In the decimal number system, we have the remarkable fact that 3 7 divides 1 1 1 and 3 + 7 = 1 0 . Is there an analogy of this in any other number system? In other words, given a base b , do there exist digits x , y such that x y b divides 1 1 1 b and x + y = 1 0 b ?
In your answer, indicate for how many bases b < 1 0 0 such a pair of digits x , y exists .
Note
A "digit" in base b has values between 0 and b − 1 (inclusive). For instance, in base 16 there are 16 digits: 0 , 1 , 2 , 3 , 4 , 5 , 6 , 7 , 8 , 9 , A ( = 1 0 1 0 ) , B , C , D , E , F ( = 1 5 1 0 ) .
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.
Could you please explain where I went wrong as I don't understand. This is my working (sorry about the layout, I'm not too good with LaTex):
(b^2 + b + 1) / (xb + y) = k (where k is an integer)
x + y = b
Therefore:
[b(b + 1) + 1] / (xb + y) = k
xb + y = Odd (as being even would make k a fraction as the numerator is odd)
Then, testing the possible combination of parities of x,b and y, and seeing if they satisfy the rules:
x, b, y, xb + y = Odd?, x + y = b?
Odd Odd Odd No
Odd Odd Even Yes Yes
Odd Even Odd Yes Yes
Even Odd Odd Yes Yes
Odd Even Even No
Even Odd Even No
Even Even Odd Yes No
Even Even Even No
As x and y are digits, they are at most 9 and at least 1, so b is at most 18 and at least 2. With these possible values of b, I worked out b^2 + b + 1. If this value was prime, then xb + y would have to equal b^2 + b + 1, which is impossible as x and y are less than b (since they are digits used in a base-b value). Therefore, I calculated the other factors of non-primes. Finally, I used these to find the multiples of b that were within 9 under these other factors, and calculated the values of x and y. The values given in my table are only inputted if they fit x + y = b.
b, b^2 + b + 1, Prime?, Other Factors, x,y given (xb + y) and (x + y = b)
2 7 Yes
3 13 Yes
4 21 No 3,7 1,3
5 31 Yes
6 43 Yes
7 57 No 3,19 2,5
8 73 Yes
9 91 No 7,13
10 111 No 3,37 3,7
11 133 No 7,19
12 157 Yes
13 183 No 3,61 4,9
14 211 Yes
15 241 Yes
16 273 No 3,7,13,21,39,91 5,11
17 307 Yes
18 343 No 7,49
Therefore, I got 5 as my answer?
Log in to reply
If b > 1 0 , a "digit" can have a greater value than 9. For instance, in base 16, the "digits" are 0 , 1 , 2 , … , 9 , A , B , … , F .
I will clarify this in the problem statement. Apologies for the confusion.
We have
x + y = b,
bx + y divides b^2 + b + 1, which means bx + b - x divides b^2 + b +1.
Since bx + b - x divides b^2 + b + 1, it also divides (b - 1)(bx + b - x) - x(b^2 +b +1) = b^2 - 3bx - b.
gcd (bx + b - x, b) = 1, therefore bx + b - y divides (b - 3x - 1).
It can be easily shown that |b - 3x - 1| < bx + b - y, which means that b - 3x - 1 = 0. So we have:
b = 3x + 1, y = 2x + 1, and 3bx + 3y = b^2 + b + 1.
I must admit that I can't fully follow Arjen's solution, so I'm just going to put my approach and results down, which is most likely not to be as succinct or as complete (and hopefully stays towards the bottom of the page, to spare much scrolling).
I looked at which integers 111 was divisible by in base 10.
Holding onto the value 3 as a likely candidate for the lowest common denominator for each b that satisfies this problem, I then established the decimal value ( d V ) necessary to create 111 in every b < 100.
Then, to test the theory, I checked all possible combinations of b digits that added up to 1 0 b for each b < 3 6 (except those that had a prime d V ).
With the results confirming a divisor of 3 , I then counted how many d V ( m o d 3 ) returned a remainder of 0 for the total count up to b 9 9 .
