2021 German Sudoku Qualifier (22.-25.01.21)

[Realshaggy]

Dear puzzle friends,

next weekend, the qualifier for our next German Sudoku Championships (DSM) happens on the Logic Masters Germany contest engine. It's open to international participants, they may not qualify for the finals though.

Here is the instruction booklet, accessible without account on the website: https://logic-masters.de/DSM/2021/raetsel.php?chlang=en

The contest is open from friday noon to monday noon (UTC+1). Solving time is 120 minutes, puzzles have been provided by Richard Stolk, Philipp Blume and Christoph Seeliger. Hope to see you there!

On behalf of the authors
Christoph Seeliger

[jupilogy]

I’m struggling with the irregular sudoku in the instruction booklet. Is there a solution?

Edit: just found them at the end, never mind

[Realshaggy]

The contest has finished, you can see the results here. Scroll down for international participants.

https://logic-masters.de/DSM/2021/q_tabelle.php


You can download the puzzles without needing an account here. I did not publish a solution booklet, please ask if you get totally stuck.

https://logic-masters.de/DSM/2021/raetsel.php


Regarding UK results, David McNeill got the top spot with 895 points, followed by Mark Goodliffe with 829 points, both very good results in a contest that turned out to be quite a bit harder than we estimated from testsolving times.
Sam helped me testsolving and finished the set in 1h39min, Tom finished too in 1h58, but of course it's always hard to compare results from a non-competitive environment. Sorry that I'm not familiar with names further down the list, I don't see the nationalities that were set for the accounts on our page. I also don't know, if potential UK solvers are part of the UKPA community or found the contest through other channels I used for promotion.

Thanks to everybody for taking part!

[david mcneill]

I thought Mark and I had both done very well. But kudos to Sam and Tom for speedy test-solving.

There were some great puzzles in the set. I particularly liked how the Average Arrow solved as I was dreading it beforehand. I struggled timewise with some of the standard variants - particularly Antiknight, Irregular and Thermo.

[detuned]

Test solving is definitely not the same as competition, particularly when there are hard puzzles with larger proportions of total points! In order to give more useful “like-for-like” solving times I will always take puzzles one at a time and reset in between solving. I’ll generally also restart the timer if I happen to make a mistake.

So all in all I’d have been very pleased to get an equivalent result to Mark it David on this one

[detuned]

You also don’t get the minimum overhead of 2 minutes + 30s per puzzle submitted accounting for printing and answer submission - that’s easily another 10-15 minutes on top of the raw testing times!

[Realshaggy]

Yes of course, testsolving times usually don't include mistakes, printing and entering answer codes (and maybe some checking too). But if you need 15 minutes for printing + answer codes, you probably shouldn't use the printer in the other wing of your castle
Maybe nobody got through flawlessly. I also got a 1h10 and a 1h15 from testsolvers, so I thought it is comparable to previous years.

[detuned]

Test solvers need to be taken with a pinch of salt I think the top solvers are getting better and better over time - or certainly more competitive (if only with themselves!). A couple of anecdotes to support this:

(1) the finishing times on the GP rounds for the top solver seem to be fairly consistent, however the tests certainly feel a lot harder than they used (and I think there are less people who are finishing all the puzzles within time)

(2) the total testing times I run for the UKSC seem to be steadily getting faster and faster each year. But I'd say, more or less, that the difficulty over the years has remained fairly stable

Edit: finally on the point about overhead time - I’m tempted to video myself have a go at a contest and timestamp the bits where I am actually solving. Even in an offline competition there will be a few seconds between finishing one puzzle and starting the next, and if there are enough puzzles that all adds up - the majority of that won’t be printing time by any means. I’ll stick by my 10-15 minutes (I guess I should pro rata to 75% for the 90 minute GP) and let you know the results!