Roboant

Suppose you have a single tube with 30 roboants on one end and 20 roboants on the other.

All the roboants on each side start out marching in a line at a uniform pace towards the other end of the tube. When two roboants bump into each other, they instantly turn around and walk back the way they came.

How many total ant collisions will there be?

The animation is courtesy TED-Ed and Artrake Studio. The animation is courtesy TED-Ed and Artrake Studio.

Note : When the roboants reach either end of the tube, they exit the tube.


The answer is 600.

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.

15 solutions

Zain Majumder
Oct 7, 2018

Instead of imagining the roboants colliding each other, imagine them passing through each other. We can do this because the result is the same: two roboants come in facing towards each other, and they leave facing away from each other.

This version of the problem is much simpler. Each of the 30 30 roboants pass by the 20 20 roboants on the other side, for a total of 30 × 20 = 600 30 \times 20 = \boxed{600} collisions.

Elegant solution, same.

Kelvin Hong - 2 years, 8 months ago

Log in to reply

If the ants turn around ...after a collision .... Won't they hitting the ants on their own side after the first collision....and last of the ants on both sides will fall...and process repeats with left 48 ants in the tube ........then leading to a time where 20-20 of ants would have fallen from both the sides and the rest 10 will fall without any collision.!..... Leading to my point that there will be 609 collisions.. precisely...

Yash Jain - 2 years, 5 months ago

Just for fun : here is a simulation of all the collisions to occur. Each "R" represents an ant going from left to right, and L for the ants going from right to left. A collision occurs when we get RL, which makes those two ants become LR. Interesting to see the propagation of the waves of collisions. At the end, we check the 600 collisions.

RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRLLLLLLLLLLLLLLLLLLLL 1 RRRRRRRRRRRRRRRRRRRRRRRRRRRRRLRLLLLLLLLLLLLLLLLLLL 2 RRRRRRRRRRRRRRRRRRRRRRRRRRRRLRLRLLLLLLLLLLLLLLLLLL 3 RRRRRRRRRRRRRRRRRRRRRRRRRRRLRLRLRLLLLLLLLLLLLLLLLL 4 RRRRRRRRRRRRRRRRRRRRRRRRRRLRLRLRLRLLLLLLLLLLLLLLLL 5 RRRRRRRRRRRRRRRRRRRRRRRRRLRLRLRLRLRLLLLLLLLLLLLLLL 6 RRRRRRRRRRRRRRRRRRRRRRRRLRLRLRLRLRLRLLLLLLLLLLLLLL 7 RRRRRRRRRRRRRRRRRRRRRRRLRLRLRLRLRLRLRLLLLLLLLLLLLL 8 RRRRRRRRRRRRRRRRRRRRRRLRLRLRLRLRLRLRLRLLLLLLLLLLLL 9 RRRRRRRRRRRRRRRRRRRRRLRLRLRLRLRLRLRLRLRLLLLLLLLLLL 10 RRRRRRRRRRRRRRRRRRRRLRLRLRLRLRLRLRLRLRLRLLLLLLLLLL 11 RRRRRRRRRRRRRRRRRRRLRLRLRLRLRLRLRLRLRLRLRLLLLLLLLL 12 RRRRRRRRRRRRRRRRRRLRLRLRLRLRLRLRLRLRLRLRLRLLLLLLLL 13 RRRRRRRRRRRRRRRRRLRLRLRLRLRLRLRLRLRLRLRLRLRLLLLLLL 14 RRRRRRRRRRRRRRRRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLLLLLL 15 RRRRRRRRRRRRRRRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLLLLL 16 RRRRRRRRRRRRRRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLLLL 17 RRRRRRRRRRRRRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLLL 18 RRRRRRRRRRRRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLL 19 RRRRRRRRRRRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRL 20 RRRRRRRRRRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLR 20 RRRRRRRRRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRR 20 RRRRRRRRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRRR 20 RRRRRRRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRRRR 20 RRRRRRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRRRRR 20 RRRRRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRRRRRR 20 RRRRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRRRRRRR 20 RRRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRRRRRRRR 20 RRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRRRRRRRRR 20 RLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRRRRRRRRRR 20 LRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRRRRRRRRRRR 19 LLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRRRRRRRRRRRR 18 LLLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRRRRRRRRRRRRR 17 LLLLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRRRRRRRRRRRRRR 16 LLLLLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRRRRRRRRRRRRRRR 15 LLLLLLRLRLRLRLRLRLRLRLRLRLRLRLRLRLRRRRRRRRRRRRRRRR 14 LLLLLLLRLRLRLRLRLRLRLRLRLRLRLRLRLRRRRRRRRRRRRRRRRR 13 LLLLLLLLRLRLRLRLRLRLRLRLRLRLRLRLRRRRRRRRRRRRRRRRRR 12 LLLLLLLLLRLRLRLRLRLRLRLRLRLRLRLRRRRRRRRRRRRRRRRRRR 11 LLLLLLLLLLRLRLRLRLRLRLRLRLRLRLRRRRRRRRRRRRRRRRRRRR 10 LLLLLLLLLLLRLRLRLRLRLRLRLRLRLRRRRRRRRRRRRRRRRRRRRR 9 LLLLLLLLLLLLRLRLRLRLRLRLRLRLRRRRRRRRRRRRRRRRRRRRRR 8 LLLLLLLLLLLLLRLRLRLRLRLRLRLRRRRRRRRRRRRRRRRRRRRRRR 7 LLLLLLLLLLLLLLRLRLRLRLRLRLRRRRRRRRRRRRRRRRRRRRRRRR 6 LLLLLLLLLLLLLLLRLRLRLRLRLRRRRRRRRRRRRRRRRRRRRRRRRR 5 LLLLLLLLLLLLLLLLRLRLRLRLRRRRRRRRRRRRRRRRRRRRRRRRRR 4 LLLLLLLLLLLLLLLLLRLRLRLRRRRRRRRRRRRRRRRRRRRRRRRRRR 3 LLLLLLLLLLLLLLLLLLRLRLRRRRRRRRRRRRRRRRRRRRRRRRRRRR 2 LLLLLLLLLLLLLLLLLLLRLRRRRRRRRRRRRRRRRRRRRRRRRRRRRR 1 LLLLLLLLLLLLLLLLLLLLRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR 0 Total 600 collisions Gerard.

