It's good to be back! This is the first RMO Mock test from Season 3 of MOMC! As always, all the credit and authorship of the problems belongs to their respective sources.
After you do the mock, please take a moment to fill out this quick form regarding this mock test. It will help us see how hard the problems really were. If you're interested, you can find the results here.
Below, I'll be discussing the solutions of the problems in the mock. So if you plan to attempt the mock, beware of the spoilers ahead!
Problem 1 (Sharygin 2020) Let be a cyclic quadrilateral. A circle passing through and meets and at points and respectively. The lines and meet at point , and the lines and meet at point . Prove that is parallel to .
Solution: Similar to last year, I wouldn't be providing solutions to any geometry problems here in this blog. Nevertheless, you can check out some amazing solutions by AoPS users here.
Problem 2 (All Russian MO 2014) Peter and Bob play a game on a chessboard. At the beginning, all squares are white apart from one black corner square containing a rook. Players take turns to move the rook to a white square and recolour the square black. The player who can not move loses. Peter goes first. Who has a winning strategy?
Solution: AoPS User IAmTheHazard provides an extremely elegant solution.
"Peter wins iff is even. If is even, his strategy is to divide the board into rectangles; whenever a rook enters a rectangle (including on the first move) he moves it to the other cell. If is odd, Bob's strategy is to divide the board except for the starting cell into rectangles and employ the same strategy. " In RMO, you should carefully reason why these strategies work.
Click here for the AoPS link of the above solution. This solution deserves your upvote!
Problem 3 (All Russian MO 1999) Prove that for all natural numbers ,Here, denotes the fractional part of .
Solution: Induction is our best friend for inequality problems like these. The base case is simple to check. The inductive step is to show the inequality: We observe that for any in the sum. We don't have to worry about that as then . So the result is equivalent to showing that or As the number of remaining numbers () is even, we do gaussian pairing. By Jensen's inequality, we have: There are such pairs. If the sum of each pair is , then we would be done. Hence it suffices to show that which after some manipulations is equivalent to .
Problem 4 (European Mathematical Cup 2022) Determine all positive integers for which there exist positive divisors , , of such that and , , are also divisors of .
Solution: This problem is quite troll as it doesn't actually involve any remainder bounding or manipulating expressions. Observe that must be divisible by because other . Hence which implies , a contradiction! Similarly, and must also divide . Therefore must divide .
A bit of hit and trial would show that work for any multiple of . So is a necessary and sufficient condition.
Problem 5 (Sharygin 2023) Let be the orthocenter of an acute-angled triangle ; , be points on respectively, such that is a parallelogram; be the common points of the line and the circumcircle of triangle ; be the point of opposite to . Prove that is the orthocenter of triangle .
Solution: We don't do geometry here. See solutions on AoPS here.
Problem 6 (Benelux MO 2019) An integer is called rich if for any positive integer , there exist positive integers such that . An integer is poor if it is not rich.
- Show that there exist infinitely many poor integers.
- Do there exist infinitely many rich integers?
Solution: Having faith in , we try and . This works as is impossible. To see this, note that the possible quadratic residues modulo are . Observe that . Hence we require such that . It is not hard to see that this has no solutions. It turns that this works for any where .
The second part of the problem is trickier. Consider where is such that is a primitive pythagorean triplet. So . Note that . Hence every odd positive integer can be represented other than as . But can easily be represented as .
Now consider . As is odd (otherwise taking in shows that must also be even, but then wouldn't be coprime), we can represent any positive integer of the form . We don't have to worry if we require as we can easily guarantee that by switching the signs to when we need to.
To represent multiples of , we multiply the triple by enough times. To be more concrete, let . Then if is represented by then is represented by . Therefore if is representable then so is for any .
So are we done? Yeah, but this is RMO. So you might want to also show that there exist infinitely many primitive pythagorean triplets which you can do by showing that the general form works.