Ratings revisited

All about puzzles in newspapers, magazines, and books

Ratings revisited

Postby Bernard Stay » Fri May 20, 2005 3:01 pm

The 'fiendish' Times puzzles yesterday and today illustrate the problem of rating Su Doku. Unless it was just me, yesterdays puzzle (148) was markedly more severe than today's (149). I polished off today's in 10 mins. in the doctor's waiting room; yesterday's took me more than an hour, in three sessions.
Bernard Stay
 
Posts: 94
Joined: 22 March 2005

Postby lunababy_moonchild » Fri May 20, 2005 6:18 pm

Wayne Gould has already said, on two occasions that I can remember reading, on this site that in any one category you will find different grades of puzzle. To use his original example, in the Fiendish category there will be puzzles that are more Fiendish than others, and puzzles that will be less so.

I came to that conclusion myself a while ago and am pleased to have it verified. I also personally think it's necessary, otherwise I'd lose interest altogether. But maybe that's just me, I seem to be between categories at the moment. That is, Medium is too easy and Fiendish just that little bit too hard. Then again, I have yet to master the more advanced solving techniques (pairings (what is that?), x-wings, swordfish etc) so I'm pretty much unsophisticated. I also solve on paper which may be holding me back, somewhat - or maybe I'm just backward!:)

Anyway, there are different degrees of difficulty within each grade. Wayne says so, it must be true - for the puzzles he sets anyway.

Luna
lunababy_moonchild
 
Posts: 659
Joined: 23 March 2005

Postby shakers » Sat May 21, 2005 9:06 am

lunababy_moonchild wrote:Then again, I have yet to master the more advanced solving techniques (pairings (what is that?), x-wings, swordfish etc) so I'm pretty much unsophisticated.


I drank too much scotch last night to try and give a lucid explanation of what I think you mean by pairings, so I'll leave that to someone else.

However, you only need to use the X-Wing technique in some Very Hard puzzles (i.e. they'll not appear in the Times), and as far as I understand it, Wayne's program will not generate puzzles needing you to use swordfish or any of the other esoteric methods discussed from time to time - so you can relax and, like me, enjoy being in ignorance of swordfishes!:)
shakers
 
Posts: 93
Joined: 10 March 2005

Postby lunababy_moonchild » Sat May 21, 2005 11:34 am

Shakers :"I drank too much scotch last night to try and give a lucid explanation of what I think you mean by pairings, so I'll leave that to someone else. "

Ha, ha:D . I have no idea what pairings are, so can't tell you what I mean, but I have seen it mentioned so I'll wait and see or post on the Solving Techniques board. I was hoping that it was simpler than x-wings, sword-fish, Milo's Rule, and Jacko's rule (if I remember correctly), of which I'm blissfully ignorant and, since I'm not at that level can, as you say, "relax and, like me, enjoy". Which is the point really.

Thanks for that:D

Luna
lunababy_moonchild
 
Posts: 659
Joined: 23 March 2005

Re: Ratings revisited

Postby The Druid » Mon May 23, 2005 5:20 pm

Bernard Stay wrote:The 'fiendish' Times puzzles yesterday and today illustrate the problem of rating Su Doku. Unless it was just me, yesterdays puzzle (148) was markedly more severe than today's (149). I polished off today's in 10 mins. in the doctor's waiting room; yesterday's took me more than an hour, in three sessions.


I'm with you on this one.

I understand perfectly that, within each category, there will be some puzzles which are harder than others. That's inevitable.

But some allegedly-"Fiendish" puzzles are dead easy, while the occasional so-called-"Difficult" puzzle is very much harder than any "Fiendish" puzzle I have ever come across.

That illogical fact is what I find problematic, in an activity where "logic is all"...!!!
The Druid
 
Posts: 33
Joined: 22 April 2005

Pairings

Postby george-no1 » Sun May 29, 2005 7:25 pm

lunababy_moonchild: "I have no idea what pairings are, so can't tell you what I mean, but I have seen it mentioned so I'll wait and see or post on the Solving Techniques board"

I think pairings must refer to the technique where you have two cells in the same row, column or 3x3 grid, which only two particular numbers can go; for instance

l 13 13 4 l 5 789 378 l 139 2 6 l

In this row, because the 1 and 3 can only go in the first two cells, they can be removed from all other possible cells in that row, meaning the 7th number is a 9 etc. This works the same for columns and cells, but I'm sure you know that all and use it regularly anyway. However, there is a very strong possibilty that I am completely wrong and that it is just a coincidence. If so, please could somebody tell me!

George
george-no1
 
Posts: 150
Joined: 20 May 2005


Return to Published puzzles