle? Well, you multiply its length times its width. Integration just takes the length times the width of a bunch of skinny rectangles and sums them. So how does that connect to the integration expression?

      Let's say you have an integral that looks like this: ∫x^2dx. There are three parts to the integral: 1) the function part, x^2, represents the length of the rectangles--you can also think of it as the height of the function, or the y-value; 2) the dx in the integral represents the width of each rectangle; 3) and the elegant elongated S-symbol, ∫, tells you to sum the areas of the rectangles.

      Watch Sal's videos on Riemann sums and it will become obvious what dx means and why you need to have it there. Without the dx, you can't have any width. And if you don't have any width, then you can't have any area. And no area means nothing to sum. And nothing to sum means no integration. And no integration means no differentiation (you get a glimpse of the link between the two in this video, but you will see it in a profoundly beautiful way in the Fundamental Theorem of Calculus) . And no differentiation means no calculus. And think of how horrible the world would be without calculus. :)
      (30 votes)
  • leaf blue style avatar for user Vasundhara Paul
    Why are these called indefinite integrals?
    (4 votes)
    Default Khan Academy avatar avatar for user
  • ohnoes default style avatar for user Nick Kuzoff
    Is there a way to solve for C in the equation, or for any antiderivative?
    (4 votes)
    Default Khan Academy avatar avatar for user
    • leaf green style avatar for user ArDeeJ
      If you've given any initial conditions, then yes. If not, then not.
      For example, suppose you're given that f´(x) = 2x and f(0) = 1. Taking the antiderivative of f´(x) gives you x^2 + C, and with the initial condition that f(0) = 1 you can solve for f(x) = x^2 + 1.
      (8 votes)
  • leaf red style avatar for user Shawn
    Holy cow, I was sitting here wondering what a "hairier expression" is. I was thinking like a Fourier or something. I get it now.
    (2 votes)
    Default Khan Academy avatar avatar for user
  • female robot ada style avatar for user Srinath. S
    Hi Sal,
    Upto my understanding Integration is the inverse of Differentiation. So if we integrate 2x, we will get x^2. It means that d/dx (x^2) is 2x. My question is why should we neglect constants in integration? why not we can integrate some constant, say integration of 8 dx and it should be 8x accordingly . Because differentiation of 8x is 8 (a constant) right? Hope you will understand my doubt. Please reply me sal
    thanks for your nobel service sal
    Srini
    (3 votes)
    Default Khan Academy avatar avatar for user
    • leaf green style avatar for user ksinelli
      You are correct that if we are told to find the indefinite integral of 8, it would be 8x (because the derivative of 8x is 8). But if you are told to find the derivative of x^2 + 8, your answer is simply 2x. What about finding the derivative of x^2 + 9? Again, your answer is 2x. So, some information about the original function is lost when looking only at its derivative. This is the reason why we write "+ C" when finding the anti-derivative, because the derivative of x^2 + any constant is 2x. If there is no constant in the original function, you could still think of it as x^2 + 0.
      (3 votes)
  • aqualine ultimate style avatar for user angel
    I would really appreciate to look at the graph of such kind of an indefinite integral as to where do you actually put that constant C in that graph, i want to see it located in that particular graph cuz in definite integrals we have area under that specific part of that curve according to the limits we put up Its really confusing about the representation of C in the graph of indefinite integrals :-0
    (3 votes)
    Default Khan Academy avatar avatar for user
  • blobby green style avatar for user Daniel Ederango
    Hi Sir! I truly admire the way you teach, but I think their is something in this video of hairy integration that had been overlooked by Sir Khan. Please correct me if I'm wrong cause I'm just starting to learn this subject. The thing you had overlooked is in the 3rd term. the exponent you had written is -3/2 instead of -5/2. thank you :D
    (3 votes)
    Default Khan Academy avatar avatar for user
    • hopper cool style avatar for user POWA
      Lets take the anti-derivative of the third term alone, but first, some algebraic manipulation:

      (18x^(1/2))/(x^3)=18x^(1/2-3)=18x^(1/2-6/2)=18x^(-5/2)

      Now we take the anti derivative of 18x^(-5/2):

      Here we raise the exponent with 1 power:

      -5/2+1=-5/2+2/2=-3/2

      And divide by the exponent:

      [18x^(-3/2)]/[-3/2]=-12x^(-3/2) (+c if you will)

      For negative constants of exponents it might seem confusing to raise a power by one, but this is how we would do it.

