View previous topic :: View next topic |
Author |
Message |
ZeroAssoluto
Joined: 05 Feb 2017 Posts: 942 Location: Rimini, Italy
|
Posted: Thu Apr 14, 2022 9:04 am Post subject: Apr 14 VH |
|
|
Hi everyone,
Code: |
+--------------+-------------+--------------+
| 4 67 3 | 27 1 26 | 5 8 9 |
| 678 678 9 | 47 456 456 | 1 2 3 |
| 5 1 2 | 8 9 3 | 6 4 7 |
+--------------+-------------+--------------+
| 39 5 48 | 34 2 1 | 489 7 6 |
| 237 237 148 | 6 34 9 | 48 15 125 |
| 269 269 14 | 5 8 7 | 49 3 12 |
+--------------+-------------+--------------+
| 1238 238 5 | 9 36 268 | 7 16 4 |
| 1389 4 7 | 13 356 568 | 2 1569 15 |
| 129 29 6 | 124 7 245 | 3 159 8 |
+--------------+-------------+--------------+
|
Play this puzzle online at the Daily Sudoku site
Quote: | XY-Wing 1,3,6 in r7c58,r8c4 and -1 in r8c89
or
XY-Chain r5c8-1-r7c8-6-r7c5-3-r8c4-1-r8c9 and -5 in r5c9,r89c8 |
Ciao Gianni |
|
Back to top |
|
|
TomC
Joined: 30 Oct 2020 Posts: 358 Location: Wales
|
Posted: Thu Apr 14, 2022 9:55 am Post subject: |
|
|
Code: |
+--------------+--------------+--------------+
| 4 67 3 | 27 1 26 | 5 8 9 |
| 678 678 9 | 47 456 456 | 1 2 3 |
| 5 1 2 | 8 9 3 | 6 4 7 |
+--------------+--------------+--------------+
| 39 5 48 | 34 2 1 | 489 7 6 |
| 237 237 148 | 6 34b 9 | 48 15a 125 |
| 269 269 14 | 5 8 7 | 49 3 12 |
+--------------+--------------+--------------+
| 1238 238 5 | 9 36c 268 | 7 16d 4 |
| 1389 4 7 | 13 356 568 | 2 1569 15 |
| 129 29 6 | 124 7 245 | 3 159 8 |
+--------------+--------------+--------------+
|
If a=1 then <48> pair in row 5 gives b=3, c=6 and d is also 1
so a must be 5 |
|
Back to top |
|
|
Mogulmeister
Joined: 03 May 2007 Posts: 1151
|
Posted: Thu Apr 14, 2022 12:40 pm Post subject: |
|
|
In Eureka is:
(5=1)r5c8-(1=48)r5c37-(4=3)r5c5-(3=6)r7c5-(6=1)r7c8-(1=5)r5c8
R5c8 < > 1; stte |
|
Back to top |
|
|
TomC
Joined: 30 Oct 2020 Posts: 358 Location: Wales
|
Posted: Thu Apr 14, 2022 12:50 pm Post subject: |
|
|
Thanks for the notation Mogulmeister, I struggle with eureka, but if you describe something I know then this helps a lot!
Interestingly, you can put any value in r5c3 containing <148> because it still shows r5c8 cannot be <1> |
|
Back to top |
|
|
dongrave
Joined: 06 Mar 2014 Posts: 568
|
Posted: Thu Apr 14, 2022 12:52 pm Post subject: |
|
|
TomC wrote: | Code: | +--------------+--------------+--------------+
| 4 67 3 | 27 1 26 | 5 8 9 |
| 678 678 9 | 47 456 456 | 1 2 3 |
| 5 1 2 | 8 9 3 | 6 4 7 |
+--------------+--------------+--------------+
| 39 5 48 | 34 2 1 | 489 7 6 |
| 237 237 148 | 6 34b 9 | 48 15a 125 |
| 269 269 14 | 5 8 7 | 49 3 12 |
+--------------+--------------+--------------+
| 1238 238 5 | 9 36c 268 | 7 16d 4 |
| 1389 4 7 | 13 356 568 | 2 1569 15 |
| 129 29 6 | 124 7 245 | 3 159 8 |
+--------------+--------------+--------------+
|
If a=1 then <48> pair in row 5 gives b=3, c=6 and d is also 1
so a must be 5 |
Hi Tom, I just wanted to mention that the sudoku expert Clement taught me to put the letters in the front so that they stand out a little bit better. (See the example below.) When you put the letters at the end, they sort of blend in with the candidates. Code: |
+--------------+---------------+---------------+
| 4 67 3 | 27 1 26 | 5 8 9 |
| 678 678 9 | 47 456 456 | 1 2 3 |
| 5 1 2 | 8 9 3 | 6 4 7 |
+--------------+---------------+---------------+
| 39 5 48 | 34 2 1 | 489 7 6 |
| 237 237 148 | 6 b34 9 | 48 a15 125 |
| 269 269 14 | 5 8 7 | 49 3 12 |
+--------------+---------------+---------------+
| 1238 238 5 | 9 c36 268 | 7 d16 4 |
| 1389 4 7 | 13 356 568 | 2 1569 15 |
| 129 29 6 | 124 7 245 | 3 159 8 |
+--------------+---------------+---------------+ |
Last edited by dongrave on Thu Apr 14, 2022 1:01 pm; edited 2 times in total |
|
Back to top |
|
|
Mogulmeister
Joined: 03 May 2007 Posts: 1151
|
Posted: Thu Apr 14, 2022 12:59 pm Post subject: |
|
|
I confess to using both over the years! |
|
Back to top |
|
|
TomC
Joined: 30 Oct 2020 Posts: 358 Location: Wales
|
Posted: Thu Apr 14, 2022 1:06 pm Post subject: |
|
|
Thanks dongrave, I see your point with the clarity |
|
Back to top |
|
|
dongrave
Joined: 06 Mar 2014 Posts: 568
|
Posted: Thu Apr 14, 2022 5:34 pm Post subject: |
|
|
TomC wrote: | Thanks for the notation Mogulmeister, I struggle with eureka, but if you describe something I know then this helps a lot!
