Psycho-Babble Social Thread 721428

Shown: posts 1 to 21 of 21. This is the beginning of the thread.

 

Anyone good at number theory?

Posted by linkadge on January 11, 2007, at 17:17:02

If you are given that the GCD(a,b)=1, what is the value of GDD(a^2 + b^2,a+b)?

I've been struggling for hours.

Linkadge


 

Re: Anyone good at number theory?

Posted by AuntieMel on January 12, 2007, at 11:36:28

In reply to Anyone good at number theory?, posted by linkadge on January 11, 2007, at 17:17:02

How fast do you need it? I aced number theory, but it's been quite a few years.

Let me think on it.

 

Re: I forget » linkadge

Posted by AuntieMel on January 12, 2007, at 11:37:57

In reply to Anyone good at number theory?, posted by linkadge on January 11, 2007, at 17:17:02

I know what GCD is, but what is GDD?

 

Re: Sorry, both should be GCD (nm)

Posted by linkadge on January 12, 2007, at 12:46:42

In reply to Anyone good at number theory?, posted by linkadge on January 11, 2007, at 17:17:02

 

Euclidian algorithms? (nm)

Posted by kid47 on January 12, 2007, at 13:18:35

In reply to Anyone good at number theory?, posted by linkadge on January 11, 2007, at 17:17:02

 

Re: Anyone good at number theory?

Posted by AuntieMel on January 12, 2007, at 14:53:31

In reply to Anyone good at number theory?, posted by linkadge on January 11, 2007, at 17:17:02

Well, the answer is 2, but getting there is another matter.

But it makes sense - for a and be to be relatively prime they must be odd and have no other factors in common. so a^2 + b^2 has to be even, and a + b has to be even - and there are no other factors in common. {but the last half of that sentence is what I still need to come up with}

 

correction

Posted by AuntieMel on January 12, 2007, at 15:12:20

In reply to Anyone good at number theory?, posted by linkadge on January 11, 2007, at 17:17:02

they don't have to be both be odd. one can be even.

so a + b can be odd...

back to the drawing board.

 

Re: Well, the answer...

Posted by AuntieMel on January 12, 2007, at 16:36:20

In reply to Anyone good at number theory?, posted by linkadge on January 11, 2007, at 17:17:02

Well, the answer seems to be 2 if both a and b are odd and 1 otherwise (one odd, one even)

It works when plugging numbers into it.

But I keep wanting the answer to be abs|(a - b)|

(a^2 + b^2) factors into (a + b) * (a - b) so it seems if you divide one by the other the remainder is always a - b

I'll pull my number theory book out of mothballs when I get home.

 

Re: Well, the answer...

Posted by linkadge on January 12, 2007, at 16:54:19

In reply to Re: Well, the answer..., posted by AuntieMel on January 12, 2007, at 16:36:20

>(a^2 + b^2) factors into (a + b) * (a - b) so it >seems if you divide one by the other the >remainder is always a - b

Thats what I was thinking at first, but a^2 + b^2 doesn't factor into (a+b)(a-b). a^2 - b^2 (minus) factors into (a+b)(a-b).

>I'll pull my number theory book out of mothballs >when I get home.

Thanks a lot. Though, no worries either way.

Linkadge

 

Re: Well, the answer...

Posted by AuntieMel on January 13, 2007, at 13:07:25

In reply to Re: Well, the answer..., posted by linkadge on January 12, 2007, at 16:54:19

Of course you are right with the factoring.

Hey - it's been 20 years!

 

Re: Well, the answer...

Posted by linkadge on January 14, 2007, at 13:26:51

In reply to Re: Well, the answer..., posted by AuntieMel on January 13, 2007, at 13:07:25

You're right about the answer though, it is either 1 or 2, but just stuck showing it.

Linkadge

 

Re: I searched my youse

Posted by AuntieMel on January 15, 2007, at 8:28:43

In reply to Re: Well, the answer..., posted by linkadge on January 14, 2007, at 13:26:51

And I can't find my number theory book. There must be some boxes that haven't been unpacked.

I'm going to keep looking, but if you find out the answer let me know.

 

Re: I can partly show it

Posted by AuntieMel on January 15, 2007, at 12:15:27

In reply to Re: Well, the answer..., posted by linkadge on January 14, 2007, at 13:26:51

We know a and b can't both be even (or the gcd would include 2)

so a^2 is odd if a is odd, even otherwise
and b^2 is odd if b is odd, even otherwise

Odd + even = odd (explains the answer of "1")
odd + odd = even (explains the answer of "2")

also put - (a^2 + b^2(mod 2)) = (a + b(mod2)) which is either 0 or 1

What I can't show is why there aren't any other, higher numbers that can fit.

 

Re: the missing bit » linkadge

Posted by AuntieMel on January 16, 2007, at 9:17:37

In reply to Anyone good at number theory?, posted by linkadge on January 11, 2007, at 17:17:02


I found a property:

GCD (a + b, b) = GCD (a, b)

So GCD (a^2 + b^2, a + b) = GCD (a^2, a + b).

