Why Only Semi-Perfect

A multiplicative perfect number is a positive integer n n who's divisors' product totals to n 2 n^2 .

For example, the factors of 22 22 are 1 , 2 , 11 , 22 , 1 2 11 22 = 484 1,2,11,22,~~1\cdot2\cdot11\cdot22=484

Define the term semi-multiplicative perfect number as a positive integer who's divisors' product (including the number itself) totals to n 3 n^3 . Find the sum of the first 8 semi-multiplicative perfect numbers.


The answer is 200.

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.

4 solutions

First we note that n = 1 n = 1 satisfies the condition, if in fact the condition is that the "divisor product" equals n 3 . n^{3}.

To achieve a "divisor product" of n 3 n^{3} for n > 1 n \gt 1 , we will need to have three distinct pairs of divisors that have a product of n . n. This implies that n n must have precisely 6 6 divisors. So we will need to look for the least n n which have prime decompositions of either the form p 5 p^{5} or p 2 q p^{2}q for primes p , q . p,q. This gives us

2 2 3 = 12 , 2 3 2 = 18 , 2 2 5 = 20 , 2 2 7 = 28 2^{2}*3 = 12, 2*3^{2} = 18, 2^{2}*5 = 20, 2^{2}*7 = 28

2 5 = 32 , 2 2 11 = 44 , 3 2 5 = 45. 2^{5} = 32, 2^{2}*11 = 44, 3^{2}*5 = 45.

The desired sum is then 1 + 12 + 18 + 20 + 28 + 32 + 44 + 45 = 200 . 1 + 12 + 18 + 20 + 28 + 32 + 44 + 45 = \boxed{200}.

I took 2 6 2^{6} 's divisors as 6 6 and hence discluded it. Brilliance at it's best. Brb gonna jump off a cliff.

Kunal Verma - 6 years ago

Oh@Damn it!I didn't take 1 and got the answer 249 because I took 50!

Adarsh Kumar - 6 years, 2 months ago

Log in to reply

Yeah, my first attempt was 249 as well and then I remembered about 1.

Brian Charlesworth - 6 years, 2 months ago

Log in to reply

Yes,but I thought about 1 but then rejected it as it has only one factor.Help!

Adarsh Kumar - 6 years, 2 months ago

Log in to reply

@Adarsh Kumar Well, 1 has no proper divisors, i.e., no (positive) divisors that are not 1, but since it is in general a divisor of itself it does qualify as a multiplicative perfect number as defined in this question.

(Note that 1 is not a perfect number as normally defined, i.e., it is not equal to the sum of its proper divisors.)

Brian Charlesworth - 6 years, 2 months ago

Log in to reply

@Brian Charlesworth Well,then I believe that the wording of the question should be changed a bit,it should be"product of the divisor(s)".Don't you think?

Adarsh Kumar - 6 years, 2 months ago

Log in to reply

@Adarsh Kumar Perhaps, but when I see the word "divisors" I think of all divisors, including the number itself, rather than just the proper divisors. Maybe the default for others, including yourself, is proper divisors; is that the case? If so, then I can understand why you discarded 1, in which case the question should probably specify that the divisor product involves all possible divisors of n . n.

Brian Charlesworth - 6 years, 2 months ago

Log in to reply

@Brian Charlesworth Actually sir,when I saw the word 'divisors',I discarded 1 because I thought that the number needs to have more than 1 divisor.

Adarsh Kumar - 6 years, 2 months ago

Log in to reply

@Adarsh Kumar Ah, o.k., in that case your suggestion of "product of the divisor(s)" might be the appropriate edit, but technically the wording "divisors" does not rule out the possibility of there being just one divisor. It's up to Trevor if he wants to make that edit.

Brian Charlesworth - 6 years, 2 months ago

Log in to reply

@Brian Charlesworth Ok sir,I got it. @Trevor Arashiro what do you think?

