75 of 100: Odd Triples

True or False:
Every odd number greater than 1 is the smallest member of a primitive Pythagorean triple.

Definitions:

  • A "Pythagorean triple" is a set of three whole numbers which can be the three side lengths of a right triangle.
  • "Primitive" means the three numbers don't have a common factor. For example, the Pythagorean triple ( 6 , 8 , 10 ) (6,8,10) is not primitive because all three numbers are divisible by 2.

Note that the statement is false if and only if there is an odd number greater than 1 that is NOT the smallest member of some primitive Pythagorean triple. For example, there are primitive Pythagorean triples where the smallest member is even, but that's not relevant to the problem.

True False

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.

24 solutions

Zach Abueg
Aug 13, 2017

Proof without words:

( 2 m + 1 ) 2 + ( 2 m 2 + 2 m ) 2 = ( 2 m 2 + 2 m + 1 ) 2 \big({\color{#D61F06}{2m + 1}}\big)^2 + \left(2m^2 + 2m\right)^2 = \left(2m^2 + 2m + 1\right)^2

Proof without algebra: an odd square is always odd. The differences between the successive square numbers are the successive odd numbers, so for every odd number, including every odd square , there are two adjacent square numbers that have that odd number as their difference. These three square numbers must correspond to a Pythagorean triple, since the sum of two of them is equal to the third.

Mark Lama - 3 years, 10 months ago

Log in to reply

Oh, very nice! A wonderful observation, thank you :)

Zach Abueg - 3 years, 10 months ago

Log in to reply

Thanks! To complete the proof, I should have noted that since adjacent numbers (greater than one) are relatively prime, the Pythagorean triple has to be a primitive one.

Mark Lama - 3 years, 10 months ago

Log in to reply

@Mark Lama I didn't understand this part. How can I know that the smaller number is relatively prime with the other two numbers?

MARCOS VINICIUS QUEIROZ ROCHA - 3 years, 10 months ago

Log in to reply

@Marcos Vinicius Queiroz Rocha For a factor to be shared by all three numbers, it would also have to be shared by every possible pair of them.

Mark Lama - 3 years, 10 months ago

This was my observation as well! Great thinking!

John Allums - 3 years, 10 months ago

I don't understand your proof.

You said for every odd number (7, say) there are two adjacent square numbers that have that odd number as their difference (so 9 and 16). You then said these three numbers must be a Pythagorean triple, but I'm not sure that 7, 9 and 16 do form a Pythagorean triple. In the last part you then said the sum of two of them is equal to the third, but isn't it only a Pythagorean triple if that last statement applies to the squares and not the numbers themselves?

I may have misinterpreted your proof, but could you please clarify how it works?

Thanks

Akaash Thao - 3 years, 10 months ago

Log in to reply

My language was a little sloppy. Notice that I said for every odd number, including every odd square . So if we have an odd square, such as 121, we know that there are two adjacent square numbers that have a difference of 121. In this case, these square numbers are 3600 and 3721. Thus, the sum of 3600 and 121 is 3721. What I should have said, to be perfectly clear, is that the roots of these three squares form a Pythagorean triple, in this case, (11, 60, 61). That is what I meant. I edited my comment to alleviate the ambiguity.

Mark Lama - 3 years, 10 months ago

Log in to reply

@Mark Lama I think I get it now, so the number you pick first has to be an odd square, then the two adjacent square numbers with that as their difference can form a Pythagorean triple with it (the roots form a pythagorean triple)?

So instead of picking 7, I should have picked something like 9, then 16 and 25 are the adjacent squares with 9 as their difference. Thus, the roots 3, 4 and 5 form a Pythagorean triple. Also, I could pick 49, which is the difference between 625 and 576. Hence, the Pythagorean triple is 7, 24 and 25.

Therefore, that means that since it works for every odd square and all the odd squares have an odd number as a root, then all odd numbers can form a Pythagorean triple. I hope I am on the right track now.

One thing that is still unclear to me is the 'including' part. Do you mean to say that 'for every odd number, there are two adjacent square numbers that have the odd number squared as their difference'? To me, 'including' makes it seem as though any odd number could be used as the difference between two adjacent squares to form a triple. Whereas, here it seems that only odd squares can be used as the difference (hence, three square numbers as opposed to two square numbers and any odd number). That's where I was confused about why 7 doesn't work. Apologies if my question is long-winded and confusing.

