I am looking for an idea to solve below system of equation. $$\begin{cases}x+y+\lfloor x\rfloor+\lfloor y\rfloor = 5.1\\x^2+y^2+\lfloor x^2\rfloor+\lfloor y^2\rfloor = 50.93\end{cases}$$ My trial was about to plotting them together ,I think two solution exist here . I think $$(-1.7,4.8),(4.8,-1.7)$$ is solution with respect to graphical solving. https://www.desmos.com/calculator/9mc2c6y0qd
I am interested if there was an analytical method...I am thankful for your Idea(s) .

$remark:$I tried also to put $$x=n_x+p_x,y=n_y+p_y$$ but it comes more complicated .

  • 1
    $\begingroup$ Notice the Desmos notice about "fine detail that has not fully been resolved." Actually I think it's worse than that: it's plotted points that don't satisfy the equations. But the graph does give you a rough idea of where to look. $\endgroup$ – David K Sep 27 '17 at 17:52
  • $\begingroup$ @DavidK:yes it works in second equation . $\endgroup$ – Khosrotash Sep 27 '17 at 17:57
  • $\begingroup$ How is this calculus? $\endgroup$ – mtheorylord Sep 27 '17 at 18:01
  • $\begingroup$ @Khosrotash Right, I noticed an error in my arithmetic right after I wrote the wrong result. Sometimes you get lucky! $\endgroup$ – David K Sep 27 '17 at 18:07
  • $\begingroup$ Look for solutions of the first equation when $x=0.05.$ Look for solutions of the second equation when $x=0.99$ or $x=1.41.$ $\endgroup$ – David K Sep 27 '17 at 18:11

We can find a range based on condition.

  1. From the first equation, we have that: $$x+y + [x]+[y] = 5.1\\ 2(x+y) - (f_x+f_y) = 5.1$$ Based on the range of $f$ which lies in $[0,1)$, we can say that: $$x+y \in \left[\frac{5.1}{2}, \frac{7.1}{2}\right)$$

  2. Similarly from second equation, one can deduce: $$x^2 + y^2 \in \left[\frac{50.93}{2}, \frac{52.93}{2}\right)$$

Basically the first condition gives two lines and second condition gives two circles. Solution lies in area trapped between these curves!

Finding the solution

What I am saying is that you can find solution analytically. Consider the first area. You can find the coordinates of intersection of circle and straight line analytically. I am graphing it to show that the range is quite small:

enter image description here

We are lucky that here, $y$ only lies in $(4,5)$ so that $[y] = 4$. But $x$ lies in $(-3,-1)$, so you must consider two intervals for fixing $[x]$.

  1. $x\in (-3, -2)$. So here, $[x] = -3$. This gives us $x+y = 4.1$, which is not possible.

  2. $x \in [-2,-1)$. So here, $[x] = -2$. This gives us $x+y = 3.1$, which is possible.

We have narrowed down to $[x] = -2$ and $[y] = 4$. Now we have more cases for $[x^2]$ and $[y^2]$! Using our conditions we get $1\le [x^2]\le 4$ and that $16 \le [y^2] \lt 25$

Using the second equation, you must have $[x^2] + [y^2] \in (24.465,25.465)$. Thus $[x^2] + [y^2] = 25$. From this we get only three possible pairs of $([x^2],[y^2])$, which are $(1,24), (2,23), (3,22)$

Thus we only deal with these three cases and you will be able to solve it analytically!

Also noting that the system is symmetric in $x,y$ so that we have an even number of roots.


I tried this : we have $$[x+y]=[x]+[y] \text{or }[x+y]=[x]+[y]+1$$ so ,if we take floor of the first equation $$x+y+[x]+[y]=5.1\\ [x+y]+[x]+[y]=5\\\begin{cases}[x]+[y]+[x]+[y]=5 & imposiible\\ [x]+[y]+1+[x]+[y]=5 \to & [x]+[y]=2\end{cases}$$ if we do the same for 2nd eq. we have $$\quad{[x^2+y^2+[x^2]+[y^2]]=[50.93]\\ [x^2+y^2]+[x^2]+[y^2]=50 \\ [x^2]+[y^2]-1+[x^2]+[y^2]=50 \text { not acceptable}\\ [x^2]+[y^2]+[x^2]+[y^2]=50\\ [x^2]+[y^2]=25 \to x^2+y^2=25.93 \text{ the only possibilities}\\}$$finally we have
If we say $s=x+y ,p=xy$
also from the first one :$x+y+[x]+[y]=5.1 ,[x]+[y]=2 \to x+y=3.1$ $$\quad{\begin{cases}x+y=3.1\\x^2+y^2=25.93\end{cases}\\\begin{cases}s=3.1\\s^2-2p=25.93\end{cases}\\z^2-3.1z-8.16=0 \\z=-1.7\\z=4.8 \\so \\(x,y)=(4.8,-1.7),(-1.7,4.8)}$$ am I right ?

  • $\begingroup$ In the other case, shouldn't it be $[x+y]=[x]+[y]+1$? $\endgroup$ – jonsno Sep 28 '17 at 8:33
  • $\begingroup$ Nice solution! Drastically simplifies the problem! $\endgroup$ – jonsno Sep 28 '17 at 8:34
  • $\begingroup$ @samjoe :Tahnk you to find the mistake , I'll correct to progress. $\endgroup$ – Khosrotash Sep 28 '17 at 8:46

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.