Adarsh Kumar - 6 years, 2 months ago

Log in to reply

@Adarsh Kumar Ya, I'll make the change. I thought it was implied because of the example (22) I gave.

Trevor Arashiro - 6 years, 2 months ago

@Trevor Arashiro Another nice question, and another "divisor product" edit suggestion. :)

Brian Charlesworth - 6 years, 2 months ago

Log in to reply

Oh, haha, I wrote it the first time but forgot to write it the second time. Thanks for the complement.

This set introduced me to many of the coolest and/or useless facts about special numbers. The most interesting one being that there is a mathematical proof for why math isn't boring.

Define a boring number as a number which has no information about it, there are no facts (whether known or not known) about it. This number is now a boring number. But wait, that's a fact about it! "It's a boring number" is a fact about the boring number and hence our definition, that number can not be boring :3. problem world?

Trevor Arashiro - 6 years, 2 months ago

Log in to reply

Haha. Now define an "interesting" number as a number that has nothing interesting about it. So then if a number is "interesting" then it is not interesting, and hence can't be "interesting", which is interesting in itself, because no number can be both interesting and not interesting at the same time, can it?! So what is the order of the set of all interesting numbers? Now what if we were to look at the set of all "interesting" sets, defined in a similar fashion. Would this be just the null set? Interesting .... not? :D

P.S.. The "divisor product" edit goes for your question "Only Partially Perfect?" as well. :)

Brian Charlesworth - 6 years, 2 months ago

Log in to reply

@Brian Charlesworth O.o Warning error code 404: "Trevor.exe" has stopped working.

Reason: mind blown.

I believe that we have achieved a paradox similar to that of Russel's paradox about the set of all sets which contain themselves. I'm not sure I fully understand this set stuff. It's TOOO complicated for me. :3

Trevor Arashiro - 6 years, 2 months ago

Log in to reply

@Trevor Arashiro Hahaha. Yeah, whenever a contradictory self-reference shows up, healthy brains tend to shut down to protect their circuits. Those that continue processing risk irreparable damage. :) For me, contemplating the art of Magritte and Escher is a more palatable way of delving into this issue.

P.S.. Sorry to be the 'grammar police', but the possessive is "whose", not "who's". :)

Brian Charlesworth - 6 years, 2 months ago

@Trevor Arashiro , u are Wrong, 1+4+8+9+12+18+20+28=100

anchal shukla - 6 years ago

Log in to reply

4 4 has divisors 1 1 , 2 2 and 4 4 whose product is 8 8 which is not equal to 64 64 aka 4 3 4^{3} . Same goes with 8 8 whose product of divisors is 64 64 and not 512 512 as well as for 9 9 whose product of divisors is 27 27 and not 729 729 Hence, the answer remains 200 200 .

Kunal Verma - 6 years ago

1,4,8,9,12,18,20,25 are right no.s sum is 100

anchal shukla - 6 years ago

Great and upvoted solution! By the way don't we need to take in some argument that prevent n > 1 n > 1 to be a perfect square? In this case there could be only 5 distinct divisors. Taht's what happens for 1, for istance.

Andrea Palma - 6 years ago
Andrea Palma
Jun 13, 2015

Using the Prime Factorization Theorem we can find a charatherization of the semi multiplicative perfect numbers. Suppose n n to have three distinct primes in its factorization: p , q , r p,q,r . Then n p , n q , n r \dfrac{n}{p}, \dfrac{n}{q}, \dfrac{n}{r} and n p q , n q r , n p r \dfrac{n}{pq}, \dfrac{n}{qr}, \dfrac{n}{pr} are distinct divisors of n n (possibly and likely not all the divisors) Their products ρ \rho equals to n 6 p 2 q 2 r 2 n 4 > n 3 \dfrac{n^6}{p^2q^2r^2} \geq n^4 > n^3 so this implies n n can't be a semi-multiplicative perfect number (the product of ALL its divisors being obviously not lower than r h o rho and hence strictly greater than n 3 n^3 ).

