It's alright to count

Find the highest power of 2 that divides 33 ! 33! .


The answer is 31.

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.

7 solutions

Chew-Seong Cheong
Aug 22, 2015

The highest power of 2 2 that divides n ! n! is given by:

p ( n ) = k = 1 n 2 k \begin{aligned} p(n) & = \sum_{k=1}^\infty \left \lfloor \frac{n}{2^k} \right \rfloor \end{aligned}

where x \lfloor x \rfloor is the greatest integer function.

For n = 33 n=33 , we have:

p ( 33 ) = k = 1 33 2 k = 33 2 + 33 4 + 33 8 + 33 16 + 33 32 = 16 + 8 + 4 + 2 + 1 = 31 \begin{aligned} p(33) & = \sum_{k=1}^\infty \left \lfloor \frac{33}{2^k} \right \rfloor \\ & = \left \lfloor \frac{33}{2} \right \rfloor + \left \lfloor \frac{33}{4} \right \rfloor + \left \lfloor \frac{33}{8} \right \rfloor + \left \lfloor \frac{33}{16} \right \rfloor + \left \lfloor \frac{33}{32} \right \rfloor \\ & = 16+8+4+2+1 \\ & = \boxed{31} \end{aligned}

The question was horribly written.

Case 1: Assuming no remainder restrictions, the answer is infinity. eg, 2^100 divides into 33, eg, 2^999999 also divides into 33. and so on.

Case 2: Assuming the remainder is restricted to be 0, the highest whole number power of 2 that divides into 33 is (-1). ie, 2^(-1) divides into 33 exactly 66 times.

The question NEVER mentioned greatest integer functions. The question should have been more carefully crafted.

Danny Speagle - 5 years, 9 months ago

Seriously awesome upvoted! Can we use the power of 2 divisible equation for any number like 3?

Department 8 - 5 years, 9 months ago

Log in to reply

Yes. The same for other numbers.

Chew-Seong Cheong - 5 years, 9 months ago

Is this a formula that had been previously proved or is it something you came up with on your own? I can't seem to understand it unfortunately...

Jeffrey Li - 5 years, 9 months ago

Log in to reply

It is not that difficult actually.

33 ! = 1 × 2 × 3 × . . . × 33 33! = 1\times 2 \times 3 \times ... \times 33 . Every even number from 1 1 to 33 33 contributes 1 1 to the power of 2 2 , therefore there are 33 2 = 16 \left \lfloor \frac{33}{2} \right \rfloor = 16 . But numbers that are divisible by 4 = 2 2 4=2^2 , each contributes another 1 1 to the power. Therefore, we have to add 33 4 = 8 \left \lfloor \frac{33}{4} \right \rfloor = 8 . Similarly, for numbers divisible by 8 8 , 16 16 and 32 32 .

Chew-Seong Cheong - 5 years, 9 months ago

I understood the solution. Isn't p(n) = p(33) instead of p(30)? [x] is not mentioned in the expression for p(n). So what is [x]?

M Menon - 5 years, 9 months ago

Log in to reply

Thanks. A typo.

Chew-Seong Cheong - 5 years, 9 months ago
Otto Bretscher
Aug 26, 2015

ν 2 ( 33 ! ) = ν 2 ( 32 ! ) = 32 1 2 1 = 31 \nu_2(33!)=\nu_2(32!)=\frac{32-1}{2-1}=\boxed{31} by Legendre's Theorem

WOW looks too good to be true. +1

Pi Han Goh - 5 years, 9 months ago

Log in to reply

It's good to hear from you, Comrade. I'm sorry I can't be more active on Brilliant these days, but... duty calls.

Otto Bretscher - 5 years, 9 months ago

Log in to reply

haha no worries. Being an adult sucks haha

Pi Han Goh - 5 years, 9 months ago
Dev Sharma
Aug 22, 2015

The highest power of p dividing any number a is given by:

[a/p] + [a/p^2] +...

use it for a = 33 and p = 2

Is there a name for this? Like, a famous Law or Rule? Or is it simply just floor function?

Rico Lee - 4 years, 8 months ago
Ian McKay
Aug 25, 2015

It's not as fancy or as ‘clean’ as how other people solved it, but an intuitive way to approach the problem (for those who don't know/understand math terminology as well)

I just considered each number in 33! and calculated how many times you could divide each by 2. 2/2 = 1, 3/2= 0, 4/2= 2, 5/2 = 0, 6/2= 1, etc… And then added those numbers up to get 31.

Moderator note:

Yes, that is the slow (and painful) way to solve this problem. What if we want 1000 ! 1000! ? Do we need to sum up 1000 (or at least 500) numbers?

It may be 'slow and painful' but at least it is a genuine solution that does not rely on invoking other theorems. Well done!

A Former Brilliant Member - 4 years, 5 months ago
Hadia Qadir
Aug 30, 2015

33/2 +33/4+33/8+33/16+33/32=16+8+4+2+1=31

Solved it computationally. Basically, compute n! mod 2^k (for k = 0..2n, say).

Richard Fink
Aug 26, 2015

I used the brute force method. Enter 33! into calc divided it by 2 until I got an odd number. I hit the return key 31 times.

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...