Kwontom Outcast 
tobiwan KwonTom Obsessive Puzzles: 3778 Best Total: 17m 44s  Posted  2006.10.26 15:58:13 Does anyone else find that the majority of people don't get kwontomloop/slitherlink? I've tried to explain it to a lot of people but they look at me blankly.
The only other person that's into it is my sister. Which leads to the unfortunate situation that whenever we get together we tend to end up talking about kwontomloop and having the piss taken by our spouses. 
Acorn KwonTom Obsessive Puzzles: 555 Best Total: 42m 58s  Posted  2006.10.26 17:10:19 I find that most people just don't get puzzles. When people come over and see the printed out puzzles I'm doing (several different kinds). They think I'm crazy. 
caelie KwonTom Obsessive Puzzles: 2504 Best Total: 24m 2s  Posted  2006.10.26 17:55:53 Maybe we're not the outcasts, maybe they are. I'm tobiwan's sister by the way. Our spouses have been laughing at us lots today for talking about kwontomloop... But I think they're just jealous! Lol. 
matth314 KwonTom Obsessive Puzzles: 2431 Best Total: 16m 52s  Posted  2006.10.26 19:08:15 No one ever seems to understand why I like it. I try to explain that it's based on the same ideas as Sudoku (which they of course love), and it's even more fun. Perhaps it just goes right over their heads. 
he_he KwonTom Obsessive Puzzles: 2001 Best Total: 14m 41s  Posted  2006.10.26 19:24:16 Our spouses have been laughing at us lots today for talking about kwontomloop...
Yeah, me too  my husband thinks I'm nuts, and my parents just don't get it. However, people at work have really gotten the bug since I introduced it to them, and I cannot now get on the computer to do the puzzles cos they're hogging it!! 
tobiwan KwonTom Obsessive Puzzles: 3778 Best Total: 17m 44s  Posted  2006.10.26 21:07:23 It's interesting that Sudoku is so widespread (particularly in England at the moment) when Kwontomloop has much more depth and longevity (I expect to be learning new patterns etc for a while).
I guess maybe that's part of the problem. It takes a little while to pick up compared to Sudoku. 
dee KwonTom Obsessive Puzzles: 1034 Best Total: 22m 21s  Posted  2006.10.26 23:29:45 I definitely think you're onto something there, tobiwan. Sudoku is comparatively easy for a lowlevel puzzler to pick up and run with, at least at the easy level. Whereas I count myself a seasoned puzzler  long history of logic puzzles, sudoku, and cryptic crosswords (and I climb, which is puzzlesolving with exercise!)  and first time I encountered the link puzzles it threw me for a (hah) loop. That extra challenge was part of what got me hooked, but I can see it would be daunting for someone whose brain wasn't wired like that. 
itchfizzix KwonTom Obsessive Puzzles: 3030 Best Total: 17m 57s  Posted  2006.10.27 00:17:59 The average person seems to think I'm nuts too, but for those who love logictype puzzles, once I explain the general directions for Kwontom, they pick it up and start to enjoy it themselves. I just wish I could find more puzzles  they have the Sudoku and Kakuro books at Barnes&Noble and other bookstores... what about Slither Link/Kwontom Loop??? 
Jankonyex KwonTom Obsessive Puzzles: 3729 Best Total: 9m 35s  Posted  2006.10.27 14:41:15 Sudoku is the best. for kwontomloop, nonreasonable rejection is the "best" solving technique for high level puzzles, other techniques like counting and setting inequalities are not popular (for 1x1 fig.). but for sudoku, you need to "think" when solving a very high level puzzles, and you may have some brilliant ideas and discover some artful solving techniqeus that "not every person can think of". solving kwontomloop need practices where solving sudoku need intelligence. 
m2e KwonTom Obsessive Puzzles: 607 Best Total: 16m 43s  Posted  2006.10.28 02:05:53 I disagree. Kwontomloop is geared towards speed, but not slither link in general. Try the japanese site's hard puzzles as well as the harder monthly beasts here (forgotten which month were the harder ones...) and they equire more intelligence. 
chairman KwonTom Obsessive Puzzles: 1395 Best Total: 17m 32s  Posted  2006.10.28 08:08:03 I have done sudoku puzzles for a while, but it could not keep my attention that long. I did not experience that sudoku's need a more variety of solution techniques that kwontom loops, on the contrary. Of course, solving kwontom loops quickly requires a lot of exercise, but hey, this is a competitive site. Don't tell me that you are not competitive. When solving monthly beasts, I take my time and don't add crosses until I'm really stuck. This is in nice counterbalance with the daily rush.