Akaash Thao - 3 years, 10 months ago

Log in to reply

@Akaash Thao Yes, you could say it that way. What I meant was that any odd number is the difference between some two adjacent square numbers, and that this is relevant to the problem when the odd number is an odd square.

Mark Lama - 3 years, 10 months ago

Log in to reply

@Mark Lama I see, it makes better sense now, thank you very much.

Akaash Thao - 3 years, 9 months ago

I interpreted the statement wrongly. All of Pythagorean triples are as follows for a 2 + b 2 = c 2 a^2+b^2=c^2 ( a , b , c ) = ( r ( p 2 q 2 ) , 2 r p q , r ( p 2 + q 2 ) ) (a, b, c)= \big(r(p^2-q^2), 2rpq, r(p^2+q^2) \big) Where p > q p>q are co-prime positive integers with different parity. Now, for primitive solutions r = 1 r=1 and ( a , b , c ) = ( p 2 q 2 , 2 p q , p 2 + q 2 ) (a, b, c)= \big(p^2-q^2, 2pq, p^2+q^2 \big) p 2 q 2 p^2-q^2 is odd and I thought the problem states that it is always the smallest! We can choose arbitrary p p and q q such that p 2 q 2 > 2 p q p^2-q^2>2pq . Write p = q + l p=q+l , then ( q + l ) 2 q 2 > 2 ( q + l ) q l 2 > 2 q 2 (q+l)^2-q^2>2(q+l)q \ \ \ \Longrightarrow \ \ \ l^2>2q^2 We just need to choose l > q 2 l> \lfloor q \sqrt{2} \rfloor .

As an example choose q = 3 q=3 . Then l > 3 2 = 4 l> \lfloor 3 \sqrt{2} \rfloor=4 . Take l = 5 l=5 . This gives p = 8 p=8 and ( a , b , c ) = ( 55 , 48 , 73 ) (a, b, c)= (55, 48, 73) .

Kazem Sepehrinia - 3 years, 10 months ago

Log in to reply

Aw, no worries man I get you. Beautiful generalization Kazem - great work, as always :)

Zach Abueg - 3 years, 10 months ago

HEEEEEEEYYYY! #Zachisback #again #upvoted Nice solution btw :P

Angel ONG - 3 years, 10 months ago

Log in to reply

Wow, thank you again Angel, you're so kind :)

Zach Abueg - 3 years, 10 months ago

Log in to reply

No, you are(kind I mean :D) for staying here and posting 'Brilliant' (sorry, my puns are bad, I know that xD) solutions for us all! :P

Angel ONG - 3 years, 10 months ago

Log in to reply

@Angel Ong Haha! Hey, I love puns myself! Go you :)

Zach Abueg - 3 years, 10 months ago

Nice work, Zach! It is interesting to note that (2m^2 + 2m)^2 = 4[(m^2 + m)/2], meaning that in this series of Pythagorean triples, the second leg of the triangles are four times the successive triangular numbers.

Mark Lama - 3 years, 10 months ago

Log in to reply

Wow! What a fascinating property Mark. And thank you :)

Zach Abueg - 3 years, 10 months ago

Dude, nice! Keep on posting solutions!

Steven Jim - 3 years, 10 months ago

Log in to reply

Thank you, Steven! You always lighten me up :)

Zach Abueg - 3 years, 10 months ago

Log in to reply

You're welcome! And please, I'd love to see more of this!

Steven Jim - 3 years, 10 months ago

Log in to reply

@Steven Jim Your wish shall be granted as long as I'm on Brilliant :)

Zach Abueg - 3 years, 10 months ago

Log in to reply

@Zach Abueg Dude, then stay XD That'd make me feel better XD

Steven Jim - 3 years, 10 months ago

Log in to reply

@Steven Jim Me too high fives you Zach's solutions are always (quite? :P) clear and easily understood(if you're not me :D)

Angel ONG - 3 years, 10 months ago

Log in to reply

@Angel Ong Haha, thank you Angel! All I intend is to enlighten - sometimes that requires rigor, and other times that requires nothing more than a little intuition.

