From vertex C of a unit equilateral triangle A B C , we draw a line segment to the midpoint m 1 of the opposite side. From vertex m 1 of triangle B C m 1 , we similarly draw a line segment to the midpoint m 2 of the opposite side. Then we draw line segment m 2 m 3 in triangle C m 1 m 2 , and so on. We travel indefinitely in this manner, getting closer and closer to a destination point we call P .
If A P + B P + C P can be expressed as b a ( e c + f d ) where a , b , c , d , e , f are positive integers, g cd ( a , b ) = g cd ( e , f ) = 1 , and c and d are distinct and square-free, input a + b + c + d + e + f as your answer.
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.
Wow, I use barycenter to solve this, quite same.
Log in to reply
Interesting! I used recurrence relation to solve this problem. I would like to see your solution. :)
Log in to reply
Okay, I will post it later! Would you like to also post your solution?
Since
m
1
and
m
2
are midpoints of
A
B
and
B
C
,
m
1
m
2
is parallel to
A
C
Similarly, m 2 m 3 is parallel to A B
m 3 m 4 is parallel to B C
m 5 m 6 or m 6 m 8 is parallel to A C
m 6 m 7 is parallel to A B
m 7 m 8 is parallel to B C
So Δ m 6 m 7 m 8 is similar and has the same orientation as Δ A B C
P in Δ m 6 m 7 m 8 is still the same so m 6 P of Δ m 6 m 7 m 8 corresponds to A P of Δ A B C
Therefore, A , m 6 , and P must be on one line
Similarly, we can get B , m 7 , and P are collinear and C , m 8 , and P are collinear
Now using coordinates with A ( 2 1 , 2 3 ) , B ( 0 , 0 ) , and C ( 1 , 0 ) ,
We get m 6 ( 2 1 , 8 3 ) , m 7 ( 3 2 1 5 , 3 2 3 3 ) , and m 8 ( 3 2 1 7 , 3 2 3 3 )
Using these we can compute the lines
A m 6 is x = 2 1
B m 7 is y = 5 3 x
and C m 8 is y = − 5 3 ( x − 1 )
They all intersect at P ( 2 1 , 1 0 3 )
So A P = 2 3 − 1 0 3 = 5 2 3
And B P = C P = ( 2 1 ) 2 + ( 1 0 3 ) 2 = 2 5 7 = 5 7 ( B P = C P since P is on the perpendicular bisector of B C which is x = 2 1 )
Therefore, A P + B P + C P = 5 2 3 + 5 2 7 = 5 2 ( 3 + 7 )
And a + b + c + d + e + f = 2 + 5 + 1 + 3 + 1 + 7 = 1 9
nice soln @Andres Rico M. III Gonzales
Relevant Wiki: Barycentric Coordinates
Using barycentric coordinates, found that
m 1 ( 0 , 0 , 1 ) , m 2 ( 2 1 , 2 1 , 0 ) , m 3 ( 0 , 2 1 , 2 1 ) .
So generalize to
m n = 2 1 ( m n − 2 + m n − 3 )
Using this, I want to do cancellation,
2 m 4 = m 2 + m 1
2 m 5 = m 3 + m 2
2 m 6 = m 4 + m 3
2 m 7 = m 5 + m 4
2 m 8 = m 6 + m 5
. . .
2 m n − 3 = m n − 5 + m n − 6
2 m n − 2 = m n − 4 + m n − 5
2 m n − 1 = m n − 3 + m n − 4
2 m n = m n − 2 + m n − 3
Summing all of this, and do some cancellation, we get
2 m n + 2 m n − 1 + m n − 2 = m 1 + 2 m 2 + 2 m 3
Here is the main point: When n approaches infinity, it includes m n = m n − 1 = m n − 2
So
5 m n = ( 5 1 , 5 2 , 5 2 )
Return to normal xy-plane, let three verticle be ( 0 , 0 ) , ( 1 , 0 ) , ( 2 1 , 2 3 )
According to the barycenter coordinate of point P, we know that P is below upper verticle 5 4 of the height of triangle.
So the ordinary coordinate of P is ( 2 1 , 1 0 3 ) .
Using simple formula we found that
A P + B P + C P = 5 2 ( 7 + 3 )
So a + b + c + d + e + f = 1 9 .
Interesting solution. :)
Problem Loading...
Note Loading...
Set Loading...
We have the recurrence relation m n = 2 1 ( m n − 2 + m n − 3 ) with the initial conditions m 0 = c , m 1 = 2 1 ( a + b ) and m 2 = 2 1 ( b + c ) , where a , b , c are the position vectors of A , B , C respectively. This recurrence relation leads to the indicial equation 0 = 2 λ 3 − λ − 1 = ( λ − 1 ) ( 2 λ 2 + 2 λ + 1 ) so the possible values of λ are − 1 , − 2 1 ( 1 + i ) , − 2 1 ( 1 − i ) , and so m n = p + ( − 2 1 ( 1 + i ) ) n q + ( − 2 1 ( 1 − i ) ) n r for some vectors p , q , r . Of course, q and r will be complex conjugates, so that m n will turn out real.
Putting n = 0 , 1 , 2 gives us three equations for p , q , r . p + q + r p − 2 1 ( 1 + i ) q − 2 1 ( 1 − i ) r p + 2 1 i q − 2 1 i r = c = 2 1 ( a + b ) = 2 1 ( b + c ) We only need to determine p = n → ∞ lim m n since that will be the position vector of the desired point P . Elementary algebra gives us p = 5 1 a + 5 2 b + 5 2 c .
Applying a coordinate system we have a = ( 2 1 3 2 1 ) , b = ( 0 0 ) , c = ( 0 1 ) and hence p = ( 1 0 1 3 2 1 ) Thus A P = 2 1 3 − 1 0 1 3 = 5 2 3 , while B P = C P = 4 1 + 1 0 0 3 = 5 1 7 , and hence A P + B P + C P = 5 2 ( 3 + 7 ) making the answer 2 + 5 + 1 + 3 + 1 + 7 = 1 9 .