|
|
An LMI Player with Sudoku ratings 500- | | 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? | Just right | | | What did you think about the puzzle quality of the test? | Very nice | | | How did you feel about the length / time limit for this test? | A bit too many puzzles | | | Of the puzzles you solved/attempted, how well did their point values reflect the difficulty/time spent? | Many puzzles were worth too much or too little | | | What was your opinion of the booklet formatting and printing? | Just right | |
|
| |
|
Posts: 152
 
Location: United Kingdom | i'll respond only briefly re NS, but I'd suggest firstly that is different for a few reasons. firstly, and most importantly in my eyes, the monthly test format gives licence to more creative rules and procedures. secondly, the hard marathon puzzle dilemma was specifically engineered so as not to be a dilemma at all - you were supposed to do the main section first. thirdly even if no marathon puzzles were solved, top solvers were still effectively ranked by how quickly they finished the regular set via bonus points for each minute saved. fourth, i'd argue there is much less "variance" attached to larger, harder nikoli puzzles than to things like the little killer here.
Ultimately, the gimmick with the marathon puzzles was that the points in the bonus section were not equivalent to the points in the main section. Given that this would affect a tiny minority of solvers, together with the fact that this tiny minority of solvers were still effectively rewarded and separated from those who didn't finish the main section, or did finish the main section but in a slower time, there was no qualitative difference in the results. Would I use such a gimmick in a more serious competition? No. Why not attempt to make the points in both sections match up? Because I wanted the 90-95% of solvers to stay on the main section and not be drawn to the marathon section - knowing full well that this group of solvers would at best only just finish the main section if that's all they focused on. In effect, I was removing the option of the big point, high risk, larger and more difficult puzzles. The guiding philosophy were that these were bonus puzzles for "after" the test for the majority of solvers, and something else to really challenge the top solvers who otherwise blasted through the main section in about half the time.
Anyhow, I fully stand by the experiment, and I think there are several reasons why it was suitable for those particular sorts of puzzles rather than sudoku. Bringing the discussion back to the Serbian test, and the spread in points difference. Lets explore your your USPC example. The mean puzzle score there is very roughly 16, and the vast majority of the puzzles are generally worth 10, 15 or 20 points, so the standard deviation isn't hugely high, although I've not taken the time to go into the details. Now take this test. The mean puzzles score is 62.5. You have 7 puzzles from 20-43, significantly lower than the mean, 6 puzzles from 51-72 roughly around the mean and the 3 skewing puzzles from 108-157.
Now I'd agree, trying to edit the difficulties of different puzzle types is going to be a pretty unforgiving task, but I think a think a good test has broadly a consistent level of puzzle, together with a few warm-up puzzles and maybe one or two more challenging puzzles. I think my reaction to my bad result has probably tempered some of my opinion, but there are certainly things worth discussing here, and I'd stand by my observation that difficulty balance was somewhat skewed for this test. Great if you had a good day, but the chances of having a bad day were also increased.
Edited by detuned 2013-04-24 4:20 PM
|
| |
|
Posts: 7
Location: Malaysia | That's very true!! |
| |
|
Posts: 7
Location: Malaysia | f4han - 2013-04-24 9:47 AM
Will author or anyone please teach how to solve little killer, moveable digits, placement shape, killer and word search for this competition? I need help..
So, no one really kind enough to teach me how to solve all these? I'm begging one more time for help please..  |
| |
|
Diagonal Vision Author
Posts: 337
Location: Switzerland | detuned - 2013-04-24 6:41 AM
Looking at the results there are more solvers (I'm not going to name names) than I'd expect who are way below where they "usually" are. Applying the risk variance thesis gives an explanation: a test with more "risk" than normal is inevitably going to throw up more casualties to variance.
I don't know who you are talking about  .
I preferred not to react immediately after the tournament, as I was a bit like Tom just after competing: frustrated of my very poor performance.
It was not a good day for me (I could argue that musicians do sometimes excess, like playing in a jam-session until 4 am while drinking too many beers, when they know they have a rehearsal at 8 am, and they'll be very tired the rest of the weekend, including monday... sometimes clichés are not far from reality).
I started quite well and fast with the first easy sudokus: classic, quadruple, tennis, extra-regions. Then the 1rst mistake arrived in the arrow sudoku, almost at the end, which I was not able not fix. I gave up this sudoku, left the word search, as I don't really like this variant. I solved the irregular and the little killer, which took me lot of time, at least I had this impression. Then, like the arrow sudoku, I broke the frameless, almost at the end, without being able to fix the mistake. I solved the renban and submitted my codes. It was quite a good surprise to see that it remained 55 minutes. I made 397 points in 65 minutes, should have been 518 without having broken 2 sudokus. But even with 397 points, I realized after the tournament that if I had gone on this way, I could theoretically have reached the 13th rank...
