International Math Olympiad | 1998 Question 4
Рет қаралды 26,753
Күн бұрынWe look at a solution to a nice number theory problem from the 1998 International Mathematical Olympiad.
Please Subscribe: kzfaq.info...
Merch: teespring.com/stores/michael-...
Personal Website: www.michael-penn.net
Randolph College Math: www.randolphcollege.edu/mathem...
Randolph College Math and Science on Facebook: / randolph.science
Research Gate profile: www.researchgate.net/profile/...
Google Scholar profile: scholar.google.com/citations?...
If you are going to use an ad-blocker, considering using brave and tipping me BAT!
brave.com/sdp793
Buy textbooks here and help me out: amzn.to/31Bj9ye
Buy an amazon gift card and help me out: amzn.to/2PComAf
Books I like:
Sacred Mathematics: Japanese Temple Geometry: amzn.to/2ZIadH9
Electricity and Magnetism for Mathematicians: amzn.to/2H8ePzL
Abstract Algebra:
Judson(online): abstract.ups.edu/
Judson(print): amzn.to/2Xg92wD
Dummit and Foote: amzn.to/2zYOrok
Gallian: amzn.to/2zg4YEo
Artin: amzn.to/2LQ8l7C
Differential Forms:
Bachman: amzn.to/2z9wljH
Number Theory:
Crisman(online): math.gordon.edu/ntic/
Strayer: amzn.to/3bXwLah
Andrews: amzn.to/2zWlOZ0
Analysis:
Abbot: amzn.to/3cwYtuF
How to think about Analysis: amzn.to/2AIhwVm
Calculus:
OpenStax(online): openstax.org/subjects/math
OpenStax Vol 1: amzn.to/2zlreN8
OpenStax Vol 2: amzn.to/2TtwoxH
OpenStax Vol 3: amzn.to/3bPJ3Bn
My Filming Equipment:
Camera: amzn.to/3kx2JzE
Lense: amzn.to/2PFxPXA
Audio Recorder: amzn.to/2XLzkaZ
Microphones: amzn.to/3fJED0T
Lights: amzn.to/2XHxRT0
White Chalk: amzn.to/3ipu3Oh
Color Chalk: amzn.to/2XL6eIJ
@tomatrix7525 3 жыл бұрын
I really find myself improving with IMO problems after watching you for a few months. Originally I always found I’d understand the solution but would never know how to approach it myself without major hints. That’s slowly changing
@MichaelPennMath 3 жыл бұрын
That is great to hear. I have to admit, I have improved with these types of problem by producing these videos!
@goodplacetostop2973 3 жыл бұрын
5:00 15:37 For anyone needing to hear this today : you matter, you are strong, you can do it. Have a great Sunday.
@adityamohan7366 3 жыл бұрын
ffs we can't beat this guy
@goodplacetostart9099 3 жыл бұрын
Good place to start at 0:00
@minh9545 3 жыл бұрын
I just failed the entrance exam to my university. Nothing can help me now.
@goodplacetostop2973 3 жыл бұрын
TNT I'm sorry to read that you failed your entrance exam. Don’t be too hard on yourself, don’t lose hope. I would suggest you take some time to relax. Then, maybe think about the weak areas on your exam and try to find academical help. It may seem a dark moment but this is the first day of a comeback.
@ianloree2784 3 жыл бұрын
Today is a great place to start
@skwbusaidi 3 жыл бұрын
We can save steps when y=1 and y=2 We can use (xy^2+y+7) | (7x-y^2) Which you find before For y=2 (4x+9)|(7x-4) (4x+9)|(7(4x+9)-4(7x-4)) (4x+9)|79 79 is prime 4x+9=1 or 4x+9=79 x= -2 or 17.5 both are not Natural number So no solutions
@CM63_France 3 жыл бұрын
Hi, For fun: 1 "and so on and so forth", 6 "go ahead and ...", including 4 "let's go ahead and ...", 3 "great", 2 "that's a good place to stop", including the "and that's a good place to stop" at the end.
@stewartcopeland4950 3 жыл бұрын
and some of "which tells us..."
@CM63_France 3 жыл бұрын
@@stewartcopeland4950 I love those "which tells us", very poetic :) , an some times "which leaves us with..."
