| View previous topic :: View next topic |
| Author |
Message |
Ema Nymton
Joined: 17 Apr 2009
Posts: 89
|
 Posted: Sun Jan 30, 2011 2:33 pm Post subject: 20110130 Hard |
 |
|
.
| Code: |
+--------------+---------+-----------------+
| 23 4 239* | 8 7 5 | 6 1 39 |
| 358 39 3589 | 1 4 6 | 359 2 7 |
| 7 6 1 | 3 9 2 | 8 45 45 |
+--------------+---------+-----------------+
| 9 8 24A | 26 5 1 | 7 3 46 |
| 1236 7 23B | 26 38 4 | 15 9 1568 |
| 136 5 34C | 7 38 9 | 14 468 2 |
+--------------+---------+-----------------+
| 58 1 7 | 59 2 3 | 459 4568 45689 |
| 4 2 3589 | 59 6 7 | 1359 58 13589 |
| 35 39 6 | 4 1 8 | 2 7 359 |
+--------------+---------+-----------------+
|
Does the ABC alignment eliminating '2' at r1c3 qualify as a 'wing'?
Ema Nymton
. |
|
| Back to top |
|
 |
peterj
Joined: 26 Mar 2010
Posts: 974
Location: London, UK
|
| Posted: Sun Jan 30, 2011 2:46 pm Post subject: |
|
|
Ema, more simply it is a "naked triple" i.e the three cells are constrained to contain only the digits 2,3,4 hence they may be removed from other cells in the same house. In this case they eliminate the 2 from r1c3 as you noted but also 3 from other cells in c3 and 23 from other cells in b4.
This triple made up of bivalues is the bulding block of a wing - but it needs to be "bent" so that the cells are in two different houses (row/column/block) with one cell (the pivot) being common to both houses. |
|
| Back to top |
|
|
ronk
Joined: 07 May 2006
Posts: 398
|
| Posted: Sun Jan 30, 2011 3:09 pm Post subject: |
|
|
| peterj wrote: | | In this case they eliminate the 2 from r1c3 as you noted but also 3 from other cells in c3 and 23 from other cells in b4. |
In which case it is oft referred to as a locked triple.
| peterj wrote: | | This triple made up of bivalues is the bulding block of a wing - but it needs to be "bent" so that the cells are in two different houses (row/column/block) with one cell (the pivot) being common to both houses. |
Fair enough, but I think there must be a few programmed solvers where the xy-wing and xyz-wing techniques subsume the naked triple and locked triple techniques ... and IMO they should. |
|
| Back to top |
|
|
peterj
Joined: 26 Mar 2010
Posts: 974
Location: London, UK
|
| Posted: Sun Jan 30, 2011 5:00 pm Post subject: |
|
|
| ronk wrote: | | Fair enough, but I think there must be a few programmed solvers where the xy-wing and xyz-wing techniques subsume the naked triple and locked triple techniques ... and IMO they should. |
I can see from a computer solver point of view why you would say that - but surely from a "soft squishy organic" solver point of view scanning a house for a triple is quite a different task to looking for an xy-wing? |
|
| Back to top |
|
|
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
