byronosaurusrex's Journal

The US Puzzle Championship: The most fun you can have being stressed out for hours on end.
Well, here we go; very late into the lifespan thus far of this journal, and I'm finally posting. A little bit of introduction, I suppose, is necessary for the unlikely stumbler-upon of my journal: My name is Byron Calver, and I'm currently on the tail end of a bout of minor celebrity status in my hometown for winning the American Sudoku Smackdown held at the American Crossword Puzzle tournament. This ranks as my proudest achievement, over a surprisingly long list of not-quite-so-prestigious hono(u)rs, which I will spare the effort of recitation.

Today, I competed in the US Puzzle Championship. The good news is that I currently estimate my best-case score as 343 out of a possible 385. The bad news is that this is a drop from my previous estimate of 363, as I made a silly mistake, and the way these things tend to go, it won't have been my only one. The very good news is that since I placed in the top ten at the previous World Puzzle Championship, my Canadian team position was already effectively set for this year in Bulgaria.

Now, a rundown of how I got to this result.

1. Battleships, traditionally the first puzzle of the USPC since 2003. 15 points (including a five-point bonus.) Usually, I can get it done in a minute or two, copying the important details onto graph paper before the first page actually prints. This time, it didn't go so well; possibly, my graph paper strategy backfired, as I was unable to connect important information from the sides of the grid to the ships placed quickly, and ended up leaving and restarting it on the test paper itself later on. This was a disaster to my confidence, which would leave this test worse than my potential score implies.

2. Sudoku, 10 points. 6x6, so probably the easiest points available to the novice solver.

3. Pentomino Puddles. 10 points, and about three or four separate efforts to try various different contents of each puddle. Eventually, I got an answer that seems satisfactory.

4. Piecework, 15 points. Separate a 7x7 grid into pieces containing two numbers, representing perimeter and area. This one was a rarity on the USPC; a puzzle where one could benefit from thinking about it in advance, since there aren't too many possible area-perimeter combinations. Otherwise, one might become a tad lost thinking about the seemingly unsatisfactory amount of information in the grid.

5. Septoku. 10 points, another fairly straightforward sudoku variant, although the hexagonal shape could prove more confusing to solvers traditionally used to squares. It's a classic way to take the solver out of their comfort zone; pull out the hex grid and watch them squirm.

6. Benign (B-9) Crisscross, 20 points. A surprisingly easy one, by USPC crisscross standards; just start in the middle where they've given you a Z, work through the options for a given space (in most if not all necessary cases, only two potential words), and voila.

7. Mice. One point for each difference, 5 bonus if you get all ten, 15 total. This was a tad more apt for testing participants' observation skills than some previous years' difference-finding ones, as it did not fall to eye-crossing due to its mirrored image. It would have been nice to get a co-ordinate grid on the mirrored image as well, however, but it was a nice break from heavy-thinking puzzles. I only found eight differences.

8. Bulgarian Provinces. 20 points: The annual word search, whose score value has oscillated wildly in championship history and reached a sort of middle ground this year. More easy points, if you have a good idea how to go about this sort.

9. Round Trip. 15 points, the first in a series of pathfinding puzzles that semi-made up for the lack of an interesting maze type (Rolling Block, Icon, Factor, etc.) This is where regular visits to http://www.kwontomloop.com/ paid off, as Fences practice helps build some of those loop-making skills that proved necessary.

10. Arrow Ring. 15 straightforward points, as the basics of loop-making effectively dealt with any uncertainty as to where to place black squares.

11. Closing the Loop. 10 points and a 5 bonus, both of which I am still as of yet unsure as to whether I got; I was left a tad boggled by the fact that B and E seemed to be visitable in either order. If there was only one, please refrain from telling me; I don't think I want to know. Despite the title, this one didn't actually involve loopmaking skills, as it was more a straightforward maze, a less interesting version of the Icon Maze, which I didn't get last year and kicked myself for. EDIT: Never mind. See update at bottom.

12. Hex House. 5 points, surprisingly easy to make a successful guess, to my delight. Not much of a fan of those type, because they tend to end up guess-heavy.

