Volume 13, Number 5 January-February, 2009

Olympiad Corner Generating Functions

The following were the problems of
the Final Round (Part 2) of the Kin Yin Li
Austrian Mathematical Olympiad
In some combinatorial problems, we For each way of changing n dollars into
2008.
may be asked to determine a certain r 1 dollar and s 2 dollar coins, we can
First Day: June 6th, 2008 sequence of numbers a0, a1, a2, a3, …. record it as xr x2s = xn. Now r and s may
We can associate such a sequence with be any nonnegative integers. Adding all
Problem 1. Prove the inequality the following series the recorded terms for all nonnegative
integers n, then factoring, we get
1 f(x) = a0 + a1x + a2x2 + a3x3 + ⋯.
a1− ab1− bc1− c ≤ ∞ ∞ ∞ ∞ ∞
3
holds for all positive real numbers a, b
This is called the generating function of
the sequence. Often the geometric series
∑x ∑x
r =0
r

s=0
2s
= ∑∑xr+2s = ∑an xn = f (x).
r =0 s=0 n=0
and c with a+b+c = 1. 1 /(1 − t ) = 1 + t + t 2 + t 3 + L for |t| < 1 On the other hand,
Problem 2. (a) Does there exist a and its square ∞ ∞
1 1 1
polynomial P(x) with coefficients in 1 /(1 − t ) 2 = (1 + t + t 2 + t 3 + L) 2 ∑x ∑x
r =0
r

s =0
2s
= ⋅
1− x 1− x 2
=
(1 − x) 2 (1 − x)
integers, such that P(d) = 2008/d holds = 1 + 2t + 3t 2 + 4t 3 + 5t 4 + L 1⎛ 1 1 ⎞
for all positive divisors of 2008? = ⎜⎜ + ⎟
will be involved in our discussions. 2 ⎝ (1 − x) 2 1 − x 2 ⎟⎠
(b) For which positive integers n does a
polynomial P(x) with coefficients in Below we will provide examples to (
= 12 (1+ 2x + 3x2 +L) + (1+ x2 + x4 +L) )
integers exists, such that P(d) = n/d illustrate how generating functions can = 1 + x + 2 x + 2 x + 3x + 3x + L.
2 3 4 5

solve some combinatorial problems. ∞

holds for all positive divisors of n? = ∑ ([n / 2] + 1) x n .
n=0
Example 1. Let a0=1, a1=1 and
Problem 3. We are given a line g with
an = 4an−1 − 4an for n ≥ 2. Therefore, an = [n/2] + 1.
four successive points P, Q, R, S, reading
from left to right. Describe a straight- Find a formula for an in terms of n. Example 3. Let n be a positive integer.
edge and compass construction yielding
Solution. Let f(x) = a0 + a1x + a2x + ⋯. 2 Find the number an of polynomials P(x)
a square ABCD such that P lies on the
Then we have with coefficients in {0,1,2,3} such that
line AD, Q on the line BC, R on the line
P(2) = n.
AB and S on the line CD. f(x) − 1 − x = a2x2 + a3x3 + a4x4 + ⋯
= (4a1−4a0)x2 + (4a2−4a1)x3 + ⋯ Solution. Let f(t) be the generating
(continued on page 4)
= (4a1x2+4a2x3+⋯)−(4a0x2+4a1x3+⋯) function of the sequence a0, a1, a2,
Editors: 張 百 康 (CHEUNG Pak-Hong), Munsang College, HK = 4x( f(x) − 1) − 4x2 f(x). a3, …. Let P(x) = c0 + c1x + ⋯ + ckxk
高 子 眉 (KO Tsz-Mei) with ci∊{0,1,2,3}. Now P(2) = n if and
梁 達 榮 (LEUNG Tat-Wing)
Solving for f(x) and taking |x| < ½, only if c0 + 2c1 + ⋯ + 2kck = n. Taking
李 健 賢 (LI Kin-Yin), Dept. of Math., HKUST t∊(−1,1), we can record this as
吳 鏡 波 (NG Keng-Po Roger), ITC, HKPU f (x) = (1−3x)/(1−2x)2
k
Artist: 楊 秀 英 (YEUNG Sau-Ying Camille), MFA, CU = 1/(1−2x)−x/(1−2x)2 t n = t c0 t 2 c1 Lt 2 ck .
∞ ∞
= ∑ (2 x) n − x ∑ n(2 x) n −1 Note 2ici is one of the four numbers 0,
Acknowledgment: Thanks to Elina Chiu, Math. Dept.,
HKUST for general assistance.
n=0 n =1 2i, 2i+1, 3·2i. Adding all the recorded
On-line: ∞
terms for all nonnegative integers n and
http://www.math.ust.hk/mathematical_excalibur/
= ∑ (2 − n2
n n −1
) xn.
The editors welcome contributions from all teachers and n=0
all possible c0, c1, …, ck ∊{0,1,2,3},
students. With your submission, please include your name, then factoring on the right, we have
address, school, email, telephone and fax numbers (if Therefore, an = 2n − n 2n−1. ∞ ∞
f (t ) = ∑ ant n = ∏ (1 + t 2 + t 2 + t 3⋅2 ).
i i +1 i
available). Electronic submissions, especially in MS Word,
are encouraged. The deadline for receiving material for the n =0 i =0
next issue is March 7, 2009. Example 2. Find the number an of ways
For individual subscription for the next five issues for the n dollars can be changed into 1 or 2 Using 1+s+s +s =(1−s4)/(1−s), we see
2 3

