Times Fiendish without candidates!

All about puzzles in newspapers, magazines, and books

Times Fiendish without candidates!

Postby CathyW » Fri Feb 03, 2006 4:06 pm

Had to share - I finally managed a Fiendish today without entering any candidates:D
CathyW
 
Posts: 316
Joined: 20 June 2005

Re: Times Fiendish without candidates!

Postby Cec » Sun Feb 05, 2006 1:58 pm

Well done. Curious to see how 'feindish' the puzzle is (oops! was!) so could you post the initial puzzle?.
Cec
Cec
 
Posts: 1039
Joined: 16 June 2005

Postby CathyW » Sun Feb 05, 2006 3:12 pm

It's from The Times of Friday, 3rd February - link is http://www.timesonline.co.uk/article/0,,18209-2021578,00.html or so you can copy and paste:
Code: Select all
... 4.. 681
... .2. ...
..4 7.. .3.
5.3 ... ...
18. ... .53
... ... 2.6
.7. ..8 5..
... .1. ...
648 ..9 ...


Usually the Friday puzzles are the hardest of the week. But perhaps this one isn't particularly "fiendish":!:
CathyW
 
Posts: 316
Joined: 20 June 2005

Postby afjt » Mon Feb 06, 2006 1:54 pm

Cathy - I too did it without candidates, but it only required singles all the way through (unless I subconsciously used locked candidates without realising it), so my gut feeling is that it was really a 'difficult' rather than a 'fiendish'.
afjt
 
Posts: 82
Joined: 07 September 2005

Postby PaulIQ164 » Mon Feb 06, 2006 4:59 pm

The fiendishes fairly often need only singles to solve. The difficulty-rating algorithm is more complex that just a bare assessment of the simplest tactic that can be used to solve it all. If it was rated fiendish then it will have been (barring an error made by The Times themselves) have been harder than a mere difficult, presuming you solve like a person and not like a robot.
PaulIQ164
 
Posts: 533
Joined: 16 July 2005

Postby afjt » Tue Feb 07, 2006 11:33 am

I always do easy/mild/difficult without pencilling in candidates, but with fiendish ones I always pencil in candidates as I go - starting off with ones where there are only 2 of a particular number in anynonet, especially where they are in the same row or column. This is because fiendish puzzles generally have locked candidates and the pencilled candidates makes if quicker (for me) to spot these. Otherwise, I find I am covering the same ground again and again looking for singles or eliminations.
Having said that, I do far fewer vanilla sudoku puzzles these days - I much prefer killer ones.
afjt
 
Posts: 82
Joined: 07 September 2005

Postby Heuresement » Tue Feb 07, 2006 2:56 pm

First of all, good for you Cathy:)

I do the Times puzzles every day, and have been successful in doing all the fiendishes from The Times since the beginning of December without pencilmarks of any type. Sometimes it has been necessary to struggle for a long period, or take a rest in order to spot where to go next. In nearly every case, my problem has been that I have missed a cell that could only have a single candidate. I must admit that the satisfaction rating for doing the Fiendishes in this way is very high for me, and is now as quick or quicker than using pencilmarks for me.

I am still trying to do the super-Fiendishes (Pappacom v.hard) in the Sunday Times without pencilmarks, but have not yet succeeded. Hopefully one weekend soon.

I know that I have mentioned this before, but it was shame that drove me to doing the Fiendishes with no pencilmarks, when someone once accussed me of solving these puzzles using the Vordermann method! (Sorry in advance to everyone with the same name!)

The other thing that helped me was that a fellow pupil in an evening class asked me to write down how I solve the Fiendish puzzles. She brought one in, and I wrote several pages on how I did it. I realised that in general, my solving techniques were too haphazard, and actually the Fiendishes only require relatively simple steps, and very persistent eyes.

Does anyone else have any tips?

Michael
Heuresement
 
Posts: 54
Joined: 19 August 2005

Postby Lindy » Tue Feb 07, 2006 7:16 pm

I generally try to get as far as I can without any pencil marks. When stuck, I jot down locked candidates in biro and see where I can go from there. If still stuck, I list possibilities in pencil so that the locked candidates stand out all bright and sparkly as a reminder to me. When all that's done, I smack my head a couple of times because inevitably singles jump out at me. I should change my name to "Doh!"

