The Student Room Group

The Proof is Trivial!

Scroll to see replies

Original post by TheMagicMan
I've posted this problem before (not on this thread)

Problem 242 *


This has been posted already :wink:
(edited 10 years ago)
Reply 1701
Original post by jack.hadamard
Problem 241 * / **

i) Find sin(π5)\displaystyle \sin \left(\frac{\pi}{5}\right).

ii) Prove that πϕ>5\displaystyle \pi \phi > 5, where ϕ\phi is the golden ratio.


I made this problem up and it is not straightforward for a single star.

Challenge accepted :wink:

Solution 241
*

First we note that cos(x)=sin(4x)cos(x)(2sin(x)1)(4sin(x)+2sin(x)1)=0cos(x)=0, sin(x)=12, sin(x)=14(1±5)\cos(x)=\sin(4x) \Rightarrow \cos(x)(2 \sin(x)-1)(4 \sin(x)+2 sin(x) -1)=0 \Rightarrow \cos(x)=0 , \ \sin(x)=\frac{1}{2}, \ \sin(x)=\frac{1}{4}(-1 \pm \sqrt{5})

Solving for x, we get x=2πn±(π24x)x=π10, π6x=2 \pi n \pm (\frac{\pi}{2}-4x) \Rightarrow x= \frac{\pi}{10} , \ \frac{\pi}{6} for acute x.

Comparing the acute angled routes we deduce that sin(π10)=14(51)sin(π5)=2sin(π10)cos(π10)[br]=2sin(π10)1sin2(π10)=5858 \begin{aligned} \sin(\frac{\pi}{10})=\frac{1}{4}(\sqrt{5}-1) \Rightarrow \sin(\frac{\pi}{5})=2 \sin(\frac{\pi}{10}) \cos( \frac{\pi}{10}) \end{aligned}[br]= 2 \sin(\frac{\pi}{10}) \sqrt{1-\sin^2(\frac{\pi}{10})}=\sqrt{ \frac{5}{8}-{\sqrt{\frac{5}{8}}}} \ \square.

Now, the arc of a regular n-gon of circumradius 1 (which is enclosed by a surrounding circle of radius 1) is clearly n2sin(2πn)n2sin(2πn<π π>10sin(π10)\frac{n}{2} \sin(\frac{2 \pi}{n}) \Rightarrow \frac{n}{2} \sin(\frac{2\pi}{n} < \pi \ \Rightarrow \pi > 10\sin(\frac{\pi}{10}) by considering the case n=20.

πϕ>10(14(51))(12(5+1))=54(51)=5 \therefore \displaystyle \pi \phi >10 (\frac{1}{4} (\sqrt{5}-1))(\frac{1}{2} (\sqrt{5}+1))=\frac{5}{4}(5-1)=5 \ \square

Edit: Btw, the problem would be improved upon by saying that values for π\pi and 5\sqrt{5} are not quotable (forcing people to find a clever method or give a long tedious derivation of π\pi to 2 decimal places whereby, if an infinite series is used, it must be derived). That or what TheMagicMan said in adding a "hence" :smile:
(edited 10 years ago)
Reply 1702
Original post by Lord of the Flies

Spoiler


Cheers dude! :biggrin: Fully satisfied! I've always looked at the difference is perimeter being because of fractal dimension but perhaps it's not? Could it perhaps be true that one of f and g are required to be discontinuous at n number of points with nn \to \infty?

I really need to start using that arc length formula more! 'Tis certainly a beast :biggrin:
(edited 10 years ago)
Original post by Jkn

Now, the arc of a polygon of circumradius 1 (which is enclosed by a surrounding circle of radius 1)...