Zach Abueg - 3 years, 10 months ago

@Steven Jim I definitely will, Steven :)

Zach Abueg - 3 years, 10 months ago

I thought that the question was asking that every primitive pythagorean triplet has the least number as even, which is false. Consider the numbers(35,12,37). For a general argument, one can take the equation:

(s+t)(s+t) = (s-t)(s-t) + 4st, where s,t are natural numbers with opposite parity and s*t is a square of some natural number.

Then, we just need to find sufficiently large s and sufficiently small t for which, s s + t t - 6st>=0. [Sorry I forgot Latext]

Bhaskar Pandey - 3 years, 10 months ago

The answer should be false.....just look at the following example: 15^2+8^2 = 17^2...15 is not the smallest member of a primitive Pythagorean triple. Also 21^2 + 20^2 = 29^2.....21 is not the smallest member of a primitive Pythagorean triple.

Ron Kowch - 3 years, 10 months ago

Log in to reply

Please read the reports to this question. As Jason writes, the question is not asking whether the smallest leg of a primitive Pythagorean triple is always odd; it is asking if every odd number is the smallest leg of some Pythagorean triple out there.

As in my example, take 2 m + 1 = 15 m = 7 ( a , b , c ) = ( 15 , 112 , 113 ) 2m + 1 = 15 \Longrightarrow m = 7 \Longrightarrow (a, b, c) = (15, 112, 113) , where gcd ( 15 , 112 , 113 ) = 1 \gcd(15, 112, 113) = 1 .

Similarly, take 2 m + 1 = 21 m = 10 ( a , b , c ) = ( 21 , 220 , 221 ) 2m + 1 = 21 \Longrightarrow m = 10 \Longrightarrow (a, b, c) = (21, 220, 221) , where gcd ( 21 , 220 , 221 ) = 1 \gcd(21, 220, 221) = 1 .

Zach Abueg - 3 years, 10 months ago

Log in to reply

@Auro Light

Zach Abueg - 3 years, 10 months ago

I featured this problem on my blog yesterday, in this post: https://aspi.blog/2017/08/14/monday-miscellany/ and today, as promised to my readers I provided the solution and I also cleared up the side issue raised by various people who answered incorrectly. The new post is: aspi.blog/2017/08/15/answer-to-the-pythagorean-problem/ Here is an excerpt: THE SIDE ISSUE

Some people on Brilliant cavilled at this because there are some Primitive Pythagorean Triples whose smallest term is even (8,15, 17, 20,21,29 and 65,72,97 were all mentioned, although none of the complainers mentioned 12,35,37, 60,91,109 or 696,697,985). The question did not state that the triple of which the odd number is the lowest term was the lowest triple to feature that number, and indeed if one looks carefully at the triangles presented as part of the problem one can see clearly that the odd number is allowed to be in another triple where it is not the lowest term:

<img src="https://aspiblog.files.wordpress.com/2017/08/pythag.png?w=840" alt="Pythag"/>

Note that the number 5 features twice (ringed in the diagram above, once as the largest term in a triple and once as the smallest).

Thomas Sutcliffe - 3 years, 9 months ago

@Mark Lama - This is a formula for prime pythagorean triplets therefore so much explanation have even proved this formula Thanks

Soham Chitnis - 3 years, 9 months ago

Hey, are there any with 45? I cant find one.

Michael Chen - 3 years, 9 months ago

Answer should be false, as primitive triplets do exist where even number is smallest, like (8, 15, 17). This is the interpretation I took on the basis of Note given below the problem.

Auro Light - 3 years, 10 months ago

Log in to reply

as the note indicates: "For example, there are primitive Pythagorean triples where the smallest member is even, but that's not relevant to the problem."

The problem is specifically about if all odd numbers are the smallest member of some Pythagorean triple. This doesn't exclude the possibility of even numbers.

