Logic Masters India announces May Puzzle Test — MAYnipulation

Author : Murat Can Tonta

Date : 14^th and 15^th May

Length : 101 minutes

IB and Submission Link : Here

Theme : All puzzles in this test have some form of diagram manipulation, e.g. rotation, superimposition, symmetry.

100 minutes or 101?

101 minutes

Is the Kropki example not uniquely solvable? It seems that you can swap 1s with 5s and 2s with 4s to get a different answer.

We discussed about this problem, but didn't fix the image. Will upload the fixed example.

Apologies, that example was prepared rather hastily. And I was supposed to fix it..

In Persistence Of Memory, does every shape need to have at least one cell as part of the snake or can some shapes remain completely blank?

Shapes need to contain at least one part of the snake. I thought it was implied by the rules, but it seems I was being vague.

Doesn't the Ambigram Skyscrapers example have two solutions? Won't 0123,3210,1302,2031 also suffice or am I missing a rule?

That won't satisfy the "looking the same when turned upside down" part.

IB with these 2 changes uploaded.

Do you have the exact UTC time window for this contest?
If it has not been set, could it the end time be at least 1am GMT/UTC (2am BST) please, to enable as many UK competitors as possible to compete? Thank-you.

This will start at 00:00:01 AM Saturday GMT (same as all LMI tests). But we are making a change to the end time based on several complaints / suggestions.
The test will be open for 48 hours, that means players can start upto 48 hours. Once a player starts, he will always have 101 minutes to submit (so in effect it will end at 01:41:01 AM GMT)

Hope it is fine. If you want a longer window, we can start earlier or end later or both. Let us know.

Further to Gareth's query, I believe that the Ambigram Skyscrapers puzzle has 3 solutions. The third solution is 0123,3210,1032,2301. The Persistence of Memory solution puzzled me and I think the rules need to be clarified. I assumed that the orientation of the shape did not matter and that, therefore, all 4 shapes were identical. Under this interpretation, the given solution does not satisfy the rule that each shape should have snake parts with the same appearance. There is at least one valid solution using my "false" interpretation.

Please clarify these points.

About Ambigram Skyscrapers

The rule says "The solution, if written in the above notation, must look the same when the page is turned upside down."

This is Gareth's alternate solution (with page turned upside down)


David's alternate solution (with page turned upside down)


Both of them don't satisfy the rule.

The given solution is

and it looks same when rotated.

Here, 2x1 block is different from 1x2 block.

Apart from 2X1 blocks being different from 1X2 blocks, the path of the snake also must be exactly same (rotation not allowed)

If some 1X1 blocks were given :

For example, this is a valid.		This is NOT valid

The reason I created the example with 1x2 blocks and 2x1 blocks was the hope that them being different would be noticed. Guess not. If two blocks have exactly the same shape (no rotation, no nothing), they have the same appearance regarding snake parts.

Ambigram: I do not expect anyone to actually solve the puzzle using that fancy font. :) However, the rule "The solution, if written in the above notation, must look the same when the page is turned upside down." helps.

hey in the Deformable Kropki i think there are 2 solutions.
the 3's and 5's are interchangeable at R4C1 R4C2 AND R5C1 R5C2.

R3C1 is 2 and R4C1 will be 3. That is not valid since there is no dot between R3C1 and R4C1.

ok
it means that reverse rule also applies in Deformable Kropki.




Edited by swaroop2011 2011-05-10 9:24 PM

That is so obvious from the rules.

In Coral rotator ,
does whil rotating the page upside down makes 2 and interchangeable.

And in Symmetric Loop
in the first regiion that is in column 1 and 2.
the loop is U shape but at R3 it is Z shape BUt the rule says it is symmetric how?

Thanks for clarification. I hope my mistakes will not be made by any competitors in the real thing.

Yes 2 becomes 3 on page turned upside down, and vice-versa

Just look at the portion of the loop inside the marked areas. It should be symmetric, i.e., if you rotate the marked area by 180 degrees, you still get the same loop portion inside the area.

Also, 4 becomes 7 and vice versa. Same with 6 and 9.
Remaining digits 0 1 5 8 don't change when turned upside down.

