dailysudoku.com

[jabejochke]

Cannot find note submitted this morning -- certainly could have been my entry.

In working the puzzle of the 23rd, I encountered a not untypical experience for me. Had break-thru at Block 7 (left side, bottom) and solved from that point. However, I had gone to Candidate Profiling; processed all 9 Rows, 9 Columns and first 6 Blocks before hitting the break-thru at Block 7. The situation at that point is shown below (I hope).

Could I have recognized the potential break-thru at Block 7 (hence, gone there earlier or used different R/C/B sequencing)? Appreciate any tips or references to recognizing typical break-thru opportunities.

[David Bryant]

It wasn't your posting, jabejochke. I wrote a reply to your post this morning, and now that's gone, too. It's probably the "instability" samgj was talking about recently -- I think he's moving the web site to a new service provider, so the forum probably got backed up before you made your post, and then restored a bit later, if that makes any sense.

Anyway, the "break through" you're talking about is no doubt the hidden pair {1, 9} lying in r7c1 & r8c3. You can see this in the puzzle at the stage you described this morning by looking at column 2, and at row 9. You'll see the values {1, 9} lying in that column and in that row, and if you notice how they intersect in the bottom left 3x3 box you'll instantly recognize the "hidden pair."

The "trick" is to think in terms of pairs of values and not just in terms of single values. This particular puzzle feature (hidden pairs) is pretty common in samgj's puzzles, and you can almost always recognize them by seeing the pair of values lying _outside_ the box in which the "hidden pair" occurs. dcb

[Guest]

David,

I do see the pair - and the source. Recognizing the source (the 1,9 in column 2 and the 1,9 in row 9) could have been the trigger to work Block 7 earlier than I did.

Thanks for the insight. It also looks like the same source coupled with the 7's in rows 8 and 9 show a triple in Block 7.

jabejochke

[samgj]