Jason Dyer Staff - 3 years, 10 months ago
Atomsky Jahid
Aug 13, 2017

We notice the following things: 3 2 = 9 = ( 5 + 4 ) ( 5 4 ) = 5 2 4 2 3^2=9=(5+4)(5-4)=5^2-4^2 5 2 = 25 = ( 13 + 12 ) ( 13 12 ) = 1 3 2 1 2 2 5^2=25=(13+12)(13-12)=13^2-12^2 7 2 = 49 = ( 25 + 24 ) ( 25 24 ) = 2 5 2 2 4 2 7^2=49=(25+24)(25-24)=25^2-24^2

Now, we try to generalize this. ( 2 n + 1 ) 2 = 4 n 2 + 4 n + 1 = ( 4 n 2 + 4 n 2 + 1 + 4 n 2 + 4 n 2 ) ( 4 n 2 + 4 n 2 + 1 4 n 2 + 4 n 2 ) (2n+1)^2=4n^2+4n+1=(\frac{4n^2+4n}{2}+1+\frac{4n^2+4n}{2})(\frac{4n^2+4n}{2}+1-\frac{4n^2+4n}{2}) = ( 4 n 2 + 4 n 2 + 1 ) 2 ( 4 n 2 + 4 n 2 ) 2 =(\frac{4n^2+4n}{2}+1)^2-(\frac{4n^2+4n}{2})^2 = ( 2 n 2 + 2 n + 1 ) 2 ( 2 n 2 + 2 n ) 2 =(2n^2+2n+1)^2-(2n^2+2n)^2

And, two consecutive integers are coprime. So, this triplet is primitive.

P.S. Maybe I need to clarify my generalization process a bit more. Take 49 49 . We'll write it as ( 48 2 + 1 ) + 48 2 (\frac{48}{2}+1)+\frac{48}{2} . That's what I did with 4 n 2 + 4 n + 1 4n^2+4n+1 . I wrote it as ( 4 n 2 + 4 n 2 + 1 ) + 4 n 2 + 4 n 2 (\frac{4n^2+4n}{2}+1)+\frac{4n^2+4n}{2} .

You know what catches my attention on this, it has words and maths and it is very clear and easy to read. It is very organised so I love this! Thanks for writing this!

William Huang - 3 years, 10 months ago

Log in to reply

Thanks for your feedback. I actually try to convey my thinking process as clearly as I can.

Atomsky Jahid - 3 years, 10 months ago

Same as the " proof without words" but actually shows the thinking and derivation of it for those who want it. Very nice!

John Allums - 3 years, 10 months ago

@Zach Abueg @Atomsky Jahid @Kazem Sepehrinia You know, I was thinking about this, and realized that for the purpose of this problem, it might be tidier to let the odd number be defined by a single variable. Call it p . Then the Pythagorean triple in question could be defined as [ p , ( p ^2 - 1)/2, ( p ^2 + 1)/2].

Mark Lama - 3 years, 10 months ago

Log in to reply

Yes, as a definition it's tidier. We can readily see that p p must be odd as p 2 1 2 \frac{p^2-1}{2} and p 2 + 1 2 \frac{p^2+1}{2} wouldn't be integers otherwise.

Atomsky Jahid - 3 years, 10 months ago

that is a beautiful proof.

Katherine barker - 3 years, 9 months ago
Avik Das
Aug 13, 2017

We know that the difference between 2 consecutive square quantities is odd number.
So,. (n+1)²—n²=(2k+1).
Now, the square of an odd number is a odd number.i.e. in the form of (2k+1).
It is obvious that, g.c.d. of (n , n+1)=1.
So, the numbers are primitive.
Hence, the statement is true.



29,420,421 is a primitive Pythagorean triple with 29 as the smallest number. The answer is not false, you just didn't find the right triple.

Eric Lucas - 3 years, 10 months ago

15, 112,113 is likewise a primitive Pythagorean triple.

Eric Lucas - 3 years, 10 months ago

Thanks for posting Avik. I like your thinking but I think there's a problem: Even though (n+1)²—n²=(2k+1), that doesn't mean that there's always a possible value of n for every value of k?

Justin Roughley - 3 years, 10 months ago

Log in to reply

@Justin Roughley n = k

Duke Garland - 3 years, 9 months ago

I agree with Nashita....as with my example (15,8,17). Nashita's is another example where (21,20,29)....21 is not the smallest member of a primitive Pythagorean triple.

Ron Kowch - 3 years, 10 months ago

Log in to reply

Look at the triples (15, 112, 113) and (21, 220, 221). The problem only states that every odd number is the smallest member of a Pythagorean triple. It doesn't state that no other possibilities exist.