09-10 academic year, send us five stamped self-addressed

dollar coins (regardless of order). For
envelopes. Send all correspondence to: 1 − t 4 1 − t 8 1 − t 16 1 − t 32
example, when n = 3, there are 2 ways, f (t ) = ⋅ ⋅ ⋅ ⋅L
Dr. Kin-Yin LI, Math Dept., Hong Kong Univ. of Science 1 − t 1 − t 2 1 − t 4 1 − t8
and Technology, Clear Water Bay, Kowloon, Hong Kong
namely three 1 dollar coins or one 1 1 1
dollar coin and one 2 dollar coin. = ⋅ .
Fax: (852) 2358 1643 1− t 1− t 2
Email: makyli@ust.hk
Solution. Let f(x) = a0 + a1x + a2x2 + ⋯.
© Department of Mathematics, Hong Kong University of As in example 2, we get an = [n/2] + 1.
Science and Technology. To study this infinite series, let |x| < 1.
Mathematical Excalibur, Vol. 13, No. 5, Jan.-Feb. 09
For certain problems, instead of
using the generating function of a0, a1,
a2, a3, …, we may consider the series
x a 0 + x a1 + x a 2 + x a3 + L.
Example 4. (1998 IMO Shortlisted
Problem) Let a0, a1, a2, … be an
increasing sequence of nonnegative
integers such that every nonnegative
integer can be expressed uniquely in
the form ai+2aj+4ak, where i, j and k
are not necessarily distinct. Determine
a1998.
Solution. For |x| < 1, let f ( x) =
The given condition implies
∞ involving a two variable generating
1
f ( x) f ( x ) f ( x ) = ∑ x =
2 4 n
. z
n=0 1− x
2
Replacing x by x , we get
1 side has a finite limit, but the right side goes
f ( x 2 ) f ( x 4 ) f ( x8 ) =
1 − x2
From these two equations, we get f(x) =
(1+x) f(x8). Repeating this recursively,
we get
2 3
f ( x) = (1 + x)(1 + x8 )(1 + x8 )(1 + x8 )L.
In expanding the right side, we see the
exponents a0, a1, a2, … are precisely
the nonnegative integers whose base 8
representations have only digit 0 or 1.
Since 1998=2+22+23+26+27+28+29+210,
so a1998=8+82+83+86+87+88+89+810.
For our next examples, we need
some identities involving p-th roots of
unity, where p is a positive integer.
These are complex numbers λ, which
are all the solutions of the equation
z p = 1 . For a real θ, we will use the
common notation eiθ = cos θ +i sin θ.
Since the equation is of degree p, there
are exactly p p-th roots of unity. We can
easily check that they are eiθ with θ = 0,
2π/p, 4π/p, …, 2(p−1)π/p.
Below let λ be any p-th root of unity,
other than 1. When we have a series
B(z) = b0 + b1z + b2z + b3z + ⋯,
2 3
sometimes we need to find the value of
bp + b2 p + b3 p + L. We can use the fact
1− λ pj
1 + λ j + λ2 j + L + λ( p −1) j =
1− λj
(for any j not divisible by p) to get
1 p −1
∑ B(λ j ) = bp + b2 p + b3 p + L.
p j =0
For p odd, we have the factorization
Page 2
1 + t p = (1 + t )(1 + λt ) L(1 + λ p−1t ) (**) 1 p −1
∑ (1 + ω j x)(1 + ω 2 j x)L(1 + ω 2 pj x).
p j =0
since both sides have −1/λi (i=0,1,…, p−1)
as roots and are monic of degree p.
For j = 0, the term is (1+x)2p. For 1 ≤ j
Example 5. Can the set ℕ of all positive ≤ p−1, using (**) with λ = ωj and t = λx,
integers be partitioned into more than one, we see the j-th term is (1 + x p ) 2 .
