Do we have online solving enabled for this test?

Sorry, forgot to post, there won't be any online solving. It is because of the last sudoku "Prison Break".
May be we should explore, if we should enable online solving even in cases where 1 or 2 sudokus are not solvable online.

Edited by Kwaka 2012-11-03 11:12 AM

15 pages of the PB? It is too much, don´t you think?
I believe that two grids per paper are always enough. Also big grids slow the solving, because it is not easy to look at the whole grid with one view.

Anyway I really enjoyed some of the puzzles. Those I have solved in time were very tricky and nice for me. But I spent just 7 minutes until my printer deal with such a huge amount of papers (because it froze I wasn´t able to rewrite the sudokus on paper).

I'm very sorry about that.
I hadn't thought of that problem because I usually print out 2 pages per sheet when I take this kind of test.

Anyway, thanks for enjoying the test.

I think password-protected-pdfs for the tests should be designed such that players should just "Print" the pdfs without thinking / trying to find out a) whether I should print the first page b) whether I should print 2 pages per sheet etc. They should just press Control + P, and hit enter.
That is the reason almost all LMI tests (except annual contests) password-protected-pdfs don't have cover page. But we do not enforce any other formatting criteria. Having just one grid per page is too much in my opinion. But as we saw in FF2, there were lot of complaints because the grid sizes were small with 2 grids per page.

Some of these, e.g. Clue, Prison Break are better on a bigger grid for markings IMO. Also, Fake Treasure and Twin Murderers are fit into 1 page. So you're left with just a few that might have been smaller, and still not too small. The size of grid differs from person to person and adding to what Deb said about last weekend's test, its difficult to set the right size. I think viewing the entire grid shouldn't be a big problem, if it does decrease the speed by a little amount, its the same for everyone.

I enjoyed the test, thanks for writing it!

I have my own habits in printing for tests.
My printer is veeeeeery sloooooow, although it is not an old one.
For that reason, I decide before I hit the start button, which puzzle I want to start with, and I print out first only the page that contains this puzzle. I choose for a draft print to increase the speed for this page. It then still takes about 45 seconds before I can start solving.
While the printer is busy, I hit the print-button again, to print the entire file, this time in a better quality. By the time I have finished my first puzzle, the entire printout is ready, or at least enough pages to choose from to continue.

I always have difficulties with the huge grids. I think mainly because I am not used to it.
When creating puzzles, I always make cells with a size of about 1 centimeter, which is not too small so you can write notes in it and it is also not too large.

When cells have a size of (about) one centimeter, you can easily fit two puzzles on a page, especially when the instructions don't ask much space, what can be done using a smaller font size.

Richard

Thanks for the very nice test.

There are some really nice sudokus and also some interesting new ideas.
Territorial conflict was great, even if I think that I should have solved it faster. I like when you have to think globally and not only locally to solve grids. This was a very good example (so simple when you understand what is the trick...).
Little multiplication killer was probably my favourite: a very clever construction, very nice solving path.
Ditto with little killer with extraregions.
The logic of jail sudoku is also very interesting.




There are lot of online tournament that have 1 grid each page. Lot of previous monthly tests; sudokucup has also actually this format, with 15 sudokus, one per page.
Last WSC had also this format, even with bigger grids.
I think there is nothing wrong with having only one sudoku on each page. This is a matter of comfort for each player, I know some players who like very big grids, others, like you, prefer to have smaller grid (2 grids per page).
Sometimes it's not convenient to have 2 grids per page, when rules are complicated it can represent a big text. When you have outside clues, you must have even smaller grids, which is ok if players don't have to write lot of pencilmarks inside the grid.
Here I see not matter to complain about the format.
The only thing is that the pdf is perhaps a bit too big, and it can also have an influence on the printing process I think.

Fred

Kwaka, don´t worry, it is just my opinion and other players should not have such a problem. The most important thing is that your puzzles are really well made and logically solvable. The question about format and sizes needs to be discussed among more players.

I agree with Prasanna that some puzzles need to be big enough in case of pencilmarks. But do we need all the instructions next to grids even when the IB is released few days before the competition?

Fred, I went through some of that contests and I talked with Karel about format of Sudokucup´s PB. Last few times I haven´t participated at SC and helped Karel with presolving and we still have different opinion on the question of format. The number of pages and the size of grids are ones of the criteria that are crucial for me when deciding whether I will participate or not.
It is different at the WSC, because I am not the person who prints the booklets (during the competition as in LMI tests). It is issue of organisers to chose the easiest and cheapes way of printing. In the USA there was only one grid per page (width of the cell was 1 cm), but the size was normal, not as the croatian ones (1,65 cm). However I think that the hungarian puzzle booklets were also clear enough even when they contain usually two grids per page (0,9 cm).

Richard, thank you for sharing your experiences. It seems to be good idea.

Edited by greenhorn 2012-11-04 12:20 AM

Well as far as the instructions go, I think many people would like the instructions there to refer to at times. Personally I do that for some of the innovative variants on tests. At most we can just advise future authors to make an instruction-less version of the PB, but only if the one with instructions has a big number of pages.

Anyway, the test was great for me too. My favorite was Little Killer with ER and the way that the Extra Regions were designed to play into it. Other favorites include basically the same as Fred's choices, but I'll add Weighted Killer in there. The theming of the Pentomino pieces in Prison Break was nice too.

Performance-wise, I forgot to submit Clue Sudoku within the time limit, so I think my ranking's taken a hit there. It still seems to be an ok score looking at the others now.

Thanks to Kwaka and Spica!

A beautiful test, and I loved the multiplication killer and the jail - very nice solving there. I finished the jail just a minute too late!

I was very frustrated to spend a lot of time on the killer with doubled cells only to end up with a mistake in my solution. There were a lot of nice things to uncover in that puzzle but it still seemed very difficult. I should have spent my time elsewhere -- or maybe now looking at it after the test I'll see a better way to think about it.

With no more participants, Fugitive Warrant is now over.

Password removed from pdf.

Thanks for enjoying our test. It was very good experience to be an author of LMI test.
Next time I'll consider more about size of puzzle.

Also, few more things.
Little Multiplication Killer : First time, there is no given number inside of the grid. It can be still solvable without it.
Little Killer with ER : Also, the number in R9C1 which is 7 was not given first time. It can be still solvable without it. (But one can know the sum of R9C1 and R1C9 using the sum of remainders which I saw at the test 'Killer Shootout'. )

Last, I have a few questions.
How do you make a 'nonconsecutive sudoku' ?
Do you use a solution which is easily gotten from other problems, or just use a computer when you make it?
Same question for 'Antiknight', 'Untouch'.
Somebody help me to improve my ability to make.


Kwak

Edited by Kwaka 2012-11-06 6:13 AM

Congratulations to Kota for his win. It is hard to believe that this his first win in an LMI test, but he has been in top 3 in each of last 7 sudoku tests, except one.

I would love some solution tips on the weighted killer. I did eventually get it but it took me a very long time!

Link to Solutions booklet

joshuazucker// Thanks for enjoy our Sudokus.
Weighted Killer(Spoiler!!) : See http://home.postech.ac.kr/~meso/sudoku/weighted_killer.html
When I made this problem, Kwaka found that solution route very quickly. That's why we give (relatively) low points on this hard problem. Sorry for that

Yeah, I got stuck after finding the 389 in the middle of the top and finding the shaded 1s and 2s in the bottom rows ... what next?

joshuazucker// See also http://home.postech.ac.kr/~meso/sudoku/weighted_killer.jpg

Thanks for all the help! You're still earlier in the solve than the spot where it got hard for me.