Gerard Boileau - 2 years, 8 months ago

Log in to reply

This solution means that, at the end, all ants coming from the left will end at he right side and vice-versa. In my view, when the ant number 20 in the grupo of 20 collides with ant number 19 that came back from its first collision, it will never returns.So, the last ant will have just one collision. The ant 19 will have 2 collisions, and so on. The ants in the group of 20 will have 1=2+3+.. collisions, which gives 210. All the ants in the group of 20 will end returning from where they came.I'm working now on what happens with the group of 30.

Arsenio Meneses - 2 years, 8 months ago

Log in to reply

No, it simply means that all the ants will be facing towards the exit at the end. The ants never actually change order.

Zain Majumder - 2 years, 8 months ago

Log in to reply

@Zain Majumder Thanks for the reply. But I still think that the last ant (number 20 ) in the group of 20, when has its first collision with ant 19 ( who had a collision with ant 18 .... ) of its group it will turn back and will never return to its original direction. Am I wrong ?

Arsenio Meneses - 2 years, 8 months ago

Log in to reply

@Arsenio Meneses You're right. As you can see in G B's comment, the last ant is facing left half the time, until it hits the second-to-last ant. Then it faces right the rest of the time.

Zain Majumder - 2 years, 8 months ago

If you draw out a gross simplification on a piece of paper with just 4 heading right and 2 heading left, you do end up with 4 at the right, and 2 at the left, but the two at the left started at the left, and of the 4 at the right, 2 started at the right and 2 started at the left. If you do with with 4 heading right and 3 heading left, it's a similar result; 4 end up at the right, and 3 at the left. The 3 originally heading left ending at the right, where they started, plus 1 more from the left. The 3 at the left started at the left. Without wanting to go to any more trouble, my best guess (I assume I can call it induction) is that with 30 heading right and 20 heading left, you will end up with 30 at the right and 20 at the left. The 20 at the left will have started at the left. Of the 30 at the right, 20 will have started there, 10 will have come from the left.

Jon Slack-Smith - 2 years, 8 months ago

After RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRLLLLLLLLLLLLLLLLLLLL 1 Should be RRRRRRRRRRRRRRRRRRRRRRRRRRRRRLRLLLLLLLLLLLLLLLLLLL 2

Giles Brennand - 2 years, 8 months ago

<?php

$ants = array();
$collisions = 0;
$out = 0;

// build the ants
for ($i = 1; $i<=30; $i++)
{
    // position, direction
    $ants[$i] = array(2*($i-1),1);
}

for ($i = 31; $i<=50; $i++)
{
    // position, direction - direction -1 = out of tube
    $ants[$i] = array(2*($i-1),0);
}

while ($out < 50)
{
    // Move all the ants
    for ($i = 1; $i<=50; $i++)
    {
        if ($ants[$i][1]) // if direction is 1
        {
            $ants[$i][0]++; // increment position by 1
        }
        else
        {
            $ants[$i][0]--; // decrement position by 1
        }
    }

    // Check for any falling out of tube...
    for ($i = 1; $i<=50; $i++)
    {
        if (($ants[$i][0] < 0) or ($ants[$i][0] > 100))
        {
            $ants[$i][0] = -1;
            $out++;
        }
    }

    // check for collision, count and reverse
    $lastpos = 1000;
    for ($i = 1; $i<=50; $i++)
    {
        if ($ants[$i][0] == -1)
        {
            // ignore it
        }
        else if ($ants[$i][0] == $lastpos)
        {
            // collision
            $collision++;

            // reverse both ants
            if ($ants[$i][1] == 0)
            {
                $ants[$i][1] = 1;
                $ants[$i-1][1] = 0;
            }
            else
            {
                $ants[$i][1] = 0;
                $ants[$i-1][1] = 1;                 
            }

            // reset
            $lastpos = 1000;
        }
        else
        {
            // No collision
            $lastpos = $ants[$i][0];
        }
    }
}

echo $collision;

?>

Jeff Snoxell - 2 years, 8 months ago

Log in to reply

IS this python, java? What language?

Taras Berestyuk - 1 year, 5 months ago

The correct has to be infinite. Hen they turn around they bump into the an which was behind them and thus must turn around ad infinitum.

Michael Moring - 2 years, 8 months ago

Log in to reply

It qualified that if they walked to the end, they walked off so that eliminates the infinite possibility.

Sherry Griffin - 2 years, 8 months ago

It is not infinite because when the last ant from the side of 20 has its first collision it turns back and never comes back again

Arsenio Meneses - 2 years, 8 months ago

It states that when an ant reaches the end of the tube it marches out, regardless of which end of the tube it's at, so the two ants at the extreme ends of the tube only collide once. They then turn around and there are no ants to collide with between them and the ends of the tube.

Nick Pretzel - 2 years, 8 months ago

The leave the tube once the hit the end, so, therefore, there cannot be infinite possibilities.

Taras Berestyuk - 1 year, 5 months ago

Perhaps I’m reading the question incorrectly. If we number the ants in each line 1-20 (we’ll ignore that there are 30 on one side for now) starting with the ants on the end being 1 and the middle ants being 20, the ants will bump their position number x2-1. For example, last ant on right will walk L and bump into his own ant to turn R and then walk off (1x2-1=1). Ant in position 2 in right line will go LRLR for 3 bumps (2x2-1=3). If we continue, we find the ant in the 20th position will bump 39 times (this is bumps into the opposing line ants plus ants in his own line). If we sum the sequence (39+1)20/2 = 400 and that is just the bumps for 1 side. Multiply that by 2 and we get 800. Accounting for the 2 ants in the 20 position bumping into each other and removing their 20 collision double count, we have 780 and that is before considering the additional 10 ants we ignored at the beginning.

Where am I going wrong?

Sherry Griffin - 2 years, 8 months ago

Log in to reply

You have counted each bump twice, but only removed the twenty last bumps? Each and every bump is between two ants.

Ole Andersen - 2 years, 8 months ago

Ok not sure I understand the question. Because if any 1 in each direction bump into each other they turn around. Then any 2 in each lines does the same. Until ant 20 in each line collides there is always another ant available to collide. But after 20 collisions, there are 10 ants going in one direction but none more for them to collide with.

