Do You Know A Cool Way To Solve This?

Calculus Level 2

lim x 1 ( 23 1 x 23 11 1 x 11 ) = ? \large \lim_{x \to 1} \left( \frac{23}{1-x^{23}}-\frac{11}{1-x^{11}} \right) = \, ?


The answer is 6.

This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try refreshing the page, (b) enabling javascript if it is disabled on your browser and, finally, (c) loading the non-javascript version of this page . We're sorry about the hassle.

10 solutions

Karthik Kannan
Jun 23, 2015

Let L = lim x 1 ( 23 1 x 23 11 1 x 11 ) . . . . . . . . . . . . . . . ( 1 ) \mathcal{L}=\displaystyle\lim_{x \to 1} \left( \dfrac{23}{1-x^{23}}-\dfrac{11}{1-x^{11}} \right)...............(1)

Now put x = 1 y x=\dfrac{1}{y}

When x x gets closer to 1 1 , y 1 y\rightarrow 1 .

L = lim y 1 ( 23 y 23 1 y 23 + 11 y 11 1 y 11 ) \therefore \mathcal{L}=\displaystyle\lim_{y \to 1} \left( \dfrac{-23y^{23}}{1-y^{23}}+\dfrac{11y^{11}}{1-y^{11}} \right)

L = lim x 1 ( 23 x 23 1 x 23 + 11 x 11 1 x 11 ) . . . . . . . . . . . . . . . ( 2 ) \therefore \mathcal{L}=\displaystyle\lim_{x \to 1} \left( \dfrac{-23x^{23}}{1-x^{23}}+\dfrac{11x^{11}}{1-x^{11}} \right)...............(2) (Since x x and y y are dummy variables)

Adding ( 1 ) (1) and ( 2 ) (2) we get

2 L = 23 11 2\mathcal{L}=23-11

L = 6 \therefore \mathcal{L}=\boxed{6}

Moderator note:

This solution assumes that the limit exists and is finite, in order to justify the steps. This is because we can perform arithmetic on the left-hand limit and the right-hand limit.

If however, the limit does not exist, or the limit is infinite, then such an approach could lead to an erroneous solution.

Will that really work (considering x and y as dummy variables)? I am not sure that you can change the variables without bothering about the relative values. Not sure whether this really going to work.

keshavendra singh - 5 years, 11 months ago

Log in to reply

That is a good question to ask.

If we assume that the limit exists and is finite, then the steps taken are valid. This is because we can perform arithmetic on the left-hand limit and the right-hand limit.

If however, the limit does not exist, or the limit is infinite, then such a approach could lead to an erroneous solution.

Calvin Lin Staff - 5 years, 11 months ago

Short and sweet.

Kushal Patankar - 5 years, 11 months ago

Peter Macgregor wrote "I like Karthik's ingenuity, but I think there is a missing step in his argument."

I completely agree, but I have an easy fix for it: call the difference of the fractions f(x) and let g(x)=(x-1)f(x).

Then it can be seen that in a neighborhood of one g is a smooth function (a difference of two smooth functions), with g(1)=1-1=0, and that the limit sought equals g'(1) and hence exists.

Peter Byers - 4 years, 8 months ago

Really good solution! +1

User 123 - 5 years, 11 months ago

Wow! very neat, very nice, thank you Karthik :-)

Kazem Sepehrinia - 5 years, 11 months ago

nice solution.....

Sourav Kumar Surya - 5 years, 11 months ago

Simply wonderful Karthik. You just underscored the beauty of Math!!

a bagchi - 4 years, 11 months ago

Iitian hoga koi ye

Aditya Sahu - 4 years, 11 months ago

Yes keshavendra is right and unless a LHL and RHL both are finite and equal we cannot apply the limit

jeevan agarwal - 4 years, 10 months ago

Awesome one bro

Satyam Tripathi - 4 years, 9 months ago

why set x = 1/y ? . .. why not set x = 3/4y or 1/2y or some arbitrary real number times y?

Terrell Bombb - 4 years, 7 months ago

Log in to reply

It's because we want y 1 y\rightarrow 1 as x 1 x\rightarrow 1 .

Karthik Kannan - 4 years, 6 months ago

I like how you introduced dummy variables gonna try this for myself. Thank You!

