Difference between revisions of "1992 IMO Problems/Problem 1"
(→Problem) |
Not detriti (talk | contribs) m (wrong number) |
||
(9 intermediate revisions by 2 users not shown) | |||
Line 2: | Line 2: | ||
Find all integers <math>a</math>, <math>b</math>, <math>c</math> satisfying <math>1 < a < b < c</math> such that <math>(a - 1)(b -1)(c - 1)</math> is a divisor of <math>abc - 1</math>. | Find all integers <math>a</math>, <math>b</math>, <math>c</math> satisfying <math>1 < a < b < c</math> such that <math>(a - 1)(b -1)(c - 1)</math> is a divisor of <math>abc - 1</math>. | ||
== Solution == | == Solution == | ||
− | {{solution}} | + | |
+ | <math>1<\frac{abc-1}{(a-1)(b-1)(c-1)}<\frac{abc}{(a-1)(b-1)(c-1)}</math> | ||
+ | |||
+ | <math>1<\frac{abc-1}{(a-1)(b-1)(c-1)}<\left(\frac{a}{a-1}\right) \left(\frac{b}{b-1}\right) \left(\frac{c}{c-1}\right)</math> | ||
+ | |||
+ | With <math>1<a<b<c</math> it implies that <math>a \ge 2</math>, <math>b \ge 3</math>, <math>c \ge 4</math> | ||
+ | |||
+ | Therefore, <math>\frac{a}{a-1}=1+\frac{1}{a-1}</math> | ||
+ | |||
+ | which for <math>a</math> gives: <math>\frac{a}{a-1} \le 1+\frac{1}{2-1}</math>, which gives :<math>\frac{a}{a-1} \le 2</math> | ||
+ | |||
+ | for <math>b</math> gives: <math>\frac{b}{b-1} \le 1+\frac{1}{3-1}</math>, which gives :<math>\frac{b}{b-1} \le \frac{3}{2}</math> | ||
+ | |||
+ | for <math>c</math> gives: <math>\frac{c}{c-1} \le 1+\frac{1}{4-1}</math>, which gives :<math>\frac{c}{c-1} \le \frac{4}{3}</math> | ||
+ | |||
+ | Substituting those inequalities into the original inequality gives: | ||
+ | |||
+ | <math>1<\frac{abc-1}{(a-1)(b-1)(c-1)}<\left( 2 \right) \left(\frac{3}{2}\right) \left(\frac{4}{3}\right)</math> | ||
+ | |||
+ | <math>1<\frac{abc-1}{(a-1)(b-1)(c-1)}<4</math> | ||
+ | |||
+ | Since <math>\frac{abc-1}{(a-1)(b-1)(c-1)}</math> needs to be integer, | ||
+ | |||
+ | then <math>\frac{abc-1}{(a-1)(b-1)(c-1)}=2</math> or <math>\frac{abc-1}{(a-1)(b-1)(c-1)}=3</math> | ||
+ | |||
+ | Case 1: <math>\frac{abc-1}{(a-1)(b-1)(c-1)}=2</math> | ||
+ | |||
+ | <math>abc-1=2(a-1)(b-1)(c-1)=2abc-2(ab+bc+ac)+2(a+b+c)-2</math> | ||
+ | |||
+ | Case 1, subcase <math>a=2</math>: | ||
+ | |||
+ | <math>2bc-1=2bc-2(b+c)+2</math> gives: <math>2(b+c)=3</math> which has no solution because <math>2(b+c)</math> is even. | ||
+ | |||
+ | Case 1, subcase <math>a=3</math>: | ||
+ | |||
+ | <math>3bc-1=4bc-4(b+c)+4</math> | ||
+ | |||
+ | <math>bc-4b-4c+5=0</math> | ||
+ | |||
+ | <math>(b-4)(c-4)=11</math> | ||
+ | |||
+ | <math>b-4=1</math> and <math>c-4=11</math> provides solution <math>(a,b,c)=(3,5,15)</math> | ||
+ | |||
+ | Case 2: <math>\frac{abc-1}{(a-1)(b-1)(c-1)}=3</math> | ||
+ | |||
+ | <math>abc-1=3(a-1)(b-1)(c-1)=3abc-3(ab+bc+ac)+3(a+b+c)-3</math> | ||
+ | |||
+ | Case 2, subcase <math>a=2</math>: | ||
+ | |||
+ | <math>2bc-1=3bc-3(b+c)+3</math> | ||
+ | |||