Graham Courtney - 2 years, 8 months ago

Log in to reply

The ends are hollow which means the ants will exit. Cutting out chances of infinite looped collisions.

Nachiketha B N - 2 years, 8 months ago

I like the simplicity of your solution, but I would explain the base case deeper.

Federico Bicciato - 2 years, 8 months ago

Make it simpler. Assume 3 ants and 2 ants marching. We end up with 6 collisions here.(3×2=6; this is not the actual proof.) Multiplying 3 ×10 = 30 and 2×10 =20 we would have 30×20 = 600.... collisions.

Nachiketha B N - 2 years, 8 months ago

I’m confused. Let the first roboants be L1 and R1. When they meet they both turn. Then R1 marches straight into R2 and he turns again to meet L1 again, who has similarly met L2 and turned? Etc...?

Roy Stone - 2 years, 8 months ago

Log in to reply

Yes, this is what actually happens. What I'm saying here is that it would also look exactly the same if you imagined the ants passing through each other. If instead, R1 passed L1 into L2, and L1 passed R1 into R2, then the situation would be different, but since it doesn't matter which ant is which when counting collisions, the result would still be the same.

Also remember that all the ants would eventually leave since the tube's ends are open.

Zain Majumder - 2 years, 8 months ago

We can create a smaller model of the problem, lets say 3 (numbers 1,2,3) coming from one side and just 1(number 4) coming from the other side. The first collision is 3 and 4. 3 returns to collide with 2 and 4 returns and never comes back . ( we have 1 collision) . Now 3 collides with 2, returns and never comes back, since 4 is gone forever ( another collision) . 2 returns and collide with 1, Both returns and never collide) (collision number 3). Total 3 collisions. If we do this again for another model, lets say 4 and 2, we will see that the result is 8. Bottom line: the result is always the product of the numbers for group 1 times group 2. In our case, 30 x 20 = 600 .

Arsenio Meneses - 2 years, 8 months ago

Log in to reply

Of course, you haven't proved that the pattern is always true. My solution (which you just commented on) can be generalized to prove that pattern.

Zain Majumder - 2 years, 8 months ago

I think you capture the essence of the information imparted, and this is interesting, and worth considering. But you do not capture the path, which is interesting as well. The path I believe is an oscillating wave of right and left orientations -- no ants actually pass another; and they are in precisely the order they were when the process began with orientation, but for the middle 10, reversed. It comes to as those marching right are now on the right and those marching left are now on the left.

Another feature in this set up not remarked upon is the need for the turns to precisely coordinated, otherwise the wave character of this function would collapse into chaos.

cornelius cooper - 2 years, 8 months ago

Log in to reply

I agree with this comment. If the bumps won't be coordinated to create a smooth turning-arounds, the modek would collapse into chaos.

Francis Dave Cabanting - 2 years, 8 months ago

THERE WILL BE AN INFINITE NUMBER OF COLLISIONS. THE FIRST ANT TO COLLIDE TURNS AROUND! HE THEN COLLIDES WITH A BROTHER AND TURNS AROUND. NO ANT DISAPPEARS.

THIS PROBLEM SUCKS!

Bob Heasley - 2 years, 8 months ago

Log in to reply

Try modeling this with two ants from each direction. The ants in the back will each collide once, the ones in the front will collide 3 times (with the opposing ant, then the one behind it, then the opposing ant again.) The total number of collisions is 2x2=4.

Jeremy Galvagni - 2 years, 8 months ago

Don't for get the note at the bottom of the problem, "When the roboants reach either end of the tube, they exit the tube."

Jason Wu - 2 years, 8 months ago

They fall out of the end of the tube, be careful to read to the very end.

Taras Berestyuk - 1 year, 5 months ago

The first middle bump will actually cause the next ant to bump the back of the ant in front and so on and so on, since they are in line, causing all ants to turn around and walk away.

Saleem Samsaud - 2 years, 8 months ago

a wonderful proof!!!

zhihao luo - 2 years, 8 months ago

Smart imagination

Rafsan Mahim - 2 years, 8 months ago

Could you please explain why the passing through scenario is equivalent to the colliding one?

A Former Brilliant Member - 2 years, 8 months ago

Log in to reply

If two ants walking in opposite directions into each other collide, and then start walking the opposite direction, it's literally exactly the same as them phasing through each other, just imagine it. Both ways, before the point of contact, there's an ant walking left and an ant walking right; After the point of contact, there's still just an ant walking left and an ant walking right, it's as if they just walked past each other. I love that I'm not the only one who thought about this like this.

Milenko Milicevic - 2 years, 8 months ago

Log in to reply

Obviously ... Clever solution!

A Former Brilliant Member - 2 years, 8 months ago

Wow wasn’t insightful enough to thought about it this way

Tony Yin - 2 years, 8 months ago

Look at the animation above. Hold up a few fingers to hide the center of the screen. You can now really imagine the two ants passing through each other since you can't see them turning around.

Jeremy Galvagni - 2 years, 8 months ago

I had assumed that at first you have 1 collision. But when a roboant turns back it hits the ant that was originally behind it (so now there’s two collisions - one on each side). On turning back the first two collide and the second in each row collide with the third in each row (3collisions). Etc... so I was thinking along the lines of 20!

Except once 20 from each side have collided there’s still 10 to go on one side. In the row of 20 one ant will turn away and keep going (no more collisions on that side)... so you have 10x20 collisions before they start to disperse again... so I made it 20! + 20x10 + 20!

I guess I read the question differently to how the author indented it to be interpreted it.

Marleen Naessens - 2 years, 8 months ago

Log in to reply

The reasoning in your first paragraph is correct, except it doesn't add up to 20!, but rather the SUM of 1+2+...+19+20, which is 10x21=210 on each side, which makes 420. This would be the final answer if there were 20 ants on each side. Since we had 30 ants on one side, we need to add an additional 18x10=180 to exhaust the extra 10 ants from the side that had 30. (It's not 20x10 because we already counted the 20th ant on each side.) Then, 180+420=600 collisions.

Alfred Mondelli - 2 years, 8 months ago

Real ants can lift a lot of weight relative to their own weight.

Jesse Otis - 2 years, 8 months ago

Amazing imagination

Akela Chana - 2 years, 8 months ago

Captivating!