In Superimposed Loop, can both the grids contain a black cell in the same position?

Indeed. Just their superimposed form is given, the two grids are actually independent. They do not put any constraints on each other.

Small clarification: Sorry for making the diagram borders bold on the Symmetric Loop example, it led to some confusion with the actual regions. Contest puzzles do not have them bolded.

Puzzle Booklet uploaded. It has 11 pages, each page has puzzles (no cover page)

Starting this test, we are making one change
If somebody doesn't want to include this test's results in LMI Ratings, they must unselect the checkbox.



After clicking on "Start", the checkbox will be invisible.

To add to what the Admin said, if a player does not uncheck the box then irrespective of whether they submit the answers or not, their scores will be taken into consideration for the ranking and that can affect their standings too.

Rishi

Thanks for a very fun test that went by too fast (I guess 101 is a rotationally symmetric number of minutes, but I think we'll all comment that we would have loved more time to be MAY-nipulated). Still, I worked a little past the test to finish everything up and enjoyed it all.

I'm really glad if it was enjoyable. About the time limit, it was adjusted from the times of the test-solvers (many thanks to them). Competitively solving must be more stressful though. I also have to say that some of the harder ones are difficult to start, then goes smoother. This may cause loss of time as well.
Anyway, thanks for the nice feedback!
Regards,
Murat

Very nice puzzles. Among the ones I could solve, I liked deformable kropki (bottom) very much - Kropki can indeed be fun with hidden black dots, although it took me quite a while to get the right idea.

Edited by rakesh_rai 2011-05-14 6:26 PM

I don't know that 1001 minutes would have saved me today. Fortunately I have a rather large set of enjoyable looking puzzles to dive into tomorrow. Thanks Murat.

Similar sentiment from myself as well I'm afraid. Too many silly mistakes, including counting the bottom line on one of the corral instead of the one above it. STUPID me!

Thankyou for a great set of puzzles, sorry I couldn't do them justice.

@standupcanada and puzzlemad: I'm sorry things didn't turn out as you had hoped. I should be the one thanking. Thanks for participating!

Wow, that test really played to my strengths. Spiral galaxies, all kinds of path puzzles, grid jumping... just the Kropki and Skyscrapers that threw me a bit. Great fun!

Thanks for the test. Very nice puzzles.

I have solved the bottom puzzle of "First Seen Corral" correctly but entered the wrong solution code. If you look at my answer you will surely see what I did wrong and that my answer key fits to the correct solution of the puzzle. Actually I can give a good reason why my (wrong) way of entering the solution is actually the correct one. :-)

As I recall from earlier events, full points were awarded when entry mistakes happened and it was obvious that this wrong solution code fitted to the correct solution (e.g. entering the wrong row in a sudoku). I kindly request you to do so for my "First Seen Corral". If you cannot see my point or doubt that I have actually solved the puzzle, please write me a PM.

Have fun,
Stefan (euklid)

MellowMelon, I'm glad it did. Thank you!

euklid, I have sent you a PM, please check.

could you please double check my solution. I entered the solution to the first deformable kropki with about 1 minute to go but the system has registered my last submission with 15 minutes to spare. Something must have gone wrong.
my server page is still open and says "An error occurred while submitting. Please try again" and the server time is 8:00:49 (local), 7:00:49 GMT.

Edited by forcolin 2011-05-15 2:30 PM

Hi forcolin,

I checked your standings. Your last submission is 99:59 and you have indeed received 35 points for the first deformable kropki.

Murat

Excellent 'difficult' puzzles. Many times, there are difficult puzzles where the toughness tends to diminish the fun of solving it. But this test had the 'other kind' of difficult puzzles in which the toughness added fun of solving. Lot of brainwork used for this test, and loved the puzzles (most of which were solved after the test ;-) ) Thanks and hope to see more from you in the future.

