Patterns in Exponent Problems
Today we’re going to tackle a couple of tough exponent problems from GMATPrep®. The two problems I’ve chosen share some interesting characteristics.
Here’s your first one; set your timer for 2 minutes and go!
What is the greatest prime factor of
-
?
(A) 2
(B) 3
(C) 5
(D) 7
(E) 11
What was your first thought here? Mine was: get those bases into the same form.
-
And then my next thought was: give me a calculator.
Okay, clearly they’re not going to do that, so there’s got to be some other way around this. How can I simplify those two terms even further?
The solution to my problem is not
. If you’re not sure (and you can’t look up the rule because you’re in a testing situation), try a pair of small numbers to see. What’s
–
? Is it
= 3? No. It’s 27 – 9 = 18. Okay, so now what?
Oh, right, we subtract the exponents when we’re given a division problem, but this is a subtraction problem. How do we solve this one? We factor: we pull a common term out of each.
In this case, our common term is
:
=
=
= 
Hmm. The
term is made up entirely of 2s, so the largest prime number out of that piece is simply 2. What about the 63? 63 = 9 × 7 = 3 × 3 × 7. Great! The largest prime number is 7.
The answer is D.
The biggest trap answer, in my opinion, is 2 (answer A). It would be easy to look at the original problem and think, well, 2 is the only prime factor of
and 2 is also the only factor of
, so 2 has to be the biggest prime factor of both. The key is that the act of subtracting can change everything.
That process in general – of factoring out a large term when we’re adding or subtracting ridiculously-high-exponent terms – is one we need to remember. When you see this kind of set up (base-with-huge exponent plus-or-minus base-with-huge-exponent), factor out the smaller of the two terms. You’re going to end up with whatever’s left out of the larger term (in this case, that was 26) minus the number 1, so you know you’re going to get some new number out of this with different factors.
Ready for the next one? You know the drill (2 minutes!).
If
=
, what is the value of x?
(A) 9
(B) 11
(C) 13
(D) 15
(E) 17
My first thought: hmm, how do I combine that stuff on the left-hand side of the equation? My second thought: yuck. Can’t I do it any other way? Then I noticed the three on the right-hand side. The left-hand side has only base-2s. Where could that 3 come from?
That one thought let me to a really nice shortcut on this one – but the shortcut requires you to have mastered exponents already. I will show it to you, but not until after we slog through this the longer way. The thing I want you to notice right now, though, is what led me to notice that 3 – it was because I didn’t simply start down the first path I noticed but asked myself whether there was an easier way because that first path was kind of annoying. I’m not a fan of annoying solution methods; nobody is! So make that work for you – when you notice something annoying, ask yourself whether there’s a way to cut out the annoying part.
Okay, so let’s go down the somewhat more annoying path.
Hmm. Somehow, I’ve got to combine the two terms on the left-hand side of the equation. There’s a subtraction sign between them, as in the last problem, but I don’t really know how to factor out when the exponents are variables like that. (It can actually be done, but most people wouldn’t know how, so I’m going to pretend that I can’t, either.) What else?
Well, let’s try to split the variables out from the numbers as much as possible.
can be written
– 
Okay, so my full equation now reads:
-
= 
Hmm. Fractions are annoying in equations, so I suppose I could multiply everything by
, or 4:
= 
So I have 4 “two to the x” terms minus 1 “two to the x” term… that’s 3 “two to the x” terms:
= 
This is starting to look a little bit better. Divide by 3:
= 
So close! Now what? I can just set the bases equal to each other now, right? So x must equal 13.
Wait, wait! Not so fast. 4 is the same thing as
; it should be part of that
. I can’t just use some of the 2s but not all of them.
= 
So now my full equation is:
= 
Voilà. Drop the bases, set the exponents equal to each other, and we get x = 15.
The correct answer is D.
Ultimately, that math wasn’t terrible, but there were multiple steps where I wasn’t exactly sure where things were going – I was just trying to manipulate and see whether that then gave me any ideas. That’s kind of unsettling. Whenever possible, I’d rather be doing something that I know is working every step of the way.
So let’s go back to my earlier thought: what’s that 3 doing there? Where did it come from? Any ideas?
Does it remind you of anything else you’ve seen recently?
Hint, hint.
That whole right-hand side structure,
, reminded me of a similar-looking term in the previous problem:
. Sure, I wrote that a little bit differently last time, but the same two components are there: a base with a ridiculously high exponent multiplied by another plain number.