Khushaal Nandwani - 2 years, 8 months ago

A truly * brilliant * answer.

Akshay Kumar Jha - 2 years, 8 months ago

I read it as the ants would continue to turn each time it collided with another ant, with one ant leaving the tube until all 20 from one side exited, letting the remaining 10 from the left all leave at once. I came up with 1101 collisions.

Rich Finnigan - 2 years, 8 months ago

Log in to reply

I’m not sure how you get 1101, but agree with your analysis that they keep turning around into each other

Bob Bolton - 2 years, 8 months ago

That's very neat and elegant, well done! I must be getting rusty, as I tried to find a formula by working with 2 & 3 ants, then 4 & 6, then, just to make sure my formula held, with 8 & 12. I did come up with the right answer, but much more laborious.

Nick Pretzel - 2 years, 8 months ago

I don't agree with solution how comes there is only 20 ants on side the other side is only 30 ants , the solution should be 20*20=400 how comes the other 10 collided while all ants on the other side exited ???

Ahmad Tajir - 2 years, 7 months ago

The solution seems a bit more complexed then 20x30. If you work from the 20 side the first any would cause 20 collisions before 1 could leave the way it came, the second would cause 1 less therefore 20! On the other side it would be 30!-10! because once the 20 from the other side had passed the final 10 would just walk through. 20!+30!-10!=2.652528E32

Trevor M - 2 years, 7 months ago

I did 20x30 and got 600 than I did 600x20. Now I'm just confused

Mane Azaryan - 2 years, 5 months ago

Instead of maybe doing this we can also try thinking about a single sequence of collisions. This can be done by thinking that if the first ants collide thereby causing a chain reaction leading to the very end any of each line to go back. Therefore there will be 1+19+29 collisions. The next collision between the first one will lead to 1+18+28 collisions and so on. The pattern becomes 49,47,45.... and will continue until the shorter line becomes zero which is at 11(1+0+10). Now by simply using the sum of arithmetic series given by (n/2) (a+l) where n is the no. of terms(=20), a is the first term(=49), and l is the last term(=11). Therefore the sum is 10 60 which is 600.

Bhavya Mehta - 2 years, 5 months ago

There is an easier way

You could just multiply both the ants in the left side with the ants in the right side. 30×20=600

Simreen Siraj - 2 years ago

Instead of imagining them bumping into each other, imagine them phasing through each other, since both are equivalent. Now consider a simpler situation: 30 robo ants and 1 robo ant. This 1 robo ant will have to go phasing through each of the 30 ants totalling 30 times, hence 30 collisions. If it's 30 and 20 ants, each of the 20 ants will be phasing through the 30 ants giving 30 collisions for each of the 20 ants. Hence it's 30x20=600 collisions.

Santhosh Paripalli - 1 year, 10 months ago

I used my fingers for this one.. 1 ant at eachs side equals 1 collision, 2 ants 4 collisions, 3 I get 9... so far easy to count and easy to spot the pattern... I just square the number of ants assuming I have equal number of ants at each end... so far I get 20 square. But what about the remaining 10, as it was 30 not 20 that were coming from the left... well now I had a situation where I get 20 marching to the left and the 10 going to the right (this from the group of 30)... I will just add 10 square to my previous result... But i'm back at the situation that as I have yet another group of 10 ants remaining... so the final collision is 2 groups of 10, I add yet again 10square... this gives me the correct answer 100+100+400... I know this way is harder, but it was not intuitive for me to imagine the ants going through each other and the result was the same

David Galavis - 1 year, 2 months ago

I think it's a good idea to solve the problem recursively. The total number of bumps is the number of bumps of the left group plus the bumps of the right group. From each group, the total bumps is the bumps of the outermost ant (1 only) plus the bumps of the inner group. Always keeping in mind that you don't have to double count bumps.

First, I count the bumps of the right group (20 ants). I start from the rightmost ant. The 20th one bumps 1 time with the 19th. The 19th bumps 1 times with the 20th (but that's counted already) and 2 times with the 18th: so it's 2 bumps to add. So on till the 1st one of the right group, that bumps 20 times with the last of the left group. Total number of bumps of the right group is the sum of i from 1 to 20, that is 210.

Then, left group. Starting from the leftmost ant, called 1st. 1st bumps once with 2nd. Then, 1 bump 2nd bumps once with 1st (already counted), twice with the 3rd. Then, 2 more bumps. So on till the 20th, bumping 20 times.

From ant 21th to 30th of the left group, they bump 20 more times each, since the left group goes away after 20 bumps. 21th bumps into 20th 20 times (already counted), then 20 times with 22th, that turns right afterwards. Then 20 more bumps. 22th does 20 more bumps vs 23th. So on till 29th, with 20 more bumps into 30th. 30th bumps with 29th (already counted ) and the 1st of the right group (already counted). So doesn't make new bumps. Total number of bumps of the right group: sum of i from 1 to 20, plus 9 times 20: then it's 210+180 = 390. Total bumps: left group bumps + right group bumps = 600.

So this is the number of bumps of the right inner group and left inner group when they are turning back, but what about the bump of the outermost roboants? Wouldn't that make 601? (disclaimer:I'm not pretty good at this)

Alexandra Capusan - 2 years, 8 months ago

Log in to reply

Vnkjfsdgjkhvxssdghjh

SHIVAM VISHWAKARMA - 2 years, 7 months ago

I like the approach of solving this recursively. I began with simple versions of the problem - 1 ant at each end - then 2, then 4. This yielded a simple solution of n^2 where n is the number of ants at each end. It gets a bit more complicated without equal numbers of ants at each end. I think the formula n^2 - n*(n-m) where m is the number of ants at the opposite end of the tube and m < n offers a general solution to the problem. Developing proofs are not my strong point, but it seems to work :)

Larry Allan - 2 years, 8 months ago

Log in to reply

@Larry Allan Look that n 2 n ( n m ) = n 2 n 2 + n m = n m n^2 - n(n-m) = n^2 - n^2 + nm = nm Look at @Zain Majumder 's solution for an easy explanation of why this happens. Hope this helps...