@R0M4ur0 3 жыл бұрын
Oh guys... that "tells us"... for a non native speaker the sound of that "s" between "tell" and "us" is like a symphony... even because 9 out of 10 times we EAASL fail to pronounce it 😅
@shangaiguarisnaque9277 3 жыл бұрын
@@R0M4ur0 Relatable af
@R0M4ur0 3 жыл бұрын
@@shangaiguarisnaque9277 that's what i meant
@wasselkun5015 3 жыл бұрын
Thank you for the video!! learning a lot from the channel.
@andrewlayton9760 9 күн бұрын
For y=2: (4x+9) | (2x^2 + x +2) Long division gives (1/2)x - 7/8 with a remainder of 79/8. For divisibility to hold, the remainder divided by (4x+9) must be an integer. 79/8 is non-integer and 4x+9 must be an integer, so the quotient cannot be an integer --> no solution. I also used this method on the y=1 case and get the same answers as shown in the video.
@mcwulf25 3 жыл бұрын
Good one. Lots of solutions.
@ravinderbhandari1974 3 жыл бұрын
from 5:30 , we could set Y-7x=(xY+y+7)m where X= x^2 , Y= y^2 and m is an integer , putting x=Y/7, we get a cubic on RHS which is fairly easy to solve
@user-nm5ge9ht3c Жыл бұрын
I love that this problem contains both an infinite family of solutions and some small ones. It is not often that you see such problems
@mukaddastaj5223 3 жыл бұрын
Wow. Love ur videos, really, u make everything look so easy, thank u sooo much!) P.S. i have no solution when y=2
@alejandrocampos7459 3 жыл бұрын
My teachers used this problem to train me and my friends for the Centroamerican Math Competition
@zafarb4219 3 жыл бұрын
It's a pretty good problem
@newkid9807 3 жыл бұрын
Could you cover harder imo problems, like number 2 or 3
@tueherlau548 3 жыл бұрын
A no-hint solution: Write m(xy^2 + y + 7) = x^2 y + x + y. Define k s.t. m y = x + k. (to get this idea, observe that mx y^2 ~ x^2 y approximately, and so k must be "small"). By inserting and rearranging we get both (i) kxy = y - 7m - k and (ii) x = (y^2 - k (y+7) )/ ( 7 + k y^2 ). Obviously (ii) have no integer solutions for k > 0; for k < 0 it must hold that k y^2 < 7 (since x is positive) and we easily get the y=1 solutions by checking k=-1,...-6 and can rule out y=2, k=-1 by insertion. For k=0 we get x = y^2 / 7 and from (i) y = 7m. QED.
@concretemathematics414 5 ай бұрын
maybe i'm confused, but for case 1, it says that y^2-7x < xy^2 + y + 7, but we also rewrite y^2-7x as d*(xy^2 + y + 7), which that value is larger than xy^2 + y + 7 for d>0.
@iridium8562 3 жыл бұрын
How do I verify these are the only solutions?
@deeznuts4333 3 жыл бұрын
For y=2 there are no solution {x=-2;x=35/2}
@anonymoususer9837 3 жыл бұрын
Nice work, I came to the same conclusion.
@suranjanlk 3 жыл бұрын
👍well done ✅
@rishavmondal8045 3 жыл бұрын
Sir bring more and more imo problems.
@rishavmondal8045 3 жыл бұрын
Sir bring questions from latest olympiads also.
@TechToppers 3 жыл бұрын
Let s(n) denote the sum of the digits of a positive integer n in base 10. If s(m) = 20 and s(33m) = 120, what is the value of s(3m)? Can someone please help me with this? It came in India. PRMO 2019 25th August Paper
@gamedepths4792 3 жыл бұрын
5:00 i got scared ngl
@jaimeduncan6167 3 жыл бұрын
I have a question: We peek a couple of of numbers and derive a condition that the solution must satify. It seem to me that it's a necesesay but maybe not sufficient contition. For example if the selection implies that the number must be multiple of K , ca we infer that all the multiples of K are solutions?
@tracyh5751 3 жыл бұрын
No, but by plugging the infinite family back into the original condition we can see that all pairs from the infinite family work out.
