Saturday, December 16, 2023

WSPC 2023


And looking back the latest championship. Detailed results can be found in the official site.


WSC
I was clearly aiming for a playoff place.

Round 1
As my usual, I started from high pointers, which means two 70s in this round. And then solved the others, mostly from the end. I did not expect finishing here, the difficulty was much easier than expected.

Round 2
Went backwards. Surprisingly good result in Classic round, even without assumptions. Most of my Sudoku practice is for easy-middle Classic, so good reward?

Round 3
Solved 95 points Thermometer first (quite tough one), then mostly backwards but skipped Pencil Marks at first. I did not plan to solve this before the event, but with the difficulty of previous rounds, I decided to try it at some point. I was lucky to hit two-degree assumption. I tried Anti-Knight in the end but broke it, I tweaked and got valid Classic solution in time but it was still incorrect with 4-cell switch.

Round 4
Points distribution had less variance here, so in order to avoid confusion, I did not prefer Math Grid and Bubble. Maybe typical strategy? Then I went backwards, but my beginning Group Sum was the biggest trouble, where I missed repetition around the clue and wasted some time. Disappointing result considering the round theme.

Round 5
Then the trouble occurred; I was sleepy after the lunch, which is rare for me. This was my first visit to that area, and apparently I did not fit the time zone. This was significant trouble for my and our team throughout the week.  I do not want to recall this round so much... broke Triple Diagonal, major error in the first Renban, and an empty cell in Windoku!


Round 6
Went backwards. I had to reset my mind here. Not the best with some erasing, but happy to recover from the previous round.

Round 11
Kota and Hideaki solved 1-5, I and Takuya for 6-9. The pairing was just because of original seat. My side was slower with a wrong step, so Kota helped us in the end.

Round 12
My first role was to decide the position of Classic pieces, together with Killer pieces; we expected the orientation would be obvious for those. No progress for a while with given Killer clues, and I noticed the remaining sums could be calculated, and... gave them to Hideaki! It must have been the best choice. It was rather smooth after that.

 So I was 9th after day 1, largely affected by one bad round. Still not giving up the final, I had to perform so well on day 2.

Round 7
Backwards, with many guesses, managed to finish in the end but made a 2-cell switch error. I did not know that at that moment but it could have been critical.

Round 8
Starting from the second Outside, and then Skyscrapers where I smoothly found the logic. Then solved Rossini kinds and went backwards from String (it was very nice btw). I did not have enough time for the cheapest one, so checked for a while in the end. The result was better than expected, maybe puzzly side?

Round 9
Better than Classic, but  I am not comfortable with these kinds. Weave was the worst one for me, I learnt 45-degree rotation from Kota but still broke the example, so I left if for last and went backwards skipping it. I had less trouble thankfully.

Round 10
Easy to expect this round would benefit me, and actually it was. Just backwards.

Round 13
Expected construction, so my role was managing the final grid. It worked well enough.

Round 14
Kota and Takuya preferred Classic, then I chose Turquoise (green), so our positions are:
Turquoise: me
Blue: Takuya
Red: Hideaki
Yellow: Kota
We struggled with #1. Sometimes I had no progress on it, or even erased my seal which I place in my previous turn! After the round I heard that our stuck point was in Turquoise, but I do not remember the exact state. One funny thing was that I made no submission here, probably thanks to Kota.

Round 15
We sat the same seat as the previous round, and the puzzle distribution was:
Set A: Kota
Set B: Hideaki
Set C: me
Set D: Takuya
I had no trouble on instructionless part, and when I finished checking and was considering second check, Takuya submitted, so I joined in. Kota came a bit later, and it was good time to know the remaining rule. I and Takuya solved top-left grid, and when I noticed that 3-digit number can be determined, I gave it to Takuya! Similar work as round 12, again it was good decision. Hideaki managed to finish his part (probably the most troublesome set), so we could check Samurai together. We finished in 39 minutes and the round was extended for 15 minutes, so I think we should get bonus for that, but no change for the ranking.

