Thanks Dave. We hope to continue improving with each test.

Pleasant as always! I believe the answer submission was the only weakness here -- it was of course a fair answer submission but required an unfortunate amount of attention in some cases, and even then mistakes could still be made. But that is what happens with internet tests, and it did not detract from the overall enjoyment. Good work, thank you Deb.

I agree with you. After some 'guessable' answer keys in BP, I re-did some of the answer keys for this test, and in that process some answer keys became much longer than probably needed, and hence error-prone.
But yes, as you mentioned, it is very important to strike a balance in the answer keys - between being guessable and too-lengthy/error-prone.

Very nice puzzles. Enjoyed a lot solving these puzzles. Submission part was tough. Lost around 103 points as I could not submit answer for one puzzle in group E in time. Big learning from this test. Thanks Deb for creating such a great set of puzzles.

Thanks for the test !

It shows to me that I'm really a "sudoku-only player", as I made a ridiculous score !

Fred

There is always a beginning to something new and as-exciting.

I cannot see my result

My connection was gone by the end of 80 mins..

What do you mean by "connection was gone"? Did you lose internet connection?
As per our database, there is no puzzle submitted by you.

Good test, thanks guys.

Unfortunately, I wasn't very well prepared for it in technical terms - at the start, I printed the test, realising that the colour toner is very low, meaning that some of the puzzles came out in a very low quality, unsuitable for solving. I started to solve those that were visible enough and started a 2nd printing, this time in greyscale - this was when the black toner gave in too! :-(

Of course, that doesn't explain the three submission mistakes I made, in each cases the error was made at the time of determining the submission codes over a correctly solved puzzle (overlooked an X, a double bridge and a turn in Anglers, respectively). Which is part of the contest, I guess, so it's a skill to work on later!

Cheers
z.

@Zormac - bad luck all the way, i guess.
Thanks for participating, I don't recall many participants from Hungary at LMI in recent times (though I can't be sure since country was blank for many players in the past). Hopefully we will see more players from Hungary.

Great test, not very successful on my results.
The presentation is excellent and very welcoming for new puzzlers.
Nice set of classic puzzles and simple-to-understand layouts.

Although you may have to triple check their solution keys, which I guess is unhelpable.
All in all, 5 stars. Congratulations

A rule that a line can't cover a 2x2 square may be needed in some puzzles, to force a unique solution.
If such a rule does not included, one may use the fact that the puzzle has a unique solution, without this rule.

Two questions:
1. In H2O, based on the examples (and chemistry), molecules are always arranged as H-O-H and not H-H-O, but such a rule is not included. Does H-H-O possible?
2. In half dominoes, does the half dominoes should be placed in the marked 3x3 boxes, without overlapping, as in the example?

Thanks, Amnon


Edited by amnon 2010-09-05 9:58 PM

@amnon
In H20 the atoms cannot be connected as H-H-O. They can only be connected as H-O-H.

In half dominoes the dominoes need to be placed in the marked 3X3 boxes without overlapping.

Just to say thanks for the test - many of the puzzles were unknown to me but I really enjoyed having a go despite my extremely low score (although I did complete another 3 of the questions correctly worth 50 points but very stupidly made mistakes in counting the number of turns in both ABC connectors and Masyu and miscounted one of the rows of Xs in the No Four in a row puzzle! ) Not to worry, at least I'll be a bit more careful submitting the results in future! A great selection of puzzles which I will now go away and finish at leisure! Need to get the brain cells ticking over a little quicker!

First I add my thanks for the nice puzzles.
The interesting bonus system requires different approach for the selection of problems to solve.

There is some problem with my submission of the Queens park puzzle. I entered the right solution in the form (I can still see it) and I submitted 3 answers after this one which were received correctly. It tells me that answers to 13 puzzles were submitted, which is the number I sent, including the Queens park.
However, in the results page I see the Queens park, but with empty solution.

amnon, check now

Any other claims from any player, please post soon before we make the results public.

It's OK now, thanks.

Congratulations to Thomas, Ulrich, Hideaki for topping in EverGreen II.
Out of 214 participants, 166 got non zero scores - both numbers are better than any other tests at LMI.
Thank you everyone for participating.

@ amnon

Thanks for your comments about the separate bonuses for the different puzzle sections and different puzzle difficulties.

We have tried to draw inspiration from the Brazilian Puzzle Championships and the WPC in Brazil where they gave extra credit to for participants attempting puzzles across different puzzle types.

@Everybody

Please let us know how you feel about these bonuses and what we can do to improve the bonus system.

Thanks again for running a great puzzle test!

Re: the bonuses, both in my planning before the test and during the test, I figured I would be able to keep track of the "rows" and use that to know if I should stick with a puzzle or not, but it was more difficult to follow the columns or game-plan in the same way since it was less obvious on a single page of the answer sheet that some set was being completed. Perhaps a more involved user-interface would help, but my general opinion is that "group" bonuses can be a good idea but doing two different sorts as here was a little more complicated than its value.

One different aspect in Brazil is that the test-setters wanted people to spread out their work, so they used variable scores for the puzzles in each part, where here the bonuses seemed to encourage people to focus within a difficulty level or puzzle column. Perhaps the right bonus is some multiplicative factor like (# of group A solved * # of group B solved * # of group C solved ...)^some exponent. Then if you want a test with arithmetic puzzles and loop puzzles and sudoku and such grouped into sections, solvers have a motivation to go between the sections more. I'll think about this idea more for the future to see if a good system avails itself.

@motris,
I agree with both your points about this test
1) grouping should be done based on puzzle type, rather than difficulty
2) 2 different bonus system was not necessarily nice, since little planning about it can be done. May be if the puzzles were grouped by 'types' and same column had different point-puzzles, it would have made little more sense. [so that every player has to solve easy puzzles and difficult puzzles to get column bonus]. But that might just make it more complicated.

Thanks to Amit - for testing all the puzzles. His timings helped me put the puzzles into different groups.
Thanks to David - for designing the banner for the test.

Also from me, thanks for the nice contest!

@motris: I disagree that keeping track of the columns was hard, you just needed to have the table next to you and cross out all the puzzles you solved.

But there is another disadvantage with those multidimensional bonuses: One mistake might be punished 3 times (with time bonus 4 times). E.g. I lost more than 100 points by not counting the given 2 in the polymino puzzle.