Community

Looking for a challenging algebra text? Preparing for MATHCOUNTS or the AMC exams?
Check out Art of Problem Solving's Introduction to Algebra by Richard Rusczyk.
Login Register Memberlist Search AoPS Blogs Contests Galleries Forum Index
The time now is Fri Dec 04, 2009 2:39 am
All times are UTC - 8
View posts since last visit
View unanswered posts
perpendicular feet
Moderators: High School Olympiad Moderators, darij grinberg, N.T.TUAN, orl, pbornsztein, pohoatza, yetti
Post new topic   Reply to topic View previous topicView next topic
8 Posts • Page 1 of 1
Author Message
barasawala
Poincare Conjecture
Poincare Conjecture

Offline
Joined: 14 Dec 2006
Posts: 125

To rate posts you must be logged in
#1
perpendicular feet

Let AA_{1},BB_{1},CC_{1} be the altitudes in acute triangle ABC, and let X be an arbitrary point. Let M,N,P,Q,R,S be the feet of the perpendiculars from X to the lines AA_{1},BC,BB_{1},CA,CC_{1},AB. Prove that MN,PQ,RS are concurrent.

PostPosted: Thu Dec 28, 2006 11:51 pm  Back to top 
  ProfilePM
darij grinberg
Birch & Swinnerton Dyer
Birch & Swinnerton Dyer


Offline
Joined: 10 Feb 2004
Posts: 5778
Location: Karlsruhe / Munich

To rate posts you must be logged in
#2
Re: perpendicular feet

barasawala wrote:
Let AA_{1},BB_{1},CC_{1} be the altitudes in acute triangle ABC, and let X be an arbitrary point. Let M,N,P,Q,R,S be the feet of the perpendiculars from X to the lines AA_{1},BC,BB_{1},CA,CC_{1},AB. Prove that MN,PQ,RS are concurrent.


This is Theorem 2 from the note "On the paracevian perspector" on my website (which currently lives at http://www.cip.ifi.lmu.de/~grinberg/ or, archived, at http://sites.google.com/site/darijgrinberg/website ). In that note, I also prove a generalization of your problem (see also http://www.artofproblemsolving.com/Forum/viewtopic.php?t=14692 for a discussion of that generalization).

I will leave to you the fun of finding the short solution of the above problem which does not apply to the generalization. Wink

Darij
_________________
Now the die is cast, the first step taken, a glimmer of hope lights up our lives
Visions of the past, dreams forsaken forming right under our eyes
We are alive...
Last edited by darij grinberg on Sun Sep 13, 2009 3:10 am; edited 3 times in total 
PostPosted: Fri Dec 29, 2006 2:35 am  Back to top 
  ProfilePMWWW
vittasko
Navier-Stokes Equations
Navier-Stokes Equations

Offline
Joined: 22 Aug 2006
Posts: 1186
Location: Athina, Hellas
GreeceEuropean Union

To rate posts you must be logged in
#3
Re: perpendicular feet

barasawala wrote:
Let AA_{1},BB_{1},CC_{1} be the altitudes in acute triangle ABC, and let X be an arbitrary point. Let M,N,P,Q,R,S be the feet of the perpendiculars from X to the lines AA_{1},BC,BB_{1},CA,CC_{1},AB. Prove that MN,PQ,RS are concurrent.


This problem also has been presented last time in Hyacinthos Forum on Aug 8, 2006, by Quqng Tuan Bui and there was an extensive discussion, about also the generalization of the problem, as it has been presented first time by Eric Daneels (''paracevian perspector ?''), on Jul 23,2004.

You can see at: http://groups.yahoo.com/group/Hyacinthos/message/10135 and http://groups.yahoo.com/group/Hyacinthos/message/13907

An alternative (than Darij’s in his website) proof, it has already been posted also in Hyacinthos Forum at
http://groups.yahoo.com/group/Hyacinthos/message/14034 and some notes at http://groups.yahoo.com/group/Hyacinthos/message/14088 for a simpler yet proof.

