From: Carl Worth Date: Thu, 22 Jan 2015 05:47:42 +0000 (-0800) Subject: Add blog post about two recently-published puzzles X-Git-Url: https://git.cworth.org/git?p=cworth.org;a=commitdiff_plain;h=ae163a272c13862988caacd78e947160169ac628 Add blog post about two recently-published puzzles My 12-Pentomino-themed Star Battle published today as well as my F is for Fiendish Pentominous published a couple of weeks ago. --- diff --git a/src/puzzles/star_battle_pentominoes.mdwn b/src/puzzles/star_battle_pentominoes.mdwn new file mode 100644 index 0000000..39b11db --- /dev/null +++ b/src/puzzles/star_battle_pentominoes.mdwn @@ -0,0 +1,86 @@ +[[!meta title="Star Battle: Pentominoes, and Pentominous: F is for Fiendish"]] + +[[!tag puzzles]] + +I just had [my fifth +puzzle](http://www.gmpuzzles.com/blog/2015/01/star-battle-carl-worth/) +published at GM Puzzles. + +This was a fun one. It's my second Star Battle puzzle, and it's +definitely more approachable than my previous dual-grid [Star +Duel](star_duel/). (Incidentally, that giant Star Duel was noted as +one among several of the [best object-placement puzzles of +2014](http://www.gmpuzzles.com/blog/2014/12/best-2014-object-placement/) +at GM Puzzles. + +I got the idea for today's Star Battle when reading that best-of +post. One of the other puzzles noted there was this lovely +[9-pentomino Star +Battle](http://www.gmpuzzles.com/blog/2014/09/star-battle-zoltan-horvath/) +by Zoltán Horváth. The post mentioned that another designer, Jiří +Hrdina, had independently designed a similar 9-pentomino Star +Battle. (I can't link directly to that one since it's not freely +available on the web. It's contained in [The Art of Puzzles: Star +Battle](http://www.gmpuzzles.com/store/the-art-of-puzzles) e-book +available for sale). Then, in the comments, Matúš Demiger mentioned +that he had also independently constructed a [third 9-pentomino Star +Battle](https://www.dropbox.com/sh/ph4ntyc2r761r4c/AABpHTRWeX3eeqh82pRltc1va/03_24h_2014%20DEMIGER%20-%20PB.pdf?dl=0) +as part of the [2014 24-Hour Puzzle +Competition](https://www.dropbox.com/sh/ph4ntyc2r761r4c/iFyCTXGE7_). + +So at least three puzzle authors all happened to construct +pentomino-themed Star Battle puzzles last year, and all three happened +to choose the fairly-standard 10x10 grid size, (forcing the puzzle to +include only 9 of the 12 possible pentomino shapes). Matúš's comment +was "I hope someone will try to include all twelve pentominoes" and I +couldn't resist the challenge. + +And it was an interesting challenge since putting 12 pentominoes into +a Star Battle requires a 13x13 grid. That's not too much of a problem +in and of itself, (my Star Duel used a 15x15 grid, for example). But +with this particular theme, as the puzzle grows the total area of the +pentominoes grows linearly, while the total puzzle area grows +quadratically. In my final puzzle there are 12 regions of size 5 and +then one giant outer region with 109 cells. But all 13 regions each +only contain two stars. So the real challenge here was to ensure that +when solving the puzzle the stars in the huge region didn't get +determined early, (causing a bunch of cells to be wasted and forcing +the user to tediously mark off all of the unused cells). + +Luckily, I think it just worked out. In all of my test-solving, the +stars in the large region are among the very last determined. + +Anyway, give this puzzle a try if you'd like. The large number of tiny +regions means there are a lot of easy steps early on in the +puzzle. But there are still a few more interesting deductions in store +later on, (but nothing ever all that difficult in this Wednesday-level +puzzle). + +I should also thank Thomas Snyder for his editorial help. He found and +fixed a small ambiguity in the first version of this puzzle that I +submitted. I've since coded up a deductive Star Battle solver just to +be able to verify uniqueness for puzzles I construct. But maybe I'll +talk about that in a future post. + +PS. If anyone is following closely, I neglected to mention [my fourth +puzzle](http://www.gmpuzzles.com/blog/2014/12/pentominous-carl-worth-2/) +when it was published a few weeks ago on a Friday. It's a Pentominous +puzzle with almost no clues other than 12 F's, and I named it "F is +for Fiendish". The title is a warning, and I think it deserves it. I +think this is the hardest puzzle I've published so far. My Star Duel +earned a longer estimated "Expert" time, (37 minutes compared to 20 +minutes for "F is for Fiendish"), but that's mostly because Star Duel +is so much bigger, (2 15x15 grids compared to a single 10x10 +grid). The deductions required here are definitely harder to find. + +My sister is really kind to do some of the initial testing of several +of my puzzles. After I handed her a copy of "F is for Fiendish" one +evening, she called me later that night to ask, "Can you email me a +fresh copy of that puzzle? My husband and I have been trying it over +and over and the paper is all disintegrating after so much erasing." That's a beautiful thing for a puzzle designer to hear---that someone is terribly frustrated with a puzzle, but still determined to stick with it and keep trying. + +So if you want a challenge, give ["F is for +Fiendish"](http://www.gmpuzzles.com/blog/2014/12/pentominous-carl-worth-2/) +a try. There are logical steps that can be found at every point to +solve the puzzle without needing any guessing or back-tracking, (but +they may not be easy to find). Good luck, and happy puzzling!