Sprint Test — LMI September Puzzle Test — 3rd and 4th Sept

Information and discussion on Puzzle Competitions
Post Reply
debmohanty
Posts: 129
Joined: Fri 09 Jul, 2010 12:36 pm

Sprint Test — LMI September Puzzle Test — 3rd and 4th Sept

Post by debmohanty » Tue 30 Aug, 2011 1:47 pm

Sprint Test — LMI September Puzzle Test — 3rd and 4th September

IB and Submission Link - http://logicmastersindia.com/M201109P/
Author - Bastien "Ours brun" Vial-Jaime
On : 3rd and 4th September
Length : 35~75 minutes


User avatar
furudo.erika
Posts: 76
Joined: Wed 10 Aug, 2011 11:55 am

Re: Sprint Test — LMI September Puzzle Test — 3rd and 4th Se

Post by furudo.erika » Sun 04 Sep, 2011 4:55 pm

I hate my printer :evil:

PuzzleScot
Site Admin
Posts: 2712
Joined: Fri 18 Jun, 2010 10:45 pm
Location: Edinburgh, Scotland

Re: Sprint Test — LMI September Puzzle Test — 3rd and 4th Se

Post by PuzzleScot » Sun 04 Sep, 2011 9:32 pm

I hate my speed-solving skills! :evil:

drsteve
Posts: 722
Joined: Sun 27 Jun, 2010 7:23 am

Re: Sprint Test — LMI September Puzzle Test — 3rd and 4th Se

Post by drsteve » Mon 05 Sep, 2011 5:37 am

I'm not entirely enamoured with a bonus structure that gives some people who didn't get them all right an overall score higher than some people who did :twisted:

Wouldn't say hate, mind you. Just my sour grapes.

emma
Posts: 311
Joined: Wed 22 Jun, 2011 4:58 pm

Re: Sprint Test — LMI September Puzzle Test — 3rd and 4th Se

Post by emma » Mon 05 Sep, 2011 4:53 pm

Presumably that would mean that if you're confident of your accuracy, you may be better to submit something random for the fifteenth problem?

dickoon
Posts: 614
Joined: Mon 21 Jun, 2010 12:52 pm
Location: Stockton-on-Tees, UK
Contact:

Re: Sprint Test — LMI September Puzzle Test — 3rd and 4th Se

Post by dickoon » Mon 05 Sep, 2011 6:40 pm

Cease the negativity (*); even though I had a big last place among UK finishers who solved even one puzzle, I loved the test! Fingers crossed for more tests with us less accomplished solvers (not quite the same thing as beginners...) in mind, as well as ones for the championship performers.

(*) ...recognising that people aren't actually being negative about the test or its puzzles.

drsteve
Posts: 722
Joined: Sun 27 Jun, 2010 7:23 am

Re: Sprint Test — LMI September Puzzle Test — 3rd and 4th Se

Post by drsteve » Mon 05 Sep, 2011 6:43 pm

Well, it's the problem with a test of this nature - personally, I'd have gone with a different bonus structure that rewarded the completers ahead of the 14/15s. The moral ought to be, check your work or take a chance. As emma points out, there is a way of not attempting one puzzle and getting a better score for it, which doesn't sound like playing the game to me.

It was an excellent competition (although I thought the difficulties varied quite a lot, not necessarily in line with the points) but very enjoyable.

motris
Posts: 77
Joined: Tue 21 Dec, 2010 5:35 pm
Contact:

Re: Sprint Test — LMI September Puzzle Test — 3rd and 4th Se

Post by motris » Tue 06 Sep, 2011 1:50 am

Having a mistake (or stopping with 14/15) was effectively a 10-15 minute penalty depending on the missing puzzle. I think that is a very large price to pay. Sure, you didn't get to beat two wpc champions in uvo or tarotaro, but they also finished 20 minutes in front of you. But you did beat 6 others who finished sooner but were not completely correct.

Imagine this test with a two week time limit. Would you agree then that someone with 15/15 after 2 weeks is better than someone with 14/15 in 45 minutes with a typo or small error in the last. Perhaps you would. But I think the system as is is a good compromise.

