The strategy I use for Skyscrapers, beyond the obvious "Where are the Ns? Where are the N-1s? ...", is to replace clues with inequalities as much as possible and cross out clues that are redundant. So for example, in Skyscraper 6 on the test, when you get the 6 on R3C2, you can write R3C6 < R3C5 < R3C4 < R3C3 and then cross out the 5 clue. You can also sometimes cross out clues after writing in candidates for some cells. Here's an example in a partial 1-4 puzzle:

____
4___ 3
____
____

Here, if you mark that R2C4 can only have 1 or 2, and that R2C3 can only have 1 or 3, then all remaining ways to resolve that row satisfy the 3 clue and you can cross it out. (This is one of the more nontrivial uses that comes up a lot, so I just know it now.)

In general, I find that I have only one or two clues that aren't crossed out when I hit any sticking points in the middle, so then I know which clues to take a harder look at.

The biggest weakness of this approach is that it doesn't deal well with clues that give information like "the first cell is the largest of all before the 6". So it won't help as much on puzzles like this one: http://www.gmpuzzles.com/blog/2013/02/dr-sudoku-prescribes-33-skysc... Also, if you're bad at Futoshiki (aka greater-than latin squares), writing in the inequalities might not be as helpful for you.

And as I didn't win the Skyscrapers portion of the contest, bear in mind that this may not be the best advice anyway.

I have read the "LITS" instruction booklet. But I don"t understand how to play, please explain how to play step by step. I think many new beginner players is waiting.

I'll just start you up on the harder example puzzle in the IB.

1. Each region must have 4 continuous shaded cells, and they cannot form a 2x2 square. So whichever extremity you try to fit from, in some smaller regions there will always be some common cells shaded.



2. Ways to eliminate cells from being shaded, could be that if a shading starts from there you can't fit in 4 continuous cells by 2x2 rule (upper 2 yellow cells), or maybe it forces two similar shapes (bottom right).



3. Region on the left has two possible ways to be filled, either an S from the top or an S from the bottom. If it's from the bottom it forces an S in the region to the right.



This should mostly help you get through this one, and the general usage of such logic should get you through all of them. Only other step is to maintain connectivity of the wall.

Anybody know if the administrators accept points claims for these 'instant grading' type of contests? I very stupidly entered the wrong row into one of the Skyscrapers answer boxes and it's looking like it will end up costing me 13.5 points total after bonuses are awarded. [Finished up with puzzle 6 at 40:47 on the clock]

Hope everybody enjoyed these two competitions as much as I did! Great work LMI and all involved!

-John in cold Buffalo NY

Edited by john_reid 2014-01-09 4:27 AM

Actually, a further question for those in the know. Does an 'Adj Time' on the results page of '040:47(+1)' mean that my time BEFORE adding the penalty minute was 40:47, or AFTER it? I had thought it meant AFTER, but I'm a little confused now after looking at some of the other results. Thanks again for the help.

The time shown (i.e. 040:47) is the last puzzle submission time in the round. Additionally, if you have any mistakes that will be shown inside the bracket.
So, the time shown is "BEFORE", to be specific.

You can use the "submissions tabs" to see the individual puzzle submission times.

We have removed your penalty since wrong row / column submissions are not penalized. It does not happen automatically, and we have to do it manually.

Edited by debmohanty 2014-01-09 6:20 AM

Ok, thanks for doing that for me. Turns out I was wrong then and had misunderstood the timing notation.

So all the lost points turn out to be my own fault for leaving the most expensive puzzle until the end and falling short by a few seconds! This is not the first time I've made that mistake either. Oh well, live and learn.

Password removed and solutions appended.

Thanks to all the participants who participated. Hope you all enjoyed the test.
Congratulations to Top 3 in Beginners as well as Seasoners Category.
I will post more details about my experience later in short time.
A special thanks to Prasanna and Deb for helping me out throughout the process.
Hope to author few more tests in future. :)



Edited by swaroop2011 2014-01-09 9:22 PM

The top 3 Beginners in the Skyscrapers section were snakeeyes (China), Timothy (France) and cnarrikkattu (USA), and in the LITS section the top 3 were tarosan (Japan), WA1729 (USA) and Yoshiap (USA). The top 3 Seasoners in the Skyscrapers section were deu (Japan), Hausigel (Germany) and misko (Germany) and in the LITS section the top 3 were MellowMelon (USA), EKBM (Japan) and nyuta (Japan). Congratulations to all the winners. Interestingly there were no repeats in either category anywhere. This is a good thing as we've covered two very different logical aspects here, benefiting a larger pool of Beginners.

