Also I would like to add few words to the competition. Before the test I thought about the system of submitting answers. I usually submit all solved puzzles and then have 5-8 minutes to "solve" the rest. The biggest disadvantage of this system is that I often missed the time limit and could not submit all the puzzles. But this time I realised that 40 minutes are enough to solve and submit everything. There was also no necessary to speculate on the order of puzzles to solve (this is my another bad habit). But after the test I was quite surprised that 8 puzzles from the top10 have submitted their answers continuosly during solving. Only I and Bastian did it after solving everything. However 6 of the puzzles at positions from 11 to 20 used the way of one submission. Whereas the differences among the top 20 (or among the 3 and 20th position) were very close it seems so that the continious submitting is more suitable. It would be interesting to know how many players of these used the online solving mode, because I suppose, that the great majority of "one submission players" use this way.

When you look at answer strings of each competitor, you can see who solved the puzzles on paper (strings with # in between) and who solved online (just one string of digits)

Díky

Thanks for organizing this contest.

Seeing the results and participation (more than 200 Indian players !), I think you can consider that it's a success.
Being administrator of a french forum, I'm happy to see a good participation of french players. Some of them are regular player here, and others are new or don't play LMI test regularly.
I hope that there will be more swiss players next month. I know some players who could have fun in taking part in these beginners' contests.

I think it should be good to have 2 separate rankings. 1 for experienced or good players ("""professional"""), another one for beginners or players being less competitive and who have perhaps less fun in the "regular" monthly test. I think it should be good to let new players choose in which category they want to play, but I think for following tests, there should be a clear criterion edicted by LMI. The limit could be based on the number of points gained in preceeding beginners' test, if the tests are all approximately the same difficulty. Perhaps having done more than 60 points (means that the player has solved all puzzles or had enough time bonus in one part to compensate some mistake or unsolved in the other part) could be an interesting limit. If you are able to solve the 8 puzzles in 40+40 minutes, you should be able to solve a few grids in "regular" monthly test and perhaps have fun with them. Of course players who didn't have a score of 60 points can also have fun in participating in "regular" monthly test, but perhaps these players will have some frustrating time on hard tests (see distri's previous message).
Perhaps you have to think also about players who solved only one part (classic or variant). If a very good player only take part to one part, that doesn't mean that he should be in the "beginners' liga" next month...

I don't know if a such system can be implemented for these beginners contests?


About solving modes:

I think it could be a bigger difference between both modes than for regular monthly test. Results could let us think that it's probably more efficient to use online mode for these short tests, especially for good players. I played on paper. Despite the fact that I made very good results using online solving on LMI tests in 2010, I've seen that I was not comfortable with the flash LMI interface (just a matter of habit).
I think it doesn't really matter, as it affects much more players who solve each part in less than say 20 minutes, and they are not the target audience of these contests. But it could be good just to mention for each player if they used "online mode", "paper mode" or mixed both.

Finally: the Classics were tougher than the Variants , BUT:
I think it's better that way: One can find tons of classic sudokus in newspaper, website, etc... so we can really practice classic sudokus everywhere, we don't need LMI to get good at classic sudokus. It's a bit different with variants, even if lot of resource can be found on internet.

Fred

Cheers LMI....I think the goal is accomplished.....we have 200 plus Indians participating....hope to see this number increasing in the next months contest....

I had suggested something similar a long time back, that on the score page, to have a tab which shows the results of ONLY those players with LMI rating <= yours (or maybe +- 10% of your rating). That way, each player can see how they performed w.r.t. players who have a rating lower than their's. Similarly, have a tab showing results of all first-timers.

This can be implemented in the LMI Monthly Tests as well.

I dont know how useful this will be, but when I view the score page, I always try to 'find' the players' results whose ratings is +- 50 of my rating.

Thanks to greenhorn, Fred, Rohan for ideas.

We shall be announcing the August Beginners' contest soon. We have decided to go with the following changes for now -
The participation will be divided into two categories, Beginners and Seasoners.

The following are Seasoners by default -
1. Those who have attained, at any point, an LMI Sudoku rating of 500+.
2. Those who have achieved a rank of 25 or better in an LMI monthly Sudoku test.
3. Those who get 2 75+ total scores in the Beginners' Contests(This can obviously take effect only from the September edition).

The rest of the participants all have an option to participate as a Beginner and have their own set of rankings.

There seems to be a couple of bugs in the new Result Table of July Contest... Hopefully, you must be aware about them..... But one of the weirdest one is that it doesn't show the details of all the individual sudokus when I click the row against my username..... It shows them for every other user that I clicked.... I wasn't logged on while viewing the page.... :)

Fixed. Please check again.

Not really. Feel free to post any other oddities that you / anyone else notice/s.

1. TSecs column shows incorrect values for most of the users...

2. When I single-click on my username - it shows multiple entries for the same sudoku... See the snapshot below... (This is again weird - just for me)

3. Problem of multiple entries was there for every user when I checked it before that post.... It has been solved now.... However, if I click on a username 2/3/4/multiple times, details of individual sudokus are retrieved 2/3/4/multiple times as well.....

Snapshot

PS - Why do I have to write the posts in html.... Even after I disable html - I have to use html tags.... Do I need to make any changes in settings?

Edited by adityasaraf007 2012-08-08 11:23 PM

Fixed.

That is because you have submitted same sudoku multiple times. This shown for other users also (check misko for example)

Are you expecting the details to be collapsed?

Not sure. This happens to few users (happens to Rakesh too I think)

Replying to the quotes in a numbered manner because of html issues:

1. Thank you...

2. OMG!!!! Does it mean I am so demanding on LMI Servers..... I used to think everyone submits all the grids for the 2nd time when they are finished solving all the grids/the test is about to over....However, In my view it should show just the last correct timestamp for all the grids....

3. Expecting it to retrieve and show just the last correct timestamp just once (even if the row is clicked multiple times)....

4. Okay.... I would have to live with these html issues....

Edited by adityasaraf007 2012-08-09 3:45 PM

We decided to keep it this way because some people may want to keep track of the multiple submissions too. As a participant I sometimes like to see such things myself. Seeing as it has no change in the scoring, I don't see it as a big issue either way, so I'm thinking we'll let it be that way.

No Problem... :)

I could not see the results before finishing Classics, which was fine. But once I finished Classics, I could see overall results for BOTH Classics and Variants. More of an oddity than an issue. :-)

Fixed. Please check this time.

In light of some performances from players with a prior reputation, we have added a 4th category as follows -
4. Those who score 100+ on the Beginners' Contest shall be bumped to Seasoner level on the spot.

We must remember that the main point of these contests is to encourage new talents who are not familiar with all the types and techniques of Sudoku.

Two of the links on the left pane of the test page extend across the width of the page.was that intended?

Edited by anurag 2012-08-11 4:43 PM

The two in question are explanations of the LMI interface in detail, with screenshots of the various features. So yes, it was intended.

LAME! While sitting here waiting for the timer to run out, I notice that I entered the same answer twice! Lucky I saw it, but I lost 12 minutes or so. :(

In the score page, I find a couple of scores (excluding Time Bonus) which are not whole numbers.... Scores like 28.60, 28.80, 17.50, etc.... Does it indicate partial points for 1 answer key being correct or something like this.... How are partial points allotted...

Partial points are allotted if just one number is wrong, or if 2 numbers adjacent to each other are swapped in the entire key, and nothing more than that. For more details you can look at the July version. I will say, we are a bit more lenient here as it is a Beginners' Contest. You may not get points for the same keys in a Monthly Test.