but still a finite number of arithmetic Using these, we have
progressions with no two having the same 1 p−1 1
common differences? ∑F j (x) = p[(1+ x)2 p + ( p −1)(1+ x p )2 ].
p j=0 ω
p
Solution. (Due to Donald J. Newman) Therefore, the coefficient of x is
Assume the set ℕ can be partitioned into
1 ⎡⎛ 2 p ⎞ ⎤.
sets S1, S2,…,Sk, where Si={ai+ndi: n∊ℕ} C= ⎢⎜⎜ ⎟⎟ + 2( p − 1) ⎥
p ⎣⎝ p ⎠
with d1 > d2 > ⋯> dk. Then for |z| < 1, ⎦
∞
∑x
i =0
ai
. ∞
∑z = ∑z n
∞
a1 + nd1
∞ ∞
+ ∑ z a2 +nd2 + L + ∑ z ak +ndk . So far all generating functions were
n =1 n=1 n =1 n=1 in one variable. For the curious mind,
next we will look at an example
Summing the geometric series, this gives
z a1 z a2 z ak function
= + +L+ . ∞ ∞
1− z 1− z d1
1− z d2
1 − zdk f ( x, y ) = ∑∑ ai , j x i y j
i = 0 j =1
Letting z tend to e 2 π i / d 1 , we see the left
of the simplest kind.
.
to infinity. That gives a contradiction.
Example 7. An a×b rectangle can be
tiled by a number of p×1 and 1×q types
Example 6. (1995 IMO) Let p be an odd of rectangles, where a, b, p, q are fixed
prime number. Find the number of positive integers. Prove that a is
subsets A of the set {1,2,…,2p} such that divisible by p or b is divisible by q.
(i) A has exactly p elements, and (Here a k×1 and a 1×k rectangles are
considered to be different types.)
(ii) the sum of all the elements in A is
divisible by p. Solution. Inside the (i, j) cell of the a×b
rectangle, let us put the term xiyj for
Solution. Consider the polynomial i=1,2,…,a and j=1,2,…,b. The sum of
the terms inside a p×1 rectangle is
Fa(x) = (1+ax)(1+a2x)(1+a3x)⋯(1+a2px)
xiyj+⋯+ xi+p−1y j= (1 + x + ⋯ + xp−1) xiyj,
When the right side is expanded, let cn,k if the top cell is at (i, j), while the sum
count the number of terms of the form of the terms inside a 1×q rectangle is
(a i1 x)(a i2 x) L (a ik x) , where i1, i2, …, ik are
integers such that 1≤ i1< i2 <⋯< ik ≤ 2p and xiy j+⋯+ xiy j+q−1= xiyj (1 + y + ⋯ + y q−1),
i1+i2 +⋯+ik = n. Then
if the leftmost cell is at (i, j). Now take
⎛ ∞ 2p
⎞ x = e 2 π i / p and y = e 2 π i / q .
Fa ( x) = 1 + ∑ ⎜ ∑ cn ,k a n ⎟x k .
k =1 ⎝ n =1 ⎠ Then both sums become 0. If the desired
tiling is possible, then the total sum of
Now, in terms of cn,k, the answer to the all terms in the a×b rectangle would be
problem is C = c p , p + c2 p , p + c3 p , p + L.
a b
(1 − x a )(1 − y b )
0 = ∑∑ xi y j = xy .
To get C, note the coefficient of x in
p
i =1 j =1 (1 − x)(1 − y )
∞ This implies that a is divisible by p or b
Fa(x) is ∑c n, p a n . By (*) above, we see is divisible by q.
n =1
For the readers who like to know
1 p −1 ∞
C= ∑∑ cn, pω nj .
p j =0 n=1
more about generating functions, we
=0 recommend two excellent references:
T. Andreescu and Z. Feng, A Path to
Now the right side is the coefficient of xp
Combinatorics for Undergraduates,
p −1
Birkhäuser, Boston, 2004.
in 1 ∑ F j ( x), which equals
(*)
p j =0 ω M. Novaković, Generating Functions,
The IMO Compendium Group, 2007
(www.imomath.com)
Mathematical Excalibur, Vol. 13, No. 5, Jan.-Feb. 09 Page 3