Utkarsh Kulshrestha - 4 years ago

Just plot the graph and you will clearly notice as x approaches 1 the limit of the function goes to 6. So simple and sweet

Victor Etse - 1 year, 1 month ago

nice joke!!

Mahfuzul Haque - 5 years ago

Can you explain how you got from 4th eqn to eqn (2) (the 3rd eqn)?

Hassan Javaid - 4 years, 11 months ago
Kazem Sepehrinia
Jun 22, 2015

Consider a more general form of this limit: L = lim x 1 ( p 1 x p q 1 x q ) L=\lim_{x \to 1} \left( \frac{p}{1-x^{p}}-\frac{q}{1-x^{q}} \right)

1) Let x = 1 + t x=1+t with t 0 t \to 0 , It follows that L = lim t 0 ( p 1 ( 1 + t ) p q 1 ( 1 + t ) q ) L=\lim_{t \to 0} \left( \frac{p}{1-(1+t)^p}-\frac{q}{1-(1+t)^q}\right) 2) According to taylor series expansions we have ( 1 + t ) a = 1 + a t + 1 2 ( a 1 ) a t 2 + O ( t 3 ) (1+t)^a=1+at+\frac{1}{2}(a-1)at^2+O(t^3) Use this series expansions to rewrite ( 1 + t ) p (1+t)^p and ( 1 + t ) q (1+t)^q neglecting O ( t 3 ) O(t^3) : L = lim t 0 ( p p t 1 2 ( p 1 ) p t 2 q q t 1 2 ( q 1 ) q t 2 ) L=\lim_{t \to 0} \left( \frac{p}{-pt-\frac{1}{2}(p-1)pt^2}-\frac{q}{-qt-\frac{1}{2}(q-1)qt^2}\right) 3) Go ahead: L = lim t 0 ( 1 t ( 1 1 1 2 ( p 1 ) t 1 1 1 2 ( q 1 ) t ) ) = lim t 0 ( 1 t ( 1 1 + 1 2 ( q 1 ) t 1 1 + 1 2 ( p 1 ) t ) ) = lim t 0 ( 1 t 1 2 ( p 1 ) t 1 2 ( q 1 ) t ( 1 + 1 2 ( q 1 ) t ) ( 1 + 1 2 ( p 1 ) t ) ) = lim t 0 ( 1 2 ( p 1 ) 1 2 ( q 1 ) ( 1 + 1 2 ( q 1 ) t ) ( 1 + 1 2 ( p 1 ) t ) ) = p q 2 \begin{array}{c}\\ L&=\lim_{t \to 0} \left( \frac{1}{t}\left( \frac{1}{-1-\frac{1}{2}(p-1)t}-\frac{1}{-1-\frac{1}{2}(q-1)t}\right)\right) \\ &=\lim_{t \to 0} \left( \frac{1}{t}\left( \frac{1}{1+\frac{1}{2}(q-1)t}-\frac{1}{1+\frac{1}{2}(p-1)t}\right)\right) \\ &=\lim_{t \to 0} \left( \frac{1}{t} \frac{\frac{1}{2}(p-1)t-\frac{1}{2}(q-1)t}{\left(1+\frac{1}{2}(q-1)t\right)\left(1+\frac{1}{2}(p-1)t\right)}\right) \\ &=\lim_{t \to 0} \left( \frac{\frac{1}{2}(p-1)-\frac{1}{2}(q-1)}{\left(1+\frac{1}{2}(q-1)t\right)\left(1+\frac{1}{2}(p-1)t\right)}\right) \\ &=\frac{p-q}{2} \end{array}

wow ! . .... beauty man! .... Generalisation is always great! .... were'd u get this ? ORIGINAL?

Abhinav Raichur - 5 years, 11 months ago

Log in to reply

Hey Abhinav, thanks man! Problem is not original, general form asked by someone in open study :-)

Kazem Sepehrinia - 5 years, 11 months ago

Nice method, Sir! +1

User 123 - 5 years, 11 months ago

Log in to reply

Thanks Ishan :-)

Kazem Sepehrinia - 5 years, 11 months ago

Excellent problem Kazem! I did this in two ways (1) factoring (1-x)^2 from numerator and denominator and taking limits of the residual polynomials in x, and (2) of course, L'Hospital's method. Now we know 4 ways of doing this problem.

a bagchi - 4 years, 11 months ago

oh my goodness. wow +1

Sean Thrasher - 3 years, 10 months ago

What is in this situation that you may just "neglect O(t^3)"?