The only possible cases for a and b are
1) a is odd, b is even
2) a is even, b is odd
3) a and b are both odd

In case 1, the GCD of (a^2, a + b) is 1
case 2, the GCD of (a^2, a + b) is 1 (a + b is odd)
case 3, the GCD of (a^2, a + b) is 2 (a + b is even)

 

Re: number theory

Posted by Dr. Bob on January 17, 2007, at 1:06:50

In reply to Re: Well, the answer..., posted by linkadge on January 12, 2007, at 16:54:19

> GCD (a + b, b) = GCD (a, b)
>
> So GCD (a^2 + b^2, a + b) = GCD (a^2, a + b).

I don't know, the first would seem to imply:

GCD (a^2 + a + b, a + b) = GCD (a^2, a + b)

> a^2 + b^2 doesn't factor into (a+b)(a-b).

I think that might be the direction to go, maybe think of it as:

GCD ((a + b)^2 - 2ab, a + b)

Bob

 

Re: number theory

Posted by crazymaisie on January 17, 2007, at 22:40:45

In reply to Re: number theory, posted by Dr. Bob on January 17, 2007, at 1:06:50

sorry, i don't post much, but i do lurk

don't know if this will be of any use
but if you think of a as the product of primes:
p(1)...p(n) and b as the product of primes q(1)..q(m) with no p(i), q(j) equal, then
a^2 + b^2 would be p(1)^2...p(n)^2 + q(1)^2...q(m)^2
where you couldn't factor out any prime cause no p(i) equals any q(j)
similarly, p(1)...p(n) + q(1)...q(m) couldn't be factored either so they should be relatively prime

i could be completely wrong...it has been a while.
good luck, though!

 

:-) » Dr. Bob

Posted by Dinah on January 18, 2007, at 2:54:14

In reply to Re: number theory, posted by Dr. Bob on January 17, 2007, at 1:06:50

You ought to be careful.

That's how I fell in love with my now husband - in advanced math.

 

Re: hey! » Dinah

Posted by karen_kay on January 18, 2007, at 8:25:25

In reply to :-) » Dr. Bob, posted by Dinah on January 18, 2007, at 2:54:14

that's about enough of that dinah.

i don't want to have to fight you over mr. bob, but if you continue 'that' talk, i will be forced to.

watch your step missy! kk's an EXTREMELY jealous type of person.

oh, but there is something attractive about a man who can do math, isn't there?

 

Re: hey! » karen_kay

Posted by Dinah on January 18, 2007, at 9:37:59

In reply to Re: hey! » Dinah, posted by karen_kay on January 18, 2007, at 8:25:25

No worries, my husband can still whisper sweet equations into my ears. AND he can do great impressions of TV characters, in context.

I am well content with my own math guy.

 

Re: Anyone good at number theory? » linkadge

Posted by Klavot on January 20, 2007, at 16:18:58

In reply to Anyone good at number theory?, posted by linkadge on January 11, 2007, at 17:17:02

> If you are given that the GCD(a,b)=1, what is the value of GDD(a^2 + b^2,a+b)?
>
> I've been struggling for hours.
>
> Linkadge

Let p be a prime dividing both a^2 + b^2 and a+b. Then a^2 + b^2 = mp and a+b = np for some integers m,n. Since a^2 + b^2 = (a+b)^2 - 2ab then mp = n^2p^2 - 2ab, i.e. 2ab = (n^2p - m)p.
Thus p = 2 or p divides a or p divides b.

Suppose p divides a. Since p divides a+b then p divides b as well, a contradiction because gcd(a,b) = 1. Hence p does not divide a, and likewise p does not divide b either. Hence p = 2.

Thus we have shown that if a prime p divides both a+b and a^2 + b^2 then p=2.

Regarding possible values of a and b, there are two cases to consider.

Case 1: a odd, b odd. Then a+b is even and a^2 + b^2 is even, so gcd(a^2 + b^2,a+b) >= 2. From the above it follows that gcd(a^2 + b^2,a+b) = 2.

Case 2: a odd, b even. Then a+b is odd and a^2 + b^2 is odd. From the above, since the only candidate prime to divide both a+b and a^2 + b^2 is 2, it follows that gcd(a^2 + b^2,a+b) = 1.

Klavot

 

Re: Thank you » Klavot

Posted by AuntieMel on January 22, 2007, at 18:12:08

In reply to Re: Anyone good at number theory? » linkadge, posted by Klavot on January 20, 2007, at 16:18:58

That's what I was trying to get out but couldn't find the words.

Sometimes my language fails me.


This is the end of the thread.


Show another thread

URL of post in thread:


Psycho-Babble Social | Extras | FAQ


[dr. bob] Dr. Bob is Robert Hsiung, MD, bob@dr-bob.org

Script revised: February 4, 2008
URL: http://www.dr-bob.org/cgi-bin/pb/mget.pl
Copyright 2006-17 Robert Hsiung.
Owned and operated by Dr. Bob LLC and not the University of Chicago.