Mark Lama - 3 years, 10 months ago

Log in to reply

@Mark Lama I don't understand your logic. Is (15,8,17) a primitive Pythagorean triple? YES Is 15 an odd number greater than 1? YES Is 8 a member of the primitive Pythagorean triple? Yes Is 15 smallest member? NO - 8 is smaller than 15. Therefore the statement..."Every odd number greater than 1 is the smallest member of a primitive Pythagorean triple." is FALSE

Ron Kowch - 3 years, 10 months ago

Log in to reply

@Ron Kowch Ron, put the emphasis on the word a . "Every odd number greater than 1 is the smallest member of a primitive Pythagorean triple." The argument is that, for every odd number greater than 1, there exists a primitive Pythagorean triple of which that odd number is the smallest member. It is not meant to exclude other possibilities, such as that, for some odd numbers, there may exist other primitive Pythagorean triples in which the number is not the smallest member. The problem's author actually added a note to the problem clarifying this. A fun side note: it works for 1, too, if you allow the triple (1, 0, 1), and eliminate the restriction that it be the smallest member. That would correspond to a triangle where one side has been shrunk to zero length. Of course, that's no longer a triangle, but the Pythagorean theorem still applies. The triple also follows the same algebraic pattern as the others.

Mark Lama - 3 years, 10 months ago

The answer should be false.....just look at the following example: 15^2+8^2 = 17^2 15 is not the smallest member of a primitive Pythagorean triple.

Ron Kowch - 3 years, 10 months ago

Log in to reply

read the statement it says "statement is false if and only if there is an odd number greater than 1 that is NOT the smallest member of some primitive Pythagorean triple" focus on the word 'some'

Ritesh Kumar - 3 years, 9 months ago

The answer should be false because 29^2=21^2+20^2 !!

Nashita Rahman - 3 years, 10 months ago

Log in to reply

However, 21 is the smallest member of the primitive Pythagorean triple (21, 220, 221). The problem does not state that it has to be the only possible Pythagorean triple using that number.

Mark Lama - 3 years, 10 months ago
Arjen Vreugdenhil
Aug 14, 2017

All primitive Pythagorean triplets can be generated by the formula a = p 2 q 2 , b = 2 p q , , c = p 2 + q 2 , a = p^2 - q^2,\ \ \ b = 2pq,\ \ \ , c = p^2 + q^2, with p , q p, q coprime and not both odd.

In order to make a = 2 n + 1 a = 2n+1 an arbitrary odd number, we choose p = n p = n and q = n + 1 q = n+1 . These are obviously coprime and not both odd. Then a = p 2 q 2 = ( p + q ) ( p q ) = ( 2 n + 1 ) 1 = 2 n + 1 a = p^2 - q^2 = (p+q)(p-q) = (2n+1)\cdot 1 = 2n+1 as desired.

Tom Van Lier
Aug 15, 2017

Suppose a is an odd positive integer.

Then to be the smallest member of a Pythagorean triple, we need b and c such that a 2 = c 2 b 2 = ( c b ) ( c + b ) a^2 = c^2 - b^2 = (c-b)(c+b) .

Take c = b + 1 c = b+1 , then a 2 1 2 = b \dfrac{a^2 -1}{2} = b . Because a is odd, b N b \in \mathbb{N} .

So a,b,c form a Pythagorean triple and a is the smallest member. Since c = b + 1 c= b+1 , is it also primitive.

Sam Campbell
Aug 14, 2017

We can test this with as many odd numbers as we like. For example, 11^2 = 121, and the two consecutive numbers that add up to 121 are 60 and 61, and 11^2+60^2 = 61^2. 13^2 = 84+85, and 13^2+84^2 = 85^2

Nansi Has
Aug 14, 2017

Answered in two words: Of course!

Syrous Marivani
Aug 14, 2017

Let the smallest side be 2n + 1 (n > 0) and the other sides be y and z, with

the largest side being z, then

z^2 = y^2 + (2n + 1)^2, (1)

Then obviously y and z have different parity. Then

(z – y)(z + y) = (2n + 1)^2

Since the gcd(z – y, z + y) | 2 gcd(z, y) = 2 and z – y and z + y are odd,

so gcd(z – y, z + y) = 1. So, (1) shows that z – y = (2m + 1)^2, z + y = (2k + 1)*2.

z = 2k(k + 1) + 2m(m + 1) + 1, y = 2k(k + 1) – 2m(m + 1) and

(2m + 1)(2k + 1) = (2n + 1). If 2n + 1 is any odd number, we can set, 2k + 1 = 2n + 1,

2m + 1 = 1,So, z = 2n(n + 1) + 1 and y = 2n(n + 1). So we have:

[2n(n + 1) + 1]^2 = [2n(n + 1)]^2 + (2n + 1)^2

This is actually what we are looking for.