Adrian Self - 3 years, 3 months ago

Log in to reply

Hi Adrian,

Its a 0 0 0-0 case limit. If I just neglect O ( t 2 ) O(t^2) the limit will be 0 0 0-0 again! Therefore, in order to get the right answer we should neglect higher orders.

Kazem Sepehrinia - 3 years, 3 months ago
Hobart Pao
Jun 24, 2015

Rewrite as lim x 1 ( 23 ) ( 1 x 11 ) ( 11 ) ( 1 x 23 ) ( 1 x 23 ) ( 1 x 11 ) \displaystyle\lim_{x\rightarrow 1} \dfrac{(23)(1-x^{11})-(11)(1-x^{23})}{(1-x^{23})(1-x^{11})}

If you try substituting 1 in for x, you'll get 0/0. L'Hopital's rule can be applied here.

Expand. lim x 1 12 23 x 11 + 11 x 23 1 x 11 x 23 + x 34 \displaystyle\lim_{x\rightarrow 1} \dfrac{12-23x^{11}+11x^{23}}{1-x^{11}-x^{23}+x^{34}}

Differentiate the numerator and denominator separately. lim x 1 23 11 x 10 + 23 11 x 22 11 x 10 23 x 22 + 34 x 33 \displaystyle\lim_{x\rightarrow 1} \dfrac{-23\cdot 11 \cdot x^{10} + 23 \cdot 11 \cdot x^{22} }{-11x^{10}-23x^{22}+34x^{33}}

Substituting 1 for x still gives 0/0, so differentiate the numerator and denominator separately, again. lim x 1 23 11 10 x 9 + 23 22 11 x 21 10 11 x 9 23 22 x 21 + 34 33 x 32 \displaystyle\lim_{x\rightarrow 1} \dfrac{-23\cdot 11 \cdot 10 \cdot x^{9} + 23 \cdot 22 \cdot 11 \cdot x^{21} }{-10\cdot 11\cdot x^{9}-23\cdot 22 \cdot x^{21}+34 \cdot 33 \cdot x^{32}}

Substituting 1 for x now gives us 23 11 10 + 23 22 11 10 11 23 22 + 34 33 \dfrac{-23\cdot 11 \cdot 10 + 23\cdot 22\cdot 11}{-10 \cdot 11 - 23 \cdot 22 + 34 \cdot 33 } , which simplifies to 6 \boxed{6} .

Thanks :-)

Kazem Sepehrinia - 5 years, 11 months ago

Very good! Simple and direct! Congratulations!

Vinicius Meza - 5 years, 5 months ago

Very nice Hobart. This I found to be the easiest method for me.

a bagchi - 4 years, 11 months ago

hopital rull 2 ok

Patience Patience - 4 years, 11 months ago

Very nice solution.

Hassan Javaid - 4 years, 10 months ago

Let the limit be L L .

L = lim x 1 ( 23 1 x 23 11 1 x 11 ) \displaystyle L = \lim_{x \to 1} \left( \dfrac{23}{1-x^{23}} - \dfrac{11}{1-x^{11}} \right)

L = lim x 1 ( 23 ( 1 x 11 ) 11 ( 1 x 23 ) ( 1 x 23 ) ( 1 x 11 ) ) \displaystyle L = \lim_{x \to 1} \left( \dfrac{23(1-x^{11}) - 11(1-x^{23})}{(1-x^{23})(1-x^{11})} \right)

L = lim x 1 ( 12 23 x 11 + 11 x 23 1 + x 34 x 23 x 11 ) \displaystyle L = \lim_{x \to 1} \left( \dfrac{12 -23x^{11} + 11x^{23}}{1 + x^{34} - x^{23} - x^{11}} \right)

This is of the form 0 0 \dfrac{0}{0} . So we will use L-Hopital's Rule.

