[HN Gopher] A new Sudoku layout with 81 uniquely shaped cells
       ___________________________________________________________________
        
       A new Sudoku layout with 81 uniquely shaped cells
        
       Author : dmit
       Score  : 80 points
       Date   : 2025-03-13 01:04 UTC (21 hours ago)
        
 (HTM) web link (danielchasehooper.com)
 (TXT) w3m dump (danielchasehooper.com)
        
       | IncreasePosts wrote:
       | I'll give it a shot, but IMHO while the logic of "runs" isn't
       | super complex, it is way too hard to see at a glance and so I
       | don't see myself giving much time to this.
        
         | jsnell wrote:
         | Yeah. It's an original idea, but it just doesn't feel playable
         | at all.
         | 
         | I feel that this sentence from the blog post should have made
         | some alarm bells ring:
         | 
         | > Sometimes I fall back on selecting cells randomly until I
         | find one with highly constraining runs.
         | 
         | Clicking through individual cells one by one sounds really
         | tedious, but this also makes it clear that playing through
         | puzzles doesn't build up some ability to see the runs at a
         | glance.
         | 
         | I think the only way you make this work is by eliminating the
         | concept of run being entirely determined by the board layout.
         | There would need to be way fewer (mostly longer) lines, they
         | should always be visible, and they should be set up by the
         | puzzle constructor to be interesting rather than with purely a
         | mechanical rule.
        
           | dhooper wrote:
           | OP Here, I'll try drawing all non-redundant lines, see if it
           | feels better. UPDATE: done. give it a shot.
        
             | jsnell wrote:
             | It's an improvement, playble enough that I finished a
             | puzzle.
             | 
             | It still wasn't particularly engaging though, but a purely
             | mechanical exercise in pencilmarking and ticking off the
             | obvious conflicts. It's answering the question of "could
             | you have an irregular layout" rather than "what kind of
             | interesting puzzles can you design only with an irregular
             | layout". There was not a single "oh, wait, this ruleset
             | actually also implies _X_ " deductive leap, no place where
             | it felt like there was some cool emergent property of the
             | ruleset.
        
         | lanthade wrote:
         | I agree. I played one game and it was more frustrating/annoying
         | than satisfying. I just didn't find it enjoyable and I usually
         | find hard sudoku pretty enjoyable. The run rule makes sense but
         | it just didn't get any more intuitive the longer I played.
        
       | angry_moose wrote:
       | It's an interesting idea but I really don't like it.
       | 
       | The runs not being visible makes it a really unsatisfying
       | experience - you just kindof click around hoping that a cell is
       | part of a long run (and therefore more constrained).
       | 
       | Most runs are length 2-3 as well and entirely within the the same
       | "region", so are therefore redundant as they're already
       | constrained by box logic.
       | 
       | Nice idea but its a miss for me.
        
         | dhooper wrote:
         | OP Here, I'm gonna try drawing all non-redundant lines to see
         | if it improves the feel. UPDATE: done.
         | 
         | I will say that it took time for me to develop an intuition for
         | where to focus solving efforts. I think if people spend a
         | little time with it you wont have to click around as much to
         | know where the next break in is.
        
           | angry_moose wrote:
           | I think it would, though there is a risk it'd be too
           | cluttered visibly. But it does add the option of printing
           | which is nice.
           | 
           | Looking at your example puzzle with all lines visible; the
           | 8742 line in the top right and 896 in the bottom right are
           | both redundant (second one doubly so since it's all givens).
           | Maybe 50% of the two-cell lines are redundant and 33% of the
           | 3-cell; so it might be manageable?
        
             | dhooper wrote:
             | try it now, I think showing the runs is an improvement. I
             | think theres a few more things I can do to improve the
             | visuals. I also updated the article to make the rules for
             | runs more general - a human could place them anywhere
             | they'd like without the even/odd side constraint.
        
               | angry_moose wrote:
               | Nice!
               | 
               | I don't have time to do a full solve but I got the first
               | few digits. That plays significantly better to me.
        
       | nkrisc wrote:
       | An interesting idea that sadly just isn't fun to play.
        
         | 90s_dev wrote:
         | I'm not sure Sudoku is really ever _fun_ to play. It 's more of
         | a litmus test to see how well you've developed your logic. And
         | it can be satisfying when you get a good answer.
        
           | ludsan wrote:
           | I think Sudoku can be very fun to play with new rules, when
           | you begin thinking in higher abstractions. I'm a big fan of
           | anti-knight rules.
           | 
           | I look forward everyday to see what new puzzles are out on
           | the Cracking-the-cryptic channel.
           | 
           | IMO, your use of the word 'satisfying' is to me the
           | definition of fun!
        
             | macintux wrote:
             | Negative constraints like anti-knight significantly improve
             | my enjoyment of Sudoku. They make it more likely that
             | puzzles don't devolve to straight Sudoku near the end.
        
