| View previous topic :: View next topic |
| Author |
Message |
tlanglet
Joined: 17 Oct 2007
Posts: 2468
Location: Northern California Foothills
|
 Posted: Mon Nov 28, 2011 5:05 am Post subject: Vanhagen Extreme November 28, 2011 |
 |
|
| Code: |
+-------+-------+-------+
| 6 . . | 7 9 1 | . . 2 |
| 9 . . | . 2 . | . . 7 |
| . . . | 6 . 4 | . . . |
+-------+-------+-------+
| . . 1 | 3 6 9 | 4 . . |
| . 6 . | . . . | . 7 . |
| . . 5 | 2 8 7 | 1 . . |
+-------+-------+-------+
| . . . | 8 . 3 | . . . |
| 5 . . | . 1 . | . . 9 |
| 3 . . | 9 7 2 | . . 4 |
+-------+-------+-------+
|
Play online
After basics:
| Code: | *-----------------------------------------------------------*
| 6 345 8 | 7 9 1 | 35 345 2 |
| 9 134 34 | 5 2 8 | 36 1346 7 |
| 127 1257 27 | 6 3 4 | 589 1589 158 |
|-------------------+-------------------+-------------------|
| 278 27 1 | 3 6 9 | 4 258 58 |
| 28 6 39 | 1 4 5 | 289 7 38 |
| 4 39 5 | 2 8 7 | 1 69 36 |
|-------------------+-------------------+-------------------|
| 127 49 49 | 8 5 3 | 267 126 16 |
| 5 278 27 | 4 1 6 | 2378 238 9 |
| 3 18 6 | 9 7 2 | 58 158 4 |
*-----------------------------------------------------------*
|
I tried some almost moves this time and found a winner.
First, an anp(34=1)r2c32-(1=8)r9c2-(8=5)r9c7-(5=3)r1c7; r2c78<>3 which set up the heavy hammer......
| Code: | ant(589=2)r395c7-(9)r5c7=(9-6)r6c8=r6c9-(6=1)r7c9-r9c8=r9c2-r23c2=r3c1-r3c89=(1-4)r2c8=r1c8-(4=5)r1c2; r1c7<>5
\
\
-(1=58)r9c87-(8)r8c7; r8c7<>8 |
Ted |
|
| Back to top |
|
 |
JC Van Hay
Joined: 13 Jun 2010
Posts: 494
Location: Charleroi, Belgium
|
| Posted: Mon Nov 28, 2011 8:33 am Post subject: |
|
|
Another one (in R7,B6) ...
Wing : ANP(16=2)r7c89-ANP(2=58)r4c89-ANP(8=36)r56c9 => +1r7c89;stte
Or
... : ANP(16=2)r7c89-ANQ(2=3568)B6 => +1r7c89;stte |
|
| Back to top |
|
|
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
| Posted: Mon Nov 28, 2011 7:16 pm Post subject: |
|
|
| JC Van Hay wrote: | Another one (in R7,B6) ...
Wing : ANP(16=2)r7c89-ANP(2=58)r4c89-ANP(8=36)r56c9 => +1r7c89;stte
Or
... : ANP(16=2)r7c89-ANQ(2=3568)B6 => +1r7c89;stte
|
Maybe JC sees his chains as being complete, but I had to add an additional SL to get them to work for me.
| Code: | First chain w/additional SL:
(16=2)r7c89 - (2=58)r4c89 - (8=3=6=1)r567c9 => r7c1,r9c8<>1
-equivalent notation-
(16=2)r7c89 - (2=58)r4c89 - (8=3)r5c9 - (3=6)r6c9 - (6=1)r7c9 => r7c1,r9c8<>1
|
Unfortunately, my solver didn't find this chain path. I believe it has to do with my solver not allowing an elimination to occur twice for a candidate in a cell; e.g., (16=2)r7c89 creates r7c89<>6 and this prohibits (6=1)r7c9 from creating r7c9<>6. Hmmm!!! |
|
| Back to top |
|
|
JC Van Hay
Joined: 13 Jun 2010
Posts: 494
Location: Charleroi, Belgium
|
| Posted: Mon Nov 28, 2011 9:55 pm Post subject: |
|
|
Danny,
I fully understand your POV. I disliked re-using the SL (6=1)r7c9 at the end of the chains, so I dropped it.