And to echo dickoon, the puzzles were great and the concept was a nice change too.

drsteve
Posts: 722
Joined: Sun 27 Jun, 2010 7:23 am

Re: Sprint Test — LMI September Puzzle Test — 3rd and 4th Se

Post by drsteve » Tue 06 Sep, 2011 6:01 am

motris wrote:Having a mistake (or stopping with 14/15) was effectively a 10-15 minute penalty depending on the missing puzzle. I think that is a very large price to pay. Sure, you didn't get to beat two wpc champions in uvo or tarotaro, but they also finished 20 minutes in front of you. But you did beat 6 others who finished sooner but were not completely correct.
This is starting to sound like sour grapes from me, which it isn't - I really enjoyed the test - but I would debate the word "finished" here - if yuo got one wrong, then in my opinion, you didn't finish...

PuzzleScot
Site Admin
Posts: 2712
Joined: Fri 18 Jun, 2010 10:45 pm
Location: Edinburgh, Scotland

Re: Sprint Test — LMI September Puzzle Test — 3rd and 4th Se

Post by PuzzleScot » Tue 06 Sep, 2011 7:29 am

Steve,
Two adages for you.
1) If a sentence contains the word 'but', you can disregard everthing that comes before it.
2) When you find yourself in a hole, stop digging.
Now, when I say "well done Steve", and you reply with "Thanks!", we're done 8-)

PuzzleScot
Site Admin
Posts: 2712
Joined: Fri 18 Jun, 2010 10:45 pm
Location: Edinburgh, Scotland

Re: Sprint Test — LMI September Puzzle Test — 3rd and 4th Se

Post by PuzzleScot » Tue 06 Sep, 2011 7:29 am

Well done Steve!

detuned
Posts: 1672
Joined: Mon 21 Jun, 2010 2:25 pm
Location: London, UK
Contact:

Re: Sprint Test — LMI September Puzzle Test — 3rd and 4th Se

Post by detuned » Tue 06 Sep, 2011 8:52 am

Haha, well having suffered greatly and extensively from an accuracy vs speed issue, I have to say I'm whole heartedly with Steve on this one. I'd be counting another two Times sudoku championships otherwise, having simul-solved the grand finals quicker than any finalist without having managed to qualify for the final myself. The point being, although I was the quickest solver there, my accuracy issues were certainly enough to have made me an unworthy champion in my own eyes. Nevertheless, I do think - and have stated in the past - that LMI is a great place to experiment with new ideas. Not all are going to be hits, but at least it provokes discussion!

I printed off the puzzles last night and had a go - began timing myself, but it became clear that losing my pencils somewhere between here and Dublin was making things difficult. Said conference in Dublin was the reason that I was unfortunately unable to have a go at this in time. I thought the puzzles were general fun and well presented. The masyu in particular caught my eye :)

I also wasn't a fan of the implementation of the heyawacky rule, although this only caused minor confusion at the end of the puzzle. The way Thomas (and others) have implemented it in the past via no 2 borders crossed (rather than no more than 2 different regions being traversed) feels to me the more natural extension of the heyawake solving "feel". I also thought the mini sudoku was a little too much trial-and-error for my tastes, although perhaps you can forgive this in 6x6. As I said I didn't have a pencil!

drsteve
Posts: 722
Joined: Sun 27 Jun, 2010 7:23 am

Re: Sprint Test — LMI September Puzzle Test — 3rd and 4th Se

Post by drsteve » Tue 06 Sep, 2011 11:45 am

PuzzleScot wrote:Well done Steve!
Grumblen grumble .... Thanks ... Grumble grumble...

Fred76
Posts: 83
Joined: Mon 09 Aug, 2010 1:53 pm
Contact:

Re: Sprint Test — LMI September Puzzle Test — 3rd and 4th Se

Post by Fred76 » Tue 06 Sep, 2011 12:40 pm

detuned wrote:I also thought the mini sudoku was a little too much trial-and-error for my tastes, although perhaps you can forgive this in 6x6.
:roll: The mini sudoku was perhaps a bit subtle, but really don't need T&E to be solved :P :

After putting obvious 3 in R2C6 and R4C2, one can see that R1C2 and R5C6 must be even numbers. So R2C1 and R6C5 are odd numbers, according to outsite clues. So you can put all 4's in the grid. Then you look at the last outside clue (the 2 in the big diagonal), and you can see that in this diagonal there can be at most one even number in the 4th region (placed in R4C4) and one in the 5th region. That gives you a pair 26 in region 4 (R4C4 and R4C6), then you can easily finish to solve the grid.

