Sinful sum

Algebra Level 5

Let N N be the number of ordered quadruples ( a , b , c , d ) (a,b,c,d) of integers, each of which are from 10 -10 to 10 10 (inclusive), such that { a + b + c + d = 0 sin a + sin b + sin c + sin d = 0. \begin{cases} a+b+c+d & =0\\ \sin a + \sin b+ \sin c +\sin d &=0. \end{cases} What are the last three digits of N ? N?

Details and assumptions

The calculations are done in radians, not degrees.


The answer is 261.

This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try refreshing the page, (b) enabling javascript if it is disabled on your browser and, finally, (c) loading the non-javascript version of this page . We're sorry about the hassle.

4 solutions

Thomas Beuman
Aug 21, 2013

First, we rewrite the second equation:

0 = sin a + sin b + sin c + sin d = 2 sin ( a + b 2 ) cos ( a b 2 ) + 2 sin ( c + d 2 ) cos ( c d 2 ) 0 = \sin a + \sin b + \sin c + \sin d \\ \ \ = 2 \sin(\frac{a+b}2) \cos(\frac{a-b}2) + 2 \sin(\frac{c+d}2) \cos(\frac{c-d}2)

From the first equation, it follows that a + b = ( c + d ) a+b = -(c+d) , hence we get:

0 = 2 sin ( a + b 2 ) ( cos ( a b 2 ) cos ( c d 2 ) ) 0 = 2 \sin(\frac{a+b}2) (\cos(\frac{a-b}2) - \cos(\frac{c-d}2))

For this to hold, one (or more) of three things must be true:

  1. sin a + b 2 = 0 a + b = 0 and c + d = 0 \sin{\frac{a+b}2} = 0 \iff a+b=0 \text{ and } c+d=0
  2. a b 2 = c d 2 a + d = 0 and b + c = 0 \frac{a-b}2 = \frac{c-d}2 \iff a+d=0 \text{ and } b+c=0
  3. a b 2 = c d 2 a + c = 0 and b + d = 0 \frac{a-b}2 = -\frac{c-d}2 \iff a+c=0 \text{ and } b+d=0

In each case, the relation a + b + c + d = 0 a+b+c+d=0 was used. (For example, in the second case, we have a b c + d = 0 a-b-c+d = 0 and a + b + c + d = 0 a+b+c+d=0 . Adding and subtracting these two relations gives the two relations given (times two).)

We now use the inclusion-exclusion principle to count all quadruples that meet at least one of these criteria.

First of all, there are 2 1 2 21^2 quadruples that meet the first criterion: a a and c c can take on any value, after which b b and d d are fixed. The same is true for the second and third criterion, giving a total of 3 2 1 2 3 \cdot 21^2 .

However, all quadruples that meet both the first and the second criterion were counted twice. We should thus subtract them. These quadruples meet the relation is a = b = c = d a = -b = c = -d , of which there are trivially 21 21 . We have the same number of quadruples that meet both the first and the third, and the second and the third criterion. In total, we subtract 3 21 3 \cdot 21 .

That leaves the single quadruple ( 0 , 0 , 0 , 0 ) (0,0,0,0) that satisfies all three criteria. We added it three times, then subtracted it three times, so it needs to be added again.

The net total is thus 3 2 1 2 3 21 + 1 = 1261 3 \cdot 21^2 - 3 \cdot 21 + 1 = \boxed{1261}

Moderator note:

Nicely done!

0 = sin a + sin b = 2 sin ( a + b 2 ) cos ( a b 2 ) 0=\sin a + \sin b = 2 \sin (\frac {a+b}{2})\cos (\frac {a-b }{2}) Similarly for c c and d d . Adding both gives: 0 = sin a + sin b + sin c + sin d = 0 = \sin a+\sin b+\sin c+\sin d\ = (using c + d = ( a + b ) c+d=-(a+b) ) = 2 sin ( a + b 2 ) ( cos ( a b 2 ) cos ( c d 2 ) ) = = 2\sin (\frac {a+b}{2})(\cos (\frac {a-b}{2})-\cos (\frac {c-d}{2})) = (using c d = ( a + c ) + ( b + c ) c-d=(a+c) + (b+c) and a b = ( a + c ) ( b + c ) a-b=(a+c) - (b+c) ) = 4 sin ( a + b 2 ) sin ( a + c 2 ) sin ( b + c 2 ) = 4\sin (\frac {a+b}{2})\sin (\frac {a+c}{2})\sin (\frac {b+c}{2}) which leads to: a = b a=-b or a = c a=-c or b = c b=-c .

