Triangle A B C is drawn with ∠ B = 9 0 ∘ and A C = 1 .
An angle bisector is drawn from A hitting B C at D . An altitude is drawn from B hitting A C at E . Finally, a median is drawn from C hitting A B at F .
Given that A D , B E , and C F are concurrent, then the area of triangle D E F can be represented by a + b c 1 where a , b , c are integers with c square-free. Find a + b + c .
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.
I think you may have gained too much information from Ceva's theorem that the solution does not arise from fundamental facts. The actual fact I find is equating areas of the wanted triangle from two different equations that shape the whole thing. It is interesting to find that AF or FB is not easy to be determined without going through equating areas, I think. I tried with many triangles with all the rules and failed surprisingly. Correct me if I am wrong.
Correct! Derivations is achieved.
Solution outline:
Find that A B ∥ D E by Ceva's Theorem. Now we can effectively ignore C since we are guaranteed to have C F be concurrent to A D and B E as long as C F passes through the midpoint of D E .
Note that A E = D E because ∠ A D E = ∠ B A D = ∠ D A E .
Let ∠ B A D = θ and B D = x . We represent every line segment in terms of x and θ , and find that we get the equation 1 + sin 2 2 θ 2 = tan 2 θ 1 which resolves into cos θ = 2 5 + 1 after some algebra.
Thus, cos 2 θ = 2 5 − 1 and so A B = 2 5 − 1 . Compute B C = 2 5 − 1 .
We compute that D E = 2 3 − 5 by some similar triangle bashing. Draw a altitude from F to D E with foot H and compute that F H = 2 5 5 − 1 1 with some more similar triangle bashing.
Thus, [ D E F ] = 2 1 ⋅ ( 2 3 − 5 ) ( 2 5 5 − 1 1 ) = 1 5 2 + 6 8 5 1 and our answer is 1 5 2 + 6 8 + 5 = 2 2 5 .
I would love to see an elegant solution, since my solution is the first thing I thought of and is the straightforward, expected solution from bashing everything at first glance.
Too bad I couldn't solve the problem. I realized that DE must // AB but couldn't think of Ceva's theorem. Silly me! But here's another solution based on the fact that DE // AB
Thought: S Δ D E F = S Δ B E D = 2 1 B E ⋅ D E ⋅ sin α
Remarks: I'm not sure if this's elegant enough, still lots of calculation. I notice that C E = sin α . There may be some nice relations out there.
Do you know how to make public a solution that I marked as private? Because I did that in this problem and I can't submit another answer or edit that one. Thanks.
Log in to reply
No, I don't know; I think it is impossible. However, you can ask a staff member to do it for you.
@Calvin Lin , can you help me?
Log in to reply
@Calvin Lin @Suyeon Khim @Peter Taylor someone oughta know how to get a solution public...
Unfortunately, there isn't an easy way for me to get at a solution which you marked as private. I'd try to hunt for it.
I used ceva's theorem as well as coordinate geometry to find coordinates of DEF and then found the area by area formula in coordinates. Totally inelegant though!
I notice that your cos θ is... greater than 1. My guess is that the concurrency of those 3 kinds of lines is impossible in this (right) triangle? Any thought?
Btw, the altitude of Δ D E F is essentially BD, you can calculate tan θ and compute BD, which yields exactly the same as your FH.
Log in to reply
My cos θ does exist... 2 5 + 1 < 1 .
Also, yes, that is how I calculated FH: by finding BD.
Log in to reply
Yeah, I guess the sqrt notation in your solution was kinda misplaced
I think that he made a little mistake, it should be cos θ = 2 5 + 1 . So the concurrence is possible as I showed in my solution that I hope that appears soon.
Log in to reply
I have proven that it is indeed concurrent at (0.185585165755868+, 0.236067977499790+) with origin at B.
I thought the same as you. 3 kinds of lines! Never have met! However, I finally found that the concurrency is indeed true!
Correct! Derivation is achieved.