Area of a regular n-gon with unit circumradius. :tongue: The two-star method I intended for the question was to first find sin(π5)\sin\left(\frac{\pi}{5}\right). It is a root of the equation x(1+4(1x2)(14x2))=0x(1 + 4(1 - x^2)(1 - 4x^2)) = 0. Hence, get sin(π10)\sin\left(\frac{\pi}{10}\right) and use it in

012ϕ1111x2 dx\displaystyle \int_{0}^{\frac{1}{2}\phi^{-1}} 1 - \frac{1}{\sqrt{1 - x^2}}\ dx

to obtain the inequality. :smile:
Reply 1704
Original post by jack.hadamard
Area of a regular n-gon with unit circumradius. :tongue: The two-star method I intended for the question was to first find sin(π5)\sin\left(\frac{\pi}{5}\right). It is a root of the equation x(1+4(1x2)(14x2))=0x(1 + 4(1 - x^2)(1 - 4x^2)) = 0. Hence, get sin(π10)\sin\left(\frac{\pi}{10}\right) and use it in

012ϕ1111x2 dx\displaystyle \int_{0}^{\frac{1}{2}\phi^{-1}} 1 - \frac{1}{\sqrt{1 - x^2}}\ dx

to obtain the inequality. :smile:

Wow, that seems rather elaborate! Did you intend that the * solution be precisely what I did?

So how are you actually obtaining the inequality in that method? What specific ** technique is being used? :tongue:
Original post by Jkn
What specific ** technique is being used? :tongue:


The 'technique' is simply if f(x)<0f(x) < 0 for a<x<ba < x < b, then abf(x) dx<0\int_{a}^{b} f(x)\ dx < 0.
Original post by joostan
By parts is reverse product rule which is C3 :lol:


Don't discourage people from posting questions. :smile: If you don't like a question, then don't do it. :tongue:
Original post by jack.hadamard
Don't discourage people from posting questions. :smile: If you don't like a question, then don't do it. :tongue:

To be fair it was an awful question :laugh:

We need some more * questions though, I just don't know any which would be of interest.
Original post by Mladenov
...


What universities are you considering (or have applied to, or have been accepted in)?
Original post by jack.hadamard
Don't discourage people from posting questions. :smile: If you don't like a question, then don't do it. :tongue:


That's not what I meant by any means, I hope I didn't discourage anyone :redface:
Original post by jack.hadamard
Yes, I did, but still thought the answer was kind of irritating. :tongue:

Spoiler



Yes that's right! Why did you think the answer is irritating? :tongue:

Spoiler

Original post by shamika
Yes that's right! Why did you think the answer is irritating? :tongue:


It's just because I find it not-quite-so intuitive. Having the same probability of the three numbers (picked from all positive integers) being the sides of a triangle as they being not the sides of a triangle is puzzling. If we try to formalise such statements, then considering the probability space (Ω,F,P)(\Omega, \mathcal{F}, \text{P}) with

i) Ω:={ 1,2,3,... }\Omega := \{\ 1, 2, 3, ...\ \}, F=P(Ω)\mathcal{F} = \mathcal{P}(\Omega),

ii) if AFA \in \mathcal{F}, then P(A)=aAP({a})=πA\displaystyle \text{P}(A) = \sum_{a \in A} \text{P}(\{a\}) = \pi |A| where π\pi is the (equal) probability of any given positive integer.

makes no sense, since P(Ω)={0 when π=0when π>0\text{P}(\Omega) = \begin{cases} 0\quad\ \text{when}\ \pi = 0 \\ \infty\quad \text{when}\ \pi > 0 \end{cases}.

Following the above, I also find other "intuitive" statements irritating. Like picking an even integer among all the positive integers, which has a 'probability' of exactly one half. :tongue: It's maybe because I am broken; I don't know. :biggrin:
Original post by jack.hadamard
It's just because I find it not-quite-so intuitive. Having the same probability of the three numbers (picked from all positive integers) being the sides of a triangle as they being not the sides of a triangle is puzzling. If we try to formalise such statements, then considering the probability space (Ω,F,P)(\Omega, \mathcal{F}, \text{P}) with

i) Ω:={ 1,2,3,... }\Omega := \{\ 1, 2, 3, ...\ \}, F=P(Ω)\mathcal{F} = \mathcal{P}(\Omega),

ii) if AFA \in \mathcal{F}, then P(A)=aAP({a})=πA\displaystyle \text{P}(A) = \sum_{a \in A} \text{P}(\{a\}) = \pi |A| where π\pi is the (equal) probability of any given positive integer.

makes no sense, since P(Ω)={0 when π=0when π>0\text{P}(\Omega) = \begin{cases} 0\quad\ \text{when}\ \pi = 0 \\ \infty\quad \text{when}\ \pi > 0 \end{cases}.

