Tuesday, October 13, 2020

Puzzles #202 & #203 - TOPOLOGY *handwaves* (Numberlink/Kaero)

Outside of the occasional Haisu, I haven't really tried making puzzles with topological arguments or solving largely topological types rigorously, and have taken to joking about handwaving any topology deductions I make. So I tried making a Numberlink and kept having trouble, and eventually turned up this puzzle.

In between playing CC2LP1, which is the first community set for Chip's Challenge 2, I tried to make a Kaero, and it shows. You should probably try CC2LP1 if you have the game, it's really enjoyable. What was less enjoyable was finding many, many alternate solutions to this puzzle, but the final solve path has some interesting logic available.

Tuesday, October 6, 2020

Puzzle #201 - VVVVVV (Pentominous)

Just a quick Pentominous puzzle I threw together this morning, inspired by a similar one using P clues by Deusovi.

puzz.link interface

I'm going to start using Swaroop's fork of penpa instead as it's being actively updated. Past links will remain as they are and the interfaces for the Puzzle Zodiac puzzles will also remain rjudman links. Speaking of, that contest starts in 3 days!

(2025 edit: removed dead git.io shortened link)

Friday, October 2, 2020

Puzzle #200 - A Major Milestone (Ripple Effect)

So... making puzzle 200 ended up taking a good bit longer than I expected. Originally I wanted to make a large Nanro puzzle, but I ended up not having any ideas for how to drive such a puzzle the way I made the giant LITS. But I really wanted a placement puzzle, so when I perused the list of genre this afternoon I realized Ripple Effect would work remarkably well. There was only one problem: I had no idea how to start!
An hour later, I was mostly done, but it ended up taking another hour later in the day to get the last few parts to actually resolve correctly. Probably would have been easier if I'd allowed myself clue numbers... anyway, for the interface the regions filled with question marks can be considered not part of the grid and do not necessarily need to have valid values. This just allowed me to essentially have a non-standard grid shape.


Also as a reminder, the instruction booklet for my upcoming LMI contest Puzzle Zodiac is currently available! You can participate anytime between October 9th and 14th, and I'll be uploading all of the puzzles I constructed for it here after the contest ends.

The variation series I mentioned a few posts ago will still happen, I'll just probably hold off on it until 251-275 instead due to the delay on this puzzle.

Saturday, September 5, 2020

Puzzle #199 - Sum Thing New (X Sums Sudoku)

I'm still not really a fan of classic Sudoku- there just aren't interesting deductions to me and it feels like the ruleset is missing a constraint to make it work well. That said, the fact that it's missing "a constraint" means that the missing constraint can be a wide range of things which makes it a flexible base for interesting puzzles. Inspired by a puzzle duel puzzle from a while back, I decided to try making an X Sums Sudoku with, of course, clue symmetry and theming. Solve as a standard Sudoku, additionally each outside clue gives the sum of the first X numbers from the clue, where X is the first number. So 425631 would give a sum of 17 from the left, and 1 from the right.

Monday, August 31, 2020

Puzzles #197 & #198 - Jagged River (Gokigen/Creek)

I have to say, when I set out to create a puzzle that would work as both a Gokigen and a Creek, I didn't expect to run into problems filling the last corner. See if you can guess where it was from the redundant clues- though I'm still pleased with the construction and aesthetic here, even if both turned out a little easier than anticipated.
And with that, I managed to post at least one puzzle every day this month!

Sunday, August 30, 2020

Puzzle #196 - Exponential Hybrid (Masyukabe)

Man, it's been a little while since I created a hybrid of multiple genres- while brainstorming some variant puzzles I want to create, I hit on Masyu and Nurikabe and decided to construct it now rather than later. Solve as a standard Masyu puzzle. Additionally, all connected groups of cells not on the loop must contain exactly one number, indicating the size of that group as in Nurikabe. The loop may not pass through numbers.


puzz.link interface      penpa interface

My current posting plan is one more pair of puzzles tomorrow, then a bit of a break until I can post my next puzzle duel construction, then puzzle #200. After that, the next 25 should all be variants and hybrids so I can focus on that area some more, and then I'll work in the puzzles from my upcoming contest on LMI (starting in about a month!). I should be able to reach 270 by the end of the year...

Saturday, August 29, 2020

Puzzle #195 - Lost in the Labyrinth (Nurimaze)

Now that I've actually learned how to find topological arguments, I can make much more interesting Nurimaze puzzles than I used to. Granted, this is only the second one that I've ever made, and the first ignored circles and triangles, but still- I can't see myself coming up with this a year ago!

puzz.link interface

Friday, August 28, 2020

Puzzle #194 - Three Out of Five (Snake)

It's kind of amazing how quickly my opinion on Snake puzzles has turned around, now that I understand how to make some deductions in them. I still end up having to try things out to see the implications a lot of the time but I'm getting better- for instance, I made a bad deduction when constructing this puzzle but could easily add a clue to both improve the theme and the solve!

puzz.link interface

Thursday, August 27, 2020

Puzzle #193 - Fourced Triangle (Nurimisaki)

Nurimisaki seems to have no end to interesting clue combinations, as the ruleset can force quite a bit from a few seeds. Guiding that process is still fascinating to me, and I feel like I've barely scratched the surface of what the type can do.

Wednesday, August 26, 2020

Puzzle #192 - Framed Yet Again (Gokigen)

I've noticed that my design style tends to eschew clues that can obviously only resolve in one way in my puzzles. So when I realized that I could use a bunch of 0 clues in a Gokigen puzzle to set up a series of interesting deductions, I jumped at the opportunity.