The Puzzle Robot #2 – Tapa (Irregular)
or solve online (using our beta test of Penpa-Edit tools; use tab to shift between shading mode and the composite Yajilin mode where left click marks cells, right click marks dots in cells or X’s on edges, left click+drag draws lines.)
This is a “Twisted Tuesday” puzzle variation.
Theme: Region shapes
Rules: Variation of Tapa rules. Additionally, some squares are combined into areas, each counting as one cell for the Tapa clues. Or see here. [Note: the no 2×2 shaded rule applies as usual based on cell count, not area count, and some areas may contribute multiple cells to such a group.]
Answer String: Enter the length in cells (not the # of areas) of each of the shaded segments from left to right for the marked rows, starting at the top. Separate each row’s entry from the next with a comma.
Time Standards (highlight to view): Grandmaster = 2:15, Master = 3:15, Expert = 6:30
Solution: PDF
01:26. How I beat the Grandmaster time by so much is beyond my comprehension. Floodfilling wins against manual shading by a large margin here 😛
Nice puzzle. Aesthetically by the outer two cells (yes I only find the outer two cells that are nice; the rest aren’t that much), logically by how the puzzle works with the weird shapes. With cells this irregular, I’m curious whether an Irregular Tapa with only one clue can work…
I was pretty sure a paint solver would crush this time. Or anyone who chooses a particular guess to make early and guesses right.
Given how set-ups like the center areas are a “this one or that one” kind of consequence, I bet you could stack a small number of nested areas up in a way that a single clue would specify a valid puzzle. But again, then flood-filling or guessing would probably yield a solution in ten seconds. Making it complicated enough probably requires seeding two paths by the one clue that don’t immediately touch.
Now that I come back and do this one on Paint, yes, it really does change the timing. On paper I was rushing to scribble in shading or Xs on the larger cells. That made things less visually obvious and it was still time consuming. So for example the time between “oh the top left block is unshaded” to “oh I can solve that 3 now” went from 5-10 seconds on paper to less than one on Paint.
Wow, took me forever to see the break in, though it was pretty straight forward after that.
I’m wondering if I got the intended breakin for this one (ROT13):
Gur “bar-bar” pyhr ba gur evtug fvqr unf bayl gjb cbffvoyr pbasvthengvbaf, naq bar pna or cebira jebat snveyl dhvpxyl, nf vg vfbyngrf n frpgvba bs gur jnyy.
It seems like the correct place to start anyway. I hope the ROT13-of spoilers is alright, since this is still a young puzzle.
You are right. That was the first step of my structure.
I started by…ernyvmvat gur obggbz evtug fdhner nebhaq gur gjb-gjb va gur obggbz evtug zhfg or juvgr, orpnhfr vs vg jrer oynpx, gura vs gur obeqre nebhaq vg jrer oynpx gura lbh’q unir n oybpx bs 1, naq vs gur obeqre jrer juvgr, gura lbh’q arrq n oybpx bs 3.
Nsgrejneqf, vs lbh nffhzr n oybpx bs 2 fgnegf nobir gung juvgr fdhner, lbh dhvpxyl trg genccrq, fb vg zhfg or juvgr nf jryy. Naq gung nyybjf lbh gb cynpr n pbhcyr bs oynpx fdhnerf nebhaq gur gjb-gjb.
So, a bit more convoluted, but in the same general area.
Ohhhhhh it’s a 10×10 grid! I didn’t realize the border and kept immediately getting contradictions.