Sudoku Players' Forum (clone)

hmm... I guess I should have double checked my post before pressing the Sumbit button...

hi Animator, thank you for your help to new triers like myself, but can you please explain the following to me?--(ref to the puzzle submitted by joolsee) simes/abailes commented that "block 4 column 2 must contain a 1 - " can see that so no probs - then "disjoint subset 26 in cells (7,4 7,5)
up dating candidates" etc. - please may i ask the bus driver to stop right there because im lost ! - i have my pencil marks for that row (seven) as 024568-24567-(9)-268-2568-256-(3) -24678 (1) and yes i can see the subset "26" in 7,4 and 7,5 but please explain to me what criteria are used to select the subset from those two cells when there is also a "26" in 7,1 - 7,2 - 7,6 and 7,7? obviously i am missing something in the correct identification of the actual subset location? the same applies to the next subset - now row eight - my pencil marks are - 12568-2356-28-12368-(7) - (4) - (9) - 268 - 268- again i can see the "268" subset referred to
but just cannot grasp why it is located in (8,3)(8,8) and (8,9) when in my tiny little mind i could suggest (8,1) to go with (8,8) and (8,9)? and to
make matters really worse i am totally confused as to the logic in "seeing" the subset "248" in block seven in the cells quoted when i also see the same numbers spread across the cells 7,1-7,2-8,1-8,2and 8,3?
9,1 - 9,2 and 9,3 obviously ive got the wrong end of the stick -likewise for block eight - please help !

oh further to my post above - please may i test my understanding by checking one other aspect? am i correct in saying that if/when i identify an xwing (say for number 4 )then i can eliminate all the 4,s that fall "along the boundaries set by the xwing but not the 4,s that form the corners of the xwing?" wow! that sounds complicated to me but i understand what im getting at - hopefully the actual question makes sense to you guys?. also if i identify a "spread" of say three numbers over three diff cells then i can eliminate all other candidates, of the same numbers, from that block,row or column? and the same applies for the identification of two numbers, confined to two cells (no other numbers present)?? am i getting the general idea as regards the reducing of candidates?

Let me first reply to your last post since that is eaiser.

If you are saying what I think you are saying then yes the note about the X-wing is correct. If you find an X-wing then you will are able to either remove candidates from two rows, or from two columns.

(If you are talking about removing candidates from both rows and columns (at the same time) then you haven't found an X-wing)

I'm not sure about the second note though... It can refer to the disjoint subset aswell... can you post an example of what you mean?


Now to reply on your previous post:

Can you post your entire grid?

Joolslee's grid has the number r7c5 filled in for example, and most likely some others too... so you are missing something else...

In your pencilmarks there simply is no subset for row 7.

When you look at Joolslee grid, then you see that these candidates r7c5: 2, 6 ; r7c6: 2, 6. That's the subset they're talking about.

Row 8 has a subset though.

Your pencilmarks:

r8c1: 1, 2, 5, 6, 8
r8c2: 2, 3, 5, 6
r8c3: 2, 8
r8c4: 1, 2, 3, 6 8
r8c8: 2, 6, 8
r8c9: 2, 6, 8

Would you see the subset if r8c3 was 2, 6, 8?

There are three cells that have exactly the same three candidates. Note though not all three numbers need to occur... the number 6 isn't in r8c3 but it still is part of the subset, because it has two out of thee candidates (and only those two).

If you don't see it, then you can always try it.

For example, what will happen on row 8 when you fill in the number 2 in r8c1? Then r8c8 has to be 6 and r8c9 has to be 6, which ofcourse is impossible.