View previous topic :: View next topic |
Author |
Message |
Earl
Joined: 30 May 2007 Posts: 677 Location: Victoria, KS
|
Posted: Sat Mar 07, 2009 3:00 pm Post subject: March 7 DB |
|
|
The first March DB comes in "like a lion." This is no one-stepper.
Earl
Code: |
+-------+-------+-------+
| . 2 . | . . . | 9 . 1 |
| . . . | . . . | 2 5 . |
| . . . | 1 . 3 | . 4 . |
+-------+-------+-------+
| 5 . 8 | 3 . . | . . . |
| 6 7 . | . . . | . 1 2 |
| . . . | . . 5 | 8 . 7 |
+-------+-------+-------+
| . 6 . | 8 . 4 | . . . |
| . 5 7 | . . . | . . . |
| 8 . 4 | . . . | . 2 . |
+-------+-------+-------+
|
Play this puzzle online at the Daily Sudoku site |
|
Back to top |
|
|
Marty R.
Joined: 12 Feb 2006 Posts: 5770 Location: Rochester, NY, USA
|
Posted: Sat Mar 07, 2009 5:17 pm Post subject: |
|
|
Three steps for me.
Quote: | Type 2 UR on 29, XY-Wing on 789 and W-Wing on 79 with pincer transport |
|
|
Back to top |
|
|
tlanglet
Joined: 17 Oct 2007 Posts: 2468 Location: Northern California Foothills
|
Posted: Sat Mar 07, 2009 5:54 pm Post subject: |
|
|
A two stepper: Quote: | UR <29> in r67c13 with x-wing <2> overlay deletes <9> in r67c1. (Marty, I see this UR as a Type 4, not Type 2)
The UR set up a w-wing on <79> that completes the puzzle.
|
Ted |
|
Back to top |
|
|
Marty R.
Joined: 12 Feb 2006 Posts: 5770 Location: Rochester, NY, USA
|
Posted: Sat Mar 07, 2009 6:03 pm Post subject: |
|
|
Ted, it was a Type 4. Thanks for pointing that out.
(Creeping [galloping?] senility.) |
|
Back to top |
|
|
Captain Pete
Joined: 09 Jun 2007 Posts: 55 Location: Oley, PA
|
Posted: Sat Mar 07, 2009 10:28 pm Post subject: |
|
|
I need a hint on this one. |
|
Back to top |
|
|
tlanglet
Joined: 17 Oct 2007 Posts: 2468 Location: Northern California Foothills
|
Posted: Sat Mar 07, 2009 10:51 pm Post subject: |
|
|
Marty R. wrote: | Ted, it was a Type 4. Thanks for pointing that out.
(Creeping [galloping?] senility.) |
It is also wide spread. I just recently responded to a post by Norm and referred to you as the person who made the post. I think I will just drink more wine and enjoy the situation since I can't fight it.
Ted |
|
Back to top |
|
|
wapati
Joined: 10 Jun 2008 Posts: 472 Location: Brampton, Ontario, Canada.
|
Posted: Sun Mar 08, 2009 12:27 am Post subject: |
|
|
Captain Pete wrote: | I need a hint on this one. |
Where you are stuck would be helpful. |
|
Back to top |
|
|
storm_norm
Joined: 18 Oct 2007 Posts: 1741
|
Posted: Sun Mar 08, 2009 4:52 am Post subject: |
|
|
Quote: | This is no one-stepper |
sorry Earl,
Code: | .------------------.------------------.------------------.
| 3 2 6 | 45 45 78 | 9 W78 1 |
| 47 48 1 | 79 689 6789 | 2 5 3 |
|W79 89 5 | 1 2 3 |w67 4 68 |
:------------------+------------------+------------------:
| 5 1 8 | 3 7 2 | 46 69 469 |
| 6 7 3 | 49 489 89 | 5 1 2 |
| 249 49 29 | 6 1 5 | 8 3 7 |
:------------------+------------------+------------------:
|E129 6 E29 | 8 35-9 4 | 137 W79 5-9 |
|E19 5 7 | 2 369 169 | 1346 689 4689 |
| 8 3 4 | 579 569 1679 | 16 2 569 |
'------------------'------------------'------------------' |
the extended W-wing {7,9}marked, elimantes the 9's in r7c5 and r7c9 solves it in one step.