On the other hand, the use of ALS hides the xyt nature of the chain :
(1=6)r7c9-(6=3)r6c9-(3=8)r5c9-(8=5)r4c9-[5(8#3)=2]r4c8-[2(6#1)=1]r7c8 => 1r7c9=1r7c8 where #1 and #3 refers to the 1st and 3rd SL respectively.
This is maybe why your solver doesn't find the chain.
JC |
|
| Back to top |
|
|
ronk
Joined: 07 May 2006
Posts: 398
|
| Posted: Mon Nov 28, 2011 10:29 pm Post subject: |
|
|
| JC Van Hay wrote: | Danny,
I fully understand your POV. I disliked re-using the SL (6=1)r7c9 at the end of the chains, so I dropped it.
On the other hand, the use of ALS hides the xyt nature of the chain ... |
That's unfortunate. Chains with endpoint-overlap are true chains and are not nettish like xyt-"chains".
Last edited by ronk on Tue Nov 29, 2011 1:01 pm; edited 1 time in total |
|
| Back to top |
|
|
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
| Posted: Wed Nov 30, 2011 6:38 pm Post subject: |
|
|
While editing variable names in my chain() routine, I ran across a section of code that hadn't been updated to handle multiple values on one side of a SL. After updating chain(), I finally managed to get JC's chain. (It's the second one listed below.)
| Code: | +--------------------------------------------------------------+
| 6 345 8 | 7 9 1 | 35 345 2 |
| 9 134 34 | 5 2 8 | 36 1346 7 |
| 127 1257 27 | 6 3 4 | 589 1589 158 |
|--------------------+--------------------+--------------------|
| 278 27 1 | 3 6 9 | 4 258 58 |
| 28 6 39 | 1 4 5 | 289 7 38 |
| 4 39 5 | 2 8 7 | 1 69 36 |
|--------------------+--------------------+--------------------|
| 127 49 49 | 8 5 3 | 267 126 16 |
| 5 278 27 | 4 1 6 | 2378 238 9 |
| 3 18 6 | 9 7 2 | 58 158 4 |
+--------------------------------------------------------------+
# 60 eliminations remain
(58=1)r34c9 - r3c1 = r7c1 - (16=2)r7c89 - (2=58)r4c89 => r5c9<>8
(16=2)r7c89 - (2=58)r4c89 - (8=3)r5c9 - (3=6)r6c9 - (6=1)r7c9 => r7c1,r9c8<>1
(16=2)r7c89 - (2=58)r4c89 - (8=3=6=1)r567c9 => r7c1,r9c8<>1
|
===== ===== ===== =====
| JC wrote: | On the other hand, the use of ALS hides the xyt nature of the chain:
(1=6)r7c9-(6=3)r6c9-(3=8)r5c9-(8=5)r4c9-[5(8#3)=2]r4c8-[2(6#1)=1]r7c8 => 1r7c9=1r7c8
|
While I have an issue with some ALS() patterns being shorthand for an embedded network (as you indicate), I've accepted what I call the ANS() partition, of the full ALS() pattern, as being consistent with the philosophy of a "chain".
In the ANS() partition, a statement like (W=XYZ)r357c9 implies that the candidates X, Y, and Z (each) occur at least twice in r357c9. It also implies that all eliminations associated with <XYZ> occur in [c9] only. These implied constraints are not present in ALS() patterns used by others.
Regards, Danny |
|
| Back to top |
|
|
|
|
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
|
Powered by phpBB © 2001, 2005 phpBB Group
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