We hope everyone enjoyed this contest, and also thank Swaroop for the nice elegant puzzles. Thank you all for participating!

Hi all sorry for late reply, last few days have not been so good for me. Anyways here it goes,
So it all started 3 months ago, when I approached LMI and asked I want author next Beginners Contest. They agreed to it and gave LITS and SKYSCRAPER to prepare. Then on 5th Dec I got the complete instructions of what should be the sizes of puzzles and info regarding exploratory side, First we thought of Wacky City or Skyscraper Sudoku as skyscraper variant. But then Deb provided with the new idea of clues shared by two rows. In LITS we thought of LITSO or LITS2. I felt LITS2 was more exciting than LITSO which of course is also good. Then I started making puzzles, starting with Skyscraper as they are little easy and fast to make as compared to LITS. Then I used to send bunch of puzzles each time to Prasanna for testing. Few would become hard, few would become non unique sometimes but none the less he was always co-operative and helped me throughout the process. I finally completed creating all puzzles by 30th Dec 2013. Funny thing is after one day I realized that I have not selected answer key rows, distributed points to the puzzles. May be because of 1st time it didn’t came into my mind. Anyways on 2nd Jan 2014 we completed that part as well and it was all set to go.
My favourites were
SKYSCRAPER: – 5, 6 and 8.
In skyscraper 7, the 4 clue on the right side edge was actually a dummy clue. I mean the puzzle could be solved even without that clue but I was told to provide more clues being beginners test.
LITS: - 2, 5, 7 and 8.
I remember the time I spent on making LITS 7, I made it and it was turning out to be non-unique. But I wanted to keep that same logic. No matter how hard I tried by changing shapes some or other problem would come. I was about to leave that puzzle and start a new, but then gave an attempt on a new morning and resolved. This was the last puzzle of my complete beginners set and a joy of relief.
I once again congratulate to the winners.
Thanks again to LMI for allowing authoring a test and helping me throughout the process.
I think this is just the beginning for me hope to have few more tests in future. :)

Thanks for the test, Swaroop. Some nice puzzles, especially in LITS.

In Spiral Galaxies, should each region definitely contain a circle? Or can there be regions whose point of symmetry is not shown?

In standard Spiral Galaxies all circles / central points will be given. That means, each region must contain exactly one of the given circles.

As noted in the IB, and as it had happened in the previous beginners' tests, some rules are deviated in the exploratory category. For those puzzles, you need to look at the given example to find out the deviation in rules, if any.

Thanks, that cleared my doubt. Where can I find solving techniques for Spiral Galaxies? There seems to be more than one way of proceeding with each puzzle, and I'm able to realize very late that I proceeded wrongly.

Basic deductions:
- Two circles that are in adjacent cells have walls between them. Note that a circle is included even if it's only partially inside a cell (at the edge or corner).
- When a circle has a wall adjacent to it, the symmetrically opposite border must also be wall.

Intermediate deductions:
- Keep track of what cells a circle contains. The above two rules apply even for non-circle cells, as long as they belong to different regions.
- Sometimes, a cell is just so far from other circles; try to figure out which circle holds this cell.
- Sometimes intuition helps scoring a quick time.

Beginners' contest is delayed - we will post a note when it starts.

Any pointers on how to proceed with this Spiral Galaxies puzzle after this?

http://oi59.tinypic.com/27zlzll.jpg

Cells C12-R8,9,10 are in the same region, for example.

also c14 r1 and r2 have only one choice .. basically find some cells which can go in unique way. :)

Okay got it, thanks to both of you for those tips :)

Edited by mathcrazy 2014-03-14 8:41 PM

How do cells C14R1 and C14R2 have only one choice?

ok think in this way to which galaxy does c14r2 cell can join, you will find only one way.

Personally, in a forty-minute contest, I think that good notation is going to make a huge difference. Every time I try a puzzle like this, I go back and forth between drawing the border and connecting the cells directly, wasting a lot of time and creating a real mess of lines across the puzzle. This may be one puzzle type where I need to break into my child's supply of color pencils...