@otakurocklee 3 жыл бұрын
As Tracy H said, we have to plug in the solutions we find into to the original equation to check that they work, for exactly the reason you mentioned. We may have extraneous solutions.
@shreyathebest100 3 жыл бұрын
Good question answered well by Tracy h
@milanchrastina8014 3 жыл бұрын
No solutions here also, cool vids man.
@123bluestorm1 3 жыл бұрын
4:40 if any of you wanna hear the notification
@user-kl8dh7nt2e 3 жыл бұрын
3：53 The formula is satisfied for all integer a, b, but you only choose some nice ones. How do you prove that this method would find all solutions?
@jadegrace1312 3 жыл бұрын
Because the divisibility property he gave is true regardless of what a and b you choose.
@otakurocklee 3 жыл бұрын
If x is a solution to our original problem, then it is a solution for every particular choice of a and b. Therefore solving for any choice of a and b will get us all the solutions we need... maybe extraneous ones also. We have to plug back into the original to check that our solutions are correct. But there's no way we will miss a solution.
@basedblueboy8770 3 жыл бұрын
nice
@thomasstokes9412 3 жыл бұрын
For y=2 there are no solutions in the natural numbers: plugging in y=2 => (4x+9)|(a(4x+9)+b(2x^2+x+2)) Picking a=x and b=-2 => (4x+9)|(7x-4) => (4x+9)|(a(4x+9)+b(7x-4)) Picking a =7 and b=-4 => (4x+9)|79 So either 4x+9=1 or 4x+9=79, none of these have solutions in the natural numbers.
@narekbojikian5885 3 жыл бұрын
Thanks this was a nice video! A quick question: shouldn’t we have plugged the solution (7m2, 7m) from the first case in the given formula to check its correctness? since we derived it from a chosen restriction of a and b and it is not necessarily valid for all values of a and b
@jaimeduncan6167 3 жыл бұрын
I am not sure about the procedure in general, it seems to me that it finds necesary contitions not suficient.
@otakurocklee 3 жыл бұрын
Yes, that's right. It's the same with all the solutions. We may have extraneous solutions so we have to plug into the original to check.
@Andrew-mx7vt 3 жыл бұрын
Wanted to watch a video from putnam exam, but decided that this one could be more interesting :) also who wants to solve maths problems ,perhaps ,we can unite and do it(just a suggestion :)
@h4z4rd28 3 жыл бұрын
Me :)
@electrovector7212 3 жыл бұрын
Ofc me
@mehdifachel2057 3 жыл бұрын
Me
@electrovector7212 3 жыл бұрын
@@h4z4rd28 your messages do not show up on my KZfaq can you comment from another device?
@h4z4rd28 3 жыл бұрын
@@electrovector7212 now?
@tgx3529 3 жыл бұрын
What does it mean more solutions? There are 2 solutions on the blackboard x=11,y=1 and x=49,y=1. And what about x=y=7?
@IanXMiller 3 жыл бұрын
That fits under case 1 where x=7m^2 and y=7m
@tgx3529 3 жыл бұрын
@@IanXMiller Thank you, I'm just looking for the end result. I wanted to think about it myself .
@ZeonLP 3 жыл бұрын
nice nice :3
@user-ht1vg5we2p 3 жыл бұрын
You can also have y equals 0 and after the simplifications get the infinite set {7k, 0} for every natural k. As for the 2 case, 4x+9 |2x^2 + x +2, 4x +9 | a (4x+9) +b(2x^2+x+2) a=x, b=-2 ======>> 4x+9 |4x^2+9x -4x^2 -2x -4=7x -4 Here you choose a=7, b=-4 : 4x +9 | 7 (4x +9)- 4 (7x- 4) = 63+16 = 79, which is prime, thus you have: 4x+9 equals 1 or 79. Case 1: 4x +9=1 =====>> x =-2, not acceptable, because -2 is not Natural. Case 2: 4x + 9= 79 =====> x is not an integer So no solutions here
@jwy4264 2 жыл бұрын
i think 0 isnt a natural number
@user-ht1vg5we2p 2 жыл бұрын
@@jwy4264 actually it is. If you want to select all integers bigger than or equal to 1, you have the set N* for that
@yuseifudo6075 7 ай бұрын
This is not true for the whole world It is a convenience and not a fact@@user-ht1vg5we2p
@DungNguyen-ti4hg 3 жыл бұрын
Wow... A "nice" sol
@MatriceMX 3 жыл бұрын
How can (X+8) divide(X^2+X+1) with x e Z ,when X^2+X+1 doesnt have integer solutions?