Quote: Originally Posted by jankonyex ... for sudoku, you need to "think" when solving a very high level puzzles, and you may have some brilliant ideas and discover some artful solving techniqeus that "not every person can think of".

Can you give an example of a sudoku that requires art or brilliancy? Is this a high level sudoku?
_5_ 18_ __6 ___ 9__ ___ 1__ __2 _78
67_ ___ ___ __2 ___ 8__ ___ ___ _53
83_ 2__ __9 ___ __7 ___ 7__ _63 _1_ 
AmyR KwonTom Obsessive Puzzles: 808 Best Total: 30m 18s  Posted  2006.11.13 04:52:28 I agree... at work people don't really get the kwontom thing, though working in a math department gives people a little more respect for puzzles. I'm excepting of course the one other avid kwon tom fan at work... you know who you are! 
Gadget1903 KwonTom Addict Puzzles: 325 Best Total: 23m 39s  Posted  2006.11.13 16:45:23
Quote: Originally Posted by chairman Can you give an example of a sudoku that requires art or brilliancy? Is this a high level sudoku?
_5_ 18_ __6 ___ 9__ ___ 1__ __2 _78
67_ ___ ___ __2 ___ 8__ ___ ___ _53
83_ 2__ __9 ___ __7 ___ 7__ _63 _1_ 
My sudoku program indicates that sudoku as Intermediate, it does not require any guessing or overly complex or artful rules...the only complex rules are: a hidden pair, two naked triples and 3 pointing/claiming patterns.
Before kwontomloop, sudoku was my favorite  partly because it inspired my first ever windows c++ program I created from scratch. My program evolved from a few very simple logic rules and regression guessing to many more complex rules (some of these rules maybe are the 'artful' ones jankonyex mentioned). There are many different logic rules there are that do not require any guessing or 'chain logic'. If I compared Sudoku's logic rules and the many 'logical patterns' in kwontom loops, there are many more patterns in kwontom loops.
This sudoku my program created as expert with no guessing  it can be solved completely with logic (a.k.a. forced patterns) but it contains several more obscure rules...
** ..2.....3 ..8.6.... .1.....5. ++ 5..3...1. .4..25... ....1.79. ++ 4....8... ..76.9..8 ...5..2.. **
This is an example of a Diabolical my program created, if you know the 'chain' logic rules, you might be able to use only only logic...I consider chain logic as an application of controlled guessing, or regression, so I never programmed it.
** .46...... 21....... ..97.4... ++ ......... ..762.8.. ..5...6.3 ++ .....5.2. .93..64.8 ...9...1. **
I think many puzzles have 'artful' or 'brilliant' ways to solve them (quickly), but if you break the solutions down into their smallest parts, it's just a matter of knowing many rules and/or methods of solving.
I think that KT loops require more patterns than Sudoku's, but when completing a more complex puzzle, detecting the KT patterns are easier than Sudoku.
If I consider that at KT faster is better, my solving strategies are geared towards speed. I have never really tried to solve Sudoku's for speed, so maybe there are speed methods I have never considered. 
chairman KwonTom Obsessive Puzzles: 1395 Best Total: 17m 32s  Posted  2006.11.14 20:01:13 Thank you for your reaction. Last night I spent a couple of hours on mine and your examples. I could not solve one of them without guessing! My program, written in Matlab, is just a bookkeeper in which the rules are implemented, no patterns.
Quote: Originally Posted by gadget1903 ...the only complex rules are: a hidden pair, two naked triples and 3 pointing/claiming patterns. 