Let d = gcd (z, 2n + 1), obviously d | (z – 2n – 1) = 2n^2. Since (2n + 1, 2n^2) = 1,

then d = 1. So obviously 2n + 1, y, and z are a primitive triple.

In general case 2n + 1, factor it as (2k + 1)(2m + 1), such that k > m, and

(2k + 1, 2m + 1) = 1, then z + y = (2k + 1)^2, z - y = (2m + 1)^2.

In this case, z = 2k(k + 1) + 2m(m + 1) + 1, and y = 2k(k + 1) - 2m(m + 1)

[2k(k + 1) + 2m(m + 1) +1]^2 = [2k(k + 1) - 2m(m + 1))^2 + (2n + 1)^2

Since k < n, then y < 2n(n + 1), which shows y < 2n + 1, so in this case 2n + 1 is not the smallest

leg, so we have to choose the first approach as the third example below shows.

1) Let 2n + 1 = 13, then k = n = 6, m = 0, z = 85, y = 84

85^2 = 84^2 + 132, 7225 = 7056 + 169

2) Let 2n + 1 = 81, then k = 40, m = 0, z = 3281, y = 3280

3281^2 = 3280^2 + 81^2, 10764961 = 10758400 + 6561

3) Let 2n + 1 = 63, then k = 4, m = 3, z = 65, y = 16

65^2 = 16^2 + 63^2, 4225 = 256 + 3969

Although this is primitive triplet, it is not what we are looking for.

So, we should let k = 31, m = 0, then z = 1985, y = 1984,

1985^2 = 1984^2 + 632 3940225 = 3936256 + 3969.

This is what we are looking for.

Aditya Khurmi
Aug 14, 2017

Every Pythagorean triplet is of the form

2 k m n , k ( m 2 n 2 ) 2kmn, k(m^{2}-n^{2}) and k ( m 2 + n 2 ) k(m^{2}+n^{2})

Allow k k to be 1 and then the second member is

( m + n ) ( m n ) = O (m+n)(m-n)=O for any odd integer O O .

One way can be

m + n = O m+n=O and m n = 1 m-n=1 . There always exists such integers m m and n n , right?

[We can also show that they are primitive.]

Roy McDonald
Aug 14, 2017

Notice that the large sides are always b and b+1. Solve:

(2a+1)^2 + b^2 = (b+1)^2

Reduces to 2a^2 + 2a = b

So for each value of a there is an odd short side (2a +1) and two long sides length 2a^2 + 2a and 2a^2 + 2a +1.

Clay Creasy
Aug 14, 2017

Using Euler's formula for constructing Pythagorean triples, instead of using two distinct values m m and n n , if we instead use m m and m + 1 m + 1 for all the natural numbers, we will see that the first term n 2 m 2 n^2 - m^2 will simplify to 2 m + 1 2m + 1 which is the general term for all the odd natural numbers.

Furthermore, subbing in m m and m + 1 m + 1 into the other two values of 2 m n 2mn and n 2 + m 2 n^2 + m^2 , we get two terms with a value of 2 m 2 + 2 m 2m^2 + 2m and 2 m 2 + 2 m + 1 2m^2 + 2m +1 , which are two consecutive natural numbers whose sum is equivalent to ( 2 m + 1 ) 2 (2m +1)^2 , therefore always making the value of the other two sides of the primitive triple add up to the square of that shortest side.

Another way of looking at this is by choosing any natural number m m and then making substitutions of a = 2 m + 1 a = 2m+1 and b = 2 m 2 + 2 m b=2m^2 + 2m , we create Pythagorean triples of the form a , b , b + 1 a, b, b+1 , where a 2 = 2 b + 1 a^2 = 2b + 1 .

Robert DeLisle
Aug 14, 2017

Here is an interesting way to find the larger two members of a Pythagorean triple of any odd number a, a >1 in terms of the sums of odd numbers. (Still working on the primitive part for this one).

