Puzzle #411 - Arafillomino

Yesterday, an Araf (Different Neighbors) of mine was posted as a Sunday Stumper on Grandmaster Puzzles. Normally these links to GMP are pretty short, but I wanted to take a bit of time to talk about the construction process and some of the issues and obstacles I ran into in creating this. Obviously, solution details will follow behind the break.

Puzzle #410 - Cave (Half and Half)

Recently, djmathman finished running Logic Showcase 31, where the requirement was to make a puzzle using the recently popularish variant "Half and Half". In addition to standard Cave rules, exactly half of the cells in each outlined area must be shaded. I almost didn't submit a puzzle for this but found myself with an hour before the deadline, an idea, and a drive to make something neat.

penpa interface

puzz.link interface (Nanro substituted, not recommended)

I did at one point have a unique puzzle without the dominoes in the corners but felt that the extra pair of regions allowed for an improvement to the logic and flow of the puzzle.

Puzzle #409 - Mirror, Mirror (Aqre)

It took me a while to actually sit down to make a serious Aqre puzzle (since rooms still don't feel right for me with the other constraints) so of course, when I did I went for a very ambitious theme of symmetric rooms, with a 5 on one side and a 2 on the other. It took surprisingly little time to find a working puzzle, but it took significantly longer to find a version that I felt had fair enough logic. It's still really subtle and I'll probably put together a walkthrough for this one to explain the intention at some point later.

puzz.link interface

Also, Puzzle Ramayan round 1 recently concluded, and I recorded my third place finish, and added a few annotations throughout. If you're interested, feel free to check it out here.

Puzzle #408 - Suspicious Clearout (Uso-One)

All the way back on August 31st, 2021, Bryce Herdt posted a 4 room Uso-One template. The theme is... obvious. I'd never tried making one before, so I gave it a shot and then sat on the puzzle until then, unsure if I was going to do anything with it or not. After all, it wasn't my theme and the solve is a bit simple. But simple isn't bad and this was the only non-meme puzzle that was made from the layout so, here it is.

puzz.link interface

What a weird type. I really need to get better at dealing with lying clues. This also concludes my backlog of puzzles that I can actually post for now- I have so many good ones waiting to be freed up :)

Puzzle #407 - Bonus Gift (NIKOJI)

MetaTerminal didn't receive a secret solver gift this year. So a bunch of people banded together to assemble a series of mostly hunt puzzles- since that's where things were headed and I don't do hunt puzzles, I didn't help out.

But then when they started streaming a solve of their gift, I felt like putting together an extra puzzle just because, and went with a NIKOJI due to available theming- this is still a weird type to construct and I ran into a lot of dead ends. Though in the end, I made something I think is pretty neat.

puzz.link interface

The extra E and the mis-spelling are unfortunate. But hey, this works.

Puzzle #406 - Quartered, To Be Drawn (Simple Loop)

I felt like making a Simple Loop one day. I made this, and no longer felt like making Simple Loop any time in the near future- it's so hard to make a puzzle work without circumventing earlier deductions that I honestly just gave up.

puzz.link interface

Small compromise aside, I do like how parts of this turned out. I just think I'd need to go much, much bigger to really get anything meaty.

Puzzles #403, #404 & #405 - Light Tapas

I actually made these three Tapa puzzles a few weeks ago. The first was the result of trying to find a neat antisymmetric design, the second came after playing around with latticed ?? clues in pzprrt (but made by hand later), and the third was just because I wanted to make a standard Tapa afterward. No extra tricks.

puzz.link interface

puzz.link interface

puzz.link interface

The 111 clue in the last one wrinkles a little bit, but I couldn't find another nice way to make the theme resolve. I normally avoid "21", instead formatting it as 12, but it just had to be 21 here.

Puzzles #401 & #402 - Unspoken Rules (TomTom)

One of the things I like to do with my puzzles sometimes is question a convention that most, or all puzzles in the genre have. Early on, I made a LITS with internal borders and thought nothing of it until people started discussing if it was a variant or not. Then early on this blog, there are the Moon or Sun puzzles that allow passing through all 0 of a symbol in a region. Now, here's a TomTom that uses negative given values, which is something I don't think I've seen before.

penpa interface      puzz.link (not recommended)

I tried making this between two other TomToms on request for GMP, actually- the first draft was broken (I missed that 3-3-2-1 = -3 as a possibility) because the idea crossed my mind. However, the other two were actually accepted for future use, and I'm really pleased with what I made for the one posted a couple weeks ago.