Fred

drsteve
Posts: 722
Joined: Sun 27 Jun, 2010 7:23 am

Re: Sprint Test — LMI September Puzzle Test — 3rd and 4th Se

Post by drsteve » Tue 06 Sep, 2011 1:19 pm

Fred76 wrote:
detuned wrote:I also thought the mini sudoku was a little too much trial-and-error for my tastes, although perhaps you can forgive this in 6x6.
:roll: The mini sudoku was perhaps a bit subtle, but really don't need T&E to be solved :P :

After putting obvious 3 in R2C6 and R4C2, one can see that R1C2 and R5C6 must be even numbers. So R2C1 and R6C5 are odd numbers, according to outsite clues. So you can put all 4's in the grid. Then you look at the last outside clue (the 2 in the big diagonal), and you can see that in this diagonal there can be at most one even number in the 4th region (placed in R4C4) and one in the 5th region. That gives you a pair 26 in region 4 (R4C4 and R4C6), then you can easily finish to solve the grid.

Fred
Thought this was a lovely puzzle myself - possibly my favourite of the lot as I had to generate new solving strategies on the hoof.

Yes, you heard me. My favourite was the sudoku.

detuned
Posts: 1672
Joined: Mon 21 Jun, 2010 2:25 pm
Location: London, UK
Contact:

Re: Sprint Test — LMI September Puzzle Test — 3rd and 4th Se

Post by detuned » Tue 06 Sep, 2011 5:49 pm

Ok, I feel a bit stupid now, that wasn't so hard to see - but thanks anyway Fred! Let's hope I can find some sort of puzzling form again in the next couple of weeks...

david mcneill
Posts: 400
Joined: Mon 21 Jun, 2010 2:02 pm

Re: Sprint Test — LMI September Puzzle Test — 3rd and 4th Se

Post by david mcneill » Tue 06 Sep, 2011 10:29 pm

I too printed the puzzles off after the event. Despite having more than one pencil, I struggled. I found the Fortress and Heyawacky puzzles particularly difficult and finished the set in 76 minutes (including writing out the answer keys). Thanks to Bastien for providing a fun challenge and well done to those who showed me a clean pair of heels.

GarethMoore
Posts: 1176
Joined: Thu 24 Jun, 2010 9:27 pm
Contact:

Re: Sprint Test — LMI September Puzzle Test — 3rd and 4th Se

Post by GarethMoore » Wed 07 Sep, 2011 12:40 am

david mcneill wrote:I too printed the puzzles off after the event. Despite having more than one pencil, I struggled. I found the Fortress and Heyawacky puzzles particularly difficult and finished the set in 76 minutes (including writing out the answer keys). Thanks to Bastien for providing a fun challenge and well done to those who showed me a clean pair of heels.
Glad it wasn't just me! :) I gave up on the Fortress in the end - you can make a few logical deductions but then it's mostly luck I think? I realise you can eliminate some options immediately, but nowhere near enough so far as I can see. Probably I wasn't thorough enough at exploring the various routes, but I spent a reasonable amount of time on it both before and after and didn't manage it.

I thought the Heyawacky variation made better sense with that rule about room crossings, although it does make it fiddlier to solve - I thought the puzzle involved significant trial and error but at least it mostly gave local contradictions when you picked the wrong options so I didn't mind it so much.

And star battle remains my nemesis - I made several deductions but couldn't get near enough to finishing it.

motris
Posts: 77
Joined: Tue 21 Dec, 2010 5:35 pm
Contact:

Re: Sprint Test — LMI September Puzzle Test — 3rd and 4th Se

Post by motris » Wed 07 Sep, 2011 3:25 am

The Fortress had a nice (medium-level) logical route, but I learned how to solve the puzzle type during the test so it is probably where I lost some time.

