Mire 'Jumps'

[Prof]

If you have non-mireable land set, and mire starts to appear on the other side, it will jump the non-mireable land.
The attached Images show an example.

I'm using version 1.0.2

[Cavemaniac]

I reported something like this during beta (though it's a bit hard to tell if it's exactly the same instance as I reported from your pix) and the response I got went something like this:

Mire isn't creeper - it doesn't 'flow' around edges, rather it radiates out from its source, and as long as it's all on the same 'landmass', it can seemingly 'jump' over void.

As it's an energy source, it's better to think of it as more closely akin to the green soylent of CW1+3.

Not sure if that's the whole answer to what you've observed though.

[Prof]

I feel it would have been better if I had shown more of the map, its that it waits until a piece of land on the same land mass is mired, then all of a sudden, it skips over the non-mireable land rather than flow through it like it was moments before. The way it happens suddenly makes it seem buggy, and unintentional. If you load the save, and run it until around 1 minute, it shows it pretty clearly.

[knucracker]

Mire spreads to the nearest edge of mire on the same land mass.  Nearest means the straight line distance.  An edge is defined as any cell adjacent to the same mire type.  In certain situations this can make the mire appear to jump gaps.  Caveman is correct that it is best not to think of it as creeper, but rather as an invisible field that radiates outward in every direction and deposits only on the same land mass and only adjacent to existing mire.

Of course one of the main reasons it works this way has to do with the implementation and performance.  The current algorithm is super efficient.  A more creeper like algorithm would have imposed too high of a computational load on the game (given the particle sim and everything else that is already going on).

[Builder17]

Use line of void there between two lines of land?

[GoodMorning]

Perhaps metallic land should be deemed to be as void for the landmass-defining calculation.