the pincer {9} on the left hand side is in r7c3 that "sees" the other pincer {9} in r7c8. so this actually forms a loop.
the extension is marked "E"
w-wing marked "W" |
|
Back to top |
|
|
Earl
Joined: 30 May 2007 Posts: 677 Location: Victoria, KS
|
Posted: Sun Mar 08, 2009 2:37 pm Post subject: March 7 DB |
|
|
Norm,
You are light years ahead of me with your extended W-wing, a one-stepper containing many steps. Perhaps you could provide further details to help me, and probably others. I assume your W wing is R3C1-R7C8 tied by 7's in C7. I do not understand the W at R1C8, nor the w at R3C7. Also, I do not get the steps of the extentsion, E. I thought extensions had to be strong pairs.
Thanks for your patience.
Earl |
|
Back to top |
|
|
storm_norm
Joined: 18 Oct 2007 Posts: 1741
|
Posted: Sun Mar 08, 2009 5:59 pm Post subject: |
|
|
Quote: | Norm,
You are light years ahead of me with your extended W-wing, a one-stepper containing many steps. Perhaps you could provide further details to help me, and probably others. I assume your W wing is R3C1-R7C8 tied by 7's in C7. I do not understand the W at R1C8, nor the w at R3C7. Also, I do not get the steps of the extentsion, E. I thought extensions had to be strong pairs.
Thanks for your patience.
Earl |
maybe a picture would help?
notice that all the strong links have thick lines and the weak links have thin lines.
the 7's in box 3 are strongly linked
the 7,9 pairs are strongly linked
and lastly the 9's in box 7 are strongly linked... the group of 9's in r78c1 and the 9 in r7c3
the blue line represents the W-wing structure. the 9's circled in green are the pincers in that W-wing. they do nothing (useless) in the regular w-wing.
but...
the 9 circled in green in r3c1 can be extended via the green thin line down to the 9's in r78c1.
which is strongly linked to the 9 in r7c3, marked by the thick line.
the 9 in r7c3 now becomes the extended pincer. this 9 along with the 9 in r7c8 can "see" each other ( they share the same row ). and we know from our W-wing definition that any 9's they see can be eliminated.
-----
as a bonus!!
if you are really paying attention to the rules of loops...
notice how the thin green line in column 1 goes through the 9 in r6c1. we know from our rules of loops that that 9 can also be eliminated.
for those who keep track of the totality of the structure!!!
bonus points. |
|
Back to top |
|
|
Marty R.
Joined: 12 Feb 2006 Posts: 5770 Location: Rochester, NY, USA
|
Posted: Sun Mar 08, 2009 6:11 pm Post subject: |
|
|
Is this a case of terminology confusion, with:
Extended W-Wing = W-Wing with pincer transport? |
|
Back to top |
|
|
storm_norm
Joined: 18 Oct 2007 Posts: 1741
|
Posted: Sun Mar 08, 2009 6:18 pm Post subject: |
|
|
Marty R. wrote: | Is this a case of terminology confusion, with:
Extended W-Wing = W-Wing with pincer transport? |
exactly. its a pincer transport through the 9's in box 7 |
|
Back to top |
|
|
Earl
Joined: 30 May 2007 Posts: 677 Location: Victoria, KS
|
Posted: Sun Mar 08, 2009 9:19 pm Post subject: March 7 db |
|
|
Norm, thanks. I now see the logic of "the transport" that if R3C1 is <9> then R7C3 must be <9>. I am still not clear about box 3. Is it not sufficient to establish the 79 W-wing with the strong pair of 7's in column 7?
Thanks
Earl |
|
Back to top |
|
|
storm_norm
Joined: 18 Oct 2007 Posts: 1741
|
Posted: Sun Mar 08, 2009 9:29 pm Post subject: Re: March 7 db |
|
|
Earl wrote: | Norm, thanks. I now see the logic of "the transport" that if R3C1 is <9> then R7C3 must be <9>. I am still not clear about box 3. Is it not sufficient to establish the 79 W-wing with the strong pair of 7's in column 7?