Problem Corner satisfying f(a1) = f(a2) = ⋯ = f(a10). Nai, Vietnam) and PUN Ying Anna
(HKU Math, Year 3).
Solution 1. Kipp JOHNSON (Valley
We welcome readers to submit their Catholic School, teacher, Beaverton, Let x = a + 1, y = b + 1 and z = c + 1.
solutions to the problems posed below Oregon, USA) and PUN Ying Anna Applying the AM-GM inequality twice,
(HKU Math, Year 3).
for publication consideration. The we have
solutions should be preceded by the The number of 10-term arithmetic x4 y4 z4
solver’s name, home (or email) address + +
progressions in S is the same as the ( y − 1) 2
( z − 1) 2
( x − 1) 2
and school affiliation. Please send number of ordered pairs (a,d) such that a,
submissions to Dr. Kin Y. Li, d are in S and a+9d ≤ 2008. Since d ≤ (a + 1) 4 (b + 1) 4 (c + 1) 4
Department of Mathematics, The Hong 2007/9=223 and for each such d, a goes = + +
Kong University of Science & b2 c2 a2
from 1 to 2008−9d, so there are at most
Technology, Clear Water Bay, Kowloon, 223 ⎛ (a + 1) 4 (b + 1) 4 (c + 1) 4 ⎞
1/ 3

Hong Kong. The deadline for sending 4 ( 2008−10 )