Here is a complete proof. Given any odd number greater than 1 a primitive Pythagorean triple can be derived directly as follows:

Take the nth odd number A = 2 n + 1 A = 2n + 1 n = 1,2,3,...

A 2 = ( 2 n + 1 ) 2 = 4 n 2 + 4 n + 1 A^2 = (2n + 1)^2 = 4n^2 + 4n + 1

Set B = n A + n = n ( 2 n + 1 ) + n = 2 n 2 + 2 n = 2 n ( n + 1 ) B = n A + n = n (2n + 1) + n = 2n^2 + 2n = 2n(n+1)

Set C = B + 1

C 2 = ( B + 1 ) 2 = B 2 + 2 B + 1 C^2 = (B + 1)^2 = B^2 + 2B + 1

C 2 B 2 = 2 B + 1 = 2 ( 2 n ( n + 1 ) ) = 4 n ( n + 1 ) + 1 = 4 n 2 + 4 n + 1 = A 2 C^2 - B^2 = 2B + 1 = 2(2n(n+1)) = 4n(n+1) + 1 = 4n^2 + 4n + 1 = A^2

Therefore for any odd number, A, there are values B and C such that C 2 = A 2 + B 2 C^2 = A^2 + B^2 , forming a Pythagorean triple.

A and B have no common factors therefore it is also a primitive Pythagorean triple.

Does not really draw much blood. The consecutive squares in the given examples provide an excellent clue that suggests the proof above.

Corwin Silverman
Aug 14, 2017

For any odd number n, you can calculate n^2, then find a such that 2a+1=n^2. Once you have this, you can write the expression n^2+a^2, and we said that n^2=2a+1, so the expression comes out to be a^2+2a+1, which factors to (a+1)(a+1), or just (a+1)^2. Thus, we have found a Pythagorean triple.

Edwin Gray
Aug 14, 2017

Let a = 2m + 1 be an arbitrary odd positive integer. Let b = 2m(m + 1), and c = m^2 + (m + 1)^2. Then a^2 + b^2 = c^2 is clearly a primitive Pythagorean Triplet. Ed Gray

Moo Cow
Aug 13, 2017

Let 1 < a < b < c 1<a<b<c , and let a a be odd as per the statement. The Pythagorean statement can be stated as a 2 = c 2 b 2 a^2=c^2-b^2 . Since a a is odd, a 2 > 1 a^2>1 is guaranteed to be odd as well.

Notice the pattern between pairs of consecutive squares: ( n + 1 ) 2 n 2 = 2 n + 1 (n+1)^2-n^2=2n+1 , which is also always odd. Obviously, a 2 = 2 n + 1 a^2=2n+1 for some positive n n . This means that we can say a 2 = ( a 2 1 2 + 1 ) 2 ( a 2 1 2 ) 2 a^2=(\frac{a^2-1}{2}+1)^2-(\frac{a^2-1}{2})^2 , forming the Pythagorean triple a 2 , a 2 1 2 , a 2 1 2 + 1 a^2,\frac{a^2-1}{2},\frac{a^2-1}{2}+1 where a 2 a^2 is both odd and the smallest member of such triple.

However, we have not yet shown that the Pythagorean triple is primitive, not sharing a factor. If just two members are relatively prime, the entire triple must also be and thus primitive. Let's examine a 2 1 2 \frac{a^2-1}{2} and a 2 1 2 + 1 \frac{a^2-1}{2}+1 . Adjacent numbers are always relatively prime (try to think of a number that divides both). We are now guaranteed the Pythagorean triple is primitive. Q.E.D.

Alex Li
Aug 13, 2017

Lemma: every odd number greater than 1 can be represented by (x^2 - (x-1)^2.)

Proof: x 2 ( x 1 ) 2 = x 2 ( x 2 2 x + 1 ) = 2 x 1 x^2 - (x-1)^2 = x^2 - (x^2-2x+1) = 2x - 1

Using the equation a 2 + b 2 = c 2 a^2+b^2=c^2 , let b equal c-1 and subtracting it to get a 2 = c 2 ( c 1 ) 2 . a^2 = c^2 - (c-1)^2. Notice the RHS can be any odd number by the Lemma above. Since a^2 is odd when a is odd, we can represent it with sides c and b. The triple must be primitive because gcd(a,c-1,c)>=gcd(c,c-1)=1.