I will post here next time the two alternative synthetic proofs I have in mind of the ''paracevian perspector '' theorem.

Kostas Vittas.

PostPosted: Fri Dec 29, 2006 9:50 am  Back to top 
  ProfilePMBlog
probability1.01
Birch & Swinnerton Dyer
Birch & Swinnerton Dyer


Offline
Joined: 06 Apr 2004
Posts: 2915
Location: J'ai oublié
ChinaUnited States

To rate posts you must be logged in
#4
I did not check the above links, but here is a way to prove it:

Dilate with a factor of 2 about X. The image of MN becomes the line A_{1}X reflected about AA_{1}, and analogous results hold for the other two sides. Since the altitudes of ABC are the angle bisectors of its orthic triangle, the images of MN, PQ, and RS concur at the isogonal conjugate of X in the orthic triangle. Thus MN, PQ, and RS themselves must also concur.
_________________
Don't hatch your chickens before they count. Otherwise, they will be far behind in middle school math and perform poorly at MathCounts.

the Search for a Blog Name Continues. . .

PostPosted: Fri Dec 29, 2006 5:16 pm  Back to top 
  ProfilePMAIMYMBlog
darij grinberg
Birch & Swinnerton Dyer
Birch & Swinnerton Dyer


Offline
Joined: 10 Feb 2004
Posts: 5778
Location: Karlsruhe / Munich

To rate posts you must be logged in
#5
probability1.01 wrote:
I did not check the above links, but here is a way to prove it:

Dilate with a factor of 2 about X. The image of MN becomes the line A_{1}X reflected about AA_{1}, and analogous results hold for the other two sides. Since the altitudes of ABC are the angle bisectors of its orthic triangle, the images of MN, PQ, and RS concur at the isogonal conjugate of X in the orthic triangle. Thus MN, PQ, and RS themselves must also concur.


This is the simple proof I spoke of Smile .

The problem also appears as problem 5.81 in Prasolov's Planimetry problems book (see http://www.math.su.se/~mleites/ and http://michaj.home.staszic.waw.pl/prasolov.html for English and http://www.mccme.ru/free-books/ for Russian versions), with the same proof.

Darij
_________________
Now the die is cast, the first step taken, a glimmer of hope lights up our lives
Visions of the past, dreams forsaken forming right under our eyes
We are alive...

PostPosted: Sat Dec 30, 2006 2:21 am  Back to top 
  ProfilePMWWW
vittasko
Navier-Stokes Equations
Navier-Stokes Equations

Offline
Joined: 22 Aug 2006
Posts: 1186
Location: Athina, Hellas
GreeceEuropean Union

To rate posts you must be logged in
#6
Re: Perpendicular feet.

THE PARACEVIAN PERSPECTOR THEOREM.

– A triangle \bigtriangleup ABC is given and let H be, an arbitrary fixed point, and let \bigtriangleup HaHbHc be, it’s cevian triangle.

- Let P be, an arbitrary also point and Pa, Pb, Pc, the traces on sidelines BC, CA, AB, from the lines through P and parallel to AHa, BHb, CHc, respectively.

- Let Ap, Bp, Cp be, the traces on segment lines AHa, BHb, CHc, from the lines through P and parallel to the sidelines BC, CA, AB, respectively.

- Prove that the lines ApPa, BpPb, CpPc, are concurrent at one point so be it, Q, as the midpoint of the segment between the point P and the H-Ceva conjugate of P.



PROOF 1. - ( In my drawing AB = 15.3,AC = 14.2, BC = 11.2, AH = 12.8, BH = 4.2, AP = 8.2, CP = 6.8 ).

1) – We denote the intersection points of sidelines of \bigtriangleup ABC, from the segment lines PAp, PBp, PCp, as follows:

- As D, D', the intersection points of AC, AB respectively, from the segment line PAp.

- As E, E', the intersection points of AB, BC respectively, from the segment line PBp.

- As F, F', the intersection points of BC, AC respectively, from the segment line PCp.

Also we denote:

- As A', the intersection point of DE', D'F.

