Tetration is defined as
n a = n a ’s a a ⋅ ⋅ a .
Find the value of
n → ∞ lim n ( 2 ) .
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.
This issue is covered in this question .
L e t u n = n ( 2 ) t h e n u 1 = 2 ≤ 2 . u n + 1 = ( 2 ) u n , s o i f u n ≤ 2
t h e n u n + 1 ≤ ( 2 ) 2 = 2 .
H e n c e u n ≤ 2 f o r a l l n ( i n d u c t i o n ) a n d s o t h e l i m i t i s 2 n o t 4
It can't be 4 , because it can't be greater than e .
If you treat it like a series and check how the ratio between a number and the next decreases you will notice it could never go higher than two
I had but it in calculator and it doesn't converge !!!!!
Here i will tag you @Sharky Kesa ... please see, and say something...
Log in to reply
Well, if you try with smaller tetrations, you see that it gets closer to 2. At infinity, it would be 2.
Log in to reply
But why not 4, that also could be !
See by even your method, think of doing for smaller ones, then think that it becomes 2.
2 2 > 2 1
2 2 2 > 2 2 1
This way, we can think that it will increase , and tend to 4.
Log in to reply
@Aditya Raut – Remember that the base is 2 , not 2 .
Log in to reply
@Sharky Kesa – No, i meant if you think it becomes 2 at some place, then start thinking about further... Forget, try this problem , the 16th part of the popular set...
Log in to reply
@Aditya Raut – But it never actually reaches 2. It always gets closer and closer. Think of it like 1 + 2 1 + 4 1 + . . . . It never actually reaches 2 but gets closer and closer.
This is how I thought about it:
2 2 2 . . . ∞ = ( 2 1 / 2 ) ( 2 1 / 2 ) ( 2 1 / 2 ) . . . ∞ = 2 1 1 1 . . . ∞ = 2
Let x = lim n → ∞ n 2 . It is clear that going one further, you get 2 x = x . 2 x = x 2 . x = 2 , 4 .
To prove x = 2 , use induction to price that n 2 < 2 .
Suppose n 2 < 2 then n + 1 2 = 2 n 2 < 2 2 . Since 2 0 < 2 it is true for all n by induction.
Let x = ∞ 2 ⇒ x = 2 x ⇒ x 2 = 2 x ⇒ x = 2
Why can't x = 4 ?
Log in to reply
Because the infinite tetration, y = ∞ x x x ⋅ ⋅ ⋅ x = ∞ x fails to converge if x > e , where e is the Euler's number.
But I guess you already know this. :D
Log in to reply
Can you Elabrote on the concept? and here Its not x^x, but its (root(2))^x . and root(2)< e !
Alrite. so how about x = − 0 . 7 6 6 6 6 4 6 9 5 9 6 2 1 2 3 0 9 3 1 1 ?
Log in to reply
@Julian Poon – You probably know that if a sequence a n has a limit when n → ∞ then this limit is unique! Of course, x 2 = 2 x has 3 solutions, which means that 2 of them do not solve our initial problem. Why can't the solution be ANYTHING negatif? Well, look at the equation x = 2 x . The RHS is certainly positif, which means that the LHS , i.e. x is positif!!
A formal proof can be written as follows, but it may be a little boring. The only result you need to produce a rigorous mathematical proof is this fact "If a sequence a n is bounded and increasing (that means that for all n , you have a n + 1 ≥ a n ) then is is convergent."
That said let's produce a formal proof :)
First of all the sequence of the exercise is best defined by recursion
a n = { 2 ( 2 ) a n − 1 if n = 1 if n ≥ 2
Now we just prove the following steps.
Step 1. For every integer n ≥ 1 we have 2 ≤ a n < 2 . This can be drawn out by induction (boring I know). For n = 1 is is clearly true. If 2 ≤ a n < 2 than we have a n + 1 = ( 2 ) a n < ( 2 ) 2 = 2 (since the function y = 2 x is strictly increasing cause the base is about 1.4 and hence greater than 1. And from the other side we have (again by the fact that the exponential function with base greater than 1 is strictly increasing) a n + 1 = ( 2 ) a n ≥ ( 2 ) 2 > ( 2 ) 1 = 2 . We got the result we wanted 2 ≤ a n + 1 < 2 and by induction we proved the step 1.