Thanks for the compliment Rohan. :) I'm glad I was able to make the harder ones enjoyable.

thank you Carton Mutant. Noticed the change, I thought it was an effect of your amendment. perhaps I checked the results only seconds after the end of the contest and the result shown did not keep into account last minutes submission.... Deb is this possible? of course then I could check again only after 30 minutes.... Thank you for checking anyway.
very good set of puzzles, very challenging and of excellent quality. thank you, and congratulation for your success

Well, that was not my doing but I'm glad you got the points you earned.
Thanks for the kind words! I'm happy you enjoyed the set.

Enjoyed the test, ACM - thanks for your time and effort! Best regards, Dave

Edited by davep 2011-05-16 4:55 AM

Welcome back Dave! Haven't see you at LMI for a while!

Enjoyed the difficulty level of this test! But... I cannot understand why my answer to Spiral Galaxy I is wrong... as far as I can tell, the answer I have is correct (although I am tired! it's late here!)

Perhaps I misunderstood the answer key?

Edit: hehe sorry, I typed the answer key with one too many repeats... will you consider a correction?
Thanks :)
Ronald

Edited by ronald 2011-05-16 5:50 AM

Yes, too many repeats of the pattern. You get points for that.
Good score. Congratulations.

As posted earlier, players will be able to start till 5:30AM, and then have full 101 minutes. So, obviously, the results will be published little late.

Thanks Deb, that's kind of you to say! :)

MAYnipulation is over.

Congratulations to Palmer - after coming extremely close several times earlier, he has won first time at LMI. He had a typo in a 40 pointer, but even with that the score is impressive.

Ulrich is 2nd with 710 points. Thomas is 3rd with 683 points, and I must mention that Shinichi Aoki also has 683 points though the submissions are 4 minutes late.

Not many players from India - the usual suspects take the top spots - Rohan, Amit, Rakesh in order.

Congratulations everyone and thank you for participating.

It is probably too early morning for Murat to post a comment at this time, but I think he will post soon about puzzles, and overall experience with his first(?) puzzle contest.

Great contest! Lots of beautiful and/or difficult puzzles!
I especially liked Deformable Kropki 2, Spiral Galaxies 1 and Disjointed Tapa.
I missed the "equal number" rule in 4x4 Diamonds and wasted about 10 minutes.
Moreover, I broke Clone Battle in the last few steps and could not fix it.
I should have been more careful.

Congratulations to Mellow Melon on his first victory in LMI contest!
It is becoming harder to leave my name on the hall of fame.

Thanks Deb - not very many free weekends lately, hope to start participating more frequently......

Edited by davep 2011-05-16 11:44 AM

First of all, I thank all participants, test-solvers and LMI team once again. All three together made this competition possible.
Congratulations to Palmer Mebane for his first victory, Ulrich Voigt, Thomas Snyder and Shinichi Aoki. Also, congratulations to everyone!
I'd like to share some "background" on the puzzles and my experience with the competition.

Negative Masyu: The first several drafts of the contest were almost completely loop-based puzzles, so a desire to put in Masyu was always there. I thought a different type of "rotation" would be nice. Was a little annoying to create, I ran into some problems when whites became black and broke the puzzle. I think they came out nice and easy, though.

Deformable Kropki: Being "deformable" was a feature I initially thought for Masyu. But after I created the Negatives, I thought "Kropki has blacks and whites, why not?". I started with the aim of making one of the puzzles all white, happily, it worked out.

First Seen Corral Rotator: Another duo which were a little challenging to create, I had to allow closed areas to make them work. Anyway, I'm happy with them too.

Symmetric Loop: This one is kind of a ripoff from Persistence of Memory. Another attempt to create a loop-based puzzle with just some regions as clues.

Ambigram Skyscrapers: I'm glad there wasn't much confusion with this big guy's rules. I wanted a perfect ambigram (ignore the arrows :) ) so placing outside clues according to that were a lot of work. Use 9-6 and the upper left and lower right regions to start off.

Spiral Galaxies: A classic genre, not much to say about that. In the first one, I went for a large area. I like the fact that in both the logic path to follow is never too obvious but I avoided anything that would lead to trial-error, they should be perfectly solvable logically.

The Persistence of Memory: A lovely genre in my opinion. Went simplistic with the first one. The second is easy-ish too, just a bit more work to realize how the snake must travel the regions.