The ratio of AE/EC is in fact the golden ratio. The length of AB has a value of golden ratio as well. You can find the lengths of all segments resulting from intersection of cevians with original triangle sides quite simply. Then you get the area of the triangle DEF = Area ( ABC) - Area (AFE) - Area (FBD) - Area ( EDC). The result has a value of 2 − 2 φ φ 3 + φ φ
Correct! The golden ratio is true. I like the expression of area in term of golden ratio.
My approach is not the same. Although it is not as preferred, I find that it is special for a numerical analysis.
The critical fact is the area wanted can be found by two different expressions that coincide for fixed dimensions. One is obtainable from the whole triangle minus 3 triangles at each corner, while another one is obtainable from knowing 3 sides, all with one unknown and looking for a known.
Let AF = FB = h. To save from complicated workings, let me just elaborate the results.
Δ = h 2 1 − 4 h 2 ( 1 − 1 + 2 h 4 h 2 ) ,
Sides of it are h , h 1 + 2 h 5 − 6 h and 2 h ( 1 − 4 h 2 ) ; Δ = s ( s − a ) ( s − b ) ( s − c ) . {Correctness to all terms are guaranteed for comfortable readings.}
Knowing a + b c 1 , analysis is done for (a, b, c) of:
(152, 68, 5), (285, 11, 3), (1055, -531, 2), (513, -63, 11), (342, -12, 10), (2672, -895, 7) and (3677, -1377, 6).
Actual Excel's proximity:
h = 0.309016994374948
a = 0.309016994374948
b = 0.430884493294898
c = 0.381966011250105
s = 0.560933749459975
Area1 = 0.0573489701224549
Area2 = 0.0573489701224549
Only 1 5 2 + 6 8 5 1 with parameters above shows that "All Three Concurrency" is true!
Area = 0.05734897012245476879526277200116+
It is interesting to meet with this question because all the 7 listed (a, b, c)'s can satisfy Area1 = Area2. Concurrency becomes the only fact to determine the sole answer such that (152, 68, 5) as all others cannot be concurrent.
Δ = h 2 1 − 4 h 2 ( 1 − 1 + 2 h 4 h 2 ) = s ( s − a ) ( s − b ) ( s − c ) is not easy to solve if not by iteration from substitutions, while other factors to determine h were not found. f (h) formed seems to be a very high ordered function. With origin at B:
F(0, 0.309016994374948-), C(0.786151377757422+, 0)
A(0, 0.618033988749896-), D(0.300283106000778-, 0)
B(0, 0), E(0.300283106000778-, 0.381966011250105+)
Lengths and angles are all determined as a result. We can draw an exact triangle to see!
152 + 68 + 5 = 2 2 5
Problem Loading...
Note Loading...
Set Loading...
By Ceva's theorem , A B ∣ ∣ D E . Δ D E F = Δ B D E and ∠ D A E = ∠ B A D = ∠ A D E -we get A E = D E
again Ceva gives, C E A E . B D C D = 1 .......... (1)
C E A E = C D B D = A C A B
∴ A C C E = A B A E = s i n C = A C A B ⇒ C E = A B
from (1) , C E A E = C D B D = A C A B ∴ C E A E = C D B D = A C A B
∴ C E A C = A C A B + A C ⇒ A C 2 = C E ( A B + A C ) ⇒ A B 2 + B C 2 = A B ( A B + A C ) ⇒ B C 2 = A B . A C = A B
∴ A B 2 + A B = 1 2 ⇒ A B = 2 5 − 1 , B C = 2 5 − 1
, ∴ D E = A E = A C − C E = 1 − A B = 1 − 2 5 − 1 = 4 3 − 5
By chasing angle bisecting theorem , B D = A B + A C A B . B C
Finally , Δ D E F = Δ B D E = 2 1 . B D . D E
= 2 1 . A B + A C A B . B C . 4 3 − 5
= 2 1 . 2 5 − 1 + 1 2 5 − 1 . 2 5 − 1 . 4 3 − 5
and it gets 6 8 5 + 1 5 2 1 , a + b + c = 2 2 5