Here is a nurikabe loop : http://anuragthefirst.blogspot.in/2012/08/clean-cruise.html

I was looking at the example for signage in IB. Can anyone share a couple of starting steps? Except the clues on the edge, where arrows obviously cannot go towards the edge, I am unable to make much of a logical headway.

Since the count starts including the clue arrow itself, the 2 clues have to reach some edge within 1 arrow. e.g. The 2 clue pointing downwards at R6C4 means R7C4 will have to be an arrow pointing leftward because it has to point directly toward an edge and there's an arrow on the right. This means you can cross out all the cells to the left, and there'll be an arrow on each cell on the right.

Thanks.
I realize my problem now. My printout was not showing any arrows. The shading of the arrows did not come out at all.

^
That problem didn't occur for me, during the test(I didn't take printout of the IB).

Anyways, great puzzles. I don't really like word puzzles/letter entry puzzles much, so a part of this test wasn't in my comfort zone, but I liked every puzzle I did solve. A power cut in between made certain I didn't even attempt some types I felt uncomfortable with :P So will have to try them out later.

Thanks for the test Serkan.

Edited by prasanna16391 2012-08-18 12:06 PM

Do I get extra points for accidentally solving the Spiral Battleships example in test time? I only noticed this when I went to enter the entry key and noticed it was the same as the example's entry key.

Only if I get extra points for solving 3 of the puzzles with the sole illumination provided by the laptop screen... I'm gonna solve under candlelight from next time. :|

Nice puzzles Serkan!

This time I solved almost all of the puzzle I had pre-selected.

Thanks a lot for a fun set with pretty innovative stuff!

I personally thought some puzzles were under/over valued.
The two hours went so fast for me I didn't get to see some pages.
A great Serkan test as always. Thank you!

A puzzle test after such a long time, but performance was not that par which i expected.
Puzzles were excellent and yes i dont like word puzzles and never gave a look to it in the test :p
Anyways Thanks Serkan!! for such a wonderful test.

Thanks Serkan for excellent puzzles and a very enjoyable test.

Like Swaroop, I also gave the word puzzles a miss. Probably we should have an exclusive test at some point for word puzzles.

I never believe this. Because it only depends on solver ability. Some solvers can be good at math puzzles, some of them can be good at word puzzles etc. So from my side there is not any under/over valued things in tests. And I don't mean any extreme examples. For example, for this test three solvers tested all puzzles. Honestly there were some huge differences for some puzzles. So when I determine the puzzle points with considering tester times, I always use average solving times, and sometimes I interfere in this average value. But in any case I follow average solving times.

That seems to be a common theme among the Indian participants. If you refer to my post above, the "uncomfortable" part did indeed refer to word puzzles(well, letter-entry puzzles too, but we know that weakness from the IPC don't we?). I'd definitely have given 2-3 of them a try if I had a little more time after finishing the ones I actually did solve though.


Edited by prasanna16391 2012-08-20 1:03 AM

Thanks for the very excellent test Serkan; almost a perfect USPC preview (if I just had 2.5 hours to finish and check everything!).

I'm glad I had almost all of the back issues (7 of 9?) to practice on these puzzle types before starting. Even those who do not win the prize should consider getting AO regularly.

hi..

just wanted to know the exact time duration of the test, and whether it is available for online solving or do we have to take a printout?

IB Sudoku Builder seems to have a solution typo. Arrows ask for R4,C5. But the given solution is for R2,R4.

When both rows and columns have solution arrows, it appears the row arrows should precede the column arrows. Is that right?

Yes. That is a typo in the IB.

That is correct. But we do manual checks for each wrong submission and you will get points if the answer keys are swapped.

120 minutes. Not available for online solving, you'll have to take a printout.

Nice puzzles, thanks! Good variety, creative, challenging (for me).

Password removed from puzzle booklet.

Solution booklet uploaded here. It also has the list of authors who wrote the corresponding puzzles for akil oyunlari magazine.

Congrats to the top 3 : Motris, Deu and Mellowmelon
Puzzles were nice. As i said earlier i am weak at Word puzzles so have to look upon it now. I liked calculations most among which i have submitted others i have to still try.
Spiral Battleships second i solved twice and was stuck in middle at the end with 18,19 and 20 and now looking at the solution it seems i missed that way.The RH factor one i solved completely
i submitted the right solution and then dont know what happened saw the second row AB as -ve one although i had written +ve over there and submitted the wrong solution just 2 seconds before my time got over.
So i guess it could have been better but at the end nice test..
Thanks to serkan and LMI for this test :)

Congratulations to Thomas. He passed Hideaki with solving 20 of the 21 puzzles in 115 minutes. They solved the same amount of puzzles with the same puzzle left out, but Thomas was 3 minutes faster than Hideaki. I think the New Style Crossword was a bit hard. Actually the test was hard in general. Thanks to all participants.

The top three are as follows:

1. Thomas Snyder (USA)
2. Hideaki Jo (JAPAN)
3. Palmer Mebane (USA)

These three people won a 1-year subscription for Akil Oyunlari magazine.

Some statistics:
- The puzzle solved by least competitors was New Style Crossword. It was solved by 9 competitors.
- The puzzle solved by most competitors was Sudoku Builder. It was solved by 124 competitors.
- The puzzle with the highest rating was Clueless Scrabble (9.66) and the puzzle with the lowest rating was Skyscrapers Rising 1 (7.08).
- The puzzle submitted with the highest accuracy was Chained Ships 2 (%100, 63/63 competitors), lowest was Signage 2 (%72.72, 16/22 competitors).

Hope to see you in the next competition!

Best

Serkan

As I've already said, this was a lovely test.

I generally do well on word puzzles so I wanted to share my thoughts on the two ones here. I think the New Style Crossword was at least 25% too large, or 25% under-connected or both. I spent about 5 minutes in the test on it, getting a few potential words in at the top but not with enough links to be sure I was on the right path. The puzzle was progressing way too slowly to invest more time on it. My guess was it would take about 20+ to finish it which at 8 points per minute was worth too little.

On the other hand, I loved the Clueless Scrabble. With lots of potential interlinks, once you had some elements of the solution (for example the top row was likely used, with three descending words, and two other crossing words), the puzzle accelerated nicely. I can't extract my exact timing as I submitted this with half another sheet done, but this was more like ~3 minutes for me.

In a construction sense, I've struggled with this on my Decathlon tests too, between building a puzzle that is too tedious versus one that rewards a good observational and logical approach. But I'm still always glad to see word puzzle styles to add more balance and offer different challenges.

Edited by motris 2012-08-22 12:58 PM

When I began to construct New Style Crossword I had a nice point. There was only two words (ÇILEKE$ - VAZIH) with first and last letter appearing only once. So there were only two places to fit these words, after that solving process was going smoothly. I designed it based on this point. But then I remembered answer key format, I mentioned to write 3 unused words in the IB. So the starting point which I planned was not useful anymore because of asking 3 unused words for submission. So I changed some words, I tried to push some logic; but it didn't go well. Because I constructed all grid to consider ÇILEKE$-VAZIH fact. At the end, this one was hard puzzle, and it needed more searching than logic. One of the testers solved this puzzle in 15 minutes, the other in 22 minutes. So it seemed to me nearly 16-18 minutes puzzle.

About Clueless Scrabble, actually there was a lovely idea. There are 16 words, and 3 of them would not be used, so I tried to fit 13 words in the grid. If a solver writes possibilities of placement for each row and column, it would appear that 6 words should be placed in columns; 7 words should be placed in rows.

Edited by yureklis 2012-08-22 4:16 PM