2013 USPC

Information and discussion on Puzzle Competitions
Post Reply
sknight
Posts: 141
Joined: Mon 01 Nov, 2010 8:46 pm

2013 USPC

Post by sknight » Sat 11 May, 2013 11:12 pm

There's an announcement on the home page of the US Puzzle Team that the 2013 USPC will take place on June 15th.
I haven't seen any email or details yet, but since all the information tends to come out pretty late, you may want to pencil this into your calendars.

emma
Posts: 314
Joined: Wed 22 Jun, 2011 4:58 pm

Re: 2013 USPC

Post by emma » Fri 14 Jun, 2013 5:15 pm

I'd nearly forgotten about this, so I'll post to remind others!

kiwijam
Posts: 699
Joined: Wed 01 Jun, 2011 10:03 am

Re: 2013 USPC

Post by kiwijam » Sat 15 Jun, 2013 1:39 am

The instruction book has just been published (2 minutes ago).

Be aware that the competition window is only 12 hours long, from 6pm Saturday to 6am Sunday.
Although it is a 2.5 hour test, so that effectively means it's best to start before 3:30am Sunday...

GarethMoore
Posts: 1209
Joined: Thu 24 Jun, 2010 9:27 pm
Contact:

Re: 2013 USPC

Post by GarethMoore » Sat 15 Jun, 2013 2:25 am

I guess they didn't send any emails out for this either. Shame, because if I'd known about it just a day earlier I could have arranged things so as to have taken apart.

dickoon
Posts: 615
Joined: Mon 21 Jun, 2010 12:52 pm
Location: Stockton-on-Tees, UK
Contact:

Re: 2013 USPC

Post by dickoon » Sat 15 Jun, 2013 8:42 am

The puzzles look like a lot of fun IMHO. I'm working the night shift tonight and my 12-hour shift fairly nearly overlaps with the 12-hour window of the test. If it's a quiet one...

dickoon
Posts: 615
Joined: Mon 21 Jun, 2010 12:52 pm
Location: Stockton-on-Tees, UK
Contact:

Re: 2013 USPC

Post by dickoon » Sun 16 Jun, 2013 12:12 pm

Well, I'm claiming a literally unbeatable 33. Bow down in defeat! :mrgreen:

PuzzleScot
Site Admin
Posts: 2739
Joined: Fri 18 Jun, 2010 10:45 pm
Location: Edinburgh, Scotland

Re: 2013 USPC

Post by PuzzleScot » Sun 16 Jun, 2013 8:07 pm

Damn. Missed it. Was going to do it tonight...

detuned
Posts: 1869
Joined: Mon 21 Jun, 2010 2:25 pm
Location: London, UK
Contact:

Re: 2013 USPC

Post by detuned » Sun 16 Jun, 2013 11:31 pm

I did it at stupid o clock and had gave a pretty bad account of myself.

I'm going through some of the puzzles I didn't do in time. I've just seen the solution to Duello. I have to say the instructions and the example were far from adequate preparation for this puzzle. It's completely non-obvious a priori that you are allowed to use 0!

And I wish I'd read the pentopia instructions more closely - I tried this a couple of times, presumably doing the right thing, but giving up each time because it quickly became apparent that all 12 pentominoes were never going to fit into the grid, like the do with most of Bram's examples that I've seen before.

Interesting to see Thomas provided a counting puzzle, giving his stance on these things. I'm not entirely sure this was an improvement on the genre. I got the wrong answer out but I freely admit to using the calculator on my laptop when it became clear the answer was going to be in the order of tens of thousands.

The gimme points for favourites like battleships/masyu/tapa/cave were all ok. Looper seemed like a bit of a slog with plenty of trial and error before finding I had broken it. Perhaps this is a reflection of having never done one of these before, but I think a loop puzzle isn't worth bothering with if the apparent spectrum of difficulty is trivial-easy-trial and error. This is (in my opinion) why masyu and slitherlink are such classics of the genre.

I went back and forced myself to logically solve the star battle, because I rely on trial and error for these far too often. I mean, it's my belief that any star battle of WPC level is going to be most quickly solved via trial and error, but I reason that I should be making myself more comfortable with some techniques to make sure that when I reach a stopping point a guess is going to have the best chance of getting to the end of the puzzle (rather than hitting another stopping point!)

My printer garbled the tomtom, and I botched both the kakuro and the bombardoku (nice idea, which used the geometry of the grid to guide the solve).