Playoff
Yeah I achieved my first target! I had to wait for 6:20 but in case of severe playoff I might had a chance. It was not the case this year and easier than expected 50 minutes, I had no chance and just solved in my pace. But it must be good opportunity, and it is positive point that I was not too slow. Btw, I had strange feeling for no break between puzzles.

Sudoku GP Final
Unusual timing this year. Online Stage was announced in short, so I had to change my schedule. My online solving included breaking easy Classic, which made Tom (the author) laughing, but I believed that Tantan would not prefer Classic. It was true and I still had a chance for win in Live Stage. I was lucky in Classic here, and had a chance  of Grand Final, with Anti-Windoku (I think there was bad communication, her intention seemed to be Non-Consecutive). Badly I made a wrong Classic step, I need to improve this.

By the way, the rule says that a bye player always loses, but I think it was not applied in that way, if the structure is same next year, I would like to clarify. The best thing is that everyone can come to Live Stage, of course.

And I would like to leave my preference list:
Arrow
Sum by X
Big Bands
Thermo
Self-Joint
Classic C
Classic B
Classic A

Unfortunately I was sicky next day, had to give up Niagara Falls tour. Maybe I spent too much energy for WSC, or GP Final ended late.... Due to the jet lag I woke up every two hours, only four hour sleep in total, and after the sleepy round I reduced my lunch, there were so many reasons to lose my condition. I even doubted if I would survive WPC, at least I had to save my energy. This is why I wore more in WPC.


WPC
Anyway it began. Due to the announced difficulty, I changed my strategy, to be much more simple order.

Round 1
Went backwards. I had counting trouble on the second Fillomino, and guessed a lot on the second Kurotto, so high pointers were hard enough, but solvable. This year I checked very carefully after each solving, one error might lead to losing hundreds of points in this situation. I did not spend additional minutes in the end instead.

Round 2
The order was #4, 5, 6, 3, 2, 1, this was typical order in many rounds with 6 puzzles. I feared the unique rule of Konkat-kuro but it was not used here. I almost missed a cell in Invisible Doubles but managed to find while checking.

Round 3
Trying to go backwards, but after 20 minutes I had one unique side and one no-solution side, I made some wrong deduction but it was not easy to fix it. Giving it up, I managed to catch two middle pointers and still had 5 minutes or so, but I had some confusion in Skyscrapers and could not finish it. There were some risky round this year and this round was one of them, so with my condition I expected such trouble round, but... I was not welcomed to Niagara at all. One positive point was that my condition got a bit better with a lot of sweat during this round. And great performance from Naoki.

Round 4
Here the order was #5, 6 then backwards. Thankfully there were intuition-friendly.

Round 5
Same order as round 2. Well themed round. I tried not to guess so much because I did not want to do severe checking, and it worked well.

Round 6
Another risky round but it was rather smooth. I had to erase a bit but not too much.

Round 7
Clearly bad performance. Going backwards, I did not check given banks in Battleships and Statue Park, wasted too much time. I guessed a lot in these genres and was forced to check carefully.

Round 11
Same order as round 4, another typical way. The last steps of #2 was not easy and even after getting solution I tried the wrong side to confirm.

Round 12
Just  backwards. I broke Heyawake partially but only erased a bit.

Round 13
Next day, getting better but not perfect yet, and bad start. My solving order was #5 to 9 and then #4 to 1, and my first trouble was in #7, which I somehow broke many times and restart again and again. I spent some time for understanding #9 - at some point I felt it was similar to Mochikoro and that was the trigger. And other troubles were in easy ones, I forgot to connect same alphabets and only trying separating different alphabets. Our team did not prepare enough for this round, our fault.

Round 14
Same order as round 4. I could not see logical way for Antisymmetry, so it was solved with case distinction.

Round 15
Went backwards. Error-prone round, and I wanted to solve logically and keep accuracy, but Masyu Triangular was done by intuitive solve.

Round 16
Solved 100+ pointers first then went backwards. 140 points Exercise was the hardest for me but managed to find the pattern. One good surprise was that Aoi, our newcomer this year, finished for the first time and get out of the room.

Round 18
Same order as round 2, I wanted to finish Diagonal soon. Tough puzzles as expected, I guessed a lot, even in a cheaper one.

