7K Aviator multiplier: reading a number still in motion
The figure on the curve is the most watched thing on the screen and the most misread. It is not a score and not a price: it is a live reading of what an accepted exit would return, and the words live and accepted are doing most of the work in that sentence.
A figure you can see is not a figure you have. Until the panel confirms an exit, the number on the curve is an offer the round has not finished making.
Why the climb feels slow and then stops feeling slow
The curve starts at 1.00 and rises continuously until the round ends. It grows by multiplying rather than by adding, so the distance between 1.00 and 2.00 takes the same sort of time as the distance between 10.00 and 20.00, while the second stretch looks far more dramatic. The clock is even. The scale is the thing accelerating.
One consequence is worth holding on to, because it is the only part of the curve that is genuinely informative: the round has to pass through every number below yours before it can reach yours. A large target is not rare because the game dislikes it. It is rare because it is the far end of a path that can end at any point along the way.
Provisional means provisional
While the plane is on screen, the reading beside it describes a result that has not happened. Your stake is committed, the round is running, and the number is the current state of something still in progress. It becomes yours at one moment and one moment only: when the exit request is accepted and the panel reports back a figure.
Two numbers exist for every cash out. The one you were looking at when you decided, and the one the game wrote down. They are usually close. They are almost never identical.
That is not a defect and it is not an operator trick. It is what happens when a reading climbs continuously and a decision has to travel somewhere before it counts.
What the delay is actually made of
Four small gaps sit between your decision and the recorded figure, and together they are the whole of the difference.
The screen redraws in steps, so the figure you read was true a fraction of a moment ago. Your own reaction takes longer than either of the others and is the largest of the four by a wide margin. The request then has to reach the server, which on mobile data is a far more variable trip than on a fixed line. And the server timestamps the request on arrival, against the curve as it stands at that instant.
Nothing in that list can be removed, which is why the only sensible response is to stop aiming at an exact figure. If a target matters, set it in the automatic field and let the request be scheduled rather than aimed: an automatic exit removes your reaction time from the chain, and it leaves the other three gaps exactly as they were.
One honest limit on that advice: an automatic target only fires if the curve reaches it. Set it above where the round stops and nothing happens at all — the schedule was correct and the round simply ended first.
Three readings of the curve that are simply wrong
That the shape of the climb says anything about the ending. It does not. The curve is drawn the same way in a round that ends at 1.08 and in one that ends at 108, right up until the moment it ends. There is no tell in the line, the colour or the animation, and the reason there is none is that the ending was fixed before the drawing started.
That a slow start means a long flight. The rate of climb is a property of the display, not a clue. Watching for a hesitation in it is watching the frame rate.
That someone else's large exit means the curve had further to go. It means they left earlier than the ending, which is the only thing any successful exit ever means. The social column reports what other accounts did, in the past tense.
The useful instruments are smaller and duller than any of that: the countdown, the amount field, the automatic target, and a figure you decided on before the round started.
Questions about the moving number
I pressed at 3.2 and got 3.1. Why?
Because the recorded figure is the one at the moment the server accepted the request, and that moment is slightly after the one you saw on screen.
Does a faster connection give a better result?
It narrows the gap between the figure you saw and the figure recorded. It changes nothing about where the curve stops.
Why does the curve seem to speed up?
Because it climbs by multiplication rather than by addition, so equal stretches of time cover ever larger numbers. The clock is steady; the scale is not.
Is a very high multiplier rarer than a low one?
Yes, and the curve shows why: it has to survive every number below yours first. The sister page on past rounds explains what that does and does not imply.
Can I see where the curve was going to stop?
Only after it has stopped. Before that the value exists in sealed form, which is a different thing from being visible.
Carry on with
The control that turns a reading into a result is the cash-out panel. What happens to the figure once the round ends is the strip of past rounds, which is also where the applications claiming to read it in advance are dealt with.