- As B', the intersection point of DE', EF'.

- As C', the intersection point of EF', D'F.

Based on the below proposition, we will prove that the lines AP, BP, CP, pass through the points A', B', C', respectively.


2)PROPOSITION 1. - A triangle \bigtriangleup ABC is given and let P be, an arbitrary point inwardly to it. We denote as D, D', the intersection points of AC, AB respectively, from the line through P and parallel to BC. Also we denote as E', F, the intersection points of BC, from the lines through P and parallel to AC, AB respectively. Prove that the lines DE', D'F, AP, are concurrent.

2a) - PROOF. - We denote as A', the intersection point of DE', D'F and as P', the one of BC, from the segment line AP.

From DD'//BC, \Longrightarrow \frac{PD}{P'C}= \frac{AP}{AP'}= \frac{PD'}{P'B}

\Longrightarrow \frac{PD}{(P'C)-(PD)}= \frac{PD'}{(P'B)-(PD')} \Longrightarrow \frac{PD}{P'E'}= \frac{PD'}{P'F} ,(1) ( because of PD = E'C and PD' = FB ).

From (1) \Longrightarrow \frac{PD}{PD'}= \frac{P'E'}{P'F} and so, the points P, P', A', are collinear. Hence the lines DE', D'F, AP, are concurrent and the proof of proposition 1, is completed.


3) – In our configuration now, we conclude that the line AP, passes through the point A' ( intersection point of DE', D'F ) and similarly the lines BP, CP, pass through the points B', C', respectively.

By the same way, from the triangle \bigtriangleup AHaC and the point P inwardly to it, we conclude that the lines ApPa, DE', AP, are concurrent. Hence, the line ApPa passes through the point A' and similarly the lines BpPb, CpPc, pass through the points B', C', respectively.


4) – We denote as K, L, M, the intersection points of the line CC', trom the segment lines AB, A'Ap, AHa respectively and as N, the intersection point of EE', from the line B'L. We will prove that the points B, M, N, are collinear.

It is enough to prove that (E, N, P, E') = (K, M, P, C).

We consider the pencils B'.C'LPT and A'.C'LPT, where T, is the intersection point of CC', A'B' and then we have that

(B'.C'LPT) = (A'.C'LPT) ,(2) ( congruent double ratios of these bundles of lines ).

(B'.C'LPT) = (E, N, P, E') ,(3) ( intersection of B'.C'LPT, from the line EE' ).

(A'.C'LPT) = (D', Ap, P, D) ,(4) ( intersection of A'.C'LPT, from the line DD' ).

from (2), (3), (4), \Longrightarrow (E, N, P, E') = (D', Ap, P, D) ,(5).

The lines CC', DD', intersect the pencil A.BHaPC and so, we have that (K, M, P, C) = (D', Ap, P, D) ,(6).

From (5), (6), \Longrightarrow (E, N, P, E') = (K, M, P, C) ,(7).

From (7), we conclude that the lines EK, MN, CE', are concurrent. So, the line MN passes through the vertex B of \bigtriangleup ABC ( as the intersection point of EK, CE' ). Hence, the points B, M, N, are collinear.


5) – We will prove now, that the lines ApPa, BpPb, CpPc, are concurrent. Let Q be, the intersection point of BpPb, CpPc and it is enough to prove that this point lies on A'L ( because of it has already been proved that the line ApPa, passes through the point A' ). We will prove that the pencils B'.C'LBpA', C'.B'LCpA', have congruent double ratios. That is, we will prove that

(B'.C'LBpA') = (C'.B'LCpA').

(B'.C'LBpA') = (E, N, Bp, E') ,(8).

(C'.B'LCpA') = (F', P, Cp, F) ,(9).

We consider the bundles of lines ( = pencils ) B.ENBpE' and C.F'PCpF. They have the sideline BC of \bigtriangleup ABC, as their common ray and the points A ( as the intersection point of BE, CF' ), M ( as the intersection point of BN, CP ) and H ( as the intersection point of BBp, CCp ), lie on the same line AHa. So, we have that

(B.ENBpE') = (C.F'PCpF) ,(10).