Ron van den Burg - 7 years, 9 months ago

@Challenge Master: I wrote a pascal program to solve it and I had a result of 985 Can you point my mistake please? http://www.mediafire.com/?32ijtlgjbw8i4tz

Minh Quân Nguyễn - 7 years, 9 months ago

how are there 21^{2} quadruples that meet the first criterion ?

sreesha kashyap - 7 years, 9 months ago
Pi Han Goh
Aug 19, 2013

We consider three cases:

  • * Case I *: four 0's

  • * Case II *: two 0's

  • * Case III *: no 0

Note that a , b , c , d |a|,|b|,|c|,|d| are positive integers and not a multiple of π \pi , sum of sines of them can never be 0 0 .

Let A A' be the addictive inverse of A A . e.g: if A = 6 A=6 , A = 6 A' = -6

For Case I : a = b = c = d = 0 a=b=c=d=0 , denote 0000 0000 as a solution. 1 solution.

For Case II : 00 A A 00AA' with A = 1 , 2 , 3 , , 10 A = 1,2,3, \ldots ,10 . So the total arrangements is 4 ! 2 ! 10 = 120 \frac {4!}{2!} \cdot 10 = 120

For Case III : A A B B AA'BB' and A A A A AAA'A' are solutions with A B A \ne B .

  • For the subcase A A B B AA'BB' , total arrangements is 4 ! ( 10 2 ) = 1080 4! \cdot {10 \choose 2} = 1080 .

  • For the subcase A A A A AAA'A' , total arrangements is 4 ! 2 ! 2 ! ( 10 1 ) = 60 \frac {4!}{2! \space 2!} \cdot {10 \choose 1} = 60

Hence, N = 1 + 120 + 1080 + 60 = 1261 N = 1 + 120 + 1080 + 60 = 1261 \Rightarrow answer is 261 \boxed{261}

It is not obvious that the sum of sines is non-zero, unless the numbers are paired up to cancel. See Qiang X's solution for the proof of that. However the count part of the problem is nicely done.

Alexander Borisov - 7 years, 9 months ago

Log in to reply

Since we know that only sin ( x ) \sin (x) can be equals to 0 0 when x x is a rational multiple of π \pi , can't we just infer that sum of sines of integers which clearly aren't rational multiple of π \pi (other than 0 0 ) can never be 0 0 ? Am I missing something?

Pi Han Goh - 7 years, 9 months ago

Log in to reply

The fact that the absolute values are positive does not guarantee that their sines are positive. And a sum of several non-zero numbers can easily be zero. As an example, sin 7 5 + sin 19 5 + sin 22 5 = 0 \sin 75^\circ +\sin 195^\circ +\sin 225^\circ=0 (HInt: one can find sin and cos of 1 5 15^\circ by writing 15 = 45 30 15=45-30 ).

Alexander Borisov - 7 years, 9 months ago

Log in to reply

@Alexander Borisov You're setting the angles to be rational multiples of π \pi . 7 5 = 5 12 π 75^\circ = \frac {5}{12} \space \pi , 19 5 = 13 12 π 195^\circ = \frac {13}{12} \space \pi , 22 5 = 5 4 π 225^\circ = \frac {5}{4} \space \pi , so the sum of their sines might be equals to 0 0 . However, when measuring in radians, 1 r , 2 r , 3 r , 1^r, 2^r, 3^r, \ldots are not rational multiples of π \pi . So the sum of sines of at least one of ( 1 r , 2 r , 3 r , ) (1^r, 2^r, 3^r, \ldots ) is nonzero.

This is true because suppose we consider there might be a solution, that is sum of sines of at least one of ( 1 , 2 , 3 , ) (1,2,3, \ldots ) (measured in radians) such that it is zero, that is for certain positive integers: a 1 , , a n a_1, \ldots , a_n

j = 1 n sin ( a j ) = j = 1 n e i a j e i a j 2 \displaystyle \sum_{j=1}^n \sin (a_j) = \displaystyle \sum_{j=1}^n \frac {e^{i \space a_j} - e^{-i \space a_j} }{2 } = 0

Simplify it we get ( e a 1 + e a 2 + + e a n ) = ( e a 1 + e a 2 + + e a n ) (e^{a_1} + e^{a_2} + \ldots + e^{a_n}) = (e^{-a_1} + e^{-a_2} + \ldots + e^{-a_n}) which does not yield any non-zero integer solution.