Thomas' word search was nice, and unusually I didn't get the other word puzzles. Solving afterwards I thought ambiguity was a bit of a slog, but follow-up was nice. That flows quite nicely after you get the initial sahib/beamish/hush/quotas/jealousy/squeeze/bonanza configuration, which is all forced. During time I got hung up trying to place wielding/wanters.

I kind of liked the thermo skyscrapers but I think the jury is still out as to whether it's the novelty of the gimmick linking grids to clues providing the entertainment vs whether there is any longevity in it. I've seen it with arrows, and odd/even shading now. Certainly there's some nice interaction with thermometers effectively giving you a guaranteed number of visible skyscrapers. Maybe me being slightly churlish about the whole thing is a reflection on my general feelings about skyscrapers though...

My highlight was definitely the sudoku - which was an absolutely staggering construction and probably took me in excess of 10 minutes to solve (perhaps I should have guessed but I was enjoying myself too much!) There can only be a handful of authors in the world who can create something of such brilliant quality. As a fairly prolific sudoku author myself I have no idea how you'd go about constructing something like this! It was like 3 or 4 highly intricate Just One Cell puzzles rolled into one!

joshuazucker
Posts: 8
Joined: Sun 10 Jul, 2011 6:53 pm

Re: 2013 USPC

Post by joshuazucker » Mon 17 Jun, 2013 1:05 am

I loved the Bombardoku -- the mechanism of "marked cells are correct clues and unmarked are incorrect" is really fabulous. I enjoy David Millar's Bigfoot puzzles for the same reason. I also had a great time with the cards kakuro but maybe that's more an artifact of how quickly I saw what was going on and got to the solution. The thermo skyscraper was quite nice as well with a bunch of nice moves of self-constraining thermometers where if the bulb gets too big then the even bigger numbers along the thermometer block its view, as well as several other elegant clue points along the way.

The sudoku was quite tough, but then it was 20 points so maybe it was about right for a vanilla sudoku for that point value. I'll have to solve it again in non-test conditions where I can spend more time appreciating its beauty. Saturday night I was on an airplane (United) and their in-flight magazine's hard sudoku was actually pretty nice too, much to my surprise.

Overall a fun test as always, leaving me with a few puzzles left to work on -- I'm pretty happy about the way the amount of leftover unsolved puzzles seems to be decreasing for me, year by year. At least, I think that's more about me learning how to solve better than it is about the test getting easier.

kiwijam
Posts: 699
Joined: Wed 01 Jun, 2011 10:03 am

Re: 2013 USPC

Post by kiwijam » Mon 17 Jun, 2013 3:17 am

I swore at the Duello.
I recognized beforehand that the rules were ambiguous about whether 0 could be included, and also the huge example didn't have a single one. But there's no chance in the USPC to ask for clarifications, or to point out other typos in the instructions... (unique to this contest I think, is this just an old tradition?)
But if I assume that a 0 might be possible when it's not then I've got twice as many options to consider, so it'll take me twice as long as it should. Anyway... I eventually reprinted the puzzle to start again.

A few other gems mixed in with some rather hard puzzles.
Siamese Fences was a strain, twice I needed a break and did a 10-pointer before finishing it off. I think the two grids made it tougher than a Colorlink to keep track of everything.
Number Tower was cute ("oh that's how it's worth 20 points"), using different shapes from the example.
My answer for the counting puzzle was 99.991% of the correct answer. Which means I get the same number of points as someone that wasted no time on it. Every other type of puzzle you usually know when you finish that the answer is correct... I'm starting to think these are a much riskier use of time than a puzzle with guaranteed points at the finish (assuming you enter the answer key correctly!)
@Tom C: I think the trick for Looper (which I took a while to find) is that the cells diagonally beside the 4-clues cannot turn 'away' from the 4. This opens up the top quickly.
With 15 minutes to go I started the tricky Tomtom, not a good move, I wish I'd looked at the Kakuro instead, it wasn't as tough as its example.

puzzlemad
Posts: 855
Joined: Mon 21 Jun, 2010 10:20 pm
Location: Oldham

Re: 2013 USPC

Post by puzzlemad » Mon 17 Jun, 2013 5:47 am

I had a quick glance through the instructions before diving in. Broke too many of them, so my score won't be brilliant either (not that it is ever THAT good).

detuned
Posts: 1869
Joined: Mon 21 Jun, 2010 2:25 pm
Location: London, UK
Contact:

Re: 2013 USPC

Post by detuned » Mon 17 Jun, 2013 7:55 am

The thing with the Duello is it's quite a nice idea - I think it uses the ripple effect constraint in a much nicer way than ripple effect does, but I'm not sure how constrained the thing is which ultimately determines how much life there is in the idea. Palmer mentioned he'd written a script to generate a list of possible rows. The implication that he could spend a good while studying this list suggests it isn't all that long....