@MatriceMX 3 жыл бұрын
that is some mombo-jumbo
@TJStellmach 3 жыл бұрын
If the one polynomial divided the other, that would just tell you that _all_ possible values of x were solutions. But we don't need that; we're looking for the particular values of x that make the two values divisible. Indeed, if you divide x^2+x+1 by x+8, you get x-7 with a remainder of 57. So that means (x^2+x+1)/(x+8) = x-7+(57/(x+8)), which is an integer when 57 is divisible by x+8. It's precisely the fact that we _do_ have a nonzero remainder that implies the specific solutions.
@maxamedaxmedn6380 3 жыл бұрын
👍👌
@puranjayjoshi5549 3 жыл бұрын
6:45 why is d >= 0 when according to definition of divisibility d should belong in the integers Edit: my bad he cleared it up near 7:55
@grecunarcis09 3 жыл бұрын
Since LHS is positive => RHS have to be positive too so d must be positive because xy²+y+7 is always positive
@puranjayjoshi5549 3 жыл бұрын
@@grecunarcis09 oh yeah, thanks!
@hectormiliotis9260 13 күн бұрын
Or just solve it with the method vieta jumping and you only need 15 lines
@sidimohamedbenelmalih7133 3 жыл бұрын
x=70/4=35/2
@laszloliptak611 3 жыл бұрын
At the end you are just doing long division to divide x^2+x+1 by x+8.
@rickshaw1641 3 жыл бұрын
It is easier than the method Mr Michael talked.
@Ideophagous 3 жыл бұрын
You missed another general solution of the form (7k, 0).
@TheCrimsonMoose1 3 жыл бұрын
0 is not a natural number.
@Ideophagous 3 жыл бұрын
@@TheCrimsonMoose1 Depends on your definition. In Francophone textbooks it's considered a natural number.
@TheCrimsonMoose1 3 жыл бұрын
@@Ideophagous In almost all textbooks I've read if you include 0 the set is called the "whole numbers" and if not it's called "natural numbers" I think that is why he did not include these solutions.
@TJStellmach 3 жыл бұрын
He says just 11 seconds in what definition of natural numbers he's using.
@maxtirdatov 3 жыл бұрын
Don't we lose anything if we just pick "nice" coefficients for the linear combination?
@mathijs1752 3 жыл бұрын
We don't lose anything by picking nice coefficients. It's a rather counter-intuitive thought why, but it goes like this: We assumed x and y 'worked' and picked nice coefficients. If we then derive 'solutions' from this picking, we know that the actual solutions are a subset of the found 'solutions'. The actual solutions need to work for all coefficients, thus the found 'solutions' do contain the actual solutions but maybe more. After plugging in your 'solutions' in the original problem, you'll see that maybe some are not valid and only work for the nice coefficients, while others are valid and thus work for all coefficients. These latter values are the actual solution. In this case Michael did forget to check whether the results of case 1 and case 2 are actually valid solutions and not some results that just pop up by the nice picking of coefficients
@stewartcopeland4950 3 жыл бұрын
If a linear combination is valid for any pair of values (a,b), then it is also valid for a pair of values judiciously chosen in order to simplify the mathematical expression, without loss of generality
@mathijs1752 3 жыл бұрын
If we check the solution (7m², 7m) we get (7³m⁴+7m+7)|(7³m^5+7m²+7m). While I'm not a number theoretic, my gut feeling is that this implies m=1 is the only valid solution. Therefore his infinite family of solution comes down to just (7, 7) and maybe one or two more
@JoseFernandes-js7ep 3 жыл бұрын
@@mathijs1752 Notice that you have p(m) l p(m).m, which is true for any integer m
@mathijs1752 3 жыл бұрын
@@JoseFernandes-js7ep that's true, thanks!
