The Indian Puzzle Championship is now officially over even for the International participants.

The Solutions booklet can be downloaded here . Congrats to uvo, EKBM and deu, the International Top 3. Ulrich was the only one to finish all 35 puzzles correctly, with 12 minutes to go.

I hope everyone enjoyed the puzzles. Congrats once again to the Official winners Amit, Rohan and Swaroop. I'll give you all a day to guess the authors and post my notes about puzzle selection, etc. tomorrow. In the meantime, please feel free to ask for hints on individual puzzles and also to give more feedback on the competition in general.

Thank you all for participating!

My guess for authors:
Thomas/Grant - C,D,F
Prasanna - I,G,E
Deb -A
Bram - B,L,K
James - J,M,H

How balanced do you think the puzzle types of this test were?	Fairly balanced
What was your opinion of the distribution of easy/hard puzzles?	Just right
What did you think about the puzzle quality of the test?	Fairly Nice
What was your opinion of the booklet formatting and printing?	Just right

I really liked the test, but I have one thing to consider.

"Match the following" is a guessing puzzle. My personal view is, that there should be no guessing puzzles in a logic puzzle test. Especially when one thinks about that A2 was too guessy even for the puzzle makers, so they gave the rules for it. Which means, "a little guessy" is ok, or what? Why do we give all the rules to all the other puzzles then anyway?

First, thanks for the thoughts. This is an interesting take on "Match the Following" because the purpose to include that is to interest participants who are new to logic puzzles and might want some familiar ground. I can see where you're coming from as these in general won't follow a "solve path" as some of the common logic puzzles, but they do not require guessing either. It can sometimes be an intuitive solve, as are many other logic puzzles.

A1 was simple equations, A2 is basically another iteration of Simple Math. The rules were given because the rules were given in the Instruction booklet too. When equations are there, "precedence" and "repeating" are common clarifications, and are such in Simple Math too which is a common logic puzzle. Out of curiosity, would you call TomTom a guessing puzzle? Because A2 is an easier version of that with digits given and without the Latin Square rule.

A4 and A5 were instructionless puzzles which are a common enough occurrence in Logic Puzzle competitions. The logic there is in thinking what the association could be. Its intuitive, but I wouldn't call it guesswork, because there's enough there to make something out of. Counting puzzles have been present in WPCs before, and Elastic Bands definitely didn't require guesswork, its a simple case of spotting the vertices.

Did the Instructions Booklet (IB) provide enough information about participating in IPC?	Yes, mostly.
Did you understand the puzzle rules, examples and answer keys from the IB?	I understood all of them.
Did you find enough easy puzzles to solve during the championship?	Yes, there were many of them.
After your experience in IPC, are you going to participate in future championshiops or contests at LMI?	Most likely

A,B,I - Prasanna
C,F,H - Thomas and Grant
G - Deb
E,L,J - James
D,K,M - Bram

Regards,
Ravi

I want to share the way I did A2. Let's see whether anyone else uses the same method...

I simply make a table for all possibilities. Like, A can only go to Y or Z (if I recall correctly), etc. Now, for each equation:

Note that operations may not be repeated in an equation. So I began by trying where the division goes, if any. This usually makes some other multiplication sign forced, and there are only two more possibilities.

After that, I tried putting the multiplication sign. As there's no more division, the multiplication occasionally blows up the result, so there aren't many ways for this to go either.

After tabulating, just do a Latin Square thingy to pick one from each equation and one from each result.

I took about 2-3 minutes here, so giving 20 points here is interesting...

---

Regarding Match the Following, the three puzzles with instructions aren't guessing at all; they are simply usual puzzles taking in a different form of matching. A4 and A5 are more inductive, as in there's no set rules and in some way you can interpret things differently, but they only worth 10 points and IPC in some way is different from usual monthly puzzle tests due to this: having inductive puzzles.

How balanced do you think the puzzle types of this test were?	Fairly balanced
What was your opinion of the distribution of easy/hard puzzles?	A bit easy
What did you think about the puzzle quality of the test?	Very nice
What was your opinion of the booklet formatting and printing?	Just right

Swaroop - 7/13.
Ravi - 5/13.

Notes about Puzzle Selection -

In general, I wanted to cover all the broad common categories for the IPC, like Number Placement, shading, loops, object placement and region division. However, along the way, I had in mind the fact that I should attempt to cover the lesser known categories as well with the variants, and remaining sections.

A - Since the Match The Following has already been talked about, I'll get to the remaining ones. A quick note once again that the A section is mainly to provide a familiar easy start for participants who are newcomers to logic puzzle competitions. I thank Deb for handling this Section, and I think the idea to keep two questions instruction-less was a great idea, to cover this kind of "figure out what to do" thinking, that is definitely necessary in the WPC.

