LMI announces the LMI February Sudoku Test - Get Overlapped !

Author: Huang Yi

Contest page: http://logicmastersindia.com/2019/02S/

Instruction Booklet: http://logicmastersindia.com/lmitests/dl.asp?attachmentid=760

In No Touch, does this also apply to cells not in the same grid (e.g. R9C3-R10C4 in the example)? Just checking as this can sometimes vary in overlapping puzzles.

Thank you for the question.
In this puzzle, the rule of No Touch does apply for all cells, including the cells R9C3-R10C4 and R3C9-R4C10.

I think there are two bugs
1) Sudokus submitted after the 100th minute are considered in the scoring
2) I didn't receive the 5 extra points for completing the Samurai sudoku

How balanced do you think the puzzle types of this test were?	A bit skewed
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
Which set of puzzles did you like the most?	Killer Sudoku

I just finished the test and I got a problem on the aggregation of my score.
I sent my solutions in the order
1-8-9-11-7-13(offline)-3-2-4
and the system said all the answers are correct by marking "(S)" and revealing the five 9's (for #13). Only first five problems are counted to my final score. Hope there is a system log that can restore my actual score.

The scoreboard will be fixed. All answers have been received.

Yes. Answers sent after 100 minutes are currently not showing in the scoreboard. This will be fixed.

The bonus points for Samurai and Treble will be added later, as mentioned earlier.

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?	Too many hard puzzles
What did you think about the puzzle quality of the test?	Very nice
Which set of puzzles did you like the most?	Killer Sudoku

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?	Too many hard puzzles
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
Which set of puzzles did you like the most?	Diagonal Sudoku

A bit too much guessing there was. It started well, but at some point when you try to solve it logically and spent a bunch of time in there, you must decide whether to keep going or make a guess. And then there is another and another. Maybe a bit easier competition would have made it more enjoyable competition.
But maybe it's just me, who is not capable of solving samurais well.

Generally the puzzles were fairly nice. The theme is very interesting though.

Puzzles like these are why I stopped competing in my country's sudoku tournaments. Computer-generated, repetitive and inappropriate difficulty level.
I was quite put off by the classics (singles and samurai) needing constant guessing - and was hoping, in vain, the variants would maybe redeem this test. It didn't.
Was this test properly test-solved to get a rough estimation of completion time? Or was the aim to throw a batch of hard puzzles at the solver and have them strategize?

I apologize if these puzzles were handmade; but if an author would spend hours making puzzles that can't be distinguished from computer generation, then I'm afraid you need a new approach at construction.
There were no (at least from my perspective); aesthetics, innovations, themes, neat solving paths and, crucially, enjoyment from slowly chipping away at these grids.

I'm sorry if this review is getting too negative, but with a decade of LMI Mock tests, Sudoku Maharabat, the GP and other interesting Sudoku tests - you'd think you've seen the end of computer-generated sudokus.
It was an unpleasant surprise to see this test pop up in this day and age.

How balanced do you think the puzzle types of this test were?	Perfectly balanced
What was your opinion of the distribution of easy/hard puzzles?	Too many hard puzzles
What did you think about the puzzle quality of the test?	Average
What was your opinion of the booklet formatting and printing?	Just right
Which set of puzzles did you like the most?	Killer Sudoku

Puzzles seemed brutally difficult, or maybe I was having a bad day. I jumped right into the overlapping grids (which I normally enjoy and seek out) and brokke three different puaales in about an hour. I then went to the classics but found them uninspiring. After solving one, I simply quit the contest 80 minutes in. Just not heving enough fun.

Hmm, I admittedly solved the Samurai after the contest away from timer pressure but I did solve it in 19 minutes without requiring any guesses.

Revisited this to logically solve it.
Took me around 35 minutes.
Still didn't enjoy it though.

How balanced do you think the puzzle types of this test were?	Perfectly balanced
What did you think about the puzzle quality of the test?	Very nice

I got through the Samurai quite well during the competition and enjoyed it. I had to guess a lot on the Treble and/or the No Touch though (but there may well have been logic I missed).

Anyway, I don't regret taking part, I had my fun. Thank you to authors and organizers.

Repost @tamz29

This is the worst time I've ever participated, the reason is " Computer-generated, repetitive and inappropriate difficulty level. "

Also for the first time, I didn't do IB, because it was too difficult to waste time.

Hope that the organizers will review sudoku in the future, thanks to organizers.

Edited by Lei 2019-02-17 6:52 PM

Hi, I mixed the style of sending answers, but the results count only my solutions from online solving. It was due to technical problems that I had with tredoku and samurai sudoku. Online it didn't appear so I had to play it with the best graphic program - painting :) from pdf. Could you look at it please? I hope for additional 45 points for samurai sudoku and some penalty for one wrong part of treble. Although (now) my position will be same, I will be more happy :)

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?	Too many hard puzzles
What did you think about the puzzle quality of the test?	Average

Somehow this was not as much fun as I had hoped. Most of the grids I tried were solvable logically, but it took a lot of brute-forcing to find one's way through, rather than having a solution path jump at you.