Where did that weird 63 come from last time? Oh, right, the minus-1 thing. We had 64 – 1 = 63 (which started out as
– 1). So, in this one, did that 3 come from 4 – 1 = 3? And, if so, that 4 would have started out in the form of
, probably, something like this:
4 = (
– 1)
Put that in our original problem instead of the 3:
= 
Do you like the problem better in this form or in the original form? I think this is much more straightforward; we just need to multiply out on the right-hand side:
= 
And bingo! We can just compare the two halves of the equation. Our unknown x equals 15 (and the correct answer is still D). Now, you wouldn’t want to figure all of that out from scratch during the test – the first method is easier than actually figuring this out. But if you already knew everything that we just discussed in the first problem, including the fact that subtracting exponential terms in this way is going to result in a “something minus 1” situation, then you might be able to do what I did when I got to the second problem: look at that 3, realize it must have come from 4 – 1, which is
– 1, and then the problem just opens right up.
Key Takeaways for “Really Big Exponent” Problems:
(1) If there are ridiculous exponents, there’s a pattern somewhere. While studying, learn what the common patterns are and how to recognize them on different problems in future. That second step is critical – you don’t want to have to figure this stuff out all over again in a timed situation.
(2) Here’s one pattern to add to your list. When you see a base with a really big exponent and some coefficient (fancy word for number) out front, such as
, recognize that the weird number probably came from factoring, and that factoring will have been in the form of “something minus 1.”
(3) Sometimes, looking at the problem “backwards” is the more efficient way to do the problem. The vast majority of students would do what I also first thought to do: try to manipulate the left-hand side of the equation in the 2nd problem. Only a few will think to go “the other way” and start with the right-hand side. Look for those opportunities during your study.
* GMATPrep® questions courtesy of the Graduate Management Admissions Council. Usage of this question does not imply endorsement by GMAC.


15 comments
Rai_90 on October 22nd, 2011 at 6:12 am
Nice.
For question 2, here is another way:
2^x-2^(x-2) = 3(2^13)
On the left hand side, factor out 2^(x-2) to get
2^(x-2) (2^-1) = 3(2^13)
2^(x-2) (4-1) = 3(2^13)
2^(x-2) (3) = 3(2^13)
2^(x-2) = (2^13)
x-2=13
x=15
Rai_90 on October 22nd, 2011 at 6:14 am
Oops. I had a typo in my post. Here is a correction:
2^x-2^(x-2) = 3(2^13)
On the left hand side, factor out 2^(x-2) to get
2^(x-2) (2^2-1) = 3(2^13)
2^(x-2) (4-1) = 3(2^13)
2^(x-2) (3) = 3(2^13)
2^(x-2) = (2^13)
x-2=13
x=15
Joy on October 22nd, 2011 at 8:14 am
@Rai_90: Good solution!
Stacey Koprince on October 22nd, 2011 at 11:34 am
Yes! That was what I was referring to in my article when I said factoring can be used with variables in the exponents but most people probably wouldn't know how, so I did it a different way. Thanks for showing the path!
(To others) If you feel totally comfortable with the math needed for Rai's solution, that is absolutely a valid and efficient way to do things. If you feel like you might be more likely to make a mistake with that technique, then use another method (either what the article discusses or something else that you come up with yourself!).
Richa on October 22nd, 2011 at 1:26 pm
Nice article Stacey,but won't it save time if we substituted the options given to us on the Left hand side of the equation. We could start with a middle value such as 13 (option C) and equate the value with the right hand value of 3* 2^13. If we find that value to be less we could try higher numbers options: options (D & E). In this case if you plug in you can find that option D gives you the correct value of X. I guess this could save time and reduce the error possibility as one can avoid the complicated algebra
Stacey Koprince on October 23rd, 2011 at 7:33 am
Yes, that's another valid solution method. It's a really good idea to try to find multiple solution methods to problems so that you have options when future problems toss "twists" at you.
It's also the case that, if you learn to recognize the shortcut I showed for the second one and you really understand how the math works, you can solve this one in about 20 seconds. I didn't even write anything down - as soon as I recognized that the 3 represented what was "left" and the 2^13 term was what was "pulled out," I looked at the left-hand side and realized that 2^(x-2) (the smaller term) was also what was "pulled out"... and, boom, that meant x had to be 15.
Now, should everybody do it that way? No - only if this is a big area of strength for you and you completely get how the math itself works and the conceptual stuff that I wrote in the previous paragraph. Then, this can be a really great path.