From (10) \Longrightarrow (E, N, Bp, E') = (F', P, Cp, F) ,(11).

From (8), (9), (11), \Longrightarrow (B'.C'LBpA') = (C'.B'LCpA') ,(12).


From (12) and because of B'C' is the common ray of pencils B'.C'LBpA' and C'.B'LCpA', we conclude that the points L ( as the intersection point of B'L, C'L ), Q ( as the intersection point of B'Bp, C'Cp ) and A' ( as the intersection point of B'A', C'A' ), are collinear. Hence, the lines ApPa, BpPb, CpPc, are concurrent at one point, so be it Q.


6) – We will prove now, that the concurrency point Q, of the lines ApPa, BpPb, CpPc, is the midpoint of the segment between the point P and the H-Ceva conjugate of P.

This result has been mentioned by Quang Tuan Bui, at last discussion in Hyacinthos Forum http://groups.yahoo.com/group/Hyacinthos/message/13938 and first time by Bernard Gibert http://groups.yahoo.com/group/Hyacinthos/message/10136

Let S be, the intersection point of AA', B'C'. So, from the parallelogram AEPF', we have that PS = SA. Hence, the line through vertex A of \bigtriangleup ABC and parallel to B'C', intersects the line PB' at a point so be it B'', such that PB' = B'B''.

Similarly the line through vertex C of \bigtriangleup ABC and parallel to A'B', intersects the line PB', at the same point B'', because of PT = TC, from the parallelogram CDPE'.

Hence, the lines through vertices A, B, C, of \bigtriangleup ABC and parallel to B'C', A'C, A'B' respectively, intersect each other at points A'', B'', C'', lie on the lines AA', BB', CC' respectively. The triangle \bigtriangleup A''B''C'', is the anticevian triangle of \bigtriangleup ABC, with respect to the point P.


6a) – We will prove now, that B''Hb\parallel BpPb. In triangle \bigtriangleup B'BBp and from PPb\parallel BBp we have that

\frac{PPb}{BBp}= \frac{B'P}{B'B} \Longrightarrow \frac{B'B''}{B'B}= \frac{BpHp}{BBp} ,(13) ( because of B'P = B'B'' and PPb = BpHb ).

From (13), \Longrightarrow B''Hb\parallel B'Bp and so, we conclude that the line B''Hb connecting the vertices B'', Hb, of the triangles \bigtriangleup A''B''C'', \bigtriangleup HaHbHc respectively, intersects the segment line PQ, at a point so be it R, such that PQ = QR ( because of PB' = B'B'' ).

By the same way we can prove that the lines A''Ha, C''Hc, are parallel to A'Ap, C'Cp respectively and intersect the segment line PQ, at the sane point R. So, the triangles \bigtriangleup A''B''C'', \bigtriangleup HaHbHc, are perspective with perspector the point R, lies on the segment line PQ and such that PQ = QR.

Because of the triangle \bigtriangleup A''B''C'', is the anticevian triangle of \bigtriangleup ABC, with respect to the point P ( since \bigtriangleup ABC is the cevian triangle of \bigtriangleup A''B''C'' wrt P ), we conclude that the point Q ( the concurrency point of ApPa, BpPb, CpPc ), is the midpoint of the segment between the point P and the point R, as the H-Ceva conjugate of P and the proof is completed.


\bullet This proof is dedicated to Babis Stergiu.

Kostas Vittas.

PS. I will post here next time, the simpler proof of the ''paracevian perspector'' theorem, which has been arisen from the discussion in Hyacinthos Forum, last summer.
t=126138(a).gif
 Description   Proposition 1.
 Filesize   7.03KB
 Viewed   1112 Time(s)

t=126138(a).gif

t=126138(b).gif
 Description   Concurrency point Q.
 Filesize   21.17KB
 Viewed   136 Time(s)

t=126138(b).gif

t=126138(c).gif
 Description   Q is H-Cena conjugate of P.
 Filesize   18.8KB
 Viewed   156 Time(s)

t=126138(c).gif


PostPosted: Sat Dec 30, 2006 2:22 pm  Back to top 
  ProfilePMBlog
vittasko
Navier-Stokes Equations
Navier-Stokes Equations

Offline
Joined: 22 Aug 2006
Posts: 1186
Location: Athina, Hellas
GreeceEuropean Union

To rate posts you must be logged in
#7
Re: Perpendicular feet.