So we can limit our search of semi-multiplicative perfect numbers to numbers like p n p^n and numbers like p n q m p^nq^m .

About the type p n p^n we have easily the list of its divisors 1 , p , p 2 , , p n 1 , p n 1, p, p^2, \ldots, p^{n-1},p^n their products being p n ( n + 1 ) 2 \displaystyle{p^{\frac{n(n+1)}{2}}} If p n p^n is a semi-multiplicative perfect number we have n ( n + 1 ) 2 = 3 n {\frac{n(n+1)}{2}} = 3n and this lead necessarily to n = 0 n = 0 or n = 5 n = 5 .

About the type a = p n q m a= p^nq^m (with m , n 1 m,n \geq 1 ) we can see that if BOTH m , n 1 m,n \geq 1 than we have among the a a 's divisors a , a p , a q a, \dfrac{a}{p}, \dfrac{a}{q} and a p q , a p 2 , a q 2 \dfrac{a}{pq}, \dfrac{a}{p^2}, \dfrac{a}{q^2} when just their products ρ \rho equals to a 6 p 4 q 4 a 6 a 2 = a 4 > a 3 \dfrac{a^6}{p^4q^4} \geq \dfrac{a^6}{a^2} = a^4 > a^3 and this means m n { m , n } mn \in \{m,n\} . If both m = n = 1 m=n=1 we have easily the list of the divisors 1 , p , q , p q 1,p,q,pq and their product is a 2 a^2 (in this case we have a multiplicative perfect number, but not semi!). \ So we have one among m m and n n is 1 1 (say m = 1 m=1 ) and the other is greater (say n 2 n \geq2 ). How greater? \ If n 3 n \geq 3 we have the following distinct divisors a , a p , a p 2 , a p 3 , a q , a p q , a p 2 q a, \dfrac{a}{p}, \dfrac{a}{p^2}, \dfrac{a}{p^3}, \dfrac{a}{q}, \dfrac{a}{pq}, \dfrac{a}{p^2q} whose product ρ \rho equals to a 7 p 9 q 3 > a 7 a 3 = a 4 > a 3 \dfrac{a^7}{p^9q^3} > \dfrac{a^7}{a^3} = a^4 > a^3 and hence necessarily n = 2 n = 2 . \ Now for naturals of the type a = p 2 q a = p^2q we have easily the list of the divisors and their products is in fact a 3 a^3 , as showed below 1 p p 2 q p q p 2 q = p 6 q 3 = a 3 . 1 \cdot p \cdot p^2 \cdot q \cdot pq \cdot p^2q = p^6q^3 = a^3.

Finally the semi-multiplicative perfect numbers are exactly the positive integers whose prime factorisation is of the type p 0 = 1 p^0=1 , p 5 p^5 or p 2 q p^2q .

Writing down the smallest numbers of this type we get the list

1 , 12 , 18 , 20 , 28 , 32 , 44 , 45 1, 12, 18, 20, 28, 32, 44, 45

and the sum is 200 200 .

Aryan Gaikwad
Apr 4, 2015
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
public static void main(String args[]){
    int sum = 0;
    for(int i = 1, n = 0; n != 8; i++)
        if(factorProduct(i) == i * i * i){
            sum += i; n++;
        }
    System.out.println(sum);
}

private static int factorProduct(int n){
    int prod = n;
    for(int i = 2; i <= n / 2; i++)
        if(n % i == 0) prod *= i;
    return prod;
}

It was tagged with Computer Science...

David Holcer
Apr 19, 2015
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
def prod(n):
    prod=1
    for i in n:
        prod*=i
    return prod

def factors(n):    
    return set(reduce(list.__add__, 
                ([i, n//i] for i in range(1, int(n**0.5) + 1) if n % i == 0)))

semip=[]
c=1
while True:
    if prod(factors(c))==c**3:
        semip.append(c)
    if len(semip)==8:
        break
    c+=1

summ=0
for i in semip:
    summ+=i
print summ

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...