OM 7.2

Post puzzles for others to solve here.

Re: OM 7.2

Postby yzfwsf » Thu Apr 22, 2021 8:46 am

Is this one?
whip8.png
whip8.png (15.08 KiB) Viewed 345 times
yzfwsf
 
Posts: 854
Joined: 16 April 2019

Re: OM 7.2

Postby denis_berthier » Thu Apr 22, 2021 9:37 am

Yes, much better.
But, without making a difference between the (llcs and rlcs) and the (z- and t-), you make it very hard to see the linear, continuous structure that is the real whip.
denis_berthier
2010 Supporter
 
Posts: 3975
Joined: 19 June 2007
Location: Paris

Re: OM 7.2

Postby yzfwsf » Thu Apr 22, 2021 10:09 am

Now my question is, for CSP-Variables b2n3, why use link n3r3c2-n3r3c6 instead of link target-n3r3c6. Isn't the latter path shorter? So I see r3n3 as a branch.
yzfwsf
 
Posts: 854
Joined: 16 April 2019

Re: OM 7.2

Postby denis_berthier » Thu Apr 22, 2021 10:23 am

yzfwsf wrote:Now my question is, for CSP-Variables b2n3, why use link n3r3c2-n3r3c6 instead of link target-n3r3c6. Isn't the latter path shorter? So I see r3n3 as a branch.

Very simple: the result would not be a whip. In a whip the order of the CSP-Variables is important.
denis_berthier
2010 Supporter
 
Posts: 3975
Joined: 19 June 2007
Location: Paris

Re: OM 7.2

Postby AnotherLife » Thu Apr 22, 2021 10:28 am

yzfwsf wrote:Is this one?
whip8.png

Hello,
You still miss in your picture that r4c5=2 is linked to r3n2=c5. I think it is more easy to correct your drawing than to make my own one with a pencil and a pen.
Bogdan
AnotherLife
 
Posts: 192
Joined: 07 January 2021
Location: Moscow, Russia

Re: OM 7.2

Postby yzfwsf » Thu Apr 22, 2021 10:37 am

AnotherLife wrote:You still miss in your picture that r4c5=2 is linked to r3n2=c5. I think it is more easy to correct your drawing than to make my own one with a pencil and a pen.

Yes,you are right.
yzfwsf
 
Posts: 854
Joined: 16 April 2019

Re: OM 7.2

Postby denis_berthier » Fri Apr 23, 2021 6:07 am

.
For people who really want to see a correct graphical representation of this whip:

whip[8]: r3n2{c6 c5} - r3n3{c5 c2} - b2n3{r3c6 r2c5} - r4c5{n3 n7} - r5n7{c5 c2} - r8c2{n7 n8} - r7c2{n8 n4} - r1c2{n4 .} ==> r3c6 ≠ 4


Code: Select all
   Z    r3n2   r3n3   b2n3   r4c5   r5n7  r8c2   r7c2  r1c2

n4r3c6—n2r3c6     
          |       
       n2r3c5—n3r3c5   
                 |         
              n3r3c2—n3r3c6
                        |
                     n3r2c5—n3r4c5
                               |
                            n7r4c5—n7r5c5
                                      |
                                   n7r5c2—n7r8c2
                                             |
                                          n8r8c2—n8r7c2
                                                    |
                                                 n4r7c2—n4r1c2
                                                           |
                                                           .

The sequence of CSP-Variables is mentioned in the first line.
Horizontal bars are links. Vertical bars are csp-links.
You will notice that none of the z- or t- candidates are present, thus making obvious the continuous, linear structure.
Displaying all the z- and t- candidates at the same time on a graphics, with all their links and with no difference in colours or boldness, is an absurd approach, hiding the fundamental structure. These z- and t- candidates are NEVER useful at the same time.
Notice that this 2D graphics is already perfectly captured by the linear representation in the nrc notation.

Also, don't forget that so long whips are extremely rare in human-solvable puzzles and that this one-step solution in W8 is an absurdity, considering that there is a solution in S+W4. The 1-step requirement often leads to such absurdities.


[Edit:]The above graphic displays correctly in Safari; in case it doesn't in your browser, here is a gif form:
Image
denis_berthier
2010 Supporter
 
Posts: 3975
Joined: 19 June 2007
Location: Paris

Previous

Return to Puzzles