PROOF 2. - ( see the figure t=126138(c) ).

It has already been proved that the segment lines PAp, PBp, PCp, as parallel to the sidelines BC, AC, AB of \bigtriangleup ABC respectively, intersect them at pairs of points D, D' and E, E' and F, F', such that the intersection points A'\equiv DE'\cap D'F, B'\equiv DE'\cap EF', C'\equiv EF'\cap D'F, lie on the segment lines AP, BP, CP, respectively.

The lines through vertices A, B, C, of \bigtriangleup ABC and parallel to B'C', A'C', A'B' respectively, intersect each other at the points A'', B'', C'', also lie on the segment lines AP, BP, CP, respectively.

So, we have the configuration of the triangle \bigtriangleup ABC, as the Cevian triangle of \bigtriangleup A''B''C'', with respect to the point P ( Hence the triangle \bigtriangleup A''B''C'', is the Anticevian triangle of \bigtriangleup ABC, wrt P ),

Because of the segment lines AHa, BHb, CHc are concurrent at one point ( here the point H ) and also the segment lines A''A, B''B, C''C are concurrent at one point ( here the point P ), based on the well known proposition 2, we conclude that the segment lines A''Ha, B''Hb, C''Hc, are concurrent at one point, so be it R, which in a such our configuration, is called the H-Ceva conjugate of P, with respect to the triangle \bigtriangleup ABC.

It has already been proved ( see the proof 1 ), that the segment lines ApPa, BpPb, CpPc, pass through the points A', B', C', as the midpoints of the segments PA'', PB'', PC'' respectively.

Because of ApPa\equiv A'Ap\parallel A''Ha and BpPb\equiv B'Bp\parallel B''Hb, and CpPc\equiv C'Cp\parallel C''Hc, we conclude that the segment lines ApPa, BpPb, CpPc, intersect the segment PR, at the same point, so be it Q, as it’s midpoint. That is the segment lines ApPa, BpPb, CpPc, are concurrent at the point Q, as the midpoint of the segment between P and R as the H-Ceva conjugate of P with respect to the triangle \bigtriangleup ABC and the proof is completed.


\bullet This proof is dedicated to Vladimir Zajic.

Kostas Vittas.

PS. PROPOSITION 2. - ( well known ) - A triangle \bigtriangleup ABC is given and let \bigtriangleup DEF be, the Cevian triangle of an arbitrary point P, in the plane. If D', E', F', are three points on the sidelines EF, DF, DE of \bigtriangleup DEF respectively, such that the segment lines DD', EE', FF', to be concurrent at one point, prove that the segment lines AD', BE', CF', are also concurrent at one point.

PostPosted: Sun Dec 31, 2006 8:57 am  Back to top 
  ProfilePMBlog
jayme
Yang-Mills Theory
Yang-Mills Theory

Offline
Joined: 20 Nov 2007
Posts: 640

To rate posts you must be logged in
#8
Dear Mathlinkers,
a new proof of "The paracevian perspector" has been put on my website
http://perso.orange.fr/jl.ayme vol. 4
Sincerely
Jean-Louis

PostPosted: Thu Dec 25, 2008 5:24 am  Back to top 
  ProfilePM
Display posts from previous:   Sort by:   
8 Posts • Page 1 of 1
Post new topic   Reply to topic View previous topicView next topic
Jump to:  

You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
You cannot attach files in this forum
You can download files in this forum
You cannot post calendar events in this forum


© Copyright 2008 AoPS Incorporated. All Rights Reserved. • FoundationPrivacyContact Us