Friday 19 August 2011

Friday Puzzles #118

So I’m very glad to be able to say that once more I’m a contributor to Gareth Moore’s Sudoku Xtra magazine! If you aren;t aware of this little gem, then permit me a moment to tell you all about it! The Sudoku part of the title should tell you that a lot of the puzzles are indeed Sudoku related, but there are plenty of Sudoku variants included to keep every enthusiast’s appetite satisfied. A personal favourite of mine is odd-pair Sudoku, which I believe is Gareth’s own invention.

However, those who are underwhelmed by the whole Sudoku craze should still pay attention as the Xtra part of the title holds equal prominence. There are several nikoli-style puzzles in each issue, along with some other WPC favourites. And not to forget arguably the most interesting bit, the community section. (Of course I can’t claim to be totally impartial here!) This features a really rich and diverse collection of logic puzzles sourced from professionals, amateurs and experts alike.

OK, that’s quite enough plugging for now! This week’s puzzle is another Suraromu, inspired the puzzle that appeared in the latest issue of Sudoku Xtra, which was about as much fun I’ve had with any logic puzzle I’ve done in a while. I’m not sure this one rises to those standards, but It’s certainly non-trivial. Actually, Suraromu as a puzzle type intrigues me somewhat because much like Numberlink, as the difficulty of a given puzzle rises, so does that puzzle’s susceptibility to uniqueness strategies. This is certainly true of this puzzle, but unlike with numberlink I think I’d suggest to the puzzler wanting most satisfaction out of the puzzle to resist temptation this time around. And I think I’ve babbled on for long enough now, so all that remains to be said is, Enjoy!
    #148 Slalom – rated medium
 P.S. I should remark that I’m not entirely happy with the presentation of this puzzle. There are some squares which share gates, and I thought that during the test solving this might be a little misleading, even if it is at least implicit that every numbered gate has both of its “posts” labelled. My solution previously had always been to employ a convention that there should be no squares sharing gates, but I suppose I got a bit lazy this week. I’d be interested to know people’s thoughts on this one…

All puzzles © Tom Collyer 2009-12.

5 comments:

  1. The shared posts certainly didn’t detract from the enjoyment of the puzzle. I might have to include a book of one of these on my next Nikoli order after you’ve now got me into them!

    ReplyDelete
  2. Nice puzzle. I find this puzzle type pretty easy to deal with, which is quite fun since I am totally ridiculous at numberlinks.

    Regarding the presentation, I must say that I prefer the one used by Palmer Mebane. Much less confusing.

    ReplyDelete
  3. Agree that Palmer’s notation is very functional and works well, but doesn’t look quite right to me from an aesthetic point of view. I was vaguely toying with the idea of splitting the cells down the diagonal, a la kakuro clues, but didn’t want to do the extra fiddling around in inkscape. I’m not totally convinced that’ll look 100% right either, but I might give it a go for the next puzzle.

    I must say, I’m generally a fan of loop puzzles in general, but suraromu is fast becoming one of my favourites. They aren’t so much work to produce relatively interesting puzzles either, which is a bit of a bonus!

    ReplyDelete
  4. So I liked that puzzle, and smiled to myself as I realised that what I was trying to ignore as something forced by uniqueness was forced anyway. I must confess to not being the hugest of numberlink fans (although the whole thing with trying to “prove” uniqueness does still intrigue me), but I sort of see Suraromu as being basically consecutive numberlink. I like that having blacked out cells provides an extra constraint. I like that the constraints that the gates provide, and I find it interesting you view these as being quite a weak constraint.

    I suppose I’ve taken quite a meta-approach to construction with the puzzles I’ve constructed recently, in that I will vaguely sketch out solutions and tweak them, adding in the gates as I go along, making sure to erase any possibility of patterns that will give me uniqueness issues. From my point of view, the gates suddenly become quite strongly constraining, particularly because they force parallel strands to diverge.

    I think there are plenty of possibilities for novel tricks here (certainly more than there are with numberlink) – and Gareth’s example I was originally talking about is a wonderful case in point. Although that brings me back to my other original point – the susceptibility to uniqueness. I’m sure you can do exactly the same thing to Gareth’s puzzle. I guess what I don’t understand is why you’d single out numberlink as being any different from Suraromu here. From that particular point of view, all Suraromu does is make things easier by providing additional constraints to fiddle out a solution.

    I’ll be keeping an eye out for your variation. I’ll also try to cook up a puzzle that can’t be so naively approached…I have a few ideas already…

    ReplyDelete
  5. I sometimes start with the circle,a few gates around it and build the puzzle around that setup.The point that you raised here also worries me.I have created over 25 suraromus of varying difficulties,both in its classic form and a few variations,and also a dozen numberlinks.Only a few of those are not solvable using that trick.The inherent vulnerability to uniqueness makes this puzzle quite different from others,even slightly from numberlink,as I felt that uniqueness can be exploited much more during design than solving with numberlink,unlike suraromu where it helps equally in both solves and constructions(as ours brun's comment about numberlinks also indicates the same).Numberlinks have a slightly larger solution space to look at,i think, and sometimes do not provide those instant starts that suraromus often do.

    ReplyDelete