Do check out some of puzzle solving techniques from one of best puzzle creator Cihan Altay

http://www.wpcstylepuzzles.com/2010/09/puzzle-solving-techniques-ci...

http://www.otuzoyun.com/puzzles/SolvingTechniques.pdf

This is the direct download link to the puzzle solving techniques by Cihan Altay

Thanks for the link, rishi.

After a flop show in the Puzzle Grand prix round, I thought of attempting the left out puzzles which unfortunately are many in number.So in this Slitherlink puzzle I raced till this point and there was no trouble .But beyond this, I am clueless on how to continue through the maze.The 2 cluster looks menacing and so does the '1 area' which looks equally threatening .I can't see any way other than an endless series of guesses and contradictions to reach the end in this puzzle.What is the logical continuation?



Edited by kishy72 2014-02-07 5:05 PM

Think of the following technique.
If two one's are coming at diagonal to each other and if for one of them two the edges around connecting point is crossed then same will be reflected (mirror image) on the other 1.
Same is true if both the edges which are not originating from connecting points gets crossed.
There are lots of such points in above puzzle which will lead it to proceed in upper right part of this puzzle.

If you can able to understand and proceed using this, I will tell you one more point which will help you to proceed.

THINK IT THIS WAY..For the '1' in R3C7 the loop will either connect at the top or right of it... so continuing for the '1' at its diagonal R2C8. it will have either loop connected either at the left or down of it... if up of R3C7 is a line thn down of R2C8 wud be a line and if right of R3C7 is a line then left of R2C8 is a line. so when two such 1's are at diagonals and if either one of its opposites sides are crossed out then for the other 1 it is mirrored at its connecting point with 1... after whch, 1 at R1C9 will have a cross at it down by basic rule and even its right will be crossed out as both the ends of the loop willl go down n due to 1 at R4C10 the loop wil close..

Thanks for the hints fellas but unfortunately I could not make much progress with those .I haven't even managed to touch the 2 area.So a stronger trick is needed ,so that I don't need to post this puzzle again! .And thanks to this very instructive link http://logicmastersindia.com/forum/forums/thread-view.asp?tid=127&p...
which helped me start and complete the magnets puzzle just like that .

There area couple of things you can do with the 2's.
The one in the very bottom right must either be both "outside" segments or the other two,
and either way the outside edge segments NEXT to the two must be shaded in.
Now we have one segment leading into the 2 in row 8, column 10 (coming up from below).
One of the two bottom-right segments in that 2 must be included, and the other not. But
that means exactly one of the top left is included. Follow this diagonally and you can
shade in the left segment of the 2 in row 6 col 8.

There's a very nice example of this sort of thing with the 2 in r7c6, which forces
the segment coming out of the top left corner of that 2 which continues to the
left ricocheting off the O's.

Also, for example, the 2 in r6 c7 must either be the two top left or the two bottom right
segments from the x's you've put on that 1. But that means one of the two bottom
left segments of the 1 in r5c10 is included, and the right segment is not (and therefore
neither is the right segment in the 1 above it).

Those are the general sort of thing that can help you get through the thicket of 2's.

Thanks a lot! I finally completed the Slitherlink puzzle but not before having to guess twice .The above mentioned techniques played a big part in attaining the solution.I am beginning to realize that Slitherlink puzzles are very difficult,totally contrary to the earlier impression I had in mind about this puzzle genre......
I moved on to another unattempted puzzle namely the Japanese sums.The one posted below seems very different from the conventional Japanese sums puzzle in that all numbers from 1-n need not necessarily get placed in each row and column as long as the 'no number repeating in a row and column rule' is not violated.This gives the puzzle an added level of difficulty since the choices to consider for each sum are so many.So could someone please help me continue from this point?





Edited by kishy72 2014-02-12 11:19 PM

Look at R7. Sum of numbers is 27. So 1 cannot be there. And the three sets must be two numbers each (27, 36, 45), which means R7C8 must be a number. This, in turn, means that R7C8, R8C8 and R9C8 must be 124. So R2C8 should be 7. and R1C8 and R1C9 should be black. You can make progress in R1 and continue.

In the 7th row, each of the three 9s have to be sum of two numbers. (2+7, 3+6, 4+5). We can deduce that the box at 7th row, 8th column contains a number. Which means that in 8th column, the second 7 = 1 + 2 + 4.
Hence, the first 7 = 7, and 3 = 3.

Other than what Rakesh and witty have said, I'll just add that your notation seems a bit counter-productive if you have to erase one of the Ns each time you fix a number. Obviously it's good to keep what suits you best. I personally circle the cells which I know will contain a number. Then I just fill in said number into the circle.

Deducing that R7C8 is a number is what I failed to notice.There was no further gimmicks in the puzzle after that and I quickly took it to completion.Thanks.....



A very valid point that you have brought up.For one I would never use this notation in a competition scenario with time constraints.My goal here was obviously to complete the puzzle and not worry about the time that it might take to complete.I don't think that I am competent yet to complete many of the puzzle genres by myself without help or spoilers.So notation is something which looked secondary at that point to me.But I understand that notations are integral to a puzzle without which one can forget about improving and I will certainly seek to improve in this puzzle type by circling the cells which have numbers !
So continuing with the posting spree ,here is another left out puzzle from the Grand prix which despite my dedicated effort seemed far too out of reach and I am not surprised since I rarely solve a pentomino puzzle .All that I could deduce in this puzzle was that the blank cell in C4 is located somewhere between R5 and R8 and similarly the blank cell in R8 is between C1 and C4.What is the precise logical sequence that needs to be used here that makes this puzzle fall apart?!



