Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Stadfradt]

SPOILER WARNING

I apologize for posting this because this topic has been covered ad nauseum, but I just can't get it to click. Please tell me if my reasoning is correct.

Spoiler (click to show/hide)

I have a 4x4 map. I took a spot an even number of steps over and up/down from the corner of the embark region. Now I mark (d - m) a two diagonals from the map corners making an X; this defines the center of the map. I draw lines to the map edges through that point and repeat with the smaller rectangles. Rinse and repeat until I have 4 rows and 4 columns of rectangles, each with a corner-to-corner X marking their centers. Any and all good stuff is going to be below exactly 4 of those center points -- or approximately below, since embark tiles have even numbers so my vertical & horizontal dividing lines have to be two-tiles wide, and I don't know how to correct for that error. And now it's simply a war of attrition as I get miner after miner drowned or fried in the third-layer lakes or magma, respectively.

So, is my locating theory correct, and now it's just an engineering problem? Or have I gotten the wrong end of the stick?

Thanks in advance!

[Thisfox]

I think you'll find that all um.... minerals.... are as randomly placed as each other. But braver souls might tell me I'm wrong.

[Bumber]

You lost me at the smaller rectangles. You've got 4 candy canes. One in each quadrant around the center of the map.

If you were to take a quadrant and divide it into 4, it's a reasonable assumption that its candy cane is contained entirely by exactly one of those 4 sub-quadrants.

However, you can make no further assumptions about where the candy cane starts within the sub-quadrant that contains it. It's not centered, and can be shared between sub-sub-quadrants, AFAIK.

[feelotraveller]

Actually only one candy cane is guaranteed, with a possibility of up to 8 more.  It depends on the offset of the grid.  To get this think about the possibility of the centre 2x2 squares being the larger cane-bearing square (of which candy-cane only occurs in one square).  Then the 4x4 embark area is completed by parts of another 8 2x2 squares.  (Yes it's possible that it might be 4 complete larger squares, guarenteeing 4 candy-canes, but the chances are against it.)  Really the only sure-fire guarantee is that any 3x3 square will contain at least one candy-cane, although it could contain up to 4.

My limited experience in finding candy-canes suggests that the actual cane square is further divided into a 2x2 square with the candy-cane being present in 1-4 of those 4 squares where the only limitation is orthogonal adjacency (no diagonals without a connecting third square).  These smaller squares (64 in total on a 4x4 embark) also vary between z-levels as to whether they contain candy or not, although once again I believe the rule of othogonal adjacency applies - but keep in mind that in practice only parts of these candy-squares actually contain the sweet stuff.

Probably the best way to find candy is to reveal as many sections of the magma-sea as possible and then work up/down and sideways from any revealed goodies.  But if one wanted to pursue a more exhaustive probe type engineering I suggest that it would be best to divide the grid one more time and investigate the centre of each of the 64 mini-squares of a 4x4 embark (or 36 of a 3x3 embark, etc.).  Just beware that it is also possible to encounter unexpected fun...

[Stadfradt]

It depends on the offset of the grid.

Kill me now. LOL. My guess was that the embark region is an 8x8 of tiles_2x2, so that a move of one tile_2x2 is two clicks of movement u-m/h-k, which would keep me on the grid. I'm not assuming that is correct, nor that my description makes sense. From that I'm guessing that I have a map that is 2x2 of tiles_2x2 if and only if I move my embark square in even steps in the region. Again, I'm guessing about that. My Dwarven Geometry* locates 16 centers of 16 squares, but my foundation isn't bedrock.

To get this think about the possibility of the centre 2x2 squares being the larger cane-bearing square.... My limited experience in finding candy-canes suggests that the actual cane square is further divided into a 2x2 square.... These smaller squares (64 in total on a 4x4 embark)....

Okay, that's still better than completely random, assuming the center-of-the-square thing is still correct. I've seen it scattered about forums, the wiki, and the reddit threads, but we'll have to see how it goes.

I set up my grid, mined it out -- thank goodness for macros -- and drilled down dead nuts into a candy cane. It's hard to argue that n=1 is a good sample set. My grid is easy to subdivide, my pop has a soft cap of 15, my visitors are capped, my invaders are capped, I'm mining as little as possible, and I'm not collecting tons of stuff, so hopefully my fps will stay high.

I can't keep track of where things are nor can I picture the geography of the magma sea, so the grid is my main hope of sanity. Like gold, the stuff I'm looking for is basically useless; if I can't make kit for 15 military dwarves, now 18 when the kids grow up, what use is it? But finding it is an itch I must scratch. Gods, friends, drink, and a little work seem to make for a happy fort. Adam Smith would be proud.

*Have you read "Taxicab Geometry"? Construct a circle in Dwarven Geometry, i.e. the set of points equidistant from a center point. Dwarven eyes may be round to them, but to us they look square. Kinda creepy when you think about it.

[feelotraveller]

Yeah it seems like a reasonable assumption that the 2x2's are aligned to the 16x16 embark grid, but I've tripped over seemingly reasonable assumptions in DF many times.  Not saying it's wrong here mind you.

My reservation with drilling down in the centre of the 16 possible areas is that puts the probe right at the edge of the possible cane areas (well unless its a lucky draw 4 out of 4).  Using a single tile probe only reveals one tile of one of the 4 sub-sub-areas, 2 of 2 and 4 of the 4th.  Can envisage this missing a cane sometimes - even assuming complete candy blocks it would only need a gem or two in the wrong place(s).  And if using say a 4 tile wide probe why not split that into 4 single tile probes centred in each sub-sub-area.  Pardon my theorycrafting.

Your Dwarven Geometry is somewhat different to actual Dwarfen Geometry methinks.  Dwarfs when travelling on a diagonal actually take 1.5 units of time and not the 2 units which Manhattan distance would result in (nor the  sqrt {2} time units which might alternately be expected).  So a dwarfen circle which extends 9 tiles horizontally/vertically extends 6 tiles at 45 degrees rather than either 4 1/2ish taxicab or a bit over 6 1/3 pythagorean.  (Even though taxicab or pythagorean distance are more 'reaonable' assumptions.)  Well at least in terms of dwarfen pathing - it is possible that other mechanics calculate it differently. 

The square grid does impose all sorts of madness on overseers commited to some sort of equidistance or circularity though. 

[Stadfradt]

Pardon my theorycraftingYour Dwarven Geometry is somewhat different to actual Dwarfen Geometry methinks.  Dwarfs when travelling on a diagonal actually take 1.5 units of time and not the 2 units which Manhattan distance would result in (nor the  sqrt {2} time units which might alternately be expected).

If dwarves move at constant speed...

So, their eyes are circle-ish, but wavy circles --

HOLY SHIT! THEY'RE HYPNOTOADS!