Then came the second hour. I solved the even sudoku (I realize just now that I forgot to submit it !!!  ), the diagonal consecutive, and the thermo (I didn't break it, not like during the 2 last LMI tournament), then I started the killer, but I was stuck (my analysis in the upper right of the grid was not complete), so I switched to the movable. And then it was a black hole: I was repeatedly stuck. I missed lot of (subtle) steps, I realized that I should have worked more on this variant before the tournament (just solved the example). I was so obstinated that I almost forgot to submit the 2 (3 in reality) sudokus I solved. When I decided to submit, there were only 2 minutes left, it could also have been possible that the tournament was finished...
So in the last 55 minutes, I only scored 111 points. I lost my concentration and my motivation, too. I was frustrated, I thought that sudokus were too hard...
But after a while, I finished the set, realized that all sudokus, if hard, can be solved without guessing. Also have seen the beautiful theme and realized that the problem was not in the set of sudokus, but in my performance.
I also would like to say, as contribution to the discussion "hard/easy sudoku, balanced set,...", that some players really don't like when all sudokus are too easy, or when there are lot of classic sudokus, and they know plenty solvers will be faster than them. We had sudokucup 9, and also german tournament, which had this profile, and I know some players who were not very comfortable with this. So I think it's good to have some tournaments with 3-4 really hard sudokus, so these players can also show us that they are strong players, perhaps a bit slower, but really consistent on hard variants.
Fred |
| |
|
2021 World Sudoku+Puzzle Convention Organizer
Posts: 1784
 
Location: India | f4han - 2013-04-25 4:24 PM
f4han - 2013-04-24 9:47 AM
Will author or anyone please teach how to solve little killer, moveable digits, placement shape, killer and word search for this competition? I need help..
So, no one really kind enough to teach me how to solve all these? I'm begging one more time for help please..
Do you mean teach how to solve from the start? Is it possible for you to upload images with the progress you have made? We can then take it one at a time. Authors might be busy processing results, etc., and its a lot of typing to write full solves for 5 Sudokus. |
| |
|
Sudoku Grand Prix - Serbian Round Author
Posts: 42
Location: Serbia | f4han - 2013-04-25 4:24 PM
f4han - 2013-04-24 9:47 AM
Will author or anyone please teach how to solve little killer, moveable digits, placement shape, killer and word search for this competition? I need help..
So, no one really kind enough to teach me how to solve all these? I'm begging one more time for help please..
Here are some notes about two of these sudokus (I hope this will be enough for you to solve).
Placement:
There are 4 figures and they overlap in center. You can place 5 in center and note that in other 4 circles in central square are 1,2,6,7.
Look at head of 4 rackets. They are need to be placed in one of corner squares. When you look at givens it is obvious that it is best to start with bottom left racket.
This racket can't be placed in corners with 3 and 4 because he has those numbers in head. This racket also can't be placed in corner with 2 (problem with duplicate number in some row/column - find).
When you place this bottom left racket and transfer numbers in appropriate circles continue with similar logic.
Killer:
First look at top left corner. You have area that sums to 14, one area with ? and single number. ? = 15,30,40 and it gives that R3C3 must be 1.
With similar logic in bottom right corner R7C7 must be 7.
All regions with 3 cells marked with ? must sum to 15. So, with 1 is 5,9 or 6,8. With 7 could be 2,6 or 3,5. Now you look upper to 5 in 6th column.
Two cells in region with 7 must include 2 or 3, so 5=1+4.
Try to identify some more sums and count on that, if you have region with 5 cells that sums to 15 numbers must be small: 1-5, and if you have region with 4 cells that sums to 30 numbers must be big: 6-9
Branko |
| |
|
Posts: 7
Location: Malaysia |  for placement one..I miss to look properly of each shape..and I thought it is unsolvable when I overlap the last circle for each shape in the center of puzzle.. 
For killer, I get it..I just don't try that logic that's why I don't get it.. 
Now I already get moveable digit and word search.. (after try to understand the rules properly and use logic like killer in word search) except little killer.. I just put all possibilities for all cells with clues outside and don't get any number yet..I think it's an embarrassment for me to show my working grid because it just like I never start anywhere..
maybe I should ask where is the first cell or group of cells that can be filled by correct number and it's reason..
Now I know this round is very good except if I still don't know how to solve little killer.. so please tell me asap because I want to know it first before I can rate this competition here..  