Hmm I'm confused, I'm pretty sure I submitted puzzle 11 (Untouch). I only had a bit of time left after my last submission and got half through the Killer, that can't have been after 60 minutes as indicated right now.

How balanced do you think the puzzle types of this test were?	Perfectly balanced
What was your opinion of the distribution of easy/hard puzzles?	Too many hard puzzles
What did you think about the puzzle quality of the test?	Fairly Nice
What was your opinion of the booklet formatting and printing?	Too many pages / too large grids
Which set of puzzles did you like the most?	1-4 Classic Sudoku

This test's grids were indeed computer generated, so the difficulty may vary; however, the types of puzzles were chosen by myself.
Anyway, it's my LMI debut as an author of a monthly test, thank you for whatever you have done for this test.

Thanks for sharing this info.I invested a considerable amount of time solving IB sudokus and I got stuck in many.Even after sharing this in a group, there were still many left where none could find a continuation.So, I was left wondering if there were givens missing or if the IB sudokus were test solved.

Now, about the contest, No touch and Irregular really killed my performance as the combined time I invested in these two is approximately 45 minutes and that involves an unsuccessful attempt at irregular and multiple guessing in no touch (after a miserable attempt to find logic).

My feedback about the contest is that if majority of sudokus are computer generated, then a caveat maybe given in the IB which states something like "Sudokus may not be logically solveable " . This has even a precedence in a WSC (2010, Philadelphia) where there was a hard classic round and apparently solvers were advised before the start of the round that sudokus might not be completely logically solveable and can require bifurcation.Personally, I am not fully against computer generated sudokus.However, just like others, I too prefer human made much much more.So with a warning like I stated, it gives solvers a huge amount of flexibility. Instead of sticking to a rigid approach of finding logic (like what I adamantly did during the contest) thinking that there must be one when there very well might not be (atleast by human standards), they can quickly take a guess too.

My favourites from the contest are Killer followed by NC.Thanks Huang Yi for your contribution! Hope to see more chinese hand crafted contests, as I am a big fan of chinese authored sudokus be it CSOCs or the ones posted in FB daily league.

How balanced do you think the puzzle types of this test were?	Perfectly balanced
What was your opinion of the distribution of easy/hard puzzles?	Too many hard puzzles
What did you think about the puzzle quality of the test?	Average
Which set of puzzles did you like the most?	Killer Sudoku

Thanks for the contest. I enjoyed the some what large sized formatting of the cells. The double and killer were my favorites. I am curious about the logic used to solve the treble.

I'm actually relieved to hear that the Instruction Booklet problems were computer-generated. I was having a very difficult time in solving them, and not enjoying it. I liked the competition puzzles much better (so far!).

How balanced do you think the puzzle types of this test were?	A bit skewed
What was your opinion of the distribution of easy/hard puzzles?	Too many hard puzzles
What did you think about the puzzle quality of the test?	Very nice
What was your opinion of the booklet formatting and printing?	Too few pages / too small grids
Which set of puzzles did you like the most?	Diagonal Sudoku

How balanced do you think the puzzle types of this test were?	Perfectly balanced
What was your opinion of the distribution of easy/hard puzzles?	Too many hard puzzles
What did you think about the puzzle quality of the test?	Average
Which set of puzzles did you like the most?	Killer Sudoku

Congratulations to Tiit Vunk, Tantan Dai and Nikola Zivanovic for taking the top three places. Amongst Indian participants, Congratulations to Prasanna Seshadri, Kishore Kumar and Swaroop Guggilam for taking the top 3 places. Of the 180 participants, 48 were from India and 24 from Japan. 1 participant completed the set within the allotted time.

Thanks Huang Yi for the hard puzzles!

Participants may share feedback on the results page and rate the puzzles on the contest page.

The scoreboard will be finalized over the weekend including "no penalty" claims and bonus for puzzles 12-13.

What was your opinion of the distribution of easy/hard puzzles?	Too many hard puzzles
Which set of puzzles did you like the most?	Nonconsecutive Sudoku

Am I the only one who don't get enough information from the feedback summary page?
Which set of puzzles did you like the most? There are 5 choices, but diagram shows only 3 colours and it's quite difficult to tell what colour is what.

TiiT

True. The colour shades have ended up as the same for three choices. While we will try to fix this, you can get more information (currently) by hovering the mouse around the pie chart.

Solutions Booklet: http://logicmastersindia.com/lmitests/dl.asp?attachmentid=766

Just to say thank you to the author, I had fun with the left-over puzzles afterwards, they all seemed to have logical solutions. Well-chosen!

I hope you don't take the criticism here to heart too strongly. While I personally have a preference for hand-constructed puzzles that many here likely share, well-edited computer-generated puzzles certainly have their place. From my experience constructing Sudokus, I also suspect that quite a few established constructors use at least some computer help, particularly when it comes to getting classical Sudokus with non-elementary logic out, or when constructing variants with strong but intractable constraints (non-consecutive, etc.).

Personally I'd be happy to take part in a seond instance.

Edited by rob 2019-02-23 12:33 AM