Javier Álvarez - 2 years, 7 months ago

I came very close to the 600 collisions. But I found only 595. Where is my mistake? The leftmost and the rightmost ant only undergo 1 collision each, the next inner ant will have 3 collisions and so on until ant no. 20 on each side has participated in 39 collisions. Now here's where I am struggelling. How many collisions do ants no. 21 to 30 on the left side (L21 - L30) perform? It should be 39 too, right? If so, the total number of collisions I counted is 1190 = (1+3+5+...+37+39) 2+39 10. Since I counted the collisions for each ant seperately, the actual number is of course only half of that, so 595. If on the other hand ants L21 to L30 perform 41 collisions, then the total number of collisions would be 1210 if counted for each ant and and 605 total. So i guess to get the 600, you need 40 collisions for ants L21 to L30, but I can't figure out how they can do 40 collisions.

Marcellus Wallace - 2 years, 8 months ago

Log in to reply

This is what makes the problem a bit more interesting. They actually collide 40 times, not 39 (qv separate entry with proof, although the proof assumes that the collisions in the larger group are always 2 times the number of the smaller group for the extra ants. I came to that conclusion by a number of trials with smaller groups; 2 & 3, 4 & 6 and 8 & 12). That's where you've lost five collisions: 10×39=390 390÷2=195 (as we've counted each collision twice). There's always a danger in making assumptions. I'm adding a separate post with a proof that the answer will always be the product of the two numbers of ants (ie 20×30 in this case.

Nick Pretzel - 2 years, 8 months ago
Jasper Eenhoorn
Oct 9, 2018

Due to a friend: start with one ant crossing 20 ants. They will have 20 collisions. Then add another ant. After analysis: they will have 40 collisions. Omitting the induction proof, the answer is 30*20=600.

Found problem very unclear. Dont each of lead robots, once they get turned around then face the line of their fellow ants . Dont the two lead robots then bump immediately into the ants following them, and then get turned around again to bump each other. Doesnt this go on ad infinitum. Instead of imagining hollow ants, think about one lane of automobile traffic with cars proceeding in both directions.

peter hakim - 2 years, 8 months ago

Log in to reply

It doesn't go on ad infinitum because once the last ant on a side is turned around they keep on walking away from the tube. (see the Note: When the roboants reach either end of the tube, they exit the tube.).

On each side the ant at the end of tube (Ant R1 and L21) will be turned round once. The ant in front (say R2) will be turned round three times before exiting the tube: Once by R3, once by R2 (to the original direction of travel), and then by R3 again. R3 will be turned five times.

So if there were 20 ants on each side the number of collisions would be 2x(1+3+5+7+ ... +39) = 20x40 = 800

The extra ten ants on one side are similarly turned 2x(1+3+ ... +19)= 10x20 = 200

Total number of collisions is 1000.

If this is correct, it shows the benefit of starting with the simplest situation (An ant at the end of the tunnel), rather than the more obvious starting point (the ants at the head of each column) who are in the first collision), but which then have many other collisions.

Giles Brennand - 2 years, 8 months ago

Log in to reply

'once by R2 (to the original direction of travel)' should be 'once by R1 (to the original direction of travel').

Giles Brennand - 2 years, 8 months ago

The correct answer is 600. There seem to be a few mistakes here, for example the first equation overcounts by a factor of two, and you assumed that you can just "add on" ten ants without affecting the states of the remaining ants.

Zain Majumder - 2 years, 8 months ago

Log in to reply

@Zain Majumder But the extra 10 ants don't really affect the states of the remaining ants.

You're right about the initial overcount, it added up all the collisions for each any, but in fact each collision was counted for each of the 2 ants it involved, that's where the "factor of 2" comes from. But fixing this, the result is 400, and the difference is exactly the 200 from the extra 10 ants.

C . - 2 years, 8 months ago

Log in to reply

@C . If you add the ten ants to the end, then it will take longer for the ants in the middle to leave. If you want to solve the problem this way, you have to take this factor into account.

Zain Majumder - 2 years, 8 months ago

Log in to reply

@Zain Majumder That's the thing: you don't add the extra ants at the end, you add them in the middle, and they would be changed exactly 20 times backwards and 20 times back to how they were originally. So 20 new collisions per each new ant. Times 10 gives that 200 i mentioned.

C . - 2 years, 8 months ago

It would be ad infinitum if there were infinitely many ants on one or both sides. But there isn't, so there has to be a finite amount of collisions.

Matthew Smith - 2 years, 8 months ago

Yes, I also think it was a very unclear problem. I thought they must have meant that as ants came toward each other and collided, both ants could both turn around and exit the tube without colliding with other ants, so for my simple understanding of the problem, I only thought there would be 20 collisions.

Carlton Schroeder - 2 years, 8 months ago

I'm not sure but aren't ant collisions limited by the lowest common number of ants? After colliding evenly all ants will be able to move without obstruction with 30 robo ants moving to the right and 20 moving to the left? thereby only accounting for 20 times 20 collisions.....

Alex Dreyer - 2 years, 8 months ago

Log in to reply

No, the number of collisions is not that small, because after "colliding evenly", there will indeed be 30 ants moving to the right and 20 to the left, but only 20 of the ants moving to the right will have their path cleared. There will be 10 more that haven't yet collided. Think of these 10 as being late to the meeting, and the early ants already collided evenly somewhere to the right.

C . - 2 years, 8 months ago

Log in to reply

i thought more of it as a bouncing particle...that only bounces when there is something behind And in front of it..... without both sides blocked all "particles can move without bouncing two way allowing an open direction....your explanation only makes sense if the ants aren't moving at the same speed...could you help me understand plz

Alex Dreyer - 2 years, 8 months ago

Log in to reply

@Alex Dreyer They're moving at the same speed in my visualization too, but think of what would happen if they are not evenly spaced... Basically this: Left-most 10 ants are 1 inch apart, then there's a foot (or a mile) of free space, then there's 20 more ants 1 inch apart, all of these facing to the right... towards the incoming formation of 20 ants 1 inch apart, facing left. The 20 vs 20 "match" will be decided (evenly) well before the last 10 ants will meet the 20 that are heading their way.