@suniltshegaonkar7809 3 жыл бұрын
I have strong doubt for step at 3:11
@Ideophagous 3 жыл бұрын
It's a basic result in elementary number theory
@suniltshegaonkar7809 3 жыл бұрын
@@Ideophagous It is like starting with the assumption that needs to be proved
@Ideophagous 3 жыл бұрын
Not really x|a and x|b => x|u*a+v*b for x, a, b, u and v integers is a basic result and can easily be proven
@TJStellmach 3 жыл бұрын
@@suniltshegaonkar7809 Except he's not doing a proof. He gets to suppose the given conditions on x,y and deduce what those conditions imply. The converse of those implications _may_ not hold, but as long as every step in the chain of reasoning is biconditional, you don't even have to check.
@tgx3529 3 жыл бұрын
If I take it as [x(xy+1)+y]/[(y(xy+1)+7]= k, I see that y=7k;x=ky, so we have first solution (x,y)=(7k^2,7k) If I take x as fixed number, there is linear function variable y/ quadratic function variable y, it's not interesting for division. If I take y as fixed number, there is quadratic function variable x/ linear function variable x. Only for y=1 is division OK.For y=1 we have (x^2+x+1)=(x+8)(x-7)+57. So (x^2+x+1)/(x+8)=[(x+8)(x-7)/((x+8) ]+57/(x+8). We have 57/(x+8). 1.)case is when x+8=57, 2.) case is when we take 57/19=3, se x=11
@mehdifachel2057 3 жыл бұрын
Slt les solutions sont (7k; 0); (11; 1); (49; 1); (7q^2; 7|q|)
@natepolidoro4565 3 жыл бұрын
leaf blower
@IgnacioIacobacci 3 жыл бұрын
A problem hunts me since more than 20 years. It's a problem from "Cono Sur" Olimpiads from 1996. Given the sequence on real numbers a_n+1 = a_n + 1 / a_n. For any a_0 real positive show that a_1996 > 63. Problem nro 2 in this link in Spanish www.oma.org.ar/enunciados/con7.htm
@leickrobinson5186 3 жыл бұрын
I think there must be an error in the original problem. It’s pretty easy to show that the sequence consists of the ratios of consecutive values from a Fibonacci sequence that starts {1, a_0, a_0+1,...}. But, this ratio converges to phi, which is much less than 63. :-(
@IgnacioIacobacci 3 жыл бұрын
@@leickrobinson5186 I'm pretty sure the problem is Ok. I have a (near) solution. You can say that there for each a_n it is true that if a_n = k, a_n < ⌈k⌉ and and there are less than k a_n between k and k+1 ... so... you sum n times each n from 63 to 1 and you reach... 2016, which is pretty close, but not close enough
@Notthatkindofdr 3 жыл бұрын
You can prove by induction that a_n > sqrt(2n) for n>0, and then check that sqrt(2*1996)>63.
@IgnacioIacobacci 3 жыл бұрын
@@Notthatkindofdr Amazing... thank you very much ...
@IgnacioIacobacci 3 жыл бұрын
@Adam Romanov Single round. The outstanding participants of each national competition of South America were participating.
@DilipKumar-ns2kl 3 жыл бұрын
x=7, y=7 is a soln.
@TJStellmach 3 жыл бұрын
Yes, it's part of the infinite family of solutions of the form (x,y)=(7m^2,7m).
@beyremmerdassi3148 3 жыл бұрын
well u forgot about (x,y) = (0,7m)
@TJStellmach 3 жыл бұрын
No, the problem specifies positive x and y.
@M-F-H 3 жыл бұрын
You somewhat cheated because in case 1 you ignored the possibility x=0 (with solution x=y=0: definitely, 0 = 0*7 is a multiple of 7) which then you recover with m = 0 \in N.
@jonross1347 3 жыл бұрын
In American mathematics, by convention, the set of natural numbers does not contain 0, but that's good to point out!
@TJStellmach 3 жыл бұрын
He says ten seconds into the video that that's the definition of natural numbers he's using.
@mdajrul8344 3 жыл бұрын
Any one from 🇮🇳🇮🇳🤔🤔
Fabiosa Best Lifehacks
Рет қаралды 38 МЛН