Puzzle 402 - "The Missing One"

Puzzles #391 - 400 - MadMahogany's Christmas Tree Conundrum (Shading and Sudoku Medley)

Like the last two years, I took part in the Secret Solver event on Puzzlers Club. This time around, I got MadMahogany, someone I mostly know is semi-active (I think?) on the CTC Discord, and some of their most memorable puzzles they've posted to their blog have been shading or sudoku puzzles. Sudoku is obviously outside my general comfort zone, but whatever, this was the perfect opportunity to go big and make something incredible. So I made a Samurai Sudoku shaped like a Christmas tree with different variants in each subgrid. Rules after the puzzle images.

penpa interface

Nurikabe | Tasquare | Canal View | Nuribou | Nurimisaki | Kurotto | Mochikoro | Cave | Kurodoko

I also decided to make a collection of 5x5 shading puzzles themed on 12/25 (as close as I could get it) for each shading puzzle variant in the big puzzle. Rules for Nurikabe, Tasquare, Canal View, Nuribou, Kurotto, Kurodoko, Nurimisaki, Mochikoro and Cave can be found linked here, as well as Sudoku. These 9 puzzles were a last minute addition when I finished with time to spare, and I thought it would be nice to have a brief reminder / demonstration of the shading rules before having to dig into the tree itself.

"penpa" link (the URL is too long for some browsers, so go to Penpa, select "Load" from the top bar, and paste the URL from the text file into that box to load the puzzle)

So, here's the tree. Standard rules apply for each puzzle type within each labelled subgrid. Some grids overlap: the contents must satisfy all rules from both grids when taken only for that grid. As an example, Kurodoko requires shaded cells to not be adjacent to other shaded cells. This rule only applies within the Kurodoko grid, and so a shaded cell on the edge could be adjacent to a shaded cell in another grid. Any cell with an outlined gray circle will have its value act as a clue for the shading puzzle in that grid. Note that most thermos and arrows have the circles coincide with shading clues, but not all. A rundown of each variant rule follows. Star Battle - 1s and 2s may not share an edge or corner with each other. Thermo - numbers must strictly increase from the circle to the end. Arrow - the sum of all numbers on the arrow must equal the number in the circle. Classic - no additional restrictions apply. XV - Cells separated by a V must sum to 5. Cells separated by an X must sum to 10. Cells separated by an XV must sum to 15. Not all Xs or Vs are given. Kropki - Cells separated by a white dot must differ by exactly 1. Cells separated by a black dot must have one number be exactly double the other number. Not all dots are given. Difference - Cells separated by a red dot must have a difference of the number inside the dot. Killer - Cells inside a cage must sum to the value in the cage, and must not repeat within a cage. Product - Cells separated by a pentagon must have a product of the number inside the pentagon.

This puzzle took me probably 30 hours across several weeks to construct, and I got to work with a couple rulesets I'd never worked with before. Like overlapping grids (somehow this was new to me- many headaches from contradictions across grids though...), most of these number placement variants (Arrows are a huge pain, and Killer is way harder than I thought it would be), and Mochikoro and Canal View (fun and easy, painful and difficult. Though some of the Canal View pain was from the other restrictions, it's hard to make unique!). Special mention goes to the Nurikabe for being an absolutely massive gigantic pain to make unique, and the Cave, and the Kurotto section, and generally huge parts of this puzzle barely work. But it all comes together beautifully and I'm very proud of what I came up with here. And my solver liked it, too. :)

I'd also like to take this space to thank Rever Man for testsolving most of this puzzle before I sent it out, Swaroop for updating Penpa to even support grids this big, and djmathman for making the gift I received, which is really quite excellent.

That's all for now, sorry about the inactivity here, more puzzles soon once I write up the posts for them because I have another 8 waiting already. Errata: added a white kropki dot I thought was unnecessary, but was made necessary due to a bad deduction in completely different section.

Thoughts on Competition Scoring Methods

With Tapa Train currently running on LMI, I've been thinking about puzzle contests again. On paper, these all measure one thing: how fast people are at solves the presented puzzles. And this tends to be borne out in the results, with the same group of people near the top every time. But the order is variable, and can vary by quite a bit depending on the format. It's not immediately obvious why, but the way results are aggregated and compared make a huge difference to how these things can feel. I'm going to go into all the formats I can think of and try to pinpoint what they're measuring, beyond the obvious. I'll be citing some example cases, often from my own experiences, and also often from cases where the format has not suited me well. That's not to say that I think I deserve a better placement from using a different system, just that I can speak to my own experiences and how the systems have affected me better than I could someone else.