Round 19
After the falling in Niagara, here I began with Fillomino. Holes were not easy to grasp, when I needed one more cell and checked the grid, there was an unused hole cell.

Round 20
In the given order, for the first time this year or I should include round 6 or 19. Normally this is guessy type but I still feared for errors.

Round 10
After saving lunch, I felt hungry at the beginning of this round, maybe lacked some energy. I started from Crisscross, which I thought safe choice, but spent 20 minutes for this, clearly out of plan. Then solved Star-Crossed and One or Two, and my next plan was Crisscross Crash but could not find the trick, so went to Piece Arch, which I was to leave for last. Only 5 minutes or so was remaining then, so gathered 10 points Wordle Bank and checked for several minutes. Hard Wordle Bank was logical, so maybe I could have get some more points in different order, but not so much. Tough round for me and our team, I am not sure about the portion solving skill and language advantage, I would like to see a number-themed round to understand the situation.

To summarize the individual rounds, it was easy, maybe too much. It tested our accuracy again and again, and most of my conscious was spent for it, but I felt strategical aspect was not tested enough. Maybe finishable difficulty is good for saving my energy, but I also love tough contests too.

Puzzle GP Final
I decided to take enough dinner. I felt sleepy, but I could have some exercise during video presentation, and felt well in the Final. My preference list were:
Skyscrapers+Kakuro
Cave
Aqre
Snake Pit
Nurikabe
Infection
Kakuro
Letter Touch
When I knew the new format, my first impression was that I did not want to play "Printing" GP. Possibly I would change this next year after checking the data, but anyway this was my principal this year. Actually I had that trouble in Kakuro, as you can see in my video. Word types cannot be my preference, so my candidates were Skyscrapers+Kakuro, Cave and Aqre. I also considered adding classic Kakuro to this list, but imagining the tournament table (not revealed), Hideaki can be my opponent in my second match, and Kota in third (in fact both came true), so I chose Number types for the first. For Cave and Aqre, I believed in Matej from the result of last year, so placed Aqre to more important spot. The rule at that moment said I might choose the remaining puzzle, so my choice were to make more accident. I would have chosen differently in the correct format.

Online Stage was ok for me, and my possible opponents in Live Stage were Naoki, Jeffrey, Thomas at that moment. So my best choice for the Grand Final was Kakuro for sure. Even if I lose my first match, Naoki would skip Kakuro, so my task was how to beat him in one of the remaining three puzzles. No lose was better, but I had at most two chances. I knew his preference to Pentominous, and he must have been accustomed to Slitherlink, so chose Cave. I gave him Pentominous but I know man could have trouble in the playoff situation, and actually it was the case for him, not critical though.

It was a pity that there were no video for the audience, on the other hand we did not have to wait for setting and the Final was quite shorter than Sudoku.

Round 8
We could choose our pairs but not the puzzle sets.
First half: me and Hideaki
Second half: Kota and Naoki
Instructions were not difficult for our part. Hideaki soon finished the Latin puzzle and helped me, but we just wasted time for it. I even heard proctors talking about my stuck.... I am still not sure about good way for it. Somehow I found the solution, only 5 minutes or so to check the final grid. Loop part had been finished then but I could not fully understand the required rule. And we had an error in the third puzzle. Just a disaster round.

Round 9
We simply had one paper each and filled from outside of the grid. I and Naoki were faster by 3-4 pieces and help the others. We were anxious about error and slow to avoid it, I am very curious about how the German team did in this round.

Round 17
I and Naoki mainly handled the combination part, not strictly logical way. Kota missed copying a digit and we needed some time to notice it, but otherwise smooth.

Round 21
I do not remember what I solved in this round. We tried to use tricks in its attached grid, like rotating partial rectangles of Statue Park, changing clues of Nonogram into Ternary form (no solution), or even considering Country Road as 25x25 grid using vertices and midpoints. At some point I noticed that Cave could be rotated, and there was rotation trick in other place. Afterwards most of my work was matching, I did not solve any single grid by myself. The trick for Scrabble was obvious but we never considered using previous solutions because no other grid did that (by the way there was another spot for "NARF" in the first stage?), and I assumed changing grid size was for Observation Puzzle. I am so sorry for having asked so much work to Hideaki but in vain. That means we also missed Country Road and Nonogram. In the end of the round three of us (except Naoki in Country Road) cooperated in Word Search.

