**Problem 5. (6 points)** Find the value of

\[\sum_{m=1}^{8192}\left\lfloor\left\lfloor\frac{13}{1+\sum_{j=2}^m\left\lfloor\frac{(j-1)!+1}{j}-\left\lfloor\frac{(j-1)!}{j}\right\rfloor\right\rfloor}\right\rfloor^{1/13}\right\rfloor\]

No vote yet

13 votes

×

Problem Loading...

Note Loading...

Set Loading...

Easy Math Editor

`*italics*`

or`_italics_`

italics`**bold**`

or`__bold__`

boldNote: you must add a full line of space before and after lists for them to show up correctlyparagraph 1

paragraph 2

`[example link](https://brilliant.org)`

`> This is a quote`

Remember to wrap math in \( ... \) or \[ ... \] to ensure proper formatting.`2 \times 3`

`2^{34}`

`a_{i-1}`

`\frac{2}{3}`

`\sqrt{2}`

`\sum_{i=1}^3`

`\sin \theta`

`\boxed{123}`

## Comments

Sort by:

TopNewestSuppose \(k\) is prime, then it's quite easy to argue that \(k \mid k!\) but \(k \not \mid n!\) for any \(n < k\), so \((k-1)! \equiv -1 \mod k\). In fact, it should be obvious that the other direction holds as well: \[ (k-1)! \equiv -1 \mod k \iff k \text{ is prime}\] Now, this suggests that only when \(k\) is prime will \(\frac{(k-1)!+1}{k}\) be an integer (since by definition \(k\) divides \((k-1)! + 1\)). Furthermore, since \(0 < \frac{(k-1)!+1}{k} - \left\lfloor \frac{(j-1)!}{j} \right\rfloor \le 1\), then this will only be one iff \(k\) is prime and zero otherwise: an indicator for primeness. In otherwords the summation on the denominator counts the number of primes up to \(m\), aka the prime \(\pi\) function.

From here it's easy sailing. Since the maximum of the expression \(\frac{13}{1+\pi(m)}\) is 13, and because \(1 < \sqrt[13]{13} < 2\), each term will contribute exactly one to the sum until \(\frac{13}{1+\pi(m)} < 1\). This will happen exactly when we hit the \(13^{th}\) prime, which happens to be \(41\), so the sum must have a value of \(p_{13} - 1 = 40\).

Log in to reply

Correct! As your solution was posted before Michael Lee's, you get the 6 points. Great solution.

Log in to reply

Aww thank you so much, but I don't feel like it's fair to award the point to me since Michael solved the problem over an hour before I did, so I insist that the points should go to him! Thanks for the fun problems :)

Log in to reply

Log in to reply

Note that the expression is undefined when \(m = 1.\)

Log in to reply

By standard math protocol, \(\sum_{i=n+1}^ni=0\).

Log in to reply

Right!

Log in to reply

That's obvious....

Log in to reply

Note : \(8192 = 2^{13}\)

Suspicious.......

Log in to reply

I noticed that. -.-

Log in to reply

The answer is 40.

Log in to reply

YES! Solution?

Log in to reply

From where do you get such questions?

Log in to reply

Cody makes them.

Log in to reply

Wow, really nice! :)

Log in to reply

999?

Log in to reply

Proof: "The result will blow your mind. -- Cody Johnson"

Log in to reply

darn at first I thought that "CMC" meant "Canada Mathematics Competition".

Log in to reply