Edited by f4han 2013-04-26 10:22 AM
|
| |
|
Posts: 199
Location: United States | detuned - 2013-04-24 3:04 AM
i'll respond only briefly re NS, but I'd suggest firstly that is different for a few reasons. firstly, and most importantly in my eyes, the monthly test format gives licence to more creative rules and procedures. secondly, the hard marathon puzzle dilemma was specifically engineered so as not to be a dilemma at all - you were supposed to do the main section first. thirdly even if no marathon puzzles were solved, top solvers were still effectively ranked by how quickly they finished the regular set via bonus points for each minute saved. fourth, i'd argue there is much less "variance" attached to larger, harder nikoli puzzles than to things like the little killer here.
I strongly disagree that the situations are different. Is there some contract or set of guidelines for Grand Prix authors that doesn't give them some license to experiment where monthly LMI puzzle test authors can? Are you implying the LMI Puzzle Ratings are less meaningful than the GP ratings? That being "fair" is more important in one context and not the other? I appreciate experiments but I think there are lessons to learn and reconsider.
If anything, what separates the two situations is that at least here solvers could start the hard puzzles before the last 20-30 minutes of the test, and therefore had a better chance to finish them and properly use the test time. There is nothing more frustrating than having 15 minutes left at the end of a test to do a 20 minute puzzle. Good test-takers do not leave themselves this situation, but you as a designer still seem to consider it a "feature" (point two). And the scoring of the marathons did affect the top 10 significantly, particularly on the LMI Puzzle Ratings system. So whether solver 100 and 105 are well rated by only being on the main test, the NS was an example of how not to use long puzzles (whether variance was high or not, and this too is debatable). |
| |
|
Posts: 152
Location: United Kingdom | I suppose that's what I'm exactly what I'm implying re the ratings. The WPF GP has much larger participation and comes with prizes at the end of the road. It was a fair old while since I've done the test, but I can almost certainly guarantee that I wasn't paying much attention to the LMI puzzle ratings when I put together that test.
Looking back at the NS results, what seemed to affect things more than anything else was submitting a wrong answer and losing the time bonus. Ignoring those results with a mistake, those who finished early and didn't get a marathon out were very effectively separated by the time bonus, and I don't think there can be any complaints with the qualitative results (i.e. the rankings).
But I get the feeling we are looking at this from very different perspectives. I have to say I don't fully understand the nature of your complaint. I don't know whether you are saying 1) there should have been no significantly longer and harder puzzles at all or 2) significantly longer and harder puzzles should not have been kept separate from the main section. In case 1) - fair enough. This is pretty much my point of view regarding significantly longer and harder puzzles that aren't separated from the main section of a test. In case 1) my only defence is that of experimentation.
If it's case 2) you are arguing, then I'd appreciate some clarification.
To say a solver who finished the main section wasn't rewarded when placed in the situation of having 15 minutes left in which to solve a 20 minute puzzle isn't true because of the time bonus awarded for finishing the main section. Now, the test valued a minute at ~3 points and the time bonus valued a minute at 1 point, and so there is a quantitative issue if you didn't happen to get a marathon puzzle out. This wouldn't change the rankings unless a solver behind you in the main section got more marathon puzzles out than you did. In which case, if there was freedom to choose to do the marathon puzzles then this other solver would still be finishing ahead of you in the rankings, although perhaps not by as many points.
For better or for worse I was not, and am not, so worried by this quantitative issue. My main aim with the marathon system was to give top solvers something to do after finishing the test, whilst keeping the original rankings in place and possibly allowing for small perturbations in those rankings if something exceptional happened. No-one can (or should) legislate for mistakes wiping out bonuses, so I don't think you can say that the scoring system didn't achieve what it set out to do.
To bring things back to this test then the difference here is that the issue of significantly longer and harder puzzles only affected the top 17, and did so in way so as to minimise the effect on the rankings. By not separating the significantly longer and harder puzzles, the variance issue was exposed to everyone, and had much greater effect on the rankings. |
| |
|
Posts: 152
Location: United Kingdom | I'd also like to repeat to the Serbian authors that I very much enjoyed the puzzles on this test, and much like Fred the frustrations I had with my own performance have faded. With good reason, the issue of how to balance the relative difficulties and timings (timings are a significant factor I haven't mentioned in my discussion with motris) have been weighing on my mind greatly in the last week or so! |
| |
WSPC 2017 • Schedule • Participation • Location • WSC • WPC • Contact •Search •Recent
WPF Sudoku Grand Prix — Serbian Round — 20th - 22nd April 2013
