A Cool Diophantine Equation

Поділитися
Вставка
  • Опубліковано 2 січ 2025

КОМЕНТАРІ • 64

  • @mjones207
    @mjones207 3 роки тому +4

    One of my favorite applications of this is when the sum is 1/2. When solving 1/x + 1/y = 1/2, the resulting (x, y)
    give dimensions of a rectangle whose area equals its perimeter, and by extension, when 1/x + 1/y + 1/z = 1/2,
    the resulting (x, y, z) give dimensions of a box whose volume equals its surface area.

  • @hamzasayyid8152
    @hamzasayyid8152 3 роки тому +4

    Wow. This was just beautiful. I had no idea you could use gcd’s like this. Really awesome!

    • @SyberMath
      @SyberMath  3 роки тому

      Glad you like it!

    • @aashsyed1277
      @aashsyed1277 3 роки тому

      HAMZA GCD IS HCF OR HIGHEST COMMON FACTOR

  • @VSN1001
    @VSN1001 3 роки тому +5

    Cool! I have yet to learn how to solve diophatine equations with general solutions. Thanks for the “intro lesson”

    • @SyberMath
      @SyberMath  3 роки тому

      Glad you liked it! You mean the video on Diophantine Equations?
      ua-cam.com/video/3U3PbYnkkAM/v-deo.html

    • @SONUKUMAR-vr2jg
      @SONUKUMAR-vr2jg 3 роки тому

      This a famous diaphontine equation,which came in many mathematics Olympiad

    • @VSN1001
      @VSN1001 3 роки тому

      Oh, I did not see this video(ua-cam.com/video/3U3PbYnkkAM/v-deo.html) yet. Thanks!

  • @diedoktor
    @diedoktor 3 роки тому

    I didn't think I could solve something like this but I think I have, I haven't watched the video yet. Without loss of generality I defined x to be less than or equal to y. For the cases where x

  • @SONUKUMAR-vr2jg
    @SONUKUMAR-vr2jg 3 роки тому

    This is a famous diophantine equation, which came in many mathematics Olympiad. Thanks for solution

    • @SyberMath
      @SyberMath  3 роки тому

      Absolutely! Glad to hear that it helped!

    • @SONUKUMAR-vr2jg
      @SONUKUMAR-vr2jg 3 роки тому

      @@SyberMath can you cover Chinese Remainder Theorem , as you explain well

    • @SyberMath
      @SyberMath  3 роки тому

      @@SONUKUMAR-vr2jg You mean in a video?

    • @SONUKUMAR-vr2jg
      @SONUKUMAR-vr2jg 3 роки тому

      @@SyberMath make a video on Chinese Remainder Theorem

  • @ezzaddin9351
    @ezzaddin9351 3 роки тому +1

    this video is great, subscribed

    • @SyberMath
      @SyberMath  3 роки тому

      Glad to hear that! Welcome to the channel!

  • @sbmathsyt5306
    @sbmathsyt5306 3 роки тому

    Loving all your number theory videos recently

  • @roman_roman_roman
    @roman_roman_roman 3 роки тому +1

    At the beginning you state that m and n should be relatively prime, but at the end it turns that m and n are just posituve integers without any other restrictions.

    • @SyberMath
      @SyberMath  3 роки тому

      m and n are relatively prime but x, y and z do not have to be. We wrote x,y, and z in terms of m,n and k.

    • @falknfurter
      @falknfurter 3 роки тому +1

      If you allow m, n to have common divisors there are different triples (k, m, n) which result in the same (x, y, z). Let's assume m, n have a common factor a, e.g. n = a*b, m = a*c. With the final equation in the video you can easily check that (k, m, n) and (k*a^2, b, c) result in the same triple (x, y, z). So, if you want (x, y, z) to have a unique representation in terms of (k, m, n) you require gcd(m,n)=1. Otherwise you don't need this condition.

  • @Qermaq
    @Qermaq 3 роки тому +1

    Easy problem in one way, difficult in another. Nice!

    • @SyberMath
      @SyberMath  3 роки тому

      The simplicity of its look and the parametric solutions we have to use!

  • @sanjaysurya6840
    @sanjaysurya6840 3 роки тому

    Beautiful sir 💐.. Keep it up 👍

  • @2false637
    @2false637 3 роки тому +1

    I did not quite understand the reason behind the gcd(mn, m+n) =1...

    • @SyberMath
      @SyberMath  3 роки тому

      You mean why it is true?

    • @Relrax
      @Relrax 3 роки тому +1

      gcd(n, m)= 1 => gcd(n, m+n) = gcd(m, m+n) = 1 => gcd(n*m, m+n) | 1 => gcd(nm, m+n) = 1

    • @SyberMath
      @SyberMath  3 роки тому

      @@Relrax Nice work!

    • @tonyhaddad1394
      @tonyhaddad1394 3 роки тому

      6:33 and there gcd must equal 1 for (m,n)

  • @tonyhaddad1394
    @tonyhaddad1394 3 роки тому +2

    This what i want , cool !!!!!!!!!