Michael's reference to writing down his solving techniques has got me thinking. Would anyone would have the time or inclination to post a walk-through of a fiendish solved without any pencil-marks? It could make an interesting thread and a learning curve for those stepping up to fiendish and higher puzzles.:)
Lindy
 
Posts: 42
Joined: 03 May 2005

Postby CathyW » Wed Feb 08, 2006 4:04 pm

Lindy wrote:Michael's reference to writing down his solving techniques has got me thinking. Would anyone would have the time or inclination to post a walk-through of a fiendish solved without any pencil-marks? It could make an interesting thread and a learning curve for those stepping up to fiendish and higher puzzles.:)


I thought that noting down one's solving process could actually be an aid to solving the puzzle i.e. something written down on another piece of paper rather than candidates in cells in the puzzle, though I'd like to think I would have remembered the locked candidates and pairs that I noted! Today's Times Fiendish (no 608) which I actually found harder than last Friday's is solved mostly with singles but I did have to note some pairs - even an x-wing though that is probably not essential. It would have been solved more quickly if I'd used some candidates but speed isn't always the objective! The starting puzzle is:
Code: Select all
92. ..5 1..
... .4. ...
1.. 6.. 7..
.6. .8. ...
3.2 ... 8.5
... .9. .4.
..7 ..3 ..8
... .5. ...
..1 7.. .52

These are the steps I took in order:
r2c7=5; 7 has to be in row 2 of box 1 => r1c5=7; 3 has to be in column 4 of box 5 => r3c5=3; r7c5=2; r9c5=6; r5c4=4;r6c2=1;r1c4=8; note pairs in box 8 => r2c6=1; r2c4=2; r3c6=9; r3c8=2; r3c9=4; r1c3=4; r2c8=8; r2c9=9; r8c1=2; r4c1=4; note x-wing 4s in boxes 7&8 => r7c7=4; note (5,8) pair in r3,6c3 (took me a while to spot this one without entering candidates!) => r4c3=9; the rest is all singles.
CathyW
 
Posts: 316
Joined: 20 June 2005

Postby Lindy » Thu Feb 09, 2006 4:01 pm

Many thanks for your steps, Cathy.

I first did the puzzle my usual way during lunch yesterday and managed to complete it with just a few jots of locked candidates. In the evening I followed your steps. They seemed a lot more compact to the way I ramble around the grid! I progressed without jotting anything down and had to think for a bit in a few places without that visual aid. It didn’t hurt too much! Inevitably I had missed a couple of singles on my earlier attempt. All in all though, very satisfying.

There was one thing that confused me – the x-wing. I’ve never used this technique before as I don’t understand it that well. Looking just at boxes 7 & 8 I could see why the x-wing would put a 4 in r7c7, if I didn’t have the 4 for box 9 as locked candidates in c7r7,8&9. It looked like there could be an x-wing with boxes 8 & 9 as well. Apologies if I sound completely daft!

Again, thank you.
Lindy
 
Posts: 42
Joined: 03 May 2005

Re: Times Fiendish without candidates!

Postby lunababy_moonchild » Thu Feb 09, 2006 11:12 pm

CathyW wrote:Had to share - I finally managed a Fiendish today without entering any candidates:D

I'm late, I know, but very well done indeed.:D

Luna
lunababy_moonchild
 
Posts: 659
Joined: 23 March 2005

Postby MCC » Fri Feb 10, 2006 3:56 pm

Having tried a few fiendish without candidates, which I found were like a hurdle race, first an easy bit then a hurdle then another easy bit, hurdle etc, before the final home run.
The frustration at each hurdle before seeing the next obvious step.

So, having a few of these I turned my hand to a very hard without candidates, the result can be seen here.
Did I do it? No, not yet.

MCC
MCC
 
Posts: 1275
Joined: 08 June 2005


Return to Published puzzles