I don't know what you mean by these and I am trying to figure it out. Does the hidden pair concern rows 2 and 4, columns 4 and 6? Is there a naked triple in row 8?
Your diabolic puzzle is definitely much more difficult than the sudoku's I've seen so far. It resembled me to a real difficult kwontom loop of Jankonyex. Before having seen this one, I thought that 9x9 kwontom loops could not be that difficult.
Apparently I am not the one to judge which of the two puzzle types is preferable. All I can say that for me, kwontom loop is more addictive!
Last edited by chairman  2006.11.14 20:01:59 
gadget1903 KwonTom Addict Puzzles: 325 Best Total: 23m 39s  Posted  2006.11.14 20:59:37
Quote: Originally Posted by chairman Quote: Originally Posted by gadget1903 ...the only complex rules are: a hidden pair, two naked triples and 3 pointing/claiming patterns. 
I don't know what you mean by these and I am trying to figure it out. Does the hidden pair concern rows 2 and 4, columns 4 and 6? Is there a naked triple in row 8? 
I found the Sudoku Eplainer tool the most descriptive sudoku solver out there, it will describe the rules I noted...and the ones in the puzzles I posted. You can find it at http://diuf.unifr.ch/people/juillera/Sudoku/Sudoku.html (note the url is case sensitive)
Quote: Originally Posted by chairman All I can say that for me, kwontom loop is more addictive! 
me too
Last edited by gadget1903  2006.11.14 22:37:48 
chairman KwonTom Obsessive Puzzles: 1395 Best Total: 17m 32s  Posted  2006.11.14 22:17:55 Thanks again! Now I know that hidden pairs, naked triples, pointing and claiming are patterns I figured out myself. I just did not spot them.
Last edited by chairman  2006.11.14 22:58:22 
tilps KwonTom Obsessive Puzzles: 4236 Best Total: 20m 22s  Posted  2006.11.15 08:51:25 My 'logical' solver  rates the first of your two puzzles as 1.1.2 second as 1.2.2 1.1.2  means if you consider two options (which are the only two options ) then the Immediate common consequences are adequate to solve the 'hardest' part of the puzzle. 1.2.2  means if you consider two options (which are again the only two options for a cell or number in row/column/box) then the secondary consequences under basic rules are needed to solve the 'hardest' part of the puzzle. Xwing patterns are an example of 1.2.2
My generator finds about 1 in 1million (minimal) puzzles can't be solved by considering the common 'basic rule' consequences of options. I am yet to find any which can't be solved by a single nesting  that is by considering common consequences of options, within options.
Edit: Just for fun I tested the 'Worlds Hardest Sudoku' which was announced recently. It can be solved using options within options. (Which my solver gives the wonderfully descriptive difficulty value of '2')
Last edited by tilps  2006.11.15 09:10:34 
Para KwonTom Obsessive Puzzles: 1911 Best Total: 19m 28s  Posted  2006.11.15 20:53:17
Quote: Originally Posted by gadget1903 This is an example of a Diabolical my program created, if you know the 'chain' logic rules, you might be able to use only only logic...I consider chain logic as an application of controlled guessing, or regression, so I never programmed it.
** .46...... 21....... ..97.4... ++ ......... ..762.8.. ..5...6.3 ++ .....5.2. .93..64.8 ...9...1. **

There is one tricky step in this puzzle but mostly just locked candidates, naked pairs, an XYwing and a few empty rectangles. Thereby the tricky step can be easily bypassed by using the unique rectangle technique.
Quote: Originally Posted by gadget1903 If I consider that at KT faster is better, my solving strategies are geared towards speed. I have never really tried to solve Sudoku's for speed, so maybe there are speed methods I have never considered.

There is a difference between solving a puzzle for speed and by logic. One of the most useful tools for speed is taking into account the puzzle has a unique solution. When trying for speed you can find some nice tricks in most logic puzzles out there that have to be true to make a puzzle unique. But it's more satisfying to prove a puzzle is unique by solving it logically.
Para 