If you're capable of learning and executing on that method without making mistakes, then that's even better than checking the answers because it's even more efficient - and if you're looking to push your scores into the stratosphere, you have to learn to get faster at the harder questions so that you have more time for the even harder questions you're going to get if you really are doing that well!
Richa on October 23rd, 2011 at 2:44 pm
I agree with you Stacey that it is always better to know 2 or more ways to solve any problem and to be honest I had solved this problem the same way Rai did during the gmat prep practice test,but it took almost 2.45 to 3 mins to solve then.
So while reviewing the questions I tried to figure out a more time-efficient technique to solve it and thought of option substitution which definitely was quicker but the short cut technique mentioned by you to manipulate the R.H.S first is even more time saving and easy. Thanks again.
Nan on October 22nd, 2011 at 8:27 pm
Hi Stacey, I really liked your article and the method you used to solve the problems. But, I didn't understand Rai's solution and would like to know more about how he solved the problem. Can you please explain his way and/or where I can look up the information about the method he used?
Thanks! Nan
Stacey Koprince on October 23rd, 2011 at 7:37 am
Rai's solution is the conceptual version of what we were already doing with real numbers in the first problem.
Let's say we have:
2^8 - 2^6
I can pull out a 2^6 term to get: 2^6 (2^2 - 1)
Right?
Okay, so substitute variables in there instead. We're going to say x = 8:
2^x - 2^(x-2)
See how that's the exact same setup as the 2^8 - 2^6 problem above?
So in the "numbers" problem, we "pulled out" the lower one (2^6). You can do the exact same thing with the "variables" problem but it's abstract, so most people find that a lot harder.
Pull out the smaller one: 2^(x-2)
and then what do you have left? Well the 2^x still has "two more" exponents that we haven't pulled out, just as 2^8 had "two more" exponents left after ew pulled out a 2^6. And the 2^(x-2) doesn't have any exponents left, just as the 2^6 didn't have any left after pulling out a 2^6.
So we get: 2^(x-2) (2^2 - 1)
But see how much more complicated that gets when it's abstract? Most people wouldn't feel comfortable doing that at high speed without making mistakes or getting confused... and if that describes you, then you shouldn't follow this particular path.
Nan on October 23rd, 2011 at 9:21 am
Thanks, Stacey. It makes complete sense now. You'r a great instructor!
Geo on October 28th, 2011 at 6:48 pm
Great article. Very useful too.
Arvind Ranganath on October 31st, 2011 at 12:52 pm
Good one Stace!
Trinity on January 14th, 2012 at 1:05 pm
THanks Stacey. I was able to solve another GMAT PREP problem ( 3^x - 3X-1 = 162) in less a minute using your method. I cringed when i saw this problem; my thought process was exactly like you stated above..
Michele on February 17th, 2012 at 1:47 pm
how would I solve 45678^98765.
You can't enter it into calculator, and I know there is a pattern here but I see it> Please help.
Stacey Koprince on February 17th, 2012 at 7:47 pm
Yeah, the answer's basically infinity. You won't see that on the GMAT.
But if you really want to do it, you can use the exact same process that was described in the article (to find the unit's digit, which is what the GMAT would ask).
When raising something to a power, the only thing that determines the unit's digit is the starting unit's digit, which is 8 in this case.
What are the possible units' digits from this starting point?
8^1 = 8
8^2 = 64
8^3 = the previous unit's digit *8 = 8*4 = 32 = a unit's digit of 2
8^4 = previous unit's digit of 2 * 8 = 16 = unit's digit of 6
8^5 = previous unit's digit of 6 * 8 = 48 = unit's digit of 8 (which matches the first one, so I've just found the pattern)
The 1st = 8
The 2nd = 4
The 3rd = 2
The 4th = 6
The 5th = 8
The 6th = 4
The 7th = 2
The 8th = 6...
So far, this would all be fair game for a real GMAT question. This next step is in theory something the GMAT would make you do but NOT with such a crazy number.
What's the power of 4 that's as close as possible to 98765 but just smaller than that number?
There's a power of 4 every 4 numbers, so it either has to be 98765, 98764, 98763, or 98762. Clearly, the ones ending in odd numbers are not divisible by 4, so it ahs to be either 98764 or 98762. If a number is divisible by 4, then its final two digits are also divisible by 4. In this case, 64 is divisible by 4, so the number 98764 is also divisible by 4.
If they had asked us to raise the number to the 98764 power, then the units digit would be 6 (because this is a multiple of 4, and the multiples of 4 have this units digit, according to the pattern). The next number up, 98765, must have a units digit of 8 (because that's the next number in the pattern).