Following the above, I also find other "intuitive" statements irritating. Like picking an even integer among all the positive integers, which has a 'probability' of exactly one half. :tongue: It's maybe because I am broken; I don't know. :biggrin:


Oh, I see what you mean. This is precisely why I posted the question because I also thought it was a really strange result. But isn't this why we study maths (i.e. to see how to think about these things properly?)
Nobody does my problems anymore :frown:.

Oh well, simple, boring things.. simply 'cause I'm interested in what proofs might emerge..

Problem 243*

Find and prove formulae for the sum of the first n odd, and first n even numbers.
Reply 1714
Original post by FireGarden
Nobody does my problems anymore :frown:.

Oh well, simple, boring things.. simply 'cause I'm interested in what proofs might emerge..

Problem 243*

Find and prove formulae for the sum of the first n odd, and first n even numbers.


Solution 243*

Odd: r=1n(2r1)=2r=1nrn=n(n+1)n=n2\displaystyle \sum^n_{r=1}(2r-1) = 2 \sum^n_{r=1}r - n = n(n+1) - n = n^2

Even: r=1n(2r)=2r=1nr=n(n+1)\displaystyle \sum^n_{r=1} (2r) = 2\sum^n_{r=1} r = n(n+1)

These problems are far too easy for this thread.
Reply 1715
Original post by Ateo
Solution 243*

Odd: r=1n(2r1)=2r=1nrn=n(n+1)n=n2\displaystyle \sum^n_{r=1}(2r-1) = 2 \sum^n_{r=1}r - n = n(n+1) - n = n^2

Even: r=1n(2r)=2r=1nr=n(n+1)\displaystyle \sum^n_{r=1} (2r) = 2\sum^n_{r=1} r = n(n+1)

These problems are far too easy for this thread.


Okay, solve it without using Mathematics. :wink:
Original post by Zakee
Okay, solve it without using Mathematics. :wink:


Solve it using bananas. (Say this in Rowan Atkinsons' voice)
Original post by jack.hadamard
What universities are you considering (or have applied to, or have been accepted in)?


Apologies for the late response, I was unable to reply earlier.

Most probably, I will take a gap year...

Apropos, as solution to your problem has already been posted, I just wanted to say that I generalized it a bit to 22n3=m32^{2n}-3=m^{3} and then to y23=m3y^{2}-3=m^{3}; the latter being a special case of Mordell's equation.

Original post by FireGarden
Nobody does my problems anymore :frown:.


It is quite difficult to trace all the problems which are posted, for the OP has not been updated recently.

Solution 231

We can define f:ACf : A \to C with f(x,y)=(xx2+y2,yx2+y2,x2+y21)\displaystyle f(x,y) = (\frac{x}{\sqrt{x^{2}+y^{2}}}, \frac{y}{\sqrt{x^{2}+y^{2}}}, \sqrt{x^{2}+y^{2}}-1) and let g:CAg : C \to A be g(x,y,z)=(x(1+z),y(1+z))g(x,y,z) = (x(1+z),y(1+z)).
It is sufficient to check that fg=IdCf \circ g = Id_{C} and gf=IdA g \circ f = Id_{A}.
(edited 10 years ago)
Reply 1718
Original post by bananarama2
Solve it using bananas. (Say this in Rowan Atkinsons' voice)



Oh yes, he pronounces his b's rather distinctively, doesn't he? :tongue:
Reply 1719
Original post by Mladenov

Most probably, I will take a gap year...

You're not just doing that for another shot at the IMO, are you? :/ I can picture you getting into a rather strange situation if you started university in the first year... perhaps you could take 2/3 years of exams all at once though hmm...

Quick Reply

Latest