Interestingly, you can put any value in r5c3 containing <148> because it still shows r5c8 cannot be <1> |
I also struggled with Eureka until one day I had this 'Aha!' moment and no longer thought about 'strong' and 'weak' links and instead began thinking of '=' as 'is equal to' and '-' as 'is not equal to'.
So, in Mogulmeister's example (5=1)r5c8-(1=48)r5c37-(4=3)r5c5-(3=6)r7c5-(6=1)r7c8-(1=5)r5c8
I simply read it as 'If r5c8 is equal to 1, then r5c37 are not 1 so they're 4 and 8, so r5c5 is not 4 so it's 3, so r7c5 is not 3 so it's 6, etc.
I saw someone explain this same approach not long ago on this site and I think that if more people were aware of this trick they'd have an easier time with Eureka! |
|
Back to top |
|
|
Mogulmeister
Joined: 03 May 2007 Posts: 1151
|
Posted: Fri Apr 15, 2022 5:59 am Post subject: |
|
|
Well remembered dongrave!
Mogulmeister wrote: | ...and of course it is a chain as well.
(-1)r7c5=(1)r3c5-r3c9=(1)r8c9 so r8c4<>1
In English:
"It isn't 1 at r7c5 so must be 1 at r3c5 isn't 1 at r3c9 must be 1 at r8c9".
The logic always alternates between the " - "(isn't) to "=" (must be).
Usually there is a bracketed number or numbers just to the left of the cell reference to let you know what is going on. The above example is simple as we are only on digit 1. Where there is no bracketed number the chain assumes you are still on the same digit. I didn't really need the (1) quite so much above.
You will see in some of the more involved chains that there is a transition to another digit as in the example below:
(1=4)r4c2-(4=3)r8c2 etc
In English:
"Not 1 but must be 4 in r4c2 so not 4 but must be 3 in r8c2 etc"
This is for when you move from one strong linked digit to a different one. |
|
|
Back to top |
|
|
Mogulmeister
Joined: 03 May 2007 Posts: 1151
|
Posted: Fri Apr 15, 2022 10:15 am Post subject: |
|
|
TomC wrote: |
Interestingly, you can put any value in r5c3 containing <148> because it still shows r5c8 cannot be <1> |
Triple proof - nice ! |
|
Back to top |
|
|
dongrave
Joined: 06 Mar 2014 Posts: 568
|
Posted: Sat Apr 16, 2022 9:21 pm Post subject: |
|
|
Mogulmeister wrote: | Well remembered dongrave!
Mogulmeister wrote: | ...and of course it is a chain as well.
(-1)r7c5=(1)r3c5-r3c9=(1)r8c9 so r8c4<>1
In English:
"It isn't 1 at r7c5 so must be 1 at r3c5 isn't 1 at r3c9 must be 1 at r8c9".
The logic always alternates between the " - "(isn't) to "=" (must be).
Usually there is a bracketed number or numbers just to the left of the cell reference to let you know what is going on. The above example is simple as we are only on digit 1. Where there is no bracketed number the chain assumes you are still on the same digit. I didn't really need the (1) quite so much above.
You will see in some of the more involved chains that there is a transition to another digit as in the example below:
(1=4)r4c2-(4=3)r8c2 etc
In English:
"Not 1 but must be 4 in r4c2 so not 4 but must be 3 in r8c2 etc"
This is for when you move from one strong linked digit to a different one. |
|
Aha! So that was you Mogulmeister?! From a post last year? I wish that people like me could buy books written by you - or Clement - or Keith - or Marty when he was alive. |
|
Back to top |
|
|
|