If I remember the Q&A session correctly, the round was quite harder than expected? With the round extension, USA should have more bonus points for 15 minutes, huge difference. But I think the round was well considered to be cultural neutral, at least better than word types. Just we lacked the ability....

Round 22
First we made the combination with 800, and optimized in the next step. It had been a while since my last folding/origami, but my hand remembered. Time setting was not severe here.

Mini Playoff
Due to my condition and some uncommon round, I did not aim for each playoff, this was just a result. Easel style has never been appeared officially since 2013, which I watched as an audience, so my only have some exhibition experience in Japan. This was unofficial thing, and I did not have to avoid errors too seriously here. Maybe I could do more roughly, though. First TomTom was not easy to start, I feared mistake with pen and used small numbers. This genre was almost decided here. In the next genre first Star Battle was again decisive. For Shading, I considered different notation on easel, but finally used my normal way. Here I made the only error during this WPC, lost another chance. The cause was more like easel style, maybe I felt some fatigue though. For Loop/Path, I wrongly left an end of the Slitherlink loop in the bottom half and had to correct it with some cross marks, and I also made a wrong deduction in Cross Border Parity Loop and took a while to notice it. And in the last round, I assumed global argument was required in Araf and wasted time, my bad tendency. It did not go well as a whole, but one good point was that I could solve all 15 puzzles, meaning less trouble.


Still much to improve physically and technically. My new year would start from ASC, maybe 24HPC, and WSPC would be in near place? And I have to prepare for next GP season too.

WSPC 2022


Assuming that one team round, playoff videos, detailed results and GP puzzles will not be released, I post this draft before mentioning the latest WSPC. Probably most of my memories are getting obscure.

----

Here is my playback of the great week, which I waited for three years. This kind of article happens only after I solve all the contest puzzles (including team ones), so it is a bit late this year.


WSC
Well, the championship which I hold for a while somehow. Now I know the results of Tawan's crystal ball contest and that many people expected my performance (more than I thought), and at least I tried doing my best.

Round 1
Left: #8, 9
Error: #1
All same points, so just in order. Managed to finish #7 in time but 2-cell switch in #1.

Round 2
Solved: #10
Then disaster began. I spent half of the round for #10, then went to #5 but broke it in the final phase. I could not finish tweaking in time.

Round 3
Solved: #2, 6, 7, 8, 9, 10
From high pointers. Much more smooth than the previous round. Two puzzles were marked as wrong at first due to bad writing.

Round 4
Solved: #2, 3
Bad work on high pointers this time. Normally I solve these to hide slow classic speed, and my performance relied too much on them this year. I had some time to try #5 but only filled 1/3.

Round 5
1 minute left.
Finishing this round or not would make big difference, so happy without any significant trouble.

Round 6
Solved: #4, 5, 7
Error: #9
5-cell error in #9 was critical. I tried to get #1 or #2 in the last minutes but got stuck soon in both.

Round 7
Solved: #1, 3, 6
Error: #2
Major error this time. I filled 1/3 of #4 but it could not be in time, so checked #6 and rushed. It was as expected, to make unique puzzle with this rule.

Round 8
I do not have our papers so not sure which ones we solved, but probably all Arrows and two Diagonals? We messed up with some letters, otherwise all Diagonals and some more could have been completed.

Round 9
Decisive round for team result. We should have guessed more?

Sudoku GP Final
My first appearance. No puzzle selecting, just solving unlike Puzzle GP. I broke my first puzzle, Thermometer, resulting in no solution. I even messed up next Non-Consecutive but managed to find the pattern. Afterwards it went much better, but I lost to Timothy twice. I even feared no-winning beforehand, so good result enough.

Round 10
Solved: #3, 4, 6, 7
#3 was marked as incorrect due to a thin writing. Barely finishing all high pointers, no time to check the others.

Round 11
Solved: #1, 5, 7, 9
My initial plan was to solve #8 but I had to give it up without any inner progress (due to wrong deduction?). And the round ended midway through #6.