Disjointed Tapa: I think placing the pieces were not that much of a problem, my aim was to make the actual solving enjoyable. I hope it worked out. Lots of choke-point action. I'm happy I managed it with 10 clues in a 10x10 grid.

4x4 Diamonds: Thought of this after creating the Superimposed Loop. Came across this variant of diamonds/minesweeper a few times and always enjoyed them. The extra rule "equal amount" is there to enhance the solving process. More tricks are available when one knows exactly how many diamonds there are in a grid. One of my personal favourites.

Superimposed Loop: A little hard on the eyes, I'm afraid. First I thought making the originals real loop puzzles (i.e. When you place all the blackened cells, then you can arrive at an unique solution) but decided against it as the given diagram would contain too many black cells. It is already enough crowded-looking. The black cell on the upper right is a good place to start dissecting.

Clone Battle: I should have made this one easier after all. The diagram/s you have to use keeps changing as you progress as do the technique. More suitable for solving when you are not stressed with a time limit, I guess. Still, rewarding to tackle I'd say. Start with UR+UL, you can eliminate C7&E7, then switch to LR+LL, you can put in three stars with certainty.


About the time limit, I guess it was a little short, as any of the high-valued puzzles may lead to being annoyingly stuck. The countdown ticking away certainly does not help with that either.
This was indeed my first time having authored a whole set. I wasn't sure how it was going to be received but now I'm encouraged, thanks for all the comments! LMI team was very helpful (patient too), the communication was smooth, I couldn't ask for more. And when it was finally competition time, it was exciting the follow the scores page. My overall experience with my first contest is a warm and fuzzy feeling. :)

One final thing, (I can't promise anything though, exams draw close!) I can create a solution steps booklet if there is demand for it. What do you think?

Once again, many thanks to everyone who made this contest a reality.
Cheers,
Murat

Edited by A Carton Mutant 2011-05-16 3:51 PM

Murat, Thanks a lot for the writeup, and more importantly providing a great set up puzzles for everyone.

It would definitely help if you can document solution steps for some least solved puzzles, especially Clone Battle

You are welcome. :) Solution steps booklet it is. Coming soon-ish! (I hope)

Murat, thanks for interesting set of puzzles. I still have some left overs to finish though :-)
I hope to see some more puzzles from you in future!

Just a question to results. I have realized that I have misunderstood the answer key for spiral galaxies. Instead of 2,1,2,1,2,1 for 5th row of the first grid I have sent 6,1,1,1 based on the requirement to enter lengths of 'different' shapes. So I have added the three parts that belong to one shape. Can you consider that as valid answer? Not that it matters much, I had fun anyway.
Thanks
Jirka

Hi jhrdina, you're welcome! I hope to return with another set someday. :)

Oh, I was afraid of that interpretation actually. It should have been "enter lengths separated by lines" or something similar for clarity. Your answer is technically correct. I'll try to get you points for that (it's just that I myself can't give points :) )

Thanks for the congratulations, and thanks again for a great test. Re deu's comment about getting on the podium: although I haven't been playing here too long, I think that's how a lot of us feel dealing with you, motris and uvo. On Twist I had felt like I had a great run upon finishing, but it wasn't enough to beat any of you. :(

For the Clone Battle, the low solving statistics may be the fact that it was at the end of the test, combined with the intimidation of four grids to consider. None of the steps were particularly tricky, although noticing them could take awhile. My memory is not precise, but I think there was one nonobvious step in the beginning involving finding two disjoint rooms in the two rightmost columns (the catch being that the rooms were in different grids) to get some X's, and everything afterwards seemed to be standard Star Battle with grid hopping. That said, it did eat up a lot of time.

Superimposed Loop is the puzzle that actually got me. I only solved it with a lucky guess - perhaps it's karma that I went on to miscount one of the grids. Ambigram Skyscrapers too, although I did get that one logically... just had to spend 20-30 minutes on it.

You should see the credits now.

Request to everyone - Please try to settle the point claims as soon as you complete the test.
Lot of manual work (e.g. LMI Ratings computation, UKPA ratings computation) is done immediately after the test is over.

Thanks guys for the points. Still at the bottom, but feeling much better now :-)) J.