           | nkrisc wrote:
           | I think regular sudoku is fun.
        
           | chungy wrote:
           | On my phone, I've got 566 hours logged in 7700 games. I think
           | it's fun :)
        
       | jamie_ca wrote:
       | Interesting idea, but for clarity the cells need to be smoothed
       | to ensure each cell edge has a minimum length.
       | 
       | eg. In the example image the top-left cluster has a given 9,
       | where the orange line marking the run goes out to the right at
       | what to me looks like a corner.
        
         | dhooper wrote:
         | yes, that improvement was implied but not explicitly spelled
         | out under future work. I'll update to clarify.
        
       | scarletphoenix wrote:
       | Thanks for showing this off!
       | 
       | I played through the daily puzzle:
       | 
       | 1. It was really hard in dark mode to see what I was doing
       | 
       | 2. Scanning was almost impossible without clicking and seeing the
       | runs -- so it was mostly fully pencil mark and find pairs and
       | triples.
       | 
       | Overall, unique, but I think some further constraints or human-
       | setters (as you mentioned) could lead to more intuitive or clever
       | paths to a solution.
       | 
       | (Perhaps instead of lines, shade the cells in the runs to avoid
       | clutter with borders and writing over numbers)
        
         | dhooper wrote:
         | I removed dark mode because it was too hard to make a dark
         | color scheme that was as legible as light.
         | 
         | the puzzle now shows all runs all the time.
        
           | tener wrote:
           | Lol and I wondered why it came back light after reload. It
           | was really fun after you got past the UX hurdles! Pity there
           | aren't more online.
        
             | dhooper wrote:
             | New puzzle every day! I have thousands queued up!
        
       | hnlmorg wrote:
       | I can't explain why but I get anxiety just looking at that.
        
       | johnea wrote:
       | Isn't this just sudoku with the squares morphed into irregular
       | shapes?
        
         | dhooper wrote:
         | No, the irregular shape allows run constraints that wouldn't be
         | possible on a grid using rows/columns/lines.
        
       | pimlottc wrote:
       | I'm confused, some of the colored lines only go through a few
       | cells, shouldn't they all go through 9 of them?
        
         | dhooper wrote:
         | "Runs are a strip of 2 to 9 cells"
        
       | bigbuppo wrote:
       | Having looked at that, I now have a headache.
        
       | o11c wrote:
       | The lack of arrow shortcuts for navigation really throws me off
       | (especially since both the mouse and the numpad are normally used
       | by the right hand). Although cells have more than 4 neighbors,
       | there's still a notion of direction; maybe just break ties by
       | what is closer to the previous cell? Or use a subgrid? SGT's
       | "bridges" and "map" puzzles may be interesting; the former
       | doesn't care about neighbors but the puzzle guarantees indirect
       | connectivity somehow, whereas the latter uses a subgrid. "Loopy"
       | which has all sorts of shapes just gives up and disables it, even
       | though good solutions exist for even for non-square grids :(.
       | 
       | Is your "at most 9" logic preventing loops?
       | 
       | I think I could probably get used to visualizing runs on my own
       | if only the shapes weren't so irregular. In particular, the fact
       | that there can be apparent "four corners" means it's impossible
       | to tell which way the run actually connects; you should probably
       | tweak the grid to ensure a minimum edge size (or maybe even
       | forcibly align it to an orthogonal-with-diagonals grid? Either
       | way, remember you don't need to _stay_ Voronoi). Actually, to
       | reduce visual clutter, maybe color the _edges_ rather than the
       | lines crossing them? Perhaps make odd-sided polygons blatantly
       | different too?
       | 
       | Serious sudoko solvers have at least 2 kinds of pencil markings -
       | I find "top" and "bottom" most convenient. Maybe insert an
       | underscore at the end of the "active" set of numbers? (or
       | overdraw for pen marks). But for puzzles complicated enough to
       | need bifurcation, I really just need an arbitrary grid where I
       | can put any number at any location in any order - or else an
       | option to fork the entire puzzle.
       | 
       | And of course there needs to be a "fill in all pencil marks so I
       | can work subtractitively" button.
        
       | dwighttk wrote:
       | I guess when I'm solving sudoku I lean on "runs" being all 9 so I
       | know which of the 9 go into certain spots
       | 
       | I'm sure I'm not seeing something that lets me advance, but I'm
       | just stuck with a lot of cells with two options.
       | 
       | it's also not clear why some "runs" just connect adjacent cells.
       | I assumed adjacency (across zones) didn't matter because of that
       | but the puzzle warned me when I broke that rule.
        
         | dhooper wrote:
         | the 2-cell long runs are explained by this in the article "Runs
         | terminate on cells with an odd number of sides"
         | 
         | runs are shown between all adjacent cells.
        
