Archive for September, 2015

USPC 2015 is over

September 5, 2015

I haven’t found much discussion online; maybe it’s on sites like Facebook where I’m not present, so let’s dump some thoughts here.

Tentatively, I finished, with 0:02:08 left on the clock. Unlike in the last two rounds of the WPF Puzzle GP (where I honestly ran out of time and had no illusions of getting a full score), I actually had time to check everything, so I’m crossing my fingers that 375 + time bonus will hold. (Update: The score held. No errors.)

Some quick comments on puzzles (and I’ve probably still forgotten to say something nice about at least one puzzle I liked):

7. Raindrops – If my solution was right, I got tricked in the way Cihan probably wanted and lost quite a bit of time. But I’m still only 99% sure I got it.

11. Intersections – This probably had my favorite logical step, using the 9348 string to deal with the 12 that had a second 12 in its row and a third 12 in its column.

12. Kakuro – Some nice logic here too about fitting 2,5,6,8 into the top two rows.

21. O’utcast – I was pretty worried about this one before the test, with good reason since it did take awhile. I really like the amount of logic that was possible at the beginning though, especially relating to Os on the edge or in intersections. The end of the test in general had some very cool types/variations that were fun to solve (even the Sudoku-ish Pinochlada and Pathfinder).

19. Pinochlada – The hardest puzzle in my opinion, but unexpectedly fun. The start required dealing with information overload, but it was extremely satisfying near the middle when squares I hadn’t even looked at before could be immediately deduced by using rank and suit together.

4. Airhead – I saved this for the end of time as usual, though unlike past years I didn’t have anything left to do when I started it. I found this one to be very difficult; I didn’t have any differences for the first 5 minutes or so, and none of the ones I found felt obvious. I did avoid long gaps between my last few though, which is probably why I still scraped by with a finished test.