C . - 2 years, 8 months ago

Log in to reply

@C . im sorry but how did you just confuse me? space isnt relavent here and as far as the question allows the area isnt blocked so how do away facing ants collide with the center ten ants......? O0o

Alex Dreyer - 2 years, 8 months ago

Log in to reply

@Alex Dreyer Space is only relevant to allow for the extras to not be close enough until the main 20v20 have had all their colliding fun:

...>.>.>.>.>.....................................>.>.>.>.>...<.<.<.<.<...

..............................>.>.>.>.>..........<.<.<.<.<...>.>.>.>.>...

Umm, why restrict the center to just 10 ants?

C . - 2 years, 8 months ago

Log in to reply

@C . not restricting canceling once their path is unobstructed they are done colliding so i canceled them out

Alex Dreyer - 2 years, 8 months ago

Log in to reply

@Alex Dreyer But the collisions don't happen from the outside inwards, they happen from the inside outwards. If you look at a group of ants, you need to have it collide with an equally-sized group or it won't be the case that all ants are now facing away from their initial direction... If you decide to have "the center ten ants" grouped-up, you will need them to go againts other 10 ants... Which will leave quite a lot more "on the table" to collide afterwards.

But, even so, it's not incorrect. However, let's simplify things: group up 10 ants at a time, so now we have ants A B and C going towards the right, and ants D an E going towards the left.

...A>...B>...C>..........<D...<E...

Can you tell me which ants collide, and how will they end up after all the bumps are done?

C . - 2 years, 8 months ago

Log in to reply

@C . 2 will exit left and 3 will exit right after 6 collisions. thank you for helping by reducing the question i didnt do that my bad. i simulated it using headache tablets. But i have to ask concerning the center collisions in the original question, when bouncing off each other wont they evenly split in both directions 5 one way 5 the other way?

Alex Dreyer - 2 years, 8 months ago

@C . Thankx to you simplifying the first problem i used sets to do the same thus 6 v 4. in step by step the collisions balance out as if they go through but balances 1 ,2 ,3 , 4 ,4 ,4 ,3 ,2 ,1 these are the number of collisions step by step balancing both sides thnx for helping me C

Alex Dreyer - 2 years, 8 months ago

Not exactly. If two ants collide and go back the way they came, it'd be the same thing as saying two ants go through each other, there is no difference when it comes to two ants meeting each other at the same point. You can start counting each ant on the left (30 of them) going through each ant on the right (20 of them). Each left ant 'goes through' 20 right ants, and there are 30 left ants, so when you add the collisions/passes you're essentially adding 20+20+20+...+20 thirty times. Which is exactly like saying 20×30. Alternatively you could start with the right 20 ants, and each of them goes through 30 ants. 30+30+30+...+30 twenty times. Which is the same as saying 30×20. This yields the same answer as 20×30 due to the commutative property of multiplication.

Matthew Smith - 2 years, 8 months ago

Log in to reply

uhhm not quite ......going through implies equal distribution. Im asking why the un even ten ants would turn around 360 degrees without obstruction from both sides. As i understand the question a remainder is supposed to leave in the same direction that they originally went .For example. Ants will bounce off each other like pool balls with 0 mass for instance, once the rear 20 from the 30 side have turned around then the opposite twenty will also have turned around leaving 10 remainder if they move at an equal speed of course. distance doesn't matter either if they are moving at the same speed.

Alex Dreyer - 2 years, 8 months ago

Log in to reply

@Alex Dreyer The reason that the rear 20 have turned around was because they bumped into the center 10... Which were flipped because they bumped into the opposite twenty.

Think of the 2v1 case: call them ants A B and C, B and C colliding and bouncing is quite similar to just B and C going through each other (and changing labels). Then A will collide, which again is just like going through.

C . - 2 years, 8 months ago

Log in to reply

@C . the center ten do carry the original 20 collisions but then their path is open not obstructed....which is where i'm confused

Alex Dreyer - 2 years, 8 months ago

Log in to reply

@Alex Dreyer Okay then... Let the leftmost 20 ants fall behind and not take part in the initial "round". Which will be the center 10 you're concerned with, going towards the right against the 20 going left.

After all the collisions are done, those 10 will be going towards the left, and will meet the 20 "late" ants, and behind those 10 will be the first 10 that were going left... which had collided with 10 on either side, and these collisions cancelled out, so they are now still heading towards the left. And the remaining 10 ants (those that were further towards the right in the initial setup) are the ones going right and not facing any opposition.

C . - 2 years, 8 months ago
Badruz Choudhury
Oct 10, 2018

Well, in my case, it turned out to be the sum of an arithmetic progression. Just imagine, after the collision, the ant turned around and collided with his fellow ants, coming from his end without turning . Now, for the first pair of ants, no. of collisions =1+19+29=49, and for the second pair of ants, it will be =1+18+28=47. Similarly proceeding, we get an AP: 49, 47, 45, ... with common difference -2. It will have 20 terms. So, we can find the total no. of collisions by finding the sum of this AP, by using the formula, S = n/2(2a+(n-1)d) =10(98-38) =600

Arthur Pons
Oct 8, 2018

We have 20 roboants that are about to meet 30. As each of the 20 roboants meets 30 roboants. As we have 20 roboants, we have the number of ants that meet one roboant time the number of roboants on the other group. 20 times 30 makes 600.

Why 20x30 (Disclaimer: I am bad at these things)?

DOMINIC FRANCIS ALVAREZ - 2 years, 8 months ago

Log in to reply

I worked a a similar but simpler problem. And used a visual approach. Four ants approaching from the left (a) and six from the right (b). Each line shows the tube and the ten ants as they move forward, suffer some number of collisions and then exit the tube.

a a a a b b b b b b a a a ab b b b b b 1 collision _a a aa bb b b b b 2 collisions _ a aa ab bb b b b 3 collisions _ aa aa bb bb b b 4 collisions a aa ab bb bb b 4 collisions a a aa bb bb bb 4 collisions _a a a ab bb bb b 3 collisions a a a a bb bb b b _ 2 collisions a a a b bb b b b 1 collision a a a b b b b b b a a a b b b b b b a a b b _ b b b b _a a b b _ b b b _ a a b b _ _ b b b

Notice the total of 24 collisions. Notice also the 4 by 6 parallelogram of collisions shown in the lines. In this example four ants entering from the left encountering six from the right produces 4x6 or 24 collisions. Try other numbers on your own. This may suggest that 20 ants encountering 30 could result in 20x30 or 600 collisions. Be open to having some fun with math and numbers.

