Andrew wrote:You say passing through metal sheets is forbidden -- but the balls are inside the metal sheet to start with. Therefore, the top one doesn't go through the metal sheet, it comes from inside it -- it has to go one way or the other. In other words, not a great rule in that situation, because in theory it's quite legal for the ball to go upwards. Or not?
I'm sorry to be a pedant, but couldn't stop myself......
Besides, this issue might be important in other challenges, and clarifying would be useful.
For me there is no doubt about it:
No, it's not (legal to go upwards). AR2 has fuzzy overlapping surfaces that only generate forces. It would thus be customary to use the centre point of the dillo, and centre line of the plates, to define relative position. The left dillo prefers to emerge into the gap, so it's centre is below that of the upper plates. To go upwards, the dillo centre would have to "pass through" the centreline of the plate, so in strict engineering terms I reckon Paolo's rule can be enforced.
What if the dillo was only overlapped 50% with the plate - would you still consider the dillo to be "inside" the metal sheet? What if the dillo was almost free, but had just a tiny overlap with the plate (like where it ends up when the solver starts)?
Where does inside stop and outside begin?
What if the dillo was only half the present size, so that it really did completely disappear within the thickness of the plate? (I think it's presently hidden behind the anchor point, and not fully obscured by a plate). What if the dillo was "full size", with large amounts showing each side of the plate?
These examples hopefully show you why the centre definitions have to be used as the absolute references for deciding what is above and below, and what "passing through" really means.
But I get what Andrew is saying, and agree the rule is somewhat misleading without a horribly technical engineering interpretation.....
I default to using the centre positions for clarity, but does anyone else?
Whichever side an initial overlap is resolved towards tells me which is "up" and "down", and persuading the dillo to move to the other side implies a "passing through" to me. Any other thoughts, anyone?