[SOLVED] What's the formula for calculating large exponents?


by slakedlime
Tags: exponents, formula, solved
slakedlime
slakedlime is offline
#1
Nov1-07, 12:27 PM
P: 76
Hi! I came across the following sum and was wondering if there was some kind of formula for solving it:

16^198

Is it some kind of binomial expansion? Do you calculate the last digits first, advancing to the previous ones later?

I can't use a calculator.

Please, please help. Sums like this might come in an assessment exam I'm sitting. Thank you very much!
Phys.Org News Partner Mathematics news on Phys.org
Researchers help Boston Marathon organizers plan for 2014 race
'Math detective' analyzes odds for suspicious lottery wins
Pseudo-mathematics and financial charlatanism
Gib Z
Gib Z is offline
#2
Nov2-07, 02:06 AM
HW Helper
Gib Z's Avatar
P: 3,353
It wants you to evaluate 16^(198) without a calculator?

Its approximately [tex]2.60469314 \cdot 10^{238}[/tex], even with an efficient method it would take an enormous amount of time to evaluate, you sure you don't mean approximate, or find the last 5 digits of it?
uart
uart is offline
#3
Nov2-07, 03:28 AM
Sci Advisor
P: 2,751
Yeah it's not actually solving it but I can make a very rough approximation because I happen to remember that log_10(2) is pretty close to 0.3. Keeping this approx figure in your head lets you do quick and nasty power of 2 to power of 10 (or visa versa) conversions without a calculator.

So in this case [tex]16^{198} = 2^{4*198} \simeq 10^{0.3 * 4 * 198}[/tex], which gets you an order of magnetude calcuation of 10^238.

slakedlime
slakedlime is offline
#4
Nov2-07, 04:32 AM
P: 76

[SOLVED] What's the formula for calculating large exponents?


I can do the approximations, but is there any way to find the last 6 digits? The question asked for both the approximation and the last 6 digits, so I thought perhaps the method would be similar and that there would be a formula.

Does this have something to do with modular mathematics?

Thank you for your help. :)
Zurtex
Zurtex is offline
#5
Nov2-07, 08:31 AM
Sci Advisor
HW Helper
P: 1,123
Quote Quote by Gib Z View Post
It wants you to evaluate 16^(198) without a calculator?

Its approximately [tex]2.60469314 \cdot 10^{238}[/tex], even with an efficient method it would take an enormous amount of time to evaluate, you sure you don't mean approximate, or find the last 5 digits of it?
Why would it take an enormous amount of time to evaluate? Just multiply it out...

26046931378436930758124421057504913270096712196546516251547882077203270 46022512527938059453465450894821456996325559859549175313146140376984516 93595794.... (Edit I've removed the last load of digits not to make it too easy for you).

Takes fractions of seconds on modern computers.

And yeah, doing it via modular arithmetic is even easier. Because if you want to know the last 6 digits you can just take it mod 1'000'000, so you can just keep knocking off any digits larger than the first 6.
Moo Of Doom
Moo Of Doom is offline
#6
Nov2-07, 08:33 AM
P: 367
Quote Quote by Zurtex View Post
Takes fractions of seconds on modern computers.
You didn't read the "without a calculator" part, didn't you?
Zurtex
Zurtex is offline
#7
Nov2-07, 09:21 AM
Sci Advisor
HW Helper
P: 1,123
Quote Quote by Moo Of Doom View Post
You didn't read the "without a calculator" part, didn't you?
Well I didn't use a calculator :P

But yeah, it looks like you have to do some clever modular arithmetic, it'd pretty easy mod 10, mod 100 or perhaps even mod 1'000 by hand, but it's a bit of a pain mod 1'000'000.


I'll think on a bit and see if I come up something (certainly nothing that immediately came to mind worked).
Count Iblis
Count Iblis is offline
#8
Nov2-07, 10:52 AM
P: 2,159
I would suggest to give the answer in the binary number system
slakedlime
slakedlime is offline
#9
Nov2-07, 11:44 AM
P: 76
Thank you all for trying. Forget 6 digits. My exam's tomorrow afternoon, so if it'd be easier to find the last two digits without using a calculator, could you guys show me how? I'd just like to have an idea even if I can't answer.
Count Iblis
Count Iblis is offline
#10
Nov2-07, 12:24 PM
P: 2,159
16^198 = 16^(2*3^2 * 11)

All computations mod 100 below:

16^2 = 256 = 56

16^(2*3) = 56^3 = 16

16^(2*3^2) = 16^3 mod(100) = 96 = -4

16^(2*3^2*11) = (-4)^11 =-2^22

now 2^10 = 1024 = 24

so 2^20 = 24^2 = 576 = 76

2^22 = 2^20 * 4 = 76*4 = 4 ------>

16^198 = -4 = 96
slakedlime
slakedlime is offline
#11
Nov2-07, 09:57 PM
P: 76
Quote Quote by Count Iblis View Post
2^22 = 2^20 * 4 = 76*4 = 4 ------>
16^198 = -4 = 96
I didn't quite catch that part.
Zurtex
Zurtex is offline
#12
Nov4-07, 02:50 PM
Sci Advisor
HW Helper
P: 1,123
Quote Quote by slakedlime View Post
I didn't quite catch that part.
Reducing it down to the results rather than the calculations:

16^198 = 16^(2*3^2 * 11)

16^(2*3^2*11) = (-4)^11 =-2^22

2^22 = 4

16^(198) = -4

-4 is the same as 96, mod 100.


Register to reply

Related Discussions
[SOLVED] Exponents.... Precalculus Mathematics Homework 4
[SOLVED] Logarithms and Exponents Question Precalculus Mathematics Homework 5
Appoximating a very large number as a small formula General Math 15
Find the last digit of the sum of two numbers with large exponents. Calculus 9
Can you give me the formula for calculating its... Calculus 4