solutions is March 7, 2009.
Problem 316. For every positive
integer n > 6, prove that in every
n-sided convex polygon A1A2…An,
there exist i ≠ j such that
1 G.R.A. 20 Problem Solving Group
| cos ∠Ai − cos ∠A j |< . more (namely 42008) functions from S to
2(n − 6)
Problem 317. Find all polynomial P(x)
with integer coefficients such that for
every positive integer n, 2n−1 is
divisible by P(n).
Problem 318. In ΔABC, side BC has
length equal to the average of the two
other sides. Draw a circle passing
through A and the midpoints of AB, AC.
Draw the tangent lines from the
centroid of the triangle to the circle.
Prove that one of the points of
tangency is the incenter of ΔABC.
(Source: 2000 Chinese Team Training
Test)
Problem 319. For a positive integer n,
let S be the set of all integers m such
that |m| < 2n. Prove that whenever
2n+1 elements are chosen from S, there
exist three of them whose sum is 0.
(Source: 1990 Chinese Team Training
Test)
Problem 320. For every positive
integer k > 1, prove that there exists a
positive integer m such that among the
rightmost k digits of 2m in base 10, at
least half of them are 9’s.
(Source: 2005 Chinese Team Training
Test)
*****************
Solutions
****************
Problem 311. Let S = {1,2,…,2008}.
Prove that there exists a function
f : S → {red, white, blue, green} such
that there does not exist a 10-term
arithmetic progression a1,a2,…,a10 in S
× 4 × ∑ ( 2008 − 9d )
d =1
= 41999×223000
functions f :S→{red, white, blue, green}
such that there exists a 10-term arithmetic
progression a1,a2,…,a10 in S satisfying
f(a1) = f(a2) = ⋯ = f(a10), while there are
{red, white, blue, green}. So the desired
function exists.
Solution 2. G.R.A. 20 Problem Solving
Group (Roma, Italy).
Replace red, white, blue, green by 0, 1, 2,
3 respectively. It can be seen by a direct
checking that f:{1,2,…,2048}→ {0,1,2,3}
given by
⎡ n − 1⎤ ⎡ n − 1⎤
f ( n) = ⎢ ⎥ + 2⎢
⎣ 8 ⎦ mod 2 ⎣ 128 ⎥⎦ mod 2
avoids any 9-term arithmetic progression
having the same value (where kmod 2 is 0 if
k is even and 1 if k is odd). The range of f
is ((0818)8(2838)8)8, where for any string x,
x8 denotes the string obtained by putting
eight copies of the string x one after
another in a row and f(n) is the n-th digit in
the specified string.
Commended solvers: LKL Problem
Solving Group (Madam Lau Kam Lung
Secondary School of Miu Fat Buddhist
Monastery).
Problem 312. Let x, y, z > 1. Prove that
x4 y4 z4
+ + ≥ 48.
( y − 1) 2 ( z − 1) 2 ( x − 1) 2
Solution. Glenier L. BELLO-BURGUET
(I.E.S. Hermanos D`Elhuyar, Spain), Kipp
JOHNSON (Valley Catholic School,
teacher, Beaverton, Oregon, USA),
Kelvin LEE (Trinity College, University
of Cambridge, Year 2), LEUNG Kai
Chung (HKUST Math, Year 2), LKL
Problem Solving Group (Madam Lau
Kam Lung Secondary School of Miu Fat
Buddhist Monastery), MA Ka Hei (Wah
Yan College, Kowloon), NGUYEN Van
Thien (Luong The Vinh High School, Dong
≥ 3⎜⎜ ⎟⎟
⎝ a 2b 2 c 2 ⎠
1/ 3
⎛ (2 a ) 4 (2 b ) 4 (2 c ) 4 ⎞
≥ 3⎜⎜ ⎟
⎟ = 48.
⎝ a 2b 2 c 2 ⎠
Commended solvers: CHUNG Ping
Ngai (La Salle College, Form 5),
(Roma, Italy), NG Ngai Fung (STFA
Leung Kau Kui College, Form 6),
Paolo PERFETTI (Dipartimento di
Matematica, Università degli studi di
Tor Vergata Roma, via della ricerca
scientifica, Roma, Italy), Dimitar
TRENEVSKI (Yahya Kemal College,
Skopje, Macedonia) and TSOI Kwok
Wing (PLK Centenary Li Shiu Chung
Memorial College, Form 6).
Problem 313. In Δ ABC, AB < AC
and O is its circumcenter. Let the
tangent at A to the circumcircle cut line
BC at D. Let the perpendicular lines to
line BC at B and C cut the
perpendicular bisectors of sides AB and
AC at E and F respectively. Prove that
D, E, F are collinear.
F
A
O
M E
N
C
B
D
Solution. Glenier L. BELLO-
BURGUET (I.E.S. Hermanos
D`Elhuyar, Spain), CHUNG Ping Ngai
(La Salle College, Form 5), Kelvin
LEE (Trinity College, University of
Cambridge, Year 2), NG Ngai Fung
(STFA Leung Kau Kui College, Form 6)
and PUN Ying Anna (HKU Math,
Year 3).
Let M be the midpoint of AB and N be
the midpoint of AC. Using ∠ ABE
= ∠ ABC − 90°, ∠ FCA = 90° − ∠ ABC
and the sine law, we have
Mathematical Excalibur, Vol. 13, No. 5, Jan.-Feb. 09 Page 4

to y2−3 = 2(2x+y) = 2(2m2+3y), which can and n−b is 11.5 or 12, hence 12 is
BE BM / cos ∠ABE
= be rearranged as (y−3)2−4m2=12. This between b and n−b inclusive.
CF CN / cos ∠FCA leads to y = 7 and m = 1. Then x = 8 and z
= 13. Since 83−73=132, this gives the only Finally, observe that after each of the
1 AB / sin ∠ABC AB 2
= 2
1 AC / sin ∠BCA
= . solution. twenty-four 90° rotations, one exposed
2 AC 2 square will be hidden and one hidden
Commended solvers: LKL Problem
From ΔDCA~ΔDAB, we see square will be exposed. So the number
Solving Group (Madam Lau Kam Lung
of black squares on the surface of C can
DA DB sin∠DAB sin∠ACB AB Secondary School of Miu Fat Buddhist
= = = = . only increase by one, stay the same or
DC DA sin∠DBA sin∠ABC AC Monastery).
decrease by one.
So Problem 315. Each face of 8 unit cubes is
painted white or black. Let n be the total Therefore, at a certain moment, there
BE AB2 DA DB DB number of black faces. Determine the will be exactly 12 black squares (and
= = ⋅ = .
CF AC 2 DC DA DC values of n such that in every way of 12 white squares) on the surface of C.
coloring n faces of the 8 unit cubes black, Commended solvers: G.R.A. 20
Then ∠ BDE = ∠ CDF . Therefore
there always exists a way of stacking the 8 Problem Solving Group (Roma, Italy)
D,E,F are collinear. unit cubes into a 2×2×2 cube C so the and LKL Problem Solving Group
numbers of black squares and white (Madam Lau Kam Lung Secondary
Commended solvers: Stefan squares on the surface of C are the same.
LOZANOVSKI and Bojan School of Miu Fat Buddhist
JOVESKI (Private Yahya Kemal Monastery).
Solution. CHUNG Ping Ngai (La Salle
College, Skopje, Macedonia). College, Form 5) and PUN Ying Anna
(HKU Math, Year 3).
Problem 314. Determine all positive
integers x, y, z satisfying x3 − y3 = z2, The answer is n = 23 or 24 or 25. First
where y is a prime, z is not divisible by notice that if n is a possible value, then so Olympiad Corner
3 and z is not divisible by y. is 48−n. This is because we can (continued from page 1)