Nikita Mahilewets
Aug 13, 2017

I remember there is a formula for Pythagorean triple

x=2*k+1

y=2 k (k+1)

z=(k+1)^2 +k^2

Where k is some natural number and x^2+y^2=z^2

So from those formula is clear than the smallest member is x, which is an odd number

The answer is false. Here a triple: (20, 21, 29). The smalles number of the triple (20) is an even number.

Boris Leprof - 3 years, 10 months ago

Log in to reply

Really, the answer is false

The problem needs to be reported

I did report, better to spawn multiple reports

Nikita Mahilewets - 3 years, 10 months ago
Kalib Perry
Apr 16, 2020

think about it people ( its 3)

Liviu Vigu-Giurea
Aug 29, 2017

n^2 + x = (n+1)^2 => x = 2n +1 so yeah not only "Every odd number greater than 1 is the smallest member of a primitive Pythagorean triple.", but the other 2 numbers are consecutive one's.

Hadas Cet
Aug 20, 2017

Kevin Tong
Aug 14, 2017

We can represent all odd numbers greater than 1 as 1 + 2 x x N + 1+2x \forall x \in \mathbb{N}^+ So, using the pythagorean theorem, we get ( 1 + 2 x ) 2 + a 2 = b 2 ( 1 + 2 x ) 2 = b 2 a 2 ( 1 + 2 x ) 2 = ( b a ) ( b + a ) (1+2x)^2+a^2=b^2 \\ (1+2x)^2=b^2-a^2 \\ (1+2x)^2=(b-a)(b+a) Because 1 + 2 x 1+2x is odd, either a a or b b is odd and the other is even. Let's assume a a is even, so we can represent a a as 2 y 2y , so we get ( 1 + 2 x ) 2 = ( b + 2 y ) ( b 2 y ) (1+2x)^2=(b+2y)(b-2y) . Based on the calculated pythagorean triples given to us, let's assume that b = a + 1 = 2 y + 1 b=a+1=2y+1 . Now, plugging this into our equation, we get ( 1 + 2 x ) 2 = ( 2 y + 2 y + 1 ) ( 2 y + 1 2 y ) ( 1 + 2 x ) 2 = ( 4 y + 1 ) ( 1 ) = 4 y + 1 (1+2x)^2=(2y+2y+1)(2y+1-2y) \\ (1+2x)^2=(4y+1)(1)=4y+1 Calculating this out, we get 4 x 2 + 4 x + 1 = 4 y + 1 4 ( x 2 + x ) + 1 = 4 y + 1 y = x 2 + x 4x^2+4x+1=4y+1 \\ 4(x^2+x)+1=4y+1 \implies y=x^2+x Because x x is a positive natural number and 2 y = 2 x 2 + 2 x > x 2y=2x^2+2x>x , we now know that for all odd numbers greater than 1, it is the smallest member of a pythagorean triple.

Karl Snowsill
Aug 14, 2017

...for every odd integer j, there exist infinitely many primitive Pythagorean triples in which the hypotenuse and the even leg differ by j^2. https://en.wikipedia.org/wiki/Pythagorean_triple

Eric Lucas
Aug 14, 2017

For any odd a, b = (a^2-1)/2 and c = b+1 form a Pythagorean triple, and as long as a>2 (which is true for all odd numbers >1), a<b<c. b and c are obviously relatively prime, and it is possible to show that a is relatively prime with both b and c.

David Fairer
Aug 14, 2017

x^2 + y^2 = z^2. Now write that differently, so x^2 = z^2 - y^2. Now we'll think only of a Pythagorean triple with the largest number only 1 greater than the 2nd number. So z = y + 1. So x^2 = (y + 1)^2 - y^2. Hence x^2 = (y^2 + 2y + 1) - y^2. Hence x^2 = 2y + 1 - which is of course odd. Hence 2y = x^2 - 1. Hence y = 1 / 2 x (x^2 - 1), And then z is one more than that. So z = 1 / 2 x (x^2 + 1). So for any value of x>1 (because when x=1, y=0 and then z=1 so this is not a triangle) y and z can be found. So for example when x=19 y = 1 / 2 (19^2-1) which is 180 and then z is 1 more than it, so z = 181.

1 pending report

Vote up reports you agree with

×

Problem Loading...

Note Loading...

Set Loading...