Step 2. For every integer n ≥ 1 we have a n + 1 > a n . Again we use induction. For n = 1 is easy to see that is is true. Suppose now a n + 1 > a n and let's try to show that a n + 2 > a n + 1 . In fact a n + 2 = ( 2 ) a n + 1 > ( 2 ) a n = a n + 1 and by induction we have shows the step 2 is true.
Now we have that a n is bounded (step 1) and is (strictly) increasing. So it converges to a finite limit we can call L .
By the equality (that holds for all integers n ≥ 2
( 2 ) a n − 1 = a n
we can derive the equality of limits
n ↦ ∞ lim ( 2 ) a n − 1 = n ↦ ∞ lim a n
by the continuity of the exponential function we have
( 2 ) lim n ↦ ∞ a n − 1 = n ↦ ∞ lim a n
so we have
( 2 ) L = L
Now it is algebra (notice that L is positive cause is the limit of a sequence of numbers greater that 1.4)
2 2 L = L
squaring both the positive members
2 L = L 2
that has two solutions easy to find
L = 2 and L = 4
But L cannot be 4, because it is a limit of a sequence with terms lower than 2.
So it is necessary that L = 2 .
Let a a a . . . = x where a = 2
The equation becomes ( 2 ) x = x
Which gives x = 0 , 2 , but we ignore x = 0 because the number is more than 0. ~~~
Actually, x = 0 doesn't work because it would end up 1 = 0 .
( 2 ) 0 = 1 , and 1 != 0 so x = 0 is not a solution
The problem is not incorrect. It is easy to see that the limit can be evaluated from equation: y = ( 2 ) y .
this is true for y = 2 and y = 4 . However, let's give a simple proof why the solution is not y = 4 . Let's prove by simple induction that n ( 2 ) < 2 for all natural numbers n .
For n = 1 , it is obvious that 2 < 2 . Assume our proposition holds for all natural numbers n , and let's prove that then it holds for n + 1 .
Let a n = n ( 2 ) < 2 by inductive hypothesis. Then a n + 1 = ( 2 ) a n < ( 2 ) 2 = 2
The inequality holds because of basic properties of exponential function, and our hypothesis. Therefore we have proven our statement for all natural numbers, so our limit can be no higher than 2, and hence it cannot be 4. Therefore L = 2 , which is the solution to the problem.
Combining the following result completes the proof: Let y(n)=(sqrt2)^(y(n-1)), y(1)=sqrt(2). y(2)=(sqrt2)^(sqrt2)=1.633>sqrt(2)=1.4142 Assume y(n)>y(n-1) Now, consider y(n+1)/y(n) = (sqrt2)^(y(n)) / (sqrt2)^(y(n-1)) = (sqrt2)^(y(n)-y(n-1)) > (sqrt2) ^0 = 1 Hence, y(n)>y(n-1) for all positive integer n. With the above results, the limit exists and can be denoted by, say, s. Hence, s=sqrt(2) ^s or s^2=2^s implies s=2. Note that by graphing, the solution is unique.
Let A=\sqrt{2}^\sqrt{2}^\sqrt{2}^{...}
inf(sqrt(2)) = (2)inf/2 *inf/2.....................=(2)inf/2(raised to inf) = (2)inf/inf
= 2
inf = infinity
infinity is not a number, so even though this gives a valid result, it is not valid mathematics
Problem Loading...
Note Loading...
Set Loading...
This problem is probably incorrect.
Let y = 2 2 2 . . .
This simply becomes, y = ( 2 ) y
Now we make this into lo g y = 2 y lo g 2
This is true at y = 2 , because ( 2 ) 2 = 2
but also at y = 4 , because ( 2 ) 4 = 2 2 = 4 , and hence 4 = ( 2 ) 4
Hence this has got 2 solutions, they're y = 2 and y = 4 . I don't know how to decide, but I got it right when I tried both, right at the time of 2 , but not satisfied.
Please say something \(\color{Purple}{\textbf{ @Sharky Kesa }}\)