      I hope this was a little helpful!
      (2 votes)
  • leaf red style avatar for user Joe Runner
    Correct me if I'm wrong, but is this the same as integration by parts?
    (2 votes)
    Default Khan Academy avatar avatar for user
  • blobby green style avatar for user Fatuma
    At 3.06, why is it that we don't take the antiderivative of x^3?
    (2 votes)
    Default Khan Academy avatar avatar for user

Video transcript

So our goal in this video is to take the antiderivative of this fairly crazy looking expression. Or another way of saying it is to find the indefinite integral of this crazy looking expression. And the key realization right over here is that this expression is made up of a bunch of terms. And the indefinite integral of the entire expression is going to be equal to the indefinite integral of each of the term. So this is going to be equal to, we could look at this term right over here, and just take the indefinite that, 7x to the third dx. And then from that, we can subtract the indefinite integral of this thing. So we could say this is, and then minus the indefinite integral of 5 times the square root of x dx. And then we can look at this one right over here. So then we could say plus the indefinite integral of 18 square root of x. Square roots of x, over x to the third dx. And then finally, I'm running out of colors here, finally I need more colors in my thing. We can take the antiderivative of this. So plus the antiderivative of x to the negative 40th power dx. So I've just rewritten this and color-coded things. So let's take the antiderivative of each of these. And you'll see that we'll be able to do it using our whatever we want to call it. The inverse of the power rule, or the anti-power rule, whatever you might want to call it. So let's look at the first one. So we have-- what I'm going to do is, I'm just going to find the antiderivative without the constant, and just add the constant at the end. For the sake of this one. Just to make sure we get the most general antiderivative. So here the exponent is a 3. So we can increase it by 1. So it's going to be x to the 4th. Let me do that same purple color, or pink color. It's going to be x to the 4th, or we're going to divide by x to the 4th. So it's x to the 4th over 4 is the antiderivative of x to the 3rd. And you just had this scaling quantity, the seven out front. So we can still just have the seven out front. So we get 7x to the 4th over 4. Fair enough. From that, we're going to subtract the antiderivative of this. Now at first this might not be obvious, that you could use our inverse power rule, or anti-power rule here. But then you just need to realize that 5 times the principal root of x is the same thing as 5 times x to the 1/2 power. And so once again, the exponent here is 1/2. We can increment it by 1. So it's going to be x to the 3/2. And then divide by the incremented exponent. So divide by 3/2. And of course we had this 5 out front, so we still want to have the 5 out front. Now this next expression looks even wackier. But once again, we can simplify a little bit. This is the same thing-- let me do it right over here-- this is the same thing as 18 times x to the 1/2 times x to the negative 3 power. x to the 3rd in the denominator is the same thing as x to the negative 3. We have the same base, we could just add the exponents. So this is going to be equal to 18 times x to the 2 and 1/2 power. Or another way of thinking about it, this is the same thing as 18 times x to the 5/2 power. Did I do that right? Yeah. Negative 3. Oh sorry, this is negative 2 and 1/2. Let me make this very clear. And this is going to be the negative 5/2 power. x to the negative 3 is the same thing as x to the negative 6/2. Negative 6/2 plus 1/2 is negative 5/2. So once again, we just have to increment this exponent. So negative 5/2 plus 1 is going to be negative 3/2. So you're going to have x to the negative 3/2. And then you divide by what your exponent is when you increment it. So divide it by negative 3/2. And then you had the 18 out front. And we obviously are going to have to simplify this. And then finally, our exponent in this term. Let me not use that purple anymore. The exponent in this term right over here is negative 40. If we increment it, we get x to the negative 39 power, all of that over negative 39. And now we can add our constant, c. And all we need to do is simplify all of this craziness. So the first one is fairly simplified. We can write it as 7/4 x to the 4th. Now this term right over here is essentially negative 5 divided by 3/2. So 5 over 3/2 is equal to 5 times 2/3, which is equal to 10 over 3. So this thing right over here simplifies to negative 10/3 x to the 3/2. And then we have all of this craziness. Now 18 divided by negative 3/2 is equal to 18 times negative 2/3. Which is equal to, well we can simplify this a little bit, this is the same thing as 6 times negative 2. Which is equal to negative 12. So this expression right over here is negative 12x to the negative 3/2. And then finally, this one right over here, we can just rewrite it as, if we want, we could, well, we could just write negative 1/39 x to the negative 39 plus c. And we're done. We've found the indefinite integral of all this craziness. And I encourage you to take the derivative of this. And you can do it using really just the power rule, to take the derivative of this. And verify that it does indeed equal this expression that we took the antiderivative of.