b | Decimal value (dV) for 111 in b | d V ( m o d 3 ) | Count of 0s |
b2 | dV7 | 1 | - |
b3 | dV13 | 1 | - |
b4 | dV21 | 0 | 1 |
b5 | dV31 | 1 | - |
b6 | dV43 | 1 | - |
b7 | dV57 | 0 | 2 |
b8 | dV73 | 1 | - |
b9 | dV91 | 1 | - |
b10 | dV111 | 0 | 3 |
b11 | dV133 | 1 | - |
b12 | dV157 | 1 | - |
b13 | dV183 | 0 | 4 |
b14 | dV211 | 1 | - |
b15 | dV241 | 1 | - |
b16 | dV273 | 0 | 5 |
b17 | dV307 | 1 | - |
b18 | dV343 | 1 | - |
b19 | dV381 | 0 | 6 |
b20 | dV421 | 1 | - |
b21 | dV463 | 1 | - |
b22 | dV507 | 0 | 7 |
b23 | dV553 | 1 | - |
b24 | dV601 | 1 | - |
b25 | dV651 | 0 | 8 |
b26 | dV703 | 1 | - |
b27 | dV757 | 1 | - |
b28 | dV813 | 0 | 9 |
b29 | dV871 | 1 | - |
b30 | dV931 | 1 | - |
b31 | dV993 | 0 | 10 |
b32 | dV1057 | 1 | - |
b33 | dV1123 | 1 | - |
b34 | dV1191 | 0 | 11 |
b35 | dV1261 | 1 | - |
b36 | dV1333 | 1 | - |
b37 | dV1407 | 0 | 12 |
b38 | dV1483 | 1 | - |
b39 | dV1561 | 1 | - |
b40 | dV1641 | 0 | 13 |
b41 | dV1723 | 1 | - |
b42 | dV1807 | 1 | - |
b43 | dV1893 | 0 | 14 |
b44 | dV1981 | 1 | - |
b45 | dV2071 | 1 | - |
b46 | dV2163 | 0 | 15 |
b47 | dV2257 | 1 | - |
b48 | dV2353 | 1 | - |
b49 | dV2451 | 0 | 16 |
b50 | dV2551 | 1 | - |
b51 | dV2653 | 1 | - |
b52 | dV2757 | 0 | 17 |
b53 | dV2863 | 1 | - |
b54 | dV2971 | 1 | - |
b55 | dV3081 | 0 | 18 |
b56 | dV3193 | 1 | - |
b57 | dV3307 | 1 | - |
b58 | dV3423 | 0 | 19 |
b59 | dV3541 | 1 | - |
b60 | dV3661 | 1 | - |
b61 | dV3783 | 0 | 20 |
b62 | dV3907 | 1 | - |
b63 | dV4033 | 1 | - |
b64 | dV4161 | 0 | 21 |
b65 | dV4291 | 1 | - |
b66 | dV4423 | 1 | - |
b67 | dV4557 | 0 | 22 |
b68 | dV4693 | 1 | - |
b69 | dV4831 | 1 | - |
b70 | dV4971 | 0 | 23 |
b71 | dV5113 | 1 | - |
b72 | dV5257 | 1 | - |
b73 | dV5403 | 0 | 24 |
b74 | dV5551 | 1 | - |
b75 | dV5701 | 1 | - |
b76 | dV5853 | 0 | 25 |
b77 | dV6007 | 1 | - |
b78 | dV6163 | 1 | - |
b79 | dV6321 | 0 | 26 |
b80 | dV6481 | 1 | - |
b81 | dV6643 | 1 | - |
b82 | dV6807 | 0 | 27 |
b83 | dV6973 | 1 | - |
b84 | dV7141 | 1 | - |
b85 | dV7311 | 0 | 28 |
b86 | dV7483 | 1 | - |
b87 | dV7657 | 1 | - |
b88 | dV7833 | 0 | 29 |
b89 | dV8011 | 1 | - |
b90 | dV8191 | 1 | - |
b91 | dV8373 | 0 | 30 |
b92 | dV8557 | 1 | - |
b93 | dV8743 | 1 | - |
b94 | dV8931 | 0 | 31 |
b95 | dV9121 | 1 | - |
b96 | dV9313 | 1 | - |
b97 | dV9507 | 0 | 3 2 |
b98 | dV9703 | 1 | - |
b99 | dV9901 | 1 | - |
Note, however, that the problem did not specify that one of the divisors should be 3. For instance, modulo 9 you have 1 1 1 9 = 9 1 1 0 , which equals 7 × 1 3 . Thus 1 4 9 is a divisor of 1 1 1 9 , and the only reason that it is not a valid solution is that 1 + 4 = 5 = 1 0 9 .
Now it happens to be true that for all solutions the divisor equals 3 (see my proof), but you cannot assume that this is the case!
Log in to reply
I suppose my solution doesn't actually detail all the steps I took in my (somewhat scientific, rather than mathematical) approach.
The divisors for 111 in base 10 are pretty big clue. The number 3 was my hypothesis, and after testing the theory by dividing 1 1 1 b by all possible combinations of b digits that summed to 1 0 b for all b < 3 6 , I had enough data to draw a conclusion that 3 was the important number - which, as you say, happens to be true.
My initial theory was d V values that weren't prime numbers, but, coincidentally, 1 1 1 9 proved that incorrect.
Here is a nice way to see the pattern in d V m o d 3 : let b = 3 n + a , with a = − 1 , 0 , + 1 . Then d V = b 2 + b + 1 = ( 3 n + a ) 2 + ( 3 n + a ) + 1 ≡ a 2 + a + 1 m o d 3 . Now try the three possible values for a : ⎩ ⎪ ⎨ ⎪ ⎧ d V ≡ ( − 1 ) 2 + 1 + ( − 1 ) = 1 m o d 3 d V ≡ 0 2 + 0 + 1 = 1 m o d 3 d V ≡ 1 2 + 1 + 1 = 3 ≡ 0 m o d 3 a = − 1 a = 0 a = + 1 So you see that your pattern of 0 , 1 , 1 , 0 , 1 , 1 , … continues forever...
Problem Loading...
Note Loading...
Set Loading...
From the second equation, y = b − x ; the first equation translates to a ( b x + y ) = b 2 + b + 1 ∴ a b x + a b − a x = b 2 + b + 1 . Considering this equation modulo b , we see that a x ≡ − 1 , i.e. a x = m b − 1 for some integer m ≥ 1 . Thus m b 2 + ( a − m − 1 ) b + 1 = b 2 + b + 1 ∴ m b + ( a − m − 1 ) = b + 1 . It follows that m = 1 and a − m − 1 = 1 , so that a = 3 . The equation a x = m b − 1 translates into b = 3 x + 1 . This is a necessary and sufficient condition for there to exist a pair of digits with the required property. Naturally, x = 0 and b = 1 is not meaningful; therefore we count how many integers x ≥ 1 there are such that 3 x + 1 < 1 0 0 ; the answer is 3 2 .
Specifically, the first few solutions are 3 ⋅ 1 3 4 = 1 1 1 4 3 ⋅ 2 5 7 = 1 1 1 7 3 ⋅ 3 7 1 0 = 1 1 1 1 0 3 ⋅ 4 9 1 3 = 1 1 1 1 3 3 ⋅ 5 B 1 6 = 1 1 1 1 6