What was your opinion of the distribution of easy/hard puzzles?	Just right
What did you think about the puzzle quality of the test?	Fairly Nice
Which set of puzzles did you like the most?	Hidden Statue
What was your opinion about the answer key extraction?	Answer keys could have been better
Of the puzzles you solved/attempted, how well did the point values reflect the difficulty?	Many puzzles were worth too much or too little
What was your opinion of the booklet formatting and printing?	Just right

Congratulations to Endo Ken, Yuki Kawabe, and Hideaki Jo for making it an all-Japan Top 3.

Amongst Indian participants, Congratulations to Prasanna Seshadri, Rohan Rao and Ashish Kumar for taking the top 3 places.

Of the 206 participants, 35 were from India, 28 from Japan and 23 from USA. Morey Garelick (USA) was the first to start the test while Esther Naef (Switzerland) was the last person to start the test. 9 participants completed the set within the allotted time. The median score of the test was 41.

Thanks Amit Sowani for the puzzles!

Participants may share feedback on the results page and rate the puzzles on the contest page.

Thanks for your feedback Matus. We are contemplating a polling mechanism for selection of puzzle types for each contest.

For this particular contest with Number and Object Placement as the theme I went with three each. In wanted number placement to be different from symbol placement so each puzzle used a different properties of numbers. Futoshiki has number comparisons, Ripple Effect uses numbers to indicate distance and Partiti has number arithmetic.

For object placement I went with puzzles which have a given set of objects which need to be placed. Most of these are classic types.

Partiti is a slow moving puzzle with local logic. I attempted to make the grids smaller and introduce symmetry to aid with solving.

Thanks for your feedback Ivan. We realized the error in Battleship formatting after the booklet had been published and the test had started. We will note this for future contests.

The answer key for Partiti was based on the need for fixed length answer keys for real time validation of answer keys.
Partiti usually solves bottom up, the smaller numbers get filled in earlier than the larger ones, hence we went with the answer key for the test.

I agree with your comment on uniqueness. It gets used quite a bit and especially in Partiti2 with the center 9 boxes missing, they have to be 1-9 to ensure uniqueness.

I hope it is the prior. Some of these puzzles required a bit of prior work in identifying strategies for solving these puzzles types.
Hope this encourages newcomers to discuss solving techniques for the new types once the IB is released for each round.

I would like to thank Tiit and Rakesh for testing the puzzles and providing valuable feedback, especially pointing out errors/non-uniqueness in 2-3 puzzles.
Their timings also aligned with my anticipation of puzzle difficulty and I was in general happy with the point distribution except for one instance where in hindsight I made the mistake of overruling their timing in Half Dominoes -3 where I expected that it would be one of the easier puzzles with all the column and row sums being given.

We were aware that the set is a lot tougher than the previous round, but the only option we had was to reduce the number of puzzles, which would mean departure from the set pattern for PR rounds. This is something we could consider in future years.

I wanted to introduce a different solving strategy in each puzzle type and succeeded for most puzzle types except Futoshiki. I was especially happy with Statue Park where each of the 5 puzzles including the exploratory variants had a different solving strategy.

Partiti was the favorite puzzle type in the set. Thanks to Thinh Van Duc Lai for introducing the puzzle type to all of us. I would also like to thank Ashish whose Partiti puzzles from WPC 2017 inspired the puzzles in this test.

The theme was obvious in some types like Statue Park and Battleship but a bit more obscure in Futoshiki and Ripple Effect. This is the 10th year of LMI so I though the theme would be appropriate for this contest.

Thanks to all the participants for adding colour to the event and providing valuable feedback. Congrats to all the participants who were able to complete the set in the allotted time. I hope others enjoyed solving the rest of the puzzles offline. Do post here if you are stuck on any of the puzzles and I would be glad to provide hints on my intended solving path.

The time count in the homepage is confusing. I wanted to start test while it was declaring the test will end in 50 minutes. But, I could not start the test in the contest page, as it said I should have started 40 minutes earlier.

Can anyone give some hints for solving Partiti-3? I was able to prove that the bottom left corner has a 2 under an 8, and the bottom right corner has a 3 under a 7 (i.e. those four boxes only have a single number in them). But past that I'm pretty stuck.

Edited by ghirsch 2018-03-01 12:59 AM

Since all the number appear in a cluster (11, 10, 9, 8 and 7) identify the possible combinations for these numbers so that can appear together in a 2X2 box of cells.

Thanks Amit, I really enjoyed the puzzle set. The Partiti puzzles in particular were a lot of fun, though they feel very slow to solve to me.

(Also I can see that it would be a tough set for for the usual PR target group.)

What's the password for this? It's not showing on the contest page itself.

PW = 2isharderthan1

nice