Thanks
Earl |
Earl, YES.
the 7's in column 7 are sufficient.
as are the 7's in box 3. both provide the strong link needed to connect the 7,9 pairs
your w-wing can be seen marked below, using the 7's in column 7
Code: | .------------------.------------------.------------------.
| 3 2 6 | 45 45 78 | 9 78 1 |
| 47 48 1 | 79 689 6789 | 2 5 3 |
|W79 89 5 | 1 2 3 |W67 4 68 |
:------------------+------------------+------------------:
| 5 1 8 | 3 7 2 | 46 69 469 |
| 6 7 3 | 49 489 89 | 5 1 2 |
| 249 49 29 | 6 1 5 | 8 3 7 |
:------------------+------------------+------------------:
| 129 6 29 | 8 359 4 |W137 W79 59 |
| 19 5 7 | 2 369 169 | 1346 689 4689 |
| 8 3 4 | 579 569 1679 | 16 2 569 |
'------------------'------------------'------------------' |
my w-wing uses the strong inference on 7's in box 3
your w-wing uses the strong inference on 7's in column 7.
either way, the same "pincer transport" works from the 9 in r3c1 |
|
Back to top |
|
|
tlanglet
Joined: 17 Oct 2007 Posts: 2468 Location: Northern California Foothills
|
Posted: Mon Mar 09, 2009 6:35 pm Post subject: |
|
|
storm_norm wrote: |
we know from our rules of loops that that 9 can also be eliminated. |
Norm, could you provide me a reference to the rules?
Thanks, Ted |
|
Back to top |
|
|
Asellus
Joined: 05 Jun 2007 Posts: 865 Location: Sonoma County, CA, USA
|
Posted: Tue Mar 10, 2009 5:26 am Post subject: |
|
|
This seems like a lot of trouble to explain an ER. The <9> at r7c9 can see the W-Wing pincer <9> at r7c8 directly and can see the pincer <9> at r3c1 via the ER in Box 7. By the way, this also eliminates the <9> at r6c1 (since it "sees through the ER" in the other direction).
The "transport" idea isn't necessary. But I would say that the <9> at r3c1 is transported to r7c3 for the r7c9 elimination; or I would say that the <9> at r7c8 is transported to r8c1 for the r6c1 elimination. |
|
Back to top |
|
|
storm_norm
Joined: 18 Oct 2007 Posts: 1741
|
Posted: Wed Mar 11, 2009 7:10 am Post subject: |
|
|
Asellus wrote: | This seems like a lot of trouble to explain an ER. The <9> at r7c9 can see the W-Wing pincer <9> at r7c8 directly and can see the pincer <9> at r3c1 via the ER in Box 7. By the way, this also eliminates the <9> at r6c1 (since it "sees through the ER" in the other direction).
The "transport" idea isn't necessary. But I would say that the <9> at r3c1 is transported to r7c3 for the r7c9 elimination; or I would say that the <9> at r7c8 is transported to r8c1 for the r6c1 elimination. |
yes, the 9 at r7c1 is eliminated because of the w-wing, my apologies for not pointing that out.
If I understood Earl's question right, he was asking why I used the strongly linked 7's in box 3 rather than the strongly linked 7's in column 7 to propogate the w-wing.
and I hope Earl sees that both are correct. |
|
Back to top |
|
|
Luke451
Joined: 20 Apr 2008 Posts: 310 Location: Southern Northern California
|
Posted: Thu Mar 12, 2009 7:43 am Post subject: |
|
|
tlanglet wrote: | storm_norm wrote: |
we know from our rules of loops that that 9 can also be eliminated. |
Norm, could you provide me a reference to the rules? |
I believe Norm is referring to the "Rules of Propagation" which (some believe) are summed up nicely in Paul's Nice Loops paper. This move could be considered a continuous nice loop as Norm mentioned and actually makes all the eliminations the extended w-wing does (yes, even the bonus one!)
Code: | (9=7)r3c1-(7)r3c7=(7)r1c8-(7=9)r7c8-(9)r7c3=gp(9)r78c1-(9)r3c1=loop
|
Now, since all extended w-wings are not continuous NLs, I have my own questions.
storm_norm wrote: | the 9 in r7c3 now becomes the extended pincer. this 9 along with the 9 in r7c8 can "see" each other ( they share the same row ). and we know from our W-wing definition that any 9's they see can be eliminated. | I was trying to get going on these "extended/transported" w-wings when I came across this thread. Nice work on the diagram, I must say. So why am I getting confused?