+ | <math>bc-3b-3c+4=0</math> | ||
+ | |||
+ | <math>(b-3)(c-3)=5</math> | ||
+ | |||
+ | <math>b-3=1</math> and <math>c-3=5</math> provides solution <math>(a,b,c)=(2,4,8)</math> | ||
+ | |||
+ | Case 2, subcase <math>a=3</math>: | ||
+ | |||
+ | <math>3bc-1=6bc-6(b+c)+6</math> | ||
+ | |||
+ | Since <math>(3bc-1</math>) mod <math>3 = -1</math> and <math>(6bc-6(b+c)+6)</math> mod <math>3 = 0</math>, then there is no solution for this subcase. | ||
+ | |||
+ | Now we verify our two solutions: | ||
+ | |||
+ | when <math>(a,b,c)=(2,4,8)</math> | ||
+ | |||
+ | <math>abc-1=(2)(4)(8)-1=63</math> and <math>(a-1)(b-1)(c-1)=(1)(3)(7)=21</math> | ||
+ | |||
+ | Since <math>21</math> is a factor of <math>63</math>, this solutions is correct. | ||
+ | |||
+ | when <math>(a,b,c)=(3,5,15)</math> | ||
+ | |||
+ | <math>abc-1=(3)(5)(15)-1=224</math> and <math>(a-1)(b-1)(c-1)=(2)(4)(14)=112</math> | ||
+ | |||
+ | Since <math>112</math> is a factor of <math>224</math>, this solutions is also correct. | ||
+ | |||
+ | The solutions are: <math>(a,b,c)=(2,4,8)</math> and <math>(a,b,c)=(3,5,15)</math> | ||
+ | |||
+ | ~ Tomas Diaz. orders@tomasdiaz.com | ||
+ | |||
+ | ==Solution 2== | ||
+ | Let <math>x = a-1, y = b-1, z = c-1</math>. So <math>1 \leq x < y < z</math>. We're asked to solve | ||
+ | <cmath> | ||
+ | (k+1)xyz = (x+1)(y+z)(z+1)-1 | ||
+ | </cmath> | ||
+ | where <math>k</math> is a non-negative integer.. Dividing by <math>xyz</math> | ||
+ | <cmath> | ||
+ | \begin{align*} | ||
+ | k+1 &= \Big(1+\frac{1}{x}\Big)\Big(1+\frac{1}{y}\Big)\Big(1+\frac{1}{z}\Big) - \frac{1}{xyz} \\ | ||
+ | &< (1+1)\Big(1+\frac{1}{2}\Big)\Big(1+\frac{1}{3}\Big) = 4 | ||
+ | \end{align*} | ||
+ | </cmath> | ||
+ | So <math>k=0,1</math> or <math>2</math>. Simplifying the first equation | ||
+ | <cmath> | ||
+ | kxyz = xy + yz + zy + x + y + z. | ||
+ | </cmath> | ||
+ | <math>k=0</math> is impossible. Case <math>k=2</math>: if <math>x \geq 2</math>, dividing the previous equation by <math>xy</math> | ||
+ | <cmath> | ||
+ | \begin{align} | ||
+ | 2z &= 1 + \frac{1}{x} + \frac{1}{y} + z\Big(\frac{1}{x} + \frac{1}{y} + \frac{1}{xy}\Big) \\ | ||
+ | &\leq 1 + \frac{1}{2}+\frac{1}{3} + z\Big(\frac{1}{2}+\frac{1}{3}+\frac{1}{6}\Big) \\ | ||
+ | &< 2 + z \nonumber | ||
+ | \end{align} | ||
+ | </cmath> | ||
+ | So <math>z < 2</math> which is impossible. If <math>x = 1</math> its easy to show <math>y=2</math> leads to a contradiction. Solving <math>(1)</math> for <math>z</math> | ||
+ | <cmath> | ||
+ | z = 2 + \frac{5}{y-2} | ||
+ | </cmath> | ||
+ | which can only work if <math>y = 3, z = 7</math>. So <math>x=1,y=3,z=7</math> is a solution. Case <math>k=1</math>: considering the version of <math>(1)</math> with <math>z</math> on the LHS instead of <math>2z</math>, if <math>x = 1</math> then | ||
+ | <cmath> | ||
+ | \begin{align*} | ||