Rows 5 and 7 have only two possible circle choices and they are in the same columns so those must use the columns and you can eliminate those above/below in the same columns. This eliminates one of the circles in the bottom row. You also need to see that you cannot use a corner circle that has another circle within 2 squares of it on the edge (you cannot turn away fast enough) so that also eliminates the two corners in the bottom row, leaving just one circle. Shade that one. In the row above the bottom row, there are 4 choices but only 1 I know how to visit with the grid as is now. That circle in column 6 eliminates all but a column 4 choice in row 3.

If you hadn't to this point, notice now that the top effectively has two enclosures. The path certainly visits R1C8 and goes through one of the circles in that row. Whichever it goes through, it cannot use the circles in row 2 in C7 or C9 in order for the path to eventually get out of one of the enclosures. This eliminates R2C7 and leaves one circle in that column, R1C7, which you can pass through and from there with 4 marked circles you have mostly single choice circles left to achieve the path.

david mcneill
Posts: 400
Joined: Mon 21 Jun, 2010 2:02 pm

Re: Sprint Test — LMI September Puzzle Test — 3rd and 4th Se

Post by david mcneill » Wed 07 Sep, 2011 12:40 pm

Yes. The row 5/7 insight was the one I didn't find quickly enough. Tough in a sprint contest.

Ours brun
Posts: 34
Joined: Thu 04 Nov, 2010 1:30 pm
Contact:

Re: Sprint Test — LMI September Puzzle Test — 3rd and 4th Se

Post by Ours brun » Tue 13 Sep, 2011 8:52 am

First, sorry for the very late answer; Sprint Test wasn't even finished that I had to travel to the other side of France to organize a paper sudoku tournament - now done - and I now have an online one to finalize before this week-end. So, very few time left for the after-Sprint Test.
drsteve wrote:It was an excellent competition (although I thought the difficulties varied quite a lot, not necessarily in line with the points) but very enjoyable.
Difficulties varied a lot indeed, due to many things (amongst which the fact that when you create a puzzle, you automatically tend to find it easier than it is, since you know how to solve it - which made me underestimate at first the Fortress and the Star Battle, for example).
Regarding the points, I am happy to say that the only ones to blame are the test-solvers, since points were attributed depending on their times. If it had depended only on me, I would certainly have attributed them better than that. :roll: Seriously, I think this is very personal. For example, tester Sylvain Caudmont solved the sudoku pretty quickly since he is a strong sudoku player, but he had trouble on the masyu although the puzzle is extremely easy for someone used to the type. The main point was to warn beginners about the tougher puzzles.
detuned wrote:The masyu in particular caught my eye :)
Honestly, since I discovered the way you treat masyus, I almost feel bad when I make one which doesn't have twisted symmetry. Just soo esthetic.
detuned wrote:I also thought the mini sudoku was a little too much trial-and-error for my tastes, although perhaps you can forgive this in 6x6. As I said I didn't have a pencil!
Some centuries ago, I would probably have challenged you to a duel for such an insult. Thanks Fred and drsteve for saving my honor. "a little too much trial-and-error"... Pooh ! Work on your subtelty, my dear. :twisted:

Thanks Thomas for detailing the solving path of the Fortress. According to me it was one of the more interesting puzzles in the set, although a bit difficult for such a test.
GarethMoore wrote:I thought the Heyawacky variation made better sense with that rule about room crossings, although it does make it fiddlier to solve - I thought the puzzle involved significant trial and error but at least it mostly gave local contradictions when you picked the wrong options so I didn't mind it so much.
I personally felt the variation more interesting this way; but it makes the puzzle much more error-prone indeed. Not sure of what you consider T&E on such a puzzle, since about everything is a question of local contradictions on heyawake-like puzzles. I can guarantee that there was an interesting logical route to follow, which didn't need what I would personnally call T&E. If I have the time I will post a description of the solving path.

Many thanks to all of you for comments such as "puzzles were great", which are a really nice reward - especially since it was my first puzzle test authored, and for most types, these puzzles were my first try ever.
http://enigm-attic.blogspot.com --> sudoku variants and other handcrafted logic puzzles.

Post Reply