Difference between revisions of "2021 AMC 12B Problems/Problem 25"
Isabelchen (talk | contribs) m (→Remark) |
Isabelchen (talk | contribs) m (→Remark) |
||
Line 202: | Line 202: | ||
==Remark== | ==Remark== | ||
− | <math>\lfloor \frac{b}{a} k \rfloor</math> is the number of lattice points on line <math>x = k</math>, on or below line <math>y = \frac{b}{a} x</math> and above the <math>x</math> axis, where <math>k</math> is a positive integer. Therefore, <math>\sum_{k=1}^{a-1} \lfloor \frac{b}{a} k \rfloor</math> is the number of lattice points inside the rectangle <math>(0,0)</math>, <math>(a,0)</math>, <math>(a, b)</math>, <math>(0, b)</math>, on or below diagonal <math>y = \frac{b}{a} x</math>. If <math>a</math> and <math>b</math> are relatively prime, <math>\sum_{k=1}^{a-1} \lfloor \frac{b}{a} k \rfloor = \frac{(a-1)(b-1)}{2}</math>, as explained in solution 6. This problem is about finding the upper and lower bound of <math>\frac{b}{a}</math>, given <math>\sum_{k=1}^{30} \lfloor \frac{b}{a} k \rfloor = 300</math>. | + | <math>\lfloor \frac{b}{a} k \rfloor</math> is the number of lattice points on line <math>x = k</math>, on or below line <math>y = \frac{b}{a} x</math> and above the <math>x</math> axis, where <math>k</math> is a positive integer. Therefore, <math>\sum_{k=1}^{a-1} \lfloor \frac{b}{a} k \rfloor</math> is the number of lattice points inside the rectangle <math>(0,0)</math>, <math>(a,0)</math>, <math>(a, b)</math>, <math>(0, b)</math>, on or below diagonal <math>y = \frac{b}{a} x</math>. If <math>a</math> and <math>b</math> are relatively prime, <math>\sum_{k=1}^{a-1} \lfloor \frac{b}{a} k \rfloor = \frac{(a-1)(b-1)}{2}</math>, as explained in solution 6. This problem is about finding the upper and lower bound of <math>\frac{b}{a}</math>, given <math>\sum_{k=1}^{30} \lfloor \frac{b}{a} k \rfloor = 300</math>. The same problem can have geometric or algebraic representation. |
~[https://artofproblemsolving.com/wiki/index.php/User:Isabelchen isabelchen] | ~[https://artofproblemsolving.com/wiki/index.php/User:Isabelchen isabelchen] |
Revision as of 22:58, 11 March 2022
- The following problem is from both the 2021 AMC 10B #25 and 2021 AMC 12B #25, so both problems redirect to this page.
Contents
Problem
Let be the set of lattice points in the coordinate plane, both of whose coordinates are integers between
and
inclusive. Exactly
points in
lie on or below a line with equation
The possible values of
lie in an interval of length
where
and
are relatively prime positive integers. What is
Solution 1
First, we find a numerical representation for the number of lattice points in that are under the line
For any value of
the highest lattice point under
is
Because every lattice point from
to
is under the line, the total number of lattice points under the line is
Now, we proceed by finding lower and upper bounds for To find the lower bound, we start with an approximation. If
lattice points are below the line, then around
of the area formed by
is under the line. By using the formula for a triangle's area, we find that when
Solving for
assuming that
is a point on the line, we get
Plugging in
to
we get:
We have a repeat every values (every time
goes through a lattice point). Thus, we can use arithmetic sequences to calculate the value above:
This means that is a possible value of
Furthermore, it is the lower bound for
This is because
goes through many points (such as
). If
was lower,
would no longer go through some of these points, and there would be less than
lattice points under it.
Now, we find an upper bound for Imagine increasing
slowly and rotating the line
starting from the lower bound of
The upper bound for
occurs when
intersects a lattice point again (look at this problem to get a better idea of what's happening: https://artofproblemsolving.com/wiki/index.php/2011_AMC_10B_Problems/Problem_24).
In other words, we are looking for the first that is expressible as a ratio of positive integers
with
For each
, the smallest multiple of
which exceeds
is
respectively, and the smallest of these is
Alternatively, see the method of finding upper bounds in solution 2.
The lower bound is and the upper bound is
Their difference is
so the answer is
~JimY
An alternative would be using Farey fractions and the mediant theorem to find the upper bound. and
gives
and so on using Farey addition.
Solution 2
I know that I want about of the box of integer coordinates above my line. There are a total of 30 integer coordinates in the desired range for each axis which gives a total of 900 lattice points. I estimate that the slope, m, is
. Now, although there is probably an easier solution, I would try to count the number of points above the line to see if there are 600 points above the line. The line
separates the area inside the box so that
of the are is above the line.
I find that the number of coordinates with above the line is 30, and the number of coordinates with
above the line is 29. Every time the line
hits a y-value with an integer coordinate, the number of points above the line decreases by one. I wrote out the sum of 30 terms in hopes of finding a pattern. I graphed the first couple positive integer x-coordinates, and found that the sum of the integers above the line is
. The even integer repeats itself every third term in the sum. I found that the average of each of the terms is 20, and there are 30 of them which means that exactly 600 above the line as desired. This give a lower bound because if the slope decreases a little bit, then the points that the line goes through will be above the line.