Edited by kishy72 2014-02-14 12:07 AM

1) You can go further with column 4. There's only one place in the column that can be empty to give 4 different pentominoes. So you can fill in all cells of column 4 at the start.
2) Your row 8 is not correct. Start with looking at that row only, you can only learn that C2 and C3 are filled. All other cells might be empty (remember there is a U-shaped pentomino for C8 and C12).
3) And your line down the bottom is too long. The central cell is filled, but you don't know yet about the cells beside it.

But the biggest hint I have for this puzzle (and other puzzle types like Camping or Battleships when missing a single clue) is that you know that the number of shaded cells must be 12x5=60, so add up the clues along the left. Whatever's left will be the clue for R7. Similarly, subtract the top clues from 60 to get the C7 clue. Easier now?

Parallel Universe II - Looks like a very tough puzzle test .Thought it would be unwise to give the test without solving the practice puzzles.So coming to the point ,can someone kindly show me the continuation in this Akichiwake Puzzle and are there any general strategies for this puzzle type?(like looking for specific clues or spots to start and proceed smoothly etc.,).It was not before spending close to 45-50 mins that I reached this stuck point....I would be glad if ideas are provided for Fill-other mino and Hypermetropic Pentominoes too.I keep breaking Fill-other mino puzzle repeatedly and the solution doesn't seem to be of much help to me in understanding how to go about the puzzle.
----Kishore----



Edited by Fred76 2014-04-25 8:35 PM

Oops, sorry !
I edited your post, thinking that it was the tournament's puzzle, and after I noticed the puzzle was from the practice material.

You seem to have missed the "consecutive boundaries" rule in many places. Maybe you missed this entirely while solving? Some examples - R2C2 (coz of the two boundaries to the right), R1C4, R1C6, R6C10, R7C3.

I hope you're considering the "base rules" on each page for all of "their universe" variants too. If so, and you're still getting stuck on the practice, feel free to post images. Much easier that way.

General strategy, hmm... Akichiwake is a new type. But its based on the two-consecutive boundaries rule/connected white area constraint of the Heyawake type. As of now you haven't encountered either of these in your solve yet, seemingly. Once you do, you'll see how these restrict together. E.g. - Once you shade in R6C10, you need to avoid that entire diagonally connected group of shaded cells from touching the grid edge anywhere else.

For Fill-Other-Mino,
The clues with 4 separate digits, or the edge clues with 3 separate digits are obviously going to be a polyomino of size "1". You also get 3/4 different polyominos adjacent to these clues. The rest of the steps are more complex, but just pointing out a basic start in case you've missed a rule here.

Yes I missed the consecutive boundaries rule completely.Thanks very much for the ideas.I will try using them in the test.

Parallel Universe II ---That was a lovely set of puzzles.This being an open forum I will cut myself short right there.
Moving on,here are 2 puzzle genres that are present in the ongoing GP Puzzle round 4.The Prime place puzzle was something where initially I wasn't too flustered with the rules,but after seeing the queries asked in the forum of World puzzle org regarding this(and there seems to be so many) ,I am sure there won't be any who is more confused with the rules now than myself.The first is a prime place puzzle and the second is a Japanese loop puzzle where I am unable to make a start (both taken from the blog maybe puzzles).So dear LMIans, if someone has a very good idea or strategy to tackle these puzzles kindly post here.

I tried that Prime Place and broke it, so maybe I'm misunderstanding the rules. I just thought the converse part of it applies too.. I guess it does. Maybe I went wrong somewhere. But remember that by the rules, since all diagonal lines are given, all isolated cells must contain a diagonal line in the direction of their isolation (e.g. the one non-parallel line is in what I'd call an isolated cell in the direction of that line) in order to contain prime digits.

Prime Place is inherently tough. It has quite a lot of rules (I see you forgot about "connected horizontal/vertical lines must sum to a prime", which immediately deduces R1C3, R3C1, R3C5, R5C3 as primes), and a lot of bruteforcing (for example, a nonprime and a nonprime that sums to a nonprime must be both 4s like R4C1 and R5C2; also when you've identified that R1C3, R2C2, R3C1 are prime, nonprime, prime in that order, and they sum up to a prime, by bruteforce you can find that R2C2 is 1 and R1C3 and R3C1 are equal). I find a rather nice path without too much bruteforcing, which resorts to counting the 3s. Those should be sufficient hints for now?

for japanese loop i can see the start at 7 7 7.
as each needs to be two cells. there is only one way to do
then it should be good to go. :)

Invariably the following sequence happens when I see a Nurikabe puzzle in a test and it is becoming more of a nuisance to me :
1)The puzzle is attempted.
2)A lot of time is spent(wasted) on the puzzle .
3)The puzzle is left incomplete .............

This is one of the toughest puzzle genres that I have seen .The puzzle below looks very cruel and I am hopelessly lost in it.I tried to see if only certain numbers can reach some spots to avoid 2X2 oceans.There is nothing of that sort here.The shaded cell connectivity rule is also not of much help.A shaded cell maybe at R3C5 using the above rule?!Not too sure on that.So what is the approach to be used to complete this puzzle and in general what line of thinking is to be adopted while solving a Nurikabe puzzle where in most cases no logical deduction looks immediately apparent?



Edited by kishy72 2014-05-18 10:53 PM

As u Kishy72 R3C5 wil be a shaded cell... if not how wil the shaded cell at R3C6 wil join the other shaded cells.. no way for it to join other... Other thing is R1C1 and R2C1 wil be shaded as not island can go to it... Aftr which R3C2 will be a part of the island 3 or else it wil form a 2*2 shaded square. and after R3C1 wil be shaded or else the shaded part above it wil be blocked.... U can try aftr tht... if at all u r stuck u can still post.. :P