L = lim x 1 ( 23.11 x 10 + 11.23 x 22 34 x 33 23 x 22 11 x 10 ) \displaystyle L = \lim_{x \to 1} \left( \dfrac{-23.11x^{10} + 11.23x^{22}}{34x^{33} - 23x^{22} - 11x^{10}} \right)

Which is again of the form 0 0 \dfrac{0}{0} .

L = lim x 1 ( 23.11.10 x 9 + 11.23.22 x 21 34.33 x 32 23.22 x 21 11.10 x 9 ) \displaystyle L = \lim_{x \to 1} \left( \dfrac{-23.11.10x^{9} + 11.23.22x^{21}}{34.33x^{32} - 23.22x^{21} - 11.10x^{9}} \right)

L = 23.11 ( 22 10 ) 11 ( 34.3 23.2 10 ) = 23.11 ( 12 ) 11 ( 46 ) = 6 L = \dfrac{23.11(22-10)}{11(34.3 - 23.2 - 10)} = \dfrac{23.11(12)}{11(46)} = \boxed{6}

Thank you.

Kazem Sepehrinia - 5 years, 11 months ago

Log in to reply

You are welcome. Anytime.

Vishwak Srinivasan - 5 years, 11 months ago
Rudra Jadon
Aug 2, 2017

Simply apply l hopitals rule

Peter Macgregor
Jun 25, 2015

I like Karthik's ingenuity, but I think there is a missing step in his argument. As it stands it could be applied to each term separately to show, for example, that

lim x 1 ( 23 1 x 23 ) = 23 2 \displaystyle\lim_{x \to 1}\left(\frac{23}{1-x^{23}}\right)=\frac{23}{2} which is clearly wrong!

To make his argument watertight he would need to show that his two limits really are equal as one approaches x = 1 x=1 from above and the other one approaches from below.

My approach was to use l'Hopital's rule. Consider the more general problem

lim x 1 ( p 1 x p q 1 x q ) = lim x 1 ( p ( 1 x q ) q ( 1 x p ) ( 1 x p ) ( 1 x q ) ) \displaystyle\lim_{x \to 1}\left(\frac{p}{1-x^{p}}-\frac{q}{1-x^{q}}\right)=\lim_{x \to 1}\left(\frac{p(1-x^q)-q(1-x^p)}{(1-x^{p})(1-x^q)}\right)

Since this is indeterminate when x = 1 x=1 we can use l'Hopital's rule to get

lim x 1 ( p q ( x p 1 x q 1 ) q x q 1 p x p 1 + ( p + q ) x p + q 1 ) \displaystyle\lim_{x \to 1}\left(\frac{pq(x^{p-1}-x^{q-1})}{-qx^{q-1}-px^{p-1}+(p+q)x^{p+q-1}}\right)

This is still indeterminate when x = 1 x=1 so we can use l'Hopital again to get

lim x 1 ( p q ( ( p 1 ) x p 2 ( q 1 ) x q 2 ) ) q ( q 1 ) x q 2 p ( p 1 ) x p 2 + ( p + q ) ( p + q 1 ) x p + q 2 ) \displaystyle\lim_{x \to 1}\left(\frac{pq((p-1)x^{p-2}-(q-1)x^{q-2}))}{-q(q-1)x^{q-2}-p(p-1)x^{p-2}+(p+q)(p+q-1)x^{p+q-2}}\right)

Now at last substituting x = 1 x=1 gives the intelligible answer p q 2 \frac{p-q}{2}

In the given problem p = 23 p=23 and q = 11 q=11 leading to the answer 6 \boxed{6}

I must admit that this is all rather messy, but I love the way l'Hopital's rule makes use of a seeming disaster!

Thanks for your solution and comments for the problem sir :-)

Kazem Sepehrinia - 5 years, 11 months ago

If we do some algebra the original expression is equal to:

lim x 1 ( 23 ( 1 x 11 ) 11 ( 1 x 23 ) ( 1 x 23 ) ( 1 x 11 ) ) \lim_{x\to\ 1}\left( \frac{23(1-x^{11})-11(1-x^{23})}{(1-x^{23})(1-x^{11})} \right)