To find the upper bound, notice that each point with an integer-valued x-coordinate is either or
above the line. Since the slope through a point is the y-coordinate divided by the x-coordinate, a shift in the slope will increase the y-value of the higher x-coordinates. We turn our attention to
which the line
intersects at
. The point (30,20) is already counted below the line, and we can clearly see that if we slowly increase the slope of the line, we will hit the point (28,19) since (28,
) is closer to the lattice point. The slope of the line which goes through both the origin and (28,19) is
. This gives an upper bound of
.
Taking the upper bound of m and subtracting the lower bound yields . This is answer
.
~theAJL
Diagram
Solution 3
An alternative approach with the same methodology as Solution 1 can be done using Pick's Theorem. Wikipedia page: https://en.wikipedia.org/wiki/Pick%27s_theorem It's a formula to find the amount of lattice points strictly inside a polygon. Approximation of the lower bound is still necessary.
Solution 4
As the procedure shown in the Solution 1, the lower bound of is
Here I give a more logic way to show how to find the upper bound of
Denote N=
as the number of lattice points in
.
Let . for
Our target is finding the minimum value of which can increase one unit of
Notice that:
When We don't have to discuss this case.
When
When Here
Denote
Obviously and
are decreasing. Let's considering the situation when
This means that the answer is just , so
. Choose
~PythZhou.
Solution 5
It's easier to calculate the number of lattice points inside a rectangle with vertices ,
,
,
. Those lattice points are divided by the diagonal
into
halves. In this problem the number of lattice points on or below the diagonal and
is
,
is the number of lattice points on the diagonal,
![]()
is the total number of lattice points inside the rectangle. Subtract the number of lattice points on the diagonal, divided by 2 is the number of lattice points below the diagonal, add the number of lattice points on the diagonal, and subtract the lattice points on the
axis, then we get the total number of lattice points on or below the diagonal and
.
There are lattice points in total.
is
of
. The
coordinate of the top-right vertex of the rectangle is
,
. I guess the
coordinate of the top-right vertex of the rectangle is
. Now I am going to verify that. The slope of the diagonal is
, there are
lattice points on the diagonal. Substitute
,
to the above formula:
Because there are lattice points on line
, if
, then the number of lattice points on or below the line is less than
. So
is the lower bound.
Now I am going to calculate the upper bound. From ,
If , I will calculate by using the rectangle with blue vertex
, then subtract lattice points on line
, which is
. There are 2 lattice points on the diagonal,
.
, same as that of
![]()
If , I will calculate by using the rectangle with red vertex
, then add lattice points on line
and
, which is
. There are 2 lattice points on the diagonal,
.
,
more than that of
![]()
When increases, more lattice points falls below the line
. Any value larger than
has more than
lattice points on or below
. So the upper bound is
.
,
.
Solution 6
The lower bound of is
. Inside the rectangle with vertices
,
,
,
and diagonal
, there are
lattice points inside, not including the edges. There are
lattice points on diagonal
inside the rectangle,
. Half of the
lattice points are below diagonal
,
. There are
lattice points on edge
,
. Once
, the
lattice points on diagonal
will be above the new diagonal, making the number of lattice points on and below the diagonal less than
.
Now we are going to calculate the upper bound by the following formula:
The number of lattice points inside rectangle,
,
,
and below diagonal
is
, where
and
are relatively prime. There are
lattice points inside the rectangle. Because
and
are relatively prime, the only lattice points on the diagonal are
and
. By symmetry, half of the lattice points are below the diagonal.
When ,
,
.
When ,
,
. Below the line
, there are
lattice points on line
,
lattice points on line
,
lattice points on line
,
.
More lattice points fall below the line as
increases. There are more than
lattice points on and below the line for any
greater than
. Therefore, the upper bound is
.
,
.
Remark
is the number of lattice points on line
, on or below line
and above the
axis, where
is a positive integer. Therefore,
is the number of lattice points inside the rectangle
,
,
,
, on or below diagonal
. If
and
are relatively prime,
, as explained in solution 6. This problem is about finding the upper and lower bound of
, given
. The same problem can have geometric or algebraic representation.
Video Solution
https://youtu.be/PC8fIZzICFg ~hippopotamus1
Video Solution by Interstigation (In-Depth, Straight-forward)
~Interstigation
See Also
2021 AMC 12B (Problems • Answer Key • Resources) | |
Preceded by Problem 24 |
Followed by Last problem |
1 • 2 • 3 • 4 • 5 • 6 • 7 • 8 • 9 • 10 • 11 • 12 • 13 • 14 • 15 • 16 • 17 • 18 • 19 • 20 • 21 • 22 • 23 • 24 • 25 | |
All AMC 12 Problems and Solutions |
2021 AMC 10B (Problems • Answer Key • Resources) | ||
Preceded by Problem 24 |
Followed by Last problem | |
1 • 2 • 3 • 4 • 5 • 6 • 7 • 8 • 9 • 10 • 11 • 12 • 13 • 14 • 15 • 16 • 17 • 18 • 19 • 20 • 21 • 22 • 23 • 24 • 25 | ||
All AMC 10 Problems and Solutions |
The problems on this page are copyrighted by the Mathematical Association of America's American Mathematics Competitions.