Game Guide
Flip Match scores you in turns, not time. Here is the turn floor for each of the five free levels and the reconnaissance rule that gets you close to it.
Flip Match is the purest scoring system in MyRin: no clock, no combo multiplier, no bonus. You clear the board, and the game tells you how many turns it took. That single number makes it a much more analysable game than it looks, because with time removed, every turn is either productive or wasted, and you can decide which before you commit.
The free browser demo has five levels. The grid resizes itself to fit the card count, so unlike a fixed width board the shape changes as you progress, which matters more than it sounds.
Here is the MyRin short for Flip Match. Watching a clean run makes the turn economy visible in a way a turn count on its own does not.
A turn is two flips. After the second flip the board locks briefly, resolves, and either keeps the pair open or turns both cards back over. Since there is no clock, the cost of a turn is exactly one unit of score and nothing else. So the only question worth asking before each turn is: what does this turn buy me?
There are only three possible answers. It closes a pair you already know about, which is pure profit. It reveals a card you have never seen, which is information you can spend later. Or it re-reveals something you already knew and forgot, which is the wasted case and the one to eliminate.
That gives a simple rule with a slightly odd feel: when you do not know where a match is, your first flip of the turn should be a card you have never seen, and your second flip should also be a card you have never seen. Two unknowns per turn is the maximum information rate. The instinct to flip a known card first, hoping to stumble onto its partner, buys you at most one new coordinate instead of two.
Four cards, a sakura and a tanabata bamboo, two of each, on a two by two grid. The floor is two turns and it is reachable by luck alone. Nothing to learn here except how the flip animation and the resolve delay feel on your device.
Six cards on a three wide grid: wave, mountain and wind chime. The floor is three turns; four or five is normal.
This is the smallest board where the reconnaissance rule pays. Open with two unknowns. If they match, excellent, and if they do not, you now know two coordinates out of six, which is a third of the board for the price of one turn. There is no scenario in which flipping a card you have already seen helps you on a board this size.
Eight cards on a three wide grid, which means the bottom row is short, and the ragged shape is worth noticing because it makes positional description harder. Sakura, bamboo, tea and swan, two each. Floor is four turns; six is a good result.
With eight cards, two reconnaissance turns give you half the board. After that you are usually in partial knowledge: you know where one of something is and not its partner. The correct play in that state is to lead with the unknown card, not the known one. Flip the unseen card first, and then, if it happens to match something you already know the location of, close it with your second flip. Leading with the unknown means a failed turn still bought you a coordinate; leading with the known means a failed turn bought you nothing.
Ten cards on a four wide grid: torii, moon, herb, bell and sakura. Floor is five turns; seven to nine is realistic.
Ten positions is where most people exceed what they can hold as a flat list, so switch to spatial chunking. With a four wide grid, the natural split is by row: three rows, with the last one short. File each revealed card into a row rather than at an exact coordinate. Searching for "the bell was in the middle row" is faster and more reliable than trying to recall "the bell was third from the left in the middle row", and for the purpose of finding it, the row is usually enough.
Twelve cards on a four wide grid, three full rows, using all six motifs from the earlier levels. Floor is six turns; eight to ten means you played it well.
The interesting decision on this board is ordering. Once you know several pair locations, do you close them immediately or keep exploring? Close them. Every pair you clear removes two cards from the field you have to keep scanning, which makes everything you still hold in memory easier to maintain. Hoarding known pairs while you explore is a common instinct and a bad one, because the memory load of tracking what you know grows faster than the value of knowing more.
First, slow down on the reveal, not on the decision. Since there is no clock, the moment a card turns over is free, and it is the only moment when the information exists. Spend it: note the position, then decide. Most turns are lost in the quarter second where a player sees a card and has already tapped the next one.
Second, say the position rather than the picture. "Bell, bottom row, right end" encodes the fragile part of the memory, the identity, against the durable part, the location. Spatial information is retained more reliably than object identity, which is why positional phrasing consistently outperforms trying to remember what you saw.
Flip Match loads visuospatial working memory and, more specifically, it trains the strategy layer on top of it: chunking by region, rehearsing coordinates, and choosing the flip that maximises information. Those are learnable in a way that raw memory span is not, and they are the reason a practised player beats an unpractised one with identical memory.
The honest limit is the same as for every game of this shape. It will make you better at boards, and it will make chunking feel more automatic when you meet a similar problem. It will not extend your memory span in general. If the five levels leave you wanting the version with real pressure, the installed game runs the same mechanic for 600 levels and adds the time constraint the demo deliberately omits.