Solution. CHUNG Ping Ngai (La
Salle College, Form 5) and PUN Ying
Anna (HKU Math, Year 3).
Suppose there is such a solution. Then
z2 = x3 − y3 =(x−y)(x2+xy+y2)
= (x−y) ((x−y)2+3xy). (*)
Since y is a prime, z is not divisible by
3 and z is not divisible by y, (*) implies
(x,y)=1 and (x−y,3)=1. Then
(x2+xy+y2, x−y)=(3xy, x−y)=1. (**)
Now (*) and (**) imply
x−y=m2, x2+xy+y2=n2 and z=mn
for some positive integers m and n.
Consequently,
4n2= 4x2+4xy+4y2=(2x+y)2+3y2.
Then 3y2=(2n+2x+y)(2n−2x−y). Since
y is prime, there are 3 possibilities:
(1) 2n+2x+y = 3y2, 2n−2x−y = 1
(2) 2n+2x+y = 3y, 2n−2x−y = y
(3) 2n+2x+y = y2, 2n−2x−y = 3.
In (1), subtracting the equations leads
to 3y2−1 = 2(2x+y) = 2(2m2+3y). Then
m2 + 1 = 3y2 − 6y − 3m2 ≡ 0 (mod 3).
However, m2 + 1 ≡ 1 or 2 (mod 3). We
get a contradiction.
In (2), subtracting the equations leads
to x = 0, contradiction.
In (3), subtracting the equations leads
interchange all the black and white
coloring and the condition can still be met
by symmetry. Hence, without loss of Second Day: June 7th, 2008
generality, we may assume n ≤ 24.
Problem 4. Determine all functions f
For the 8 unit cubes, there are totally 24 mapping the set of positive integers to
pairs of opposite faces. In each pair, no the set of non-negative integers
matter how the cubes are stacked, there is satisfying the following conditions:
one face on the surface of C and one face (1) f(mn) = f(m)+f(n),
hidden. (2) f(2008) = 0, and
(3) f(n) = 0 for all n ≡ 39 (mod 2008).
If n ≤ 22, there is a coloring that has [n/2]
pairs with both opposite faces black. Problem 5. Which positive integers
Then at least [n/2] black faces will be are missing in the sequence {an}, with
hidden so that there can be at most n−[n/2]
≤ 11 black faces on the surface of C. This an = n + [ n ] + [ 3 n ]
contradicts the existence of a stacking for all n ≥ 1? ([x] denotes the largest
with 12 black and 12 white squares on the integer less than or equal to x, i.e. g
surface of C. So only n = 23 or 24 is with g ≤ x < g+1.)
possible.
Problem 6. We are given a square
Next, start with an arbitrary stacking. Let
ABCD. Let P be a point not equal to a
b be the number of black squares on the
corner of the square or to its center M.
surface of C. For each of the 8 unit cubes,
For any such P, we let E denote the
take an axis formed by the centers of a pair
common point of the lines PD and AC,
of opposite faces and rotate the cube about
if such a point exists. Furthermore, we
that axis by 90°. Then take an axis formed
let F denote the common point of the
by the centers of another pair of opposite
lines PC and BD, if such a point exists.
faces of the same cube and rotate the cube
about the axis by 90° twice. These three
All such points P, for which E and F
90° rotations switch the three exposed
exist are called acceptable points.
faces with the three hidden faces. So after
Determine the set of all acceptable
doing the twenty-four 90° rotations for the
points, for which the line EF is parallel
8 unit cubes, there will be n−b black
to AD.
squares on the surface of C.
For n = 23 or 24 and b ≤ n, the average of b