+ | z &= 1 + \frac{1}{x} + \frac{1}{y} + z \Big(\frac{1}{x} + \frac{1}{y} + \frac{1}{xy}\Big) \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ (3) \\ | ||
+ | &= 2 + \frac{1}{y} + z\Big(1+\frac{1}{y}+\frac{1}{xy}\Big) | ||
+ | \end{align*} | ||
+ | </cmath> | ||
+ | which is impossible. Similarly to <math>(2)</math>, if <math>x \geq 3</math> then | ||
+ | <cmath> | ||
+ | z \leq 1 + \frac{1}{3} + \frac{1}{4} + z \frac{8}{12} | ||
+ | </cmath> | ||
+ | i.e. <math>z < 5</math> which is impossible since <math>3 \leq x < y < z</math>. So <math>x= 2</math>. Its easy to show <math>y=3</math> leads to a contradiction. Solving <math>(3)</math> for <math>z</math> with <math>x=2</math> gives | ||
+ | <cmath> | ||
+ | z = 3 + \frac{11}{y-3} | ||
+ | </cmath> | ||
+ | which can only work if <math>y=4, z=14</math>. So <math>x=2,y=4,z=14</math> is a solution. Our two solutions give <math>a = 2,b=4,c=8</math> and <math>a=3,b=5,c=15</math>. | ||
+ | |||
+ | ~not_detriti | ||
+ | |||
+ | {{alternate solutions}} | ||
+ | |||
+ | ==See Also== | ||
+ | |||
+ | {{IMO box|year=1992|before=First Question|num-a=2}} | ||
+ | [[Category:Olympiad Geometry Problems]] | ||
+ | [[Category:3D Geometry Problems]] |
Latest revision as of 09:45, 11 July 2025
Contents
Problem
Find all integers ,
,
satisfying
such that
is a divisor of
.
Solution
With it implies that
,
,
Therefore,
which for gives:
, which gives :
for gives:
, which gives :
for gives:
, which gives :
Substituting those inequalities into the original inequality gives:
Since needs to be integer,
then or
Case 1:
Case 1, subcase :
gives:
which has no solution because
is even.
Case 1, subcase :
and
provides solution
Case 2:
Case 2, subcase :
and
provides solution
Case 2, subcase :
Since ) mod
and
mod
, then there is no solution for this subcase.
Now we verify our two solutions:
when
and
Since is a factor of
, this solutions is correct.
when
and
Since is a factor of
, this solutions is also correct.
The solutions are: and
~ Tomas Diaz. orders@tomasdiaz.com
Solution 2
Let . So
. We're asked to solve
where
is a non-negative integer.. Dividing by
So
or
. Simplifying the first equation
is impossible. Case
: if
, dividing the previous equation by
So
which is impossible. If
its easy to show
leads to a contradiction. Solving
for
which can only work if
. So
is a solution. Case
: considering the version of
with
on the LHS instead of
, if
then
which is impossible. Similarly to
, if
then
i.e.
which is impossible since
. So
. Its easy to show
leads to a contradiction. Solving
for
with
gives
which can only work if
. So
is a solution. Our two solutions give
and
.
~not_detriti
Alternate solutions are always welcome. If you have a different, elegant solution to this problem, please add it to this page.
See Also
1992 IMO (Problems) • Resources | ||
Preceded by First Question |
1 • 2 • 3 • 4 • 5 • 6 | Followed by Problem 2 |
All IMO Problems and Solutions |