Palmer has more thoughts up here. He's picked up on some common themes...

update on the counting problem: So i had 8 instead of 10 for middle to far corner. Adding 7x powers of 10 appears to be a little easier than adding 7x powers of 8. But this still raises an issue in my mind - something like this can only be nice if the solution uses a gimmick like powers of 10. And once you are aware of this kind of meta knowledge, the puzzle idea kind of becomes dead. my bad for miscounting!

kiwijam
Posts: 699
Joined: Wed 01 Jun, 2011 10:03 am

Re: 2013 USPC

Post by kiwijam » Mon 17 Jun, 2013 9:48 pm

detuned wrote:The thing with the Duello is it's quite a nice idea - I think it uses the ripple effect constraint in a much nicer way than ripple effect does, but I'm not sure how constrained the thing is which ultimately determines how much life there is in the idea. Palmer mentioned he'd written a script to generate a list of possible rows. The implication that he could spend a good while studying this list suggests it isn't all that long....
I handwrote the list, it didn't take too long when you recognise there must be an even number of odd-pairs. Not slower than programming anyway. There were only 4 options (excluding 0s) of size 8 or less.
I agree Duello has potential, there's some neat interactions possible. This particular grid idea breaks it down into a little puzzle for each row and column, there might be a better way to make it two-dimensional. Maybe if each digit only has to be paired to one other (horiz or vert) identical digit, at the appropriate distance?

Janix
Posts: 381
Joined: Tue 22 Jun, 2010 11:59 am

Re: 2013 USPC

Post by Janix » Wed 19 Jun, 2013 5:53 pm

Dickoon

I got wrapped up in things and only submitted three answers, one incorrect so you'll be pleased to know that the support has support...

Mainly writing to say that it was refreshing to see such a wide range of new puzzles although the absence of the instruction booklet until only ? hours before the event did not help - I didn't see it.

Disprove of a plain Sudoku in there and out of choice I refuse to tackle such puzzles on the grounds that there can be no guaranteed transparency here.

sf2l
Posts: 147
Joined: Fri 03 Sep, 2010 12:43 pm

Re: 2013 USPC

Post by sf2l » Sat 22 Jun, 2013 6:30 am

the results have been published. http://wpc.puzzles.com/uspc2013/results.asp?mode=intl
congratulations to Kiwijam and Murray Hughes for their majestic performance.
i was very lucky. I did a stupid mistake that costed me 10 points in the battleship. On the other hand, i broke the siamese fence but it was correct in the two rows that counted.....can't complain.....

dickoon
Posts: 615
Joined: Mon 21 Jun, 2010 12:52 pm
Location: Stockton-on-Tees, UK
Contact:

Re: 2013 USPC

Post by dickoon » Sat 22 Jun, 2013 8:01 am

Some world class scores on an evidently very difficult test; congratulations!

PuzzlerNickG
Posts: 64
Joined: Mon 21 Jun, 2010 6:22 pm

Re: 2013 USPC

Post by PuzzlerNickG » Sat 22 Jun, 2013 9:57 am

I should have scored 140, but I made some silly mistakes. I was close to finishing Ambiguity when time ran out.

kiwijam
Posts: 699
Joined: Wed 01 Jun, 2011 10:03 am

Re: 2013 USPC

Post by kiwijam » Sat 22 Jun, 2013 12:00 pm

sf2l wrote:congratulations to Kiwijam and Murray Hughes for their majestic performance.
Thanks!
I was a bit grumpy with myself at the time for breaking a couple of puzzles and submitting a wrong answer, but it looks like everyone found this one difficult.
And congratulations to you Stefano also, an excellent score and comfortably top of Italy (which means you qualify for the WPC team I guess?)

sf2l
Posts: 147
Joined: Fri 03 Sep, 2010 12:43 pm

Re: 2013 USPC

Post by sf2l » Sat 22 Jun, 2013 1:24 pm

qualification is not an issue, there are more places available than competitors willing to fill them.
the important bit is that this was announced to be valid as Italian championship. :D
btw, i had the same problem as you
kiwijam wrote: With 15 minutes to go I started the tricky Tomtom, not a good move, I wish I'd looked at the Kakuro instead, it wasn't as tough as its example.
the start of the tomtom was obvious with 20x and the 105x underneath. i went the wrong way when putting the 7 next to the 8/2 above. otherwise it would have been feasible.

Post Reply