Phillip Wagoner - 2 years, 8 months ago

(Second attempt at formatting.) I worked a a similar but simpler problem. And used a visual approach. Four ants approaching from the left (a) and six from the right (b). Each line shows the tube and the ten ants as they move forward, suffer some number of collisions and then exit the tube.

a a a a b b b b b b a a a ab b b b b b 1 collision _a a aa bb b b b b 2 collisions _ a aa ab bb b b b 3 collisions _ aa aa bb bb b b 4 collisions a aa ab bb bb b 4 collisions a a aa bb bb bb 4 collisions _a a a ab bb bb b 3 collisions a a a a bb bb b b _ 2 collisions a a a b bb b b b 1 collision a a a b b b b b b a a a b b b b b b a a b b _ b b b b _a a b b _ b b b _ a a b b _ _ b b b

Notice the total of 24 collisions. Notice also the 4 by 6 parallelogram of collisions shown in the lines. In this example four ants entering from the left encountering six from the right produces 4x6 or 24 collisions. Try other numbers on your own. This may suggest that 20 ants encountering 30 could result in 20x30 or 600 collisions. Be open to having some fun with math and numbers.

Phillip Wagoner - 2 years, 8 months ago

(Third try is a charm?) I worked a a similar but simpler problem. And used a visual approach. Four ants approaching from the left (A) and six from the right (B). Each line shows the tube and the ten ants as they move forward, suffer some number of collisions and then exit the tube. (Each “o” represents an empty space in the tube.)

AooAooAooAooBooBooBooBooBooB oAooAooAooABooBooBooBooBooBo 1 collision ooAooAooAAooBBooBooBooBooBoo 2 collisions oooAooAAooABooBBooBooBooBooo 3 collisions ooooAAooAAooBBooBBooBooBoooo 4 collisions oooAooAAooABooBBooBBooBooooo 4 collisions ooAooAooAAooBBooBBooBBoooooo 4 collisions oAooAooAooABooBBooBBooBooooo 3 collisions AooAooAooAooBBooBBooBooBoooo 2 collisions ooAooAooAooBooBBooBooBooBooo 1 collision oAooAooAooBooBooBooBooBooBoo AooAooAooBooBooooBooBooBooBo ooAooAooBooBooooooBooBooBooB oAooAooBooBooooooooBooBooBoo AooAooBooBooooooooooBooBooBo

Notice the total of 24 collisions. Notice also the 4 by 6 parallelogram of collisions shown in the lines. In this example four ants entering from the left encountering six from the right produces 4x6 or 24 collisions. Try other numbers on your own. This may suggest that 20 ants encountering 30 could result in 20x30 or 600 collisions. Be open to having some fun with math and numbers.

Phillip Wagoner - 2 years, 8 months ago

Log in to reply

Can't you edit comments?

Edit: You can.

Zain Majumder - 2 years, 8 months ago

I got 1200 collisions experienced but forgot to divide by 2 to work out the number of collisions. Then I checked the answer to see how far off i was. Gutted

Jacob Dalberg - 2 years, 8 months ago

Log in to reply

I did the exact same thing, each ant got a collision as opposed to a single collision for 2 ants...duh

Mike M - 2 years, 8 months ago
P B
Oct 13, 2018

If you draw successive states as a series of rows, it becomes apparent that the collisions form an m × n m \times n rectangle.

This is excellent visualization!!!

Mahdi Raza - 5 months, 2 weeks ago

A solution without assuming the ants pass each other (even though that one is "better" (faster)): m = 30 , n = 20 ; m > n m=30, n=20; m>n

First iteration: Two ants make one collision, call them left 1 _1 and right 1 _1

Second iteration: The two ants collide with other two ants, call them left 2 _2 and right 2 _2

Third iteration: Left 2 _2 collides with nother ant, left 3 _3 , and right 2 _2 with some other ant, right 3 _3 ; while left 1 _1 and right 1 _1 collide with each other again. In total 3 collisions in this iteration.

Generalizing, there will be n ( n + 1 ) 2 \frac{n(n+1)}{2} collisions up to iteration n n . That is to say, there are k k collisions for iteration k k . But, after that, in iteration n + 1 n+1 there will be only n n collisions again, since the collision that should be added, is lost by an ant that goes out the tube since she doesn't have another ant to collide into..

This will continue up to iteration m m . Let's consider though, only up to iteration m 1 m-1 , since iterations m m and onward are a mirror of iterations 1 1 to n n .

So in iterations 1 1 to n n , we have n ( n + 1 ) 2 \frac{n(n+1)}{2} . In iterations m m to m + n m+n we have i = m m + n 1 i = n ( n + 1 ) 2 \sum_{i=m}^{m+n-1}i=\frac{n(n+1)}{2} . And in iterations n n to m 1 m-1 we have n n times m ( n + 1 ) m-(n+1) .

That is to say, summing all: 2 n ( n + 1 ) 2 + n ( m n 1 ) = n ( n + 1 + m n 1 ) = n m 2 \frac{n(n+1)}{2} + n(m-n-1) = n(n+1+m-n-1)=nm .

If n = 20 n=20 and m = 30 : n m = 600 m=30: nm=\boxed{600} .

K T
Oct 9, 2018

Imagine an x-t diagram, where t runs vertically downward. 30 lines from NW to SE cross 20 lines from NE to SW. Each cross point is a collision where two ants meet and change direction. You could track them individually, but there's no need to do that to get the answer: 600 cross points.

Thumbs up for graphically solving this. I did the same, took me 4 sets of experimenting, my teams of ants where 2x3, 3x4, 3x5 and 4x7. In the first two I was thinking in terms of arithmetic progression and formulas. Third one was intuitive, in the fourth one it was obvious.

Xrysanthi Bekta - 2 years, 8 months ago
Lokesh Shingadi
Oct 9, 2018

Let us imagine there are only 2 ant on either side,then the number of collsions is 1.Now if there is 1 ant on one side and 2 on other side then number of collisions are 2.Now if 2 on one and 2 on other side then 4.If 1 and 3 then answer is 3,if 3 and 3 then 9,if 1and 4 then 4.We see that the number of collisions is the product of number of ants on either sides.So the number of collisions here 30*20=600

Can you prove this pattern continues?