13. Packing For Bulgaria. 25 points, and this is more like it! Word-placement, combined with the "snaky tiles" concept, makes a fun, challenging logical solve. Intimidating to look at, but entirely fair for giving openings to the solver. This is the sort of puzzle that makes a great test for a solver.

14. Welcome To Bulgaria. Hey, sensing a theme? 15 points for observation. The key to this kind of puzzle is to try and separate your thinking into tiles that share a characteristic; last year and the year before, it was the number and placement of the groundhogs/rats, and this year, it was order of names on the sign and the first direction listed. That separates the elements into more manageable sets to scan.

15. Suspects. 10 points. Not a type I'm especially fond of; seemed to succumb, however, to a bottom-up trial-and-error method. That's why I'm not fond of it; trial and error isn't nearly as fun as a solve where you can put in things with a little more confidence as you solve. It's still a form of logic, but not nearly so rewarding to the solver.

16. Traffic Signs, another 15 points, another one that I pretty much guessed my way through. Twice, actually, since originally I thought the example's lower-right sign was a typo. I solved it for what I thought the signs meant, then realized what they actually meant and solved for that. My problem was that I initially interpreted the right/left turn signs as straights, because they were straight arrows as opposed to the bent ones one might expect for a turn.

17. Atomic Fusion. 15 points, didn't solve. Didn't have much in the way of an illusion of solving. Even moreso than Observers last year, this was my Achilles' heel. Rotational symmetry is a lot more fun when you know where the centers of rotation are.

18. Sudoku, 25 points, nice and easy. Good for calming the early case of nerves that all these not-solved-on-first-try puzzles brought on. Another good thing for the casual solver to attack, as it fell to nice straightforward solving and wasn't very different from the usual sort of sudoku.

19. Tetris Crisscross. 20 points, and with a nice intuitive solve, as placing the had-to-be-black squares left the positions of the A's fairly easy to determine, leaving a pretty straightforward crisscross.

20. Slide Weight. 15 points, or in my case, -5, as I made my traditional really stupid error; in this case, I misread the top fulcrum, and multiplied the load on the left balance from the lower weights by 2 instead of 3, causing me to get an answer only one switch of weights away from the correct one. That one was the vicious breed of puzzle where the options seem overwhelming, and the only way to solve is to exhaustively search them. I don't enjoy it, but I had the right solution method.

21. Fences Variation. 20, and another 5 for the bonus makes 25 points that I very nearly screwed up. All my Kwon-Tom Loop solves, and I manage to make a mistake on my first attack at the fences variant. Fortunately, I got the better of it on my second look, but it still got me on a confidence level. But not as badly as...

22. Plus/Minus Kakuro. 35 + 5 bonus = 40 points, making this one a real gamebreaker, but one I was excited to take on since I consider Kakuro/Cross Sums a personal specialty; however, it proved more challenging to me than I had imagined it, and I only got about halfway through on pure logic, leaving myself hung up on a "It's this one or that one, but I'm not sure how to tell" position. I went with one, and it seemed to allow the rest of the puzzle to fall out elegantly, but I'm still not confident. It took the wind right out of my sails.

23. Distance. An oddly anticlimactic 15 points, but a surprisingly enjoyable solve for said points, as I tend not to have much fun with the Distance type; however, there weren't too many distances not used in the course of this, making it a sort of wander through every path the numbers could take up to the point where they can't be put in proper distance. It's always a nice feeling to get a puzzle type that you don't feel all that confident in and then surprise yourself.

All in all though, some surprising solves on puzzles I don't commonly enjoy didn't make up for my disappointing performance on the ones I do, so while 343 is a nice big number, I am not left with confidence that it's going to hold firm.

Update:...And it won't. Believe it or not, I have managed to both botch royally the rules of a puzzle and misread a diagram in the same USPC. I read Closing The Loop's fairly unambiguous "pass through every light grey 1x1 square once and avoid the dark rooms" rules with the thought in mind that we'd have 1x1 rooms somehow. This morphed the rule for me into "pass through every light grey room once" by the time the test came around. Call it 323 now, with potential to go lower.

The worst part: I only realized it when my brother started solving the puzzle and making sure to follow that rule.

Current Music: The US Open: The USPC of golf.

Advertisement