Pi Han Goh - 7 years, 9 months ago

Log in to reply

@Pi Han Goh Consider three angles: α 1 = sin 1 1 3 , \alpha_1 = \sin^{-1} \frac{1}{3}, α 2 = sin 1 2 7 , \alpha_2= \sin ^{-1} \frac{2}{7}, and α 3 = π sin 1 13 21 \alpha_3=\pi -\sin ^{-1} \frac{13}{21} . Neither of them is a rational multiple of π , \pi, and all of them are positive. Yet the sum of the sines is zero. Of course, these angles are not integer multiples of the same angle, like in the original example, but your original argument does not really use that. Your latest argument can be validated by proving that sin 1 \sin 1 is transcendental, but that is a very deep result.

Alexander Borisov - 7 years, 9 months ago

Log in to reply

@Alexander Borisov I see I see. Thank you!

Pi Han Goh - 7 years, 9 months ago

Typo:

Note that if a , b , c , d |a|, |b|, |c|, |d| are positive integers, then they are not a multiple of π \pi ...

Pi Han Goh - 7 years, 9 months ago

I used this way to find 1261. \\ No. of ways = n ( ( a + b = c + d = 0 ) ( a + c = b + d = 0 ) ( a + d = b + c = 0 ) ) = n ( W 1 W 2 W 3 ) ( s a y ) = n ( W 1 ) + n ( W 2 ) + n ( W 3 ) n ( W 1 W 2 ) n ( W 2 W 3 ) n ( W 3 W 1 ) + n ( ( W 1 W 2 W 3 ) = 441 + 441 + 441 3 ( 21 ) + 1 = 1261. n(( a+b=c+d=0) \cup (a+c=b+d=0) \cup (a+d=b+c=0)) \\ = n(W_{1} \cup W_{2} \cup W_{3}) (say) \\ = n(W_{1}) + n(W_{2}) + n(W_{3}) - n(W_{1} \cap W_{2}) - n(W_{2} \cap W_{3}) - n(W_{3} \cap W_{1}) + n((W_{1} \cap W_{2} \cap W_{3}) \\ = 441 + 441 + 441 - 3(21) +1 = 1261.

jatin yadav - 7 years, 9 months ago

I think you should work more on the sum of sines part, not the couting part

Cuong Doan - 7 years, 9 months ago
Qiang Xiao
Aug 18, 2013

Since a + b + c + d = 0 a+b+c+d=0 , we have a + b = ( c + d ) a+b=-(c+d) ; hence sin ( a ) + sin ( b ) + sin ( c ) + sin ( d ) = 2 sin ( ( a + b ) / 2 ) cos ( ( a b ) / 2 ) + 2 sin ( ( c + d ) / 2 ) cos ( ( c d ) / 2 ) = 2 sin ( ( a + b ) / 2 ) [ cos ( ( a b ) / 2 ) cos ( ( c d ) / 2 ) ] = 2 sin ( ( a + b ) / 2 ) 2 sin ( ( a b + c d ) / 4 ) sin ( ( a b c + d ) / 4 ) \sin(a)+\sin(b)+\sin(c)+\sin(d)=2\sin((a+b)/2)\cos((a-b)/2)+2\sin((c+d)/2)\cos((c-d)/2)=2\sin((a+b)/2)[\cos((a-b)/2)-\cos((c-d)/2)]=2\sin((a+b)/2)*2\sin((a-b+c-d)/4)\sin((a-b-c+d)/4) ;

Noting that a , b , c a,b,c ,and d d are integers, we obtain that a + b = c + d = 0 a+b=c+d=0 , or a + c = b + d = 0 a+c=b+d=0 , or a + d = b + c = 0 a+d=b+c=0 . And the total number of solutions is 1261;

This solution should be viewed together with the Pi Han's solution, that explains the count of 1261.

Alexander Borisov - 7 years, 9 months ago
Bill Bell
Apr 30, 2016

Before you're tempted to disdain a computer solution notice that it exposes a lovely pattern. Generate all possible tuples then look for those where values 'cancel'. No need to calculate values of the sine function.

1
2
3
4
5
6
7
8
9
count=0
for a,b,c,d in ((a,b,c,d) for a in range(-10,11) \
    for b in range(-10,11) for c in range(-10,11) \
    for d in range(-10,11) if a+b+c+d==0):

    m,n,p,q=sorted([a,b,c,d])
    count+=m==-q and n==-p

print (count)

Yes, computer programs are useful for identifying the solutions, after which we should still use our grey matter to verify that these statements are indeed true/false.

Calvin Lin Staff - 5 years, 1 month ago

Log in to reply

I sometimes wonder who you think your audience is.

Bill Bell - 5 years, 1 month ago

@Bill Bell The programming language you used looks like Python but it has an unfamiliar structure. It worked perfectly (unaltered) on cloud.sagemath.com - thanks for including the code here !!

Bob Kadylo - 5 years, 1 month ago

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...