Zain Majumder - 2 years, 8 months ago
Affan Morshed
Oct 12, 2018

You need to recognize that the bumping is basically equivalent to one passing through another (as if they were ghosts) for this calculations. After that it is fairly simple, 1 ant in the 20 ant group goes through 30 ants in the other group, and there are 20 ants, so the total number of ants going through is 30*20=600 (remember that the going through is technically a collision). I know this solution is already there but I am also trying to improve my explanation skills and I did use this method.

Aparna Phadke
Nov 18, 2018

Imagine there are 3 robot ants on the right side and 2 on the left side arranged as:

R3 R2 R1 L1 L2

•R1 will collide with L1 (collision 1) Causing L1 to change direction & collide with L2 changing directions again. (Collision 2)

•R1 will collide with R2 leading to R2 colliding with R3. (Collision 3,4) Since R3 and L2 have changed directions, they leave the tube.

•this repeats again causing R3 and L2 to leave the tube.(Collision 5,6)

Total collisions = 6 = 2x3 Same way, collisions for 30 ants on the left side and 20 on the right side = 20x30 = 600

Nick Pretzel
Oct 14, 2018

Whenever I try to post a thorough solution it gets too long and won't let me post, so I'll have to be brief. For two groups of n and n+m ants the n ants in the smaller group (as well as the outermost n is the larger) have 1,3,5,...,2n-1 collisions. Adding 1 to 2n-1 gives 2n and we can continue to do this n/2 times. The remaining m ants each have 2n collisions. We only want half of these collisions as they count each ants' therefore counting each collision twice. This gives 2 n × n 2 + 2 n × m 2 2n×\frac n2 + 2n×\frac m2 (we don't count the 2n × n/2 collisions from the larger group). So we get 2 n 2 2 + 2 n m 2 \frac{2n^2}{2} + \frac{2nm}{2} . Simplifying gives n 2 + n m n^2 + nm and factorising gives n ( n + m ) n(n +m) ie the product of the number of ants in the two groups Q E D \boxed{QED} .

It wouldn't let me add this so I've had to do it as a comment. It may be my phone, but it seems to object whenever I use multiple paragraphs. Anyway, if anyone has a proof that the n collisions are 2n×n/2 or, more particularly as the former just seems self evident, that the remaining m ants each have 2n collisions, I'd be most interested and grateful.

Nick Pretzel - 2 years, 8 months ago
Akshay Kumar Jha
Oct 12, 2018

Let's view this situation in terms of collision stage .

In each stage, the number of collision sites increases by 1 and additional collision sites are created at either side of tube as long as roboants are there . Also the adjacent collision sites turn on alternately as one can observe/verify. This will happen as long as 🐜 are there on either side. After that collision sites will recede at end will lower 🐜 (🐜 start moving one side) but increase at end with higher no. of 🐜, which cumulatively will constantiate no. of collisions. Then, even the end with higher 🐜 will see 🐜 moving one side and collisions will start receding till it reaches to one collision. Based on this reasoning the number of collisions,

N= S 1 + S 2 + S 3 S1 + S2 + S3

S1 = No. of collisions where no 🐜 are moving to ends=

1 + 2 + 3 + + 20 1 + 2 + 3 + \ldots + 20

S2 = No. of collisions where 🐜 at side with lower 🐜 only at moving at respective ends

= ( 21 1 ) + ( 22 2 ) + + ( 30 10 ) 21 - 1) + (22 - 2) + \ldots + (30 - 10)

= 20 × 10 20 \times 10

S3 = No. of collisions where 🐜 at both side move to respective end

= ( 31 11 1 ) + ( 32 12 2 ) + + 1 31 - 11 - 1) + (32 - 12 - 2) + \ldots + 1

Summing the three up, we get,

N = 600.

Nisse Que
Oct 11, 2018

I tried it out with smaller numbers and found a pattern. If a and b are the numbers of ants on respective sides, with b being the smaller number you will get 1 collision first turn, 2 second turn, 3 third and so on up to a maximum of b collisons. You will get maximum number of collisions for a-b+1 turns and then go down from b to 1 again. This gives a complicated formula of b(b+1)+b(a-b-1)+b(b+1). This can be simplified to ab!! Now I will go on to read other much better solutions. :)

Ryan Chen
Oct 11, 2018

If you draw down the ants as dots, then assume the y-axis as time. Then draw diagonal lines parallel with each side (of ants) under the dots you’ll get the paths of the ants. And wherever the lines meet is a turning point so you’ll find out that the number they meet will be the amount of points the diagonal lines meet, which is 30*20=600

Ok, so I looked from the other side and calculated how many hits, coming from the last going to first, each ant will have before it will go of the pipe. So on the 20 side last will hit once turn right and go of. 2nd before it hits last gets obviously hit by 3rd so it goes right and hit last, comes back, hit 3rd last again and then go of the pipe to the right. And so on but that's the wright pattern. Giving you numbers coming from the last one as first, = 1st -1hit 2nd - 3hits 3rd - 6hits 4th - 10hits 5ft - 15hits 6th - 21hits 7th - 28hits 8th - 36hits 9th - 45hits 10th - 55hits 11th - 66hits 12. - 78hits 13. - 91hits 14. - 105hits 15. - 120hits 16. - 136hits 17. - 153hits 18. - 171hits 19. - 190hits 20. - 210hits

Summarising TOT of 1540 hits on the right (20ants) side. Thats obviously dubles on the left giving you another 1540 so is already 3080 hits plus whatever amount last 10 ants out of 30 left side row will hit before they will disappear remembering through the both sides of the pipe, as the right hand side 20s will be already gone, or last ones just going away, but not hitting anything anymore as they just leaving the pipe going right direction one after another. As last 10 is being considered I did not calculated that yet, to much for one had on one day, at least for now so leave it to you and your judgement of my solution. Pretty sure is good.

Had a pic with sort of diagram but I can't upload, no option to share it or paste it from clipboard, at lest from my phone, but you should get the pattern, happy to share with anyone interested any other way. Hopefully speak to you soon big heads ! Bye

Jacek Senkow - 1 year, 12 months ago

1 pending report

Vote up reports you agree with

×

Problem Loading...

Note Loading...

Set Loading...