Round 12
Solved: #1, 2, 9-12
Third set first, then first set half. Confused a bit for copying.

Round 13
So I was third in our team, relatively easy position in this round but expected to win all. I did not note our opponent team in the first round, probably we met Germany but I cannot recall others, sorry. But opponent did not matter here, time was the real opponent. The round concept is good in that many people can experience playoff-like things, but I wonder how they felt with this difficulty.

We won all matches in the first round, but second Antiknight was 1-0 (only me) and fourth Windoku was also 1-0 (Kota, this was the only one I missed). Thus second seed next to Czech.

I had a note for the second round, it says we met France, USA, UN with Vincent and UN of Tantan and Thais. Semifinal was 0-0 draw (again only one I missed) but we managed to win the tournament. Good reward.

Playoff
Believing my note, the selection was like this:
Fourth round
Tiit: Sudoku Fractions for first
Kota: Arrows for second
Tantan: Liar Frame Sum for fourth
Tiit: Diagonal (second) for third
Kota: Killer for fifth

Third round
Jakub: Outside for second
Jan: Irregular for third
Hideaki: Consecutive Sequence for first
Jakub: Kropki for fourth

Second round
Jakub: Sequence Top-Bot for first
Ken: Skyscrapers for second
Thomas: Antidiagonal for third

First round
Nikola: Fortress for first
Jan: Antiknight for second
Sinchai: Integral for third

I chose Skyscrapers as a puzzler, of course Thomas was there but I thought this was the best choice. Actually I did not do it well but I earned time in the first Sequence Top-Bot and kept the lead.

My next round, after waiting for 200 seconds, I had to start checking my instructions; it was Sequence in fact! The lines pass the border so it was quite strange, but I could not believe in myself because I had not checked this type in previous round. Jakub somehow solved it and the other puzzles, and Jan happened to see the next Outside, so keeping Jakub's time for Irregular and Kropki, reselection happened:
Irregular for first, Kropki for second (original choice)
Jakub: Classic (first) for fourth
Hideaki: Squares in squares for third
This Hideaki's choice really benefitted me and himself. I could start the final Classic but time disadvantage affected here, so my defending game ended.

I did not felt so much fatigue, as if I could not use my energy well. This was somewhat true, I could not adapt to the difficulty till the end. I need more skills to choose different strategy.


WPC
And my main target to stop the gap of five years....

Round 1
#1 to #3 and then went back from the last. Writing all numbers in Nanro is not my usual, so I wanted to do it early. After solving all puzzles, I checked 5 minutes or so, because irregular grids have more possibility of errors. But except Ripple Effect; it takes much time so I believed in my solving. Then I submitted with 28 minutes left.... And error in this! 4-cell error and no partial bonus, so my possible 730 points became exactly half. I heard this in lunch break, worst feeling in the week. But as a result, this was the only error in this WPC.


Round 2
8 minutes left.
Just went backwards.

Round 3
13 minutes left.
I cannot recall my order, but it was genre by genre, based on my preference and total points. I made some guesses, especially on loop types. And with many puzzles, I spent several minutes for checking before submission.

Round 4
15 minutes left.
Solved in the given order. I had some trouble in latter half but not critical.

Round 5
Left: Spiral Snake
Then more trouble occured. Nothing was smooth, the worst thing was Easy as Snake, which I broke in the last steps and had to fix by tweaking. I solved this round from the end but skipping Spiral Snake with insufficient remaining time, so I had to get this 100 points even ignoring 30+50. I managed to do that, and hurried for 30 points. I lost this round but this could have been much worse.

Round 6
Left: Slitherlink (Crossing + Toroidal + Liar)
First I solved two 115s and then checked 135 points for several minutes, but no progress. It seems I persisted too much on global argument. Giving up and tried the others, but I had to restart Suguru and lost time, so I could not return to Slitherlink. But strategically better choice at least.

Round 7
2 minutes left.
Same feeling as WSC, we should have guessed more? I was almost working on construction, and we were too much logical I think. We lost so much points here.