= lim x 1 ( 23 ( 1 x 11 ) 11 ( 1 x 23 ) ( 1 x ) 2 ( 1 + x + x 2 + . . . + x 22 ) ( 1 + x + x 2 + . . . + x 10 ) ) \lim_{x\to\ 1}\left( \frac{23(1-x^{11})-11(1-x^{23})}{(1-x)^2(1+x+x^2+...+x^{22})(1+x+x^2+...+x^{10})} \right)

Let's factor 1 x 1-x from the numerator and we get:

= lim x 1 ( ( 1 x ) ( 23 ( 1 + x + x 2 + . . . + x 10 ) 11 ( 1 + x + x 2 + . . . + x 22 ) ( 1 x ) 2 ( 1 + x + x 2 + . . . + x 22 ) ( 1 + x + x 2 + . . . + x 10 ) ) \lim_{x\to\ 1}\left( \frac{(1-x)(23(1+x+x^2+...+x^{10})-11(1+x+x^2+...+x^{22})}{(1-x)^2(1+x+x^2+...+x^{22})(1+x+x^2+...+x^{10})} \right)

We can cancel 1 x 1-x from the top. Now if we look at the polynomial on the numerator (let's call it p ( x ) p(x) ) we see that:

p ( 1 ) = 23 ( 1 + 1 + 1 2 + . . . + 1 10 ) 11 ( 1 + 1 + 1 2 + . . . + 1 22 ) = 23 11 11 23 = 0 p(1)=23(1+1+1^2+...+1^{10})-11(1+1+1^2+...+1^{22}) = 23 \cdot 11 -11 \cdot 23 =0 then x 1 x-1 is a factor of p ( x ) p(x) .

Let's say p ( x ) = q ( x ) ( x 1 ) p(x)=q(x)(x-1) then by long division we get that

q ( x ) = 11 x 21 11 ( 2 ) x 20 + . . . + ( 11 ) ( 11 ) x 11 + ( 11 ) ( 12 ) x 10 + 12 ( 10 ) x 9 + 12 ( 9 ) x 8 + . . . + 12 ( 2 ) x 1 + 12 ( 1 ) q(x) = -11x^{21}-11(2)x^{20}+...+(-11)(11)x^{11}+(-11)(12)x^{10} + 12(-10)x^{9} +12(-9)x^{8}+...+12(-2)x^1+12(-1) cool :)

Now the original limit is just:

lim x 1 ( q ( x ) s ( x ) ) = q ( 1 ) s ( 1 ) \lim_{x\to\ 1}\left( \frac{q(x)}{s(x)} \right) = \frac{q(1)}{s(1)}

Here s ( x ) = ( 1 + x + x 2 + . . . + x 22 ) ( 1 + x + x 2 + . . . + x 10 ) s(x)= -(1+x+x^2+...+x^{22})(1+x+x^2+...+x^{10})

And the result is just 6 6 . :)

(I would like to know if that method can be generalized for any p p and q q as in the other solutions)

Amrit Bahl
Mar 22, 2020

Hey guys just cross multiply and make the limit into one fn which it u put X=1, you get 0/0 form therefore use l'hospital rule twice to get a non 0/0 answer which is 6.

Bert Seegmiller
Nov 1, 2018

Tested various selected values of x x and noted where values headed as x 1 x \Rightarrow 1 and saw the difference was approaching (relatively quickly) 6 \boxed{6} .

I know that's cheating, so to speak, but it worked. 😏

Sarvesh Dubey
Jun 22, 2015

Lhopital rule always works... Double differentiate both the numerator and denominator....

"Lhopital rule always works..."(sic) is in general a wrong statement.

Kartik Sharma - 5 years, 11 months ago

L'Hopital rule can be applied only if lim x 0 or p q \displaystyle\lim_{x \to 0 \text{ or } \infty} \frac{p}{q} becomes either 0 0 o r \frac{0}{0} or \frac{\infty}{\infty} .Other indeterminate forms can be solved using L ' Hopitals rule after some tweaking ( check out Ben Whitmore's solution in this problem this for an example of how L'hopital can be used sometimes for other indeterminate cases )

Athiyaman Nallathambi - 5 years, 9 months ago

Right now, the fraction 23 1 x 23 \frac{23}{1-x^{23} } is not of indeterminate form, so you need to be more specific about what you're doing.

Calvin Lin Staff - 5 years, 11 months ago

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...