Nikoli Selection - LMI July 2011 Puzzle Test #1

[debmohanty]

Nikoli Selection - LMI July 2011 Puzzle Test #1

Author : TOM “detuned” COLLYER

Dates : 9th and 10th July

IB and Submission Link : http://logicmastersindia.com/M201107P1/

[detuned]

I reckon this one will be alright

More seriously, if you enjoy the puzzles on my blog, then you will enjoy this test. There will be plenty of the twists (only in the puzzles themselves, this time) to keep even the most seasoned puzzlers on their toes!

[PuzzleScot]

There will be plenty of the twists (only in the puzzles themselves, this time)...

You hope!

Seriously, I hope it goes well, and plenty puzzlers play. Good luck!

[detuned]

we've had the apparent luxury of time to test it properly and extensively this time round - it has the McNeill stamp of approval as well!

[debmohanty]

Nikoli Selection has started

[PuzzleScot]

On 'Ripple Effect', the instructions say "Any given number m must be at least m cells away..."
However, the example implies this must be "... at least m+1 cells away". Othersise 1s could be adjacent to ('1 cell away from') another '1'.
I presume that's the interpretation to use in the actual puzzle...

[Mephisto]

From my site: "Write a number into each cell of the diagram so that every room of size N contains each number from 1 to N exactly once. If the same number X is written twice in a row or column, there must be at least X other numbers between these two numbers. For example, between two cells with the number 2 there must be at least two cells with other numbers."

[PuzzleScot]

Nice little contest Tom - Thanks! . Perhaps unsurprisingly, I didn't even get to open the Marathon part before the timer was up!

[joshuazucker]

Thanks for a fun contest! Like most participants, I suspect, I didn't get a chance to look at the marathon puzzles yet. I have a few left from the main test, too -- two of them being puzzle types I haven't practiced much, so I didn't really try to solve them during the test, and one that I just completely got stuck hunting for the next logical step!

I'm looking forward to spending more time with that one, especially. I bet I'll really learn something from it.

Some of these puzzles were really beautiful. I particularly enjoyed the LITS, Yajilin, and (even though I had a pretty tough time with it) the Shikaku. The Masyu was pretty interesting too, and I thought this was a much better than average Kakuro. The sudoku was pretty nice too. As you can tell, I had a lot of fun with these!

Thanks again,
--Joshua

[gareth]

I really enjoyed all of these too - thanks Tom. I particularly liked the Suraromu - I often find you have to deliberately not use uniqueness on these so as not to make them trivial, and while it was a good technique on this one it far from solved it in a single move.

[detuned]

Well the meta-clue for the Suraromu was with its deliberate pairing with numberlink. There is a clear logical path to the puzzle, but I thought if people were in the numberlink mindset but apparently stuck on the suraromu, then they'd at least have a fighting chance of getting it!

Thanks all for the comments, glad you all enjoyed it, and even more glad that everything went by without a hitch this time

[drsteve]

Well the meta-clue for the Suraromu was with its deliberate pairing with numberlink. There is a clear logical path to the puzzle, but I thought if people were in the numberlink mindset but apparently stuck on the suraromu, then they'd at least have a fighting chance of getting it!

Thanks all for the comments, glad you all enjoyed it, and even more glad that everything went by without a hitch this time

Lovely Suraromu as normally these are fun but trivial. Spent a while on it getting nowhere. After throwing the Kakuro at the wall, had another go with the notion of there being a unique solution and bashed it out very quickly - not quite quick enough to submit the solution though...

[dickoon]

Can someone go into more detail about how to use the existence of uniqueness to solve Kakuro, please? I thought "ooh good, I can usually do Kakuro" but spent a good half-hour on it, getting the stripes in the middle and about a third of the rest. For me, it came as a reminder of how different hand-generated and croco-puzzle examples of familiar formats can be :-/

I did enjoy the test, though, particularly "Ripple Effect" which was a new type to me but which solved very smoothly. Also a word of thanks for keeping the test relatively short at 90 minutes; 2.5-hour monsters are fun in their place, but there's definitely room for shorter ones as well.

[detuned]

I would like to know more about kakuro and uniqueness too!

My intended solution starts in the bottom right, with the 7 clue consisting of three cells. This must solve (from top to bottom) as 1,4,2, since 44 can't have a 1.

Similarly the 4 clue (two cells) is 3,1 and the 11 clue next to it is 2,4,5. We now move along the long 44 clue.

Skipping ahead slightly, the 10 clue (two cells) must be 6,4 otherwise we won't be able to add up to the crossing 21 clue. This leaves the remaining numbers in that 21 as an 8/9 pair.

Now have a look at the crossing 17. it crosses the 44 in either 5/7/8/9. Except it can't cross with either 8 or 9 because then the 17 clue will fill up with a sum strictly greater than 17! A little analysis on the lower diagonal means that 12 crosses the 44 with a 5/7 too, which then leaves a further 8/9 pair at the end of 44 clue.

I believe once you have got all this going, the rest of the puzzle isn't so hard. You might be tempted to try and skip ahead by setting the 44 clue as 23456789, but it might be a while before you work out whether you have the right answer or not

[gareth]

I would like to know more about kakuro and uniqueness too!

I read Steve as saying that after the Kakuro he returned to the Suraromu and used uniqueness on that, where of course it helps a lot.

I normally avoid Kakuro in competitions but trying Tom's example beforehand persuaded me that he'd use very Nikoli-esque logic for these, so I gave it a go and indeed he had. When I had about 5 squares left I guessed at the end for speed, but I probably missed something and other than that I found it was all very clean logic.

[drsteve]

I would like to know more about kakuro and uniqueness too!

I read Steve as saying that after the Kakuro he returned to the Suraromu and used uniqueness on that, where of course it helps a lot.

I normally avoid Kakuro in competitions but trying Tom's example beforehand persuaded me that he'd use very Nikoli-esque logic for these, so I gave it a go and indeed he had. When I had about 5 squares left I guessed at the end for speed, but I probably missed something and other than that I found it was all very clean logic.

So did I! Uniqueness of Suraromu, not Kakuro. Apologies for my English not good being.

[motris]

While it wasn't true on this kakuro, there are many uniqueness opportunities with them. They pop up exactly as they do in sudoku puzzles where you're looking for forbidden rectangles. For example, if you have two vertical clues that are a 17 (two-cells) and a 24 (three-cells) with the cells of the 17 seeing two of the cells in the 24, then the third cell in the 24 cannot be a 7.

[PuzzleScot]

Those 3 marathon puzzles kept me busy on holiday - they seemed to be increasing in hardness between 1 and 3 hours each!
Did you really create them yourself? Wow.