2
$\begingroup$

This problem is basically equivalent to the nontrivial case of the problem of finding a product of two factorials that is also a factorial, but that problem seems to be open as well. The keyword here is seems: I have no idea whether it has any solution yet. Thanks in advance for any help!

Sorry, I meant numbers. I am just extremely, extremely sorry.

  • 1
    We have, for example, $6!=8\times 9 \times 10$.2017-02-17
  • 0
    consecutive *digits*? so the numbers whose factorial you have to verify are from $1$ to $8$?2017-02-17
  • 0
    Do you really meant _digits_ in the title? Or numbers?2017-02-17
  • 0
    I would be very interested in your approaches so far.2017-02-17

2 Answers 2

0

The factorial of the number is equal to consecutive digits, greater than the number. So $n\leq8$. We have to check those 8 cases

for $n=1$, $n=2$ there aren't solutions

for $n=3$, $3!=6 \leq 4 * 5$

for $n=4$, $4!=24 \leq 5* 6$

for $n\geq5$, the product does to have divisble with 5. The digits are greater than the number, but there aren't any digits greater than 5 divisble with 5.

So, there aren't any solutions. If you didn't mean digits and you wanted to say numbers, we'll have to look for a different approach.

  • 0
    Sorry. I meant numbers. I wonder what happens to me sometimes.2017-02-17
  • 0
    It's okay. But I can't delete my answer. I flaged it, that's all I can do2017-02-17
0

I have no solution but I think the problem is very interesting. So I state how far i came (note: I am not a number theorist).

At first, the number of occurences of the prime factor $p$ in $n!$ is $$c_p(n)=\sum_{i=1}^{\infty}\left\lfloor\frac n{p^i} \right\rfloor.$$ (I can elaborate this, but I think this is not the important part.) This is roughly $$ \sum_{i=1}^{\infty}\left\lfloor\frac n{p^i} \right\rfloor \approx \sum_{i=1}^{\infty}\frac n{p^i}=\frac{n}{p-1}. $$ For $n!\times m!=K!$ it must hold $c_p(n)+c_p(m)=c_p(K)$. By the above estimation this gives $n+m\approx K$. I guess one can do much better results by more precise estimations of $c_p(n)$.


Assume $K\geq n+m$, espcially $m\leq K-n$. Then we have $$\frac{K!}{n!\cdot m!} \geq\frac{K!}{n!\cdot(K-n)!}={K\choose n}>1$$ for $n\not=0,K$. So we have $K