  • @jamesmanguilimotan2485
    @jamesmanguilimotan2485 3 роки тому

    Can you find the integer solution of 1/x + 1/y= ¼ please sir solve it

    • @SyberMath
      @SyberMath  3 роки тому

      Make a common denominator
      (x+y)/xy=1/4
      xy=4x+4y
      xy-4x-4y=0
      xy-4x-4y+16=16
      x(y-4)-4(y-4)=16
      (x-4)(y-4)=16
      Check divisors/factors of 16...
      The rest is trivial

    • @jamesmanguilimotan2485
      @jamesmanguilimotan2485 3 роки тому

      @@SyberMath woww thanks for the reply and the answer it was unexpected

    • @SrisailamNavuluri
      @SrisailamNavuluri 3 місяці тому

      Easy way is 1/4=1/5+1/20
      1/z=1/(z+1)+1/z(z+1) for integer, z>1

  • @CipriValdezate
    @CipriValdezate 3 роки тому

    Awesome!

  • @joaquingutierrez3072
    @joaquingutierrez3072 3 роки тому

    Nice problem!! But almost any integers m, n, k work, right? Not only the positive ones. The only constraints will be m, n, k =/= 0 and m + n =/= 0.

    • @SyberMath
      @SyberMath  3 роки тому

      Sure! You can just negate everything. I just wanted to keep it simple and restrict it to positives.

  • @36sufchan
    @36sufchan 3 роки тому

    I did work on this a while ago, but I had assumed that 'z' was a known natural number and we're looking for naturals x and y that satisfy 1/x + 1/y = 1/z
    for this I found that all solutions are of the form (x, y) = (z + f, z + z²/f) where f is a divisor of z² (obviously) leading to a number of solutions equal to number of divisors of z²
    But hey, this one's pretty neat as well. Kudos.

  • @yhamainjohn4157
    @yhamainjohn4157 3 роки тому

    m, n, k in N-{0} but gcd(m,n)= 1. Isn't it ? ok I see it's for all positive integer : great job ! Bravo

  • @shreyan1362
    @shreyan1362 3 роки тому

    1/(2n) + 1/(2n) =1/n
    There are infinite solutions of this so it has inf solns

    • @SyberMath
      @SyberMath  3 роки тому

      They are not the only solutions

  • @Germankacyhay
    @Germankacyhay 3 роки тому

    👍 Дякую.

    • @SyberMath
      @SyberMath  3 роки тому

      нема проблем

  • @deepjyoti5610
    @deepjyoti5610 3 роки тому

    Going to subscribe you

    • @SyberMath
      @SyberMath  3 роки тому +1

      Welcome to the channel!

  • @comingshoon2717
    @comingshoon2717 3 роки тому

    so fun

  • @Tomaplen
    @Tomaplen 3 роки тому

    (x,y,z)=(4,4,2)

    • @SyberMath
      @SyberMath  3 роки тому

      Sure. Are there other solutions?

  • @BCS-IshtiyakAhmadKhan
    @BCS-IshtiyakAhmadKhan 3 роки тому +1

    So easy Anyone could tell easily that this equation has infinite solution

  • @srijanbhowmick9570
    @srijanbhowmick9570 3 роки тому

    Hey dude just wanted to say something regarding your announcement ...
    You don't have to go for views ok?
    What I mean that when you first opened this channel you probably wanted to share your interest regarding geometry puzzles with other people , right ? So you made your own geometry puzzles to amuse people . If you keep on following your passion then views will automatically increase ! And I personally stayed with your channel bcoz of them ! So my solution is that keep on making videos on geometry puzzles but perhaps make them a bit more interesting , like maybe change your outlining shape to a heart , pentagon and other weird shapes etc . In short be a bit more creative 🙂
    And an another piece of advice , Micheal Penn recently opened a Google form to take viewer suggested problems , so why don't you ? Or you can even increase the difficulty of problems ! So that's it from my side and I hope my advices help you in the near future !

    • @SyberMath
      @SyberMath  3 роки тому

      Srijan, I agree with you. This channel started with my passion of solving problems that I like and sharing it with the world. It's not all about views, of course. Geometry is not my strength but I like solving puzzles that involve inscribed figures because they are fun imo and also not very hard to solve! If you check out my twitter, you will notice a great number of people that are very strong in geometry. They can basically tackle very hard problems with ease, such as the ones that appear in olympiads. I'm not one of them. I take your advice to heart and will work to come up with more interesting puzzles. Thanks for the feedback!

    • @srijanbhowmick9570
      @srijanbhowmick9570 3 роки тому

      @@SyberMath Yeah thanks for taking my feedback with a positive mindset !

    • @SyberMath
      @SyberMath  3 роки тому

      @@srijanbhowmick9570 Of course!

    • @xavier3153
      @xavier3153 3 роки тому

      can i be of any help?