B & C - This is the Number Placement section. I wanted Easy As ABC or Skyscrapers, to cover the conceptual part of things, which James took up. In general, I wanted the rules to have simple changes for most of the variants, so his Spiral variant suggestion was perfect both in this sense and in the fact that the spiral grid variation is a different kind of thinking that got covered here. I wanted a math number placement puzzle too. I could think of Kakuro but had totally forgotten TomTom actually, until Thomas suggested it, and I thought it was perfect because it had some Sudoku logic which is familiar to most of the target audience, and at the same time covered the Math portion of things. Coded TomTom as a variant meant that Cipher/substitution logic was covered well too. The bigger TomTom here was one I really liked personally, and both me and chaotic iak noted that this was a bit different from Grant's usual constructions.

D & E - This is the Object placement section. Bram contributed the Battleships, and had actually sent in a harder Observer variant at the start, which is an excellent puzzle that should appear online elsewhere in the future. The bigger classic Battleships puzzle was hard too but I let that stay because on balance I liked that some classics be hard. I wrote the Star Battles, and since we had a 9x9 with 6 regions last year (which I totally predicted in the practice puzzle set I had released ), I wanted to do something similar, but with different calculations. 9x9 with 3 regions worked well, but then I was wondering what I could fit in with 3 regions, seeing as that would only accommodate 2 letters. At this point I already had the 2014 themed normal Star Battle ready, so I just went with a "UK" theme since that's where the WPC is in 2014. It surprised a few testers

F & G - This is the Shading puzzles section. Thomas and Grant contributed to both F & G. Since shading and loops are common types, I wanted each section to have one common puzzle and one uncommon puzzle with easily understandable rules. Cross the Streams was perfect to fit the uncommon puzzle type with Nurikabe being the common one. Since CTS was uncommon, I thought two classics would be enough here. With Nurikabe, by the time Thomas had agreed to write the variant, I knew what the other authors were doing already, so I wanted word puzzle logic to be covered somewhere. A Word Nurikabe seemed the best bet, and Thomas sent in a very impressive puzzle using Indian states, as can be seen.

H & I - This is the Region Division section. The Division puzzles are mine, and I just thought that since "Pentomino thinking" is slightly different from "Tetromino thinking", one of each should do well here. The Arafs, and most notably, the highest pointer, the Different Neighbors variant, were contributed by Bram. I wasn't entirely sure whether to include the variant, even though it was my favorite puzzle of the set, because it starts with global logic that might be difficult to see. I requested another variant to choose from, and he was nice enough to send it in, but I still wanted to use the first one, so I decided that if 2 testers solve it in a reasonable time (after I already had), it should be fine to include. This is what happened, and I'm happy some solvers attempted and solved it, and were able to enjoy it within competition. For those of you willing to try it and looking for a hint - count the 4s.

J & K - This is the Loop section. As I mentioned with shading puzzles, I wanted one common loop type and one uncommon one. Slitherlink was a favorite for the common one because its so easy to understand and also because its there on a few newspapers here in India. Bram suggested the Toroidal variant, and since this covered yet another form of thinking, I loved it, even though I knew the variant in general would be hard. Indeed, here too I requested an easier version after the original Bram sent in (really, there's at least 4 high quality puzzles out there now including an extra Classic Slitherlink that will all appear elsewhere). For the uncommon loop, James suggested Maximal Lengths which had appeared in the 4th Puzzle GP and I liked it. Though I'm not particularly strong at these, I thought its a good fit for the "easy rules" requirement. He then suggested the Minimal Lengths variant which I personally liked for the competition a bit more than Maximal, because it flowed in an easier way (seemingly). There was a hard Maximal Lengths that was discarded in favor of having two Minimal Lengths. The hard one is a really nice puzzle, but had a very narrow path, and James will probably showcase it elsewhere.

L - The Snakes were written by me, and it was only because I thought that the Snake type is a fairly common one which wasn't covered on the set yet.

M - I wanted a special puzzle, but wasn't exactly sure what. James first suggested a connected puzzle with two separate grids, which I was ok with, but eventually he liked Tiki Tours better, and once he submitted it I liked it better too because it covered a double variant and it also had Object Placement and Looping which I consider a bit more common than other categories, except Shading (which kinda sorta got covered by Snake too).

Overall, I was really happy with the set and sincerely thank all the authors for their contributions and the testers for their feedback. I hope everyone had something to attempt in the competition and had fun with it.

Just to summarize -
A - Deb.
B, J, M - James.
C, F, G - Grant/Thomas.
D, I, K - Bram.
E, H, L - me.

Awesome :)
I had one suggestion that for Answer Key of Battleshipts or Star Battle can't numbers be preferred then Alphabets. It's easy to enter numbers than to note down alphabets. would like to here others take on this?

But there are 26 alphabets while only 10 digits. I can see why Prasanna prefers letters; so he doesn't need to deal with "if the column number has two digits, enter the last digit only" and stuffs.

Also Battleships already have digits around the grid. Putting another set of digits might be confusing.

For the Araf variation, it helps to have seen this puzzle from a familiar blog before:

http://www.gmpuzzles.com/blog/2014/05/araf-serkan-yurekli/



Edited by uvo 2014-05-28 10:59 PM

ok got it :)

It was kind of funny actually. I had considered this particular variant before this puzzle had been posted. So after that puzzle was posted any thought of being original quickly disappeared.