PEN-2007-5

Discussion on Bangladesh National Math Camp
User avatar
Tahmid Hasan
Posts:665
Joined:Thu Dec 09, 2010 5:34 pm
Location:Khulna,Bangladesh.
PEN-2007-5

Unread post by Tahmid Hasan » Mon Apr 02, 2012 6:14 pm

Let $x,y$ be positive integers such that $xy \mid x^2+y^2+1$.Show that $\frac {x^2+y^2+1}{xy}=3$. :oops:
Hint:
ever heard heard of root flipping? 8-)
বড় ভালবাসি তোমায়,মা

User avatar
zadid xcalibured
Posts:217
Joined:Thu Oct 27, 2011 11:04 am
Location:mymensingh

Re: PEN-2007-5

Unread post by zadid xcalibured » Mon Apr 02, 2012 6:21 pm

people heard it called vieta jumping :twisted:

User avatar
zadid xcalibured
Posts:217
Joined:Thu Oct 27, 2011 11:04 am
Location:mymensingh

Re: PEN-2007-5

Unread post by zadid xcalibured » Mon Apr 02, 2012 6:30 pm

root flipping exercise(for me too)
1.if a,b,$q=(a^2+b^2)/(ab+1)$are integers then q is a perfect square.
2.if a,b,$q=(a^2+ab+c^2)/(ab+1)$are integers then q is a perfect square
3.if a,b,$q=(a^2+b^2)/(ab-1)$are integers then q=5

User avatar
Tahmid Hasan
Posts:665
Joined:Thu Dec 09, 2010 5:34 pm
Location:Khulna,Bangladesh.

Re: PEN-2007-5

Unread post by Tahmid Hasan » Mon Apr 02, 2012 6:48 pm

here's another solution without flipping roots whatsoever but using only descent and inequality.
let $\frac {x^2+y^2+1}{xy}=k$.if $x=y$....(1)
then $x^2 \mid 2x^2+1 \Rightarrow x^2 \mid 1$,hence $x=y=1,k=3$
now let $x>y$,and let $(x,y)$ be a solution of (1) with $y$ minimal.
it is easy to check that $(y,ky-x)$ is also a solution.
now we claim $ky-x<y \Leftrightarrow k<\frac {x+y}{y} \Leftrightarrow \frac {x^2+y^2+1}{xy}< \frac {x+y}{y}$
$\Leftrightarrow b^2+1<ab \Leftrightarrow b(a-b)>0$ which is actually true contradicting the minimality of $y$.
so our assumption $x>y$ was wrong implying $a=b=1,k=3$ :mrgreen:
Last edited by Tahmid Hasan on Mon Apr 02, 2012 6:57 pm, edited 1 time in total.
বড় ভালবাসি তোমায়,মা

User avatar
Tahmid Hasan
Posts:665
Joined:Thu Dec 09, 2010 5:34 pm
Location:Khulna,Bangladesh.

Re: PEN-2007-5

Unread post by Tahmid Hasan » Mon Apr 02, 2012 6:56 pm

zadid xcalibured wrote:
root flipping exercise(for me too)
1.if a,b,$q=(a^2+b^2)/(ab+1)$are integers then q is a perfect square.
2.if a,b,$q=(a^2+ab+c^2)/(ab+1)$are integers then q is a perfect square
3.if a,b,$q=(a^2+b^2)/(ab-1)$are integers then q=5
well these problems can also be solved using the same method i've shown without flipping roots.you should try it too ;) .And i learned this technique from "An Introduction to Diophantine Equations" by Titu Andreescu and Dorin Andrica. :mrgreen:
বড় ভালবাসি তোমায়,মা

User avatar
FahimFerdous
Posts:176
Joined:Thu Dec 09, 2010 12:50 am
Location:Mymensingh, Bangladesh

Re: PEN-2007-5

Unread post by FahimFerdous » Mon Apr 02, 2012 7:14 pm

@Tahmid: your proof is called Vieta Jumping if you don't know then.
Your hot head might dominate your good heart!

User avatar
zadid xcalibured
Posts:217
Joined:Thu Oct 27, 2011 11:04 am
Location:mymensingh

Re: PEN-2007-5

Unread post by zadid xcalibured » Mon Apr 02, 2012 7:25 pm

he knows actually.

User avatar
Tahmid Hasan
Posts:665
Joined:Thu Dec 09, 2010 5:34 pm
Location:Khulna,Bangladesh.

Re: PEN-2007-5

Unread post by Tahmid Hasan » Mon Apr 02, 2012 7:26 pm

well Fahim vaiya,as far as i know if the equation $x^2+ax+b=0$ has real roots $p,q$ then by vieta's formula $pq=b,-a=p+q$ and the principle of vieta jumping is based on this and descent.And vieta jumping is also called root flipping accordong to wiki, and the phrase 'root flipping' sounds more 'cruxy' to me :lol:
Last edited by Tahmid Hasan on Mon Apr 02, 2012 7:28 pm, edited 1 time in total.
বড় ভালবাসি তোমায়,মা

User avatar
zadid xcalibured
Posts:217
Joined:Thu Oct 27, 2011 11:04 am
Location:mymensingh

Re: PEN-2007-5

Unread post by zadid xcalibured » Mon Apr 02, 2012 7:28 pm

root flipping is just one kind of descent.@tahmid just using vieta formula. :D :) ;) :( :shock: :? 8-) :lol: :x :P :oops: :cry: :evil: :twisted: :roll: :!: :?: :ugeek: :geek: :mrgreen: :| :arrow: :idea:

Post Reply