Using this example:
* With the strong link on (7) in box 3 the w-wing is established and any (9) that can see both the (79) cells can be eliminated, so r7c1<>9.
* For the "extended" part (I figured), any strong link on (9) that can see both the (79) cells creates a new pincer, in this case the (9) in r7c3. If r7c8 is the other pincer, there's the cited eliminations.
This can't be right. The group (9)'s in box 7 can see the (79) in box 1, so why don't the the group (9)'s also constitute a pincer, resulting in r6c1<>9?
I dug a hole, could you reach down and pull me out, please?
Edit=typo
Last edited by Luke451 on Thu Mar 12, 2009 1:28 pm; edited 1 time in total |
|
Back to top |
|
|
Asellus
Joined: 05 Jun 2007 Posts: 865 Location: Sonoma County, CA, USA
|
Posted: Thu Mar 12, 2009 10:06 am Post subject: |
|
|
It seems to me that this W-Wing discussion has gotten muddled. First of all, the original W-Wing is not "useless": it eliminates <9> from r7c1.
As for pincer transports...
Considering first the pincer <9> at r3c1, if it is true, then so is the <9> at r7c3. So, the <9> at r7c3 can be considered the "transported" pincer from r3c1. Together with the pincer at r7c8, it eliminates <9>s from r7c59 (the <9> at r7c1 already having been eliminated by the W-Wing without transport.)
Next, considering the pincer <9> at r7c8, if it is true, then so is the <9> at r8c1. So, the <9> at r8c1 can be considered the "transported" pincer from r7c8. Together with the pincer at r7c8, it eliminates <9> from r6c1.
As for the continuous loop... you can get into trouble if you don't notate the ER in a bidirectional manner. To me, it is not correct to write:
"...(9)r7c3=(9)r78c1..." since this only makes sense read left to right. It is better, in my opinion, to write:
"...(9)r7c13=(9)r78c1...", including the ERI cell r7c1 on both sides of the strong inference.
Now, once the continuous loop has been established, it must be possible for either side of this strong inference to be false. That is only possible if r7c1 is false. So, r7c1 must be false. But then, this is also a necessary consequence of the W-Wing, which is really saying the same thing. In fact, if you first perform the W-Wing elimination of the <9> in r7c1, then you avoid this whole ERI problem in the continous loop notation. |
|
Back to top |
|
|
daj95376
Joined: 23 Aug 2008 Posts: 3854
|
Posted: Thu Mar 12, 2009 2:13 pm Post subject: |
|
|
Well, if we're locked into supporting a one-step solution, consider:
Code: | [r7c1]= 9 => [r6c1],[r7c59]<>9
[r7c1]<>9 => XY-Loop a-b-c-d-e-f-g-h-a => [r6c1],[r7c59]<>9
+--------------------------------------------------------------+
| 3 2 6 | 5 4 78 | 9 b78 1 |
| 47 48 1 | 79 689 6789 | 2 5 3 |
| e79 89 5 | 1 2 3 | d67 4 c68 |
|--------------------+--------------------+--------------------|
| 5 1 8 | 3 7 2 | 46 69 469 |
| 6 7 3 | 49 89 89 | 5 1 2 |
| 24-9 49 29 | 6 1 5 | 8 3 7 |
|--------------------+--------------------+--------------------|
| g129 6 h29 | 8 35-9 4 | 137 a79 5-9 |
| f19 5 7 | 2 369 169 | 1346 689 4689 |
| 8 3 4 | 79 569 1679 | 16 2 569 |
+--------------------------------------------------------------+
# 56 eliminations remain
|
It doesn't support Asellus' ERI elimination [r7c1]<>9, but it does get the job done.
I like Asellus' (implied) suggestion that it should be a two-step solution with the W-Wing being performed first. Then the XY-Loop or norm's extended W-Wing could be performed w/o grouping.
Asellus (edited) wrote: | As for the continuous loop, you can get into trouble if you don't notate the ER in a bidirectional manner.
To me, it is not correct to write "...(9)r7c3=(9)r78c1..." since this only makes sense read left to right.
It is better to write "...(9)r7c13=(9)r78c1..." and include the ERI cell r7c1 on both sides of the strong inference.
|
Yes!
Last edited by daj95376 on Thu Mar 12, 2009 4:47 pm; edited 1 time in total |
|
Back to top |
|
|
|