           | dwighttk wrote:
           | >That's a confusing explanation for something I hope you
           | intuitively understand from the image above.
           | 
           | heh... not quite intuitive to me I guess
           | 
           | my confusion (as best as I can tell through the fog of
           | confusion) is that adjacent cells seem to not be allowed to
           | have the same number... and also there are a bunch of
           | obviously adjacent cells with lines connecting only those two
           | cells (but also a lot of obviously adjacent cells with no
           | lines connecting them.) Those lines connecting only two
           | plainly adjacent cells are confusing me.
        
             | dhooper wrote:
             | You don't need to understand the rules for how the
             | generator placed the runs to solve, you can just look at
             | the runs.
        
               | dwighttk wrote:
               | well, I need to understand something to make any
               | progress... I'm obviously confused about what I'm
               | confused about
        
       | culi wrote:
       | It would be mighty helpful to color-code the colored run-lines by
       | length
       | 
       | I.e. if red was always 2 cells long, green was always 3, etc
       | 
       | The length of the lines plays a significant role. Especially if
       | you get one that's 9 long
        
       | rezmason wrote:
       | This was hard, but fun!
       | 
       | One suggestion: make empty spaces visually distinct from filled
       | spaces. Why? Because cell 73 in the tenth puzzle (today's) is
       | small enough that I completely overlooked it, and used faulty
       | logic to deduce all but two spaces before I discovered it. I was
       | ready to file a bug report!
       | 
       | I'll meet you halfway-- if you'll please consider my feature
       | request, I'll go visit my optometrist. :D
        
       | bbasic wrote:
       | I liked it thanks, i thought it was fun to play, i got an
       | intuition for things after a while and i found it relaxing.
       | 
       | The instructions took a moment or two to understand for me.
        
       | jfultz wrote:
       | Solved, and it was fun! But...some of the colored lines were hard
       | to see. Particularly the light yellow ones. I had one "alternate"
       | solution that it was marking incorrect for reasons it took me
       | over a minute to spot...and it was just a super light yellow line
       | I missed without really scrunching up to my screen.
       | 
       | Improving the color choice might help, but slightly thicker lines
       | I think would also be of significant help.
       | 
       | Also, it's common for sudoku solvers to allow you to press
       | Shift+number to do a one-off toggle for the "note" checkmark.
        
       | moate wrote:
       | Hi there!
       | 
       | As someone who's had some amount of professional work on
       | games/toys I have bad news: I hate this (it's okay, I hate lots
       | of perfectly good things!) from a UI/UX standpoint.
       | 
       | "Nobody has changed the layout of Sodoku..." there might be a
       | reason for that. When you're trying to create organized
       | lists/sequences, a messy visual is going to be frustrating. The
       | geometry/symmetry of a Sodoku is part of the appeal. Think of a
       | crossword: does the shape affect the fun? You might see circles
       | vs squares, or some people building shapes by arranging the word
       | lengths creatively, but nobody is trying to have the across words
       | in one column and the down words floating in a nebulous space
       | nearby, because that's not a crossword.
       | 
       | What value does this change bring? Why do you think there's more
       | fun in the difficult to parse shapes over the clean rows/lines of
       | standard sodoku? The changes seem to make it much harder to gronk
       | how the game works and provide little added "fun" value. Think of
       | it this way: you can play sudoku blindfolded and have someone
       | fill the grid in for you. This would make it much more
       | challenging (with essentially a UI change like your change) but
       | not necessarily more fun for most people because the challenge
       | isn't what the game is about. It would be similar to playing
       | baseball in ankle deep water, or Halo using a DDR pad: novel and
       | quirky, but missing what makes the game compelling in the first
       | place.
       | 
       | I think this is a cool thought experiment and a great way to
       | illustrate some concepts of game design (sadly, as a way of what
       | NOT to do) and for that you should be really proud! Hope you're
       | able to take something productive from my post.
        
       | null_name wrote:
       | having fun with this. I like the longer runs better but it feels
       | kind of crowded, I wonder if you could optimize for fewer+longer?
       | also just had the thought you could fit this into arbitrary
       | bounding shapes, not just a square...
        
         | null_name wrote:
         | completed the day's puzzle! it's a bit hard to keep the runs in
         | my head. what if instead of run lines, the polygons had a
         | background, like stripes for one run, checkers for another, and
         | half/half for ones with both? I don't think this idea is quite
         | right, but it might spark something from someone else.. I'm
         | colorblind, which may have something to do with how hard it was
         | to keep track of the runs. maybe they could be kept as lines,
         | but with some as dashes, some dots, some solid, etc?
        
       | Malipeddi wrote:
       | This is quite nice! Thank you for sharing it. I think i came here
       | after the good feedback had been incorporated. So i did not find
       | anything confusing. It was quite fun to solve.
        
       ___________________________________________________________________
       (page generated 2025-03-13 23:00 UTC)