Round 8
10 minutes left.
Due to the points distribution, solved in the given order. First one was as expected, but I know the use of 0, I was careful about it. I felt my solving was a bit rough, and my error in the first day occurred in the morning, so I spent several minutes for checking before submission.

Round 9
27 minutes left.
Solved in the order. I wanted to be logical with this rule, but I hit an assumption for the last one.

Round 10
Left: Hidato (Knight) (second)
Crazy round as expected. I love the concept, but to be honest I do not prefer solving this in official competition; it often requires re-reading of the instructions, it is difficult to translate precisely in advance. Actually I heard that Hideaki had some confusion.
Due to the risk in this round, my solving order was complicated. I wanted go get all 70+ points, but at the same time I did not want to take too much risk, so I went from front page but skipping some uneasy puzzles like Hidato and Overlapping Square, and cheap ones like Meandering Numbers and Country Road. I was threatened by the gimmick many times but managed reach the last one, and returned to the skipped ones. Hidato twins were my last, with good remaining time, but I was too slow for the first one, I had to guess the second one in the last minutes but with no success.

Round 11
14 minutes left.
Probably did in the order, but maybe skipped cheap ones. I could not find logical way for Fuzuli during competition, thankfully I had plenty of time and decided to use brute force.

Round 12
6 minutes left.
If I remember correctly, went forward starting from Slitherlink, and returned to the skipped ones. I thought the second Slitherlink had two solutions but it was just wrong. Not stable work.

Round 13
1 minute left.
First gathered 110s, then 130, and the remaining ones. Kropki was my final puzzle, here I had to hurry for unexpected finish.

Round 14
6 minutes left.
I can see erasing in several puzzles, so I think it was not smooth for all of us.

Puzzle GP Final
I wanted to avoid lottery so I skipped Star Battle first, then used Japanese Sums considering my opponent. Next skip was Parking Lot, which I though also had some risk of lucky guess. Then Double Choco. Of course this also has some risk, but here I managed to left four stable genres, and as Japanese I might had some advantage to Thomas. It did not go as I wanted, but one lose yet.
Here a trouble happened, next puzzle should have been chosen by Kota, but Wei-Hwa mistakenly asked to Thomas. This was the earliest lose of top seed in this format, so possible confusion. Kota also wanted to choose Fillomino (and Thomas also for Yajilin in next round), so no difference luckily.
Fillomino and Yajilin were okay, but true problem was in classic Slitherlink; I broke it and could not fix it. This was out of plan, much worse than Double Choco. I was quite good in final Slitherlink (corners) (though missed a segment thus one error), so my largest regret was in losers final.

Round 15
I have a note for this side:
Pot 4: Greece
Pot 6: bye
Pot 2: UK
Pot 3: Bulgaria
Pot 5: Finland
All 4-0, so first seed. Tournament opponents were France, Czech, Netherlands, USA.
I remember Kota lost in Skyscrapers and Yuki lost in another match, but there were all. I faced many good players but did well enough. Winning both tournaments was nice.

Playoff
Round 1: Set 4
Round 2: Set 5
Round 3: Set 1
Round 4: Set 2
I was waiting in the hall, not seriously watching but checking the solving time to understand the difficulty. Set 2 and 3 were left for the final, and I just avoided the word type. At least one other finalist chose the same, but I know it was not Yuki. And there is no mention about time advantage in the instruction; it was based on the ratio of preliminary points but multiplied by 0.8. So my advantage to Thomas was (2400 seconds)*(1500/7560 points advantage)*0.8=(381 seconds). But I knew even 7 minutes had not been enough.
Before to start the final, I asked if the last puzzle type would be revealed beforehand. I did not want to re-read the instruction again. When we knew that... oh. I had bad experience in the previous round.
Smooth solve in Chained Block. Statue Park too. Context was not, I just hit an assumption. Skyscrapers was okay again. The problem was in Hidato. Due to the irregular gird, I missed some easy steps, and it was still hard even with that. Finally I hit three-degree assumption or so.

So, with relatively easy rounds, accuracy was more important this year. Large error in beginning but only that, I think I could fit the situation. And finally saved my position in the playoff. I was not solid at some point, but I believe there is some improvement. Still going.