Puzzle Description:

...in the mirror after working on this puzzle!

#1: Tom O'Connell (sensei69) on Mar 22, 2010 [HINT]

Right & bottom edge logic is required.

I used edge logic on the bottom row to get the 7 and 5 clues placed. I also used edge logic on row 1 to dot R1C17 and R1C18. I do not see how I can use edge logic on the right. What am I missing?

I'm right where Gator is, and am not seeing anything on the right either.

Hmm. After dotting R1C17 and R1C18, we know R1C8 and R1C9 are black, being part of the five. We can do an either-or thing there. We know that one (or both) of the squares on either side of that (R1C7 and R1C10) must be black. If either one is black, hen R2C7 must be black. So we can mark it black. That means R3C7 must be black too...and then nothing.

Here's a little more on row 1. Imagine trying place both the 3 and 5 clues in row 1 all the way to the left. Notice how this will make row 3 invalid. So we know that the 5 clue will at least have to be one more to the right. So we fill in R1C10. More line logic.

Here's some more...
If you try to fill in R3C6, to the left would be all dots. The result would make row 1 invalid as R1C2-R1C5 would have to be filled in. So R3C6 is a dot.

EDIT: And more...
Now try to extend the 5 clue in row 1 to columns 11 and 12. This will make row 2 invalid. So R1C12 cannot be part of the 5 clue, therefore R1C7 gets filled in. More line logic

EDIT2: Now you can use edge logic on the 6 clue in row 4 to see that R4C8 will be a dot.

EDIT3: Here's a better one. Look at the 6 clue in column 3. Starting at row 6 and then down, the 6 clue will cross rows 9-11. The next step will make it impossible to fill in columns 2 and 4. If you follow this down, R6C3 - R9C3 will be dots. More line logic.

Starting a new post...

Look at column 2. If the 3 2 clues are to both be as high as they can go, then the 2 clue would have to be in rows 10 and 11. This results in column 3 being invalid, so the 2 clue must go to the lower part. A lot more line logic.

EDIT: I was missing one line logic dot. It solves out from there.

Jan, would you verify all of this and see if it all pans outs?

there's only one place that the 6 on the right frames the 2,2's. what makes it work is the 1 on the extreme right has to fit. then the top can be done ...etc etc

Tom - I worked all of the puzzle. I was just waiting for Jan to confirm what I wrote up as it is rather complex.

I tried following what you said about the placement of the 6, but it seems that there were multiple places it could go without looking a lot of moves ahead, unless I missed something. I'll look at it again.

New version published by sensei69.

New version published by sensei69.

i concede Gator ..there is 2 places one could put the 6 on the right side ....sorry

Unpublished by sensei69

Republished without change by sensei69

Thank you for this very hard but solveable puzzle. It contributed greatly to my pbn education! :)

New version published by sensei69.

I believe I fixed the error! let me know

I solved this with only logic ... (eventually).

Thanks for taking the time to fix it!

Well, I seem to have solved a previous version, or so it tells me.

Gator, I was following your logic, but I had a question...

In #5, EDIT 3, you basically state that the 6 from column 3 can not start in row 9. I followed how it can not start in rows 6-8, but I was able to go a long way with it in row 9 (at least 10 steps, maybe more). How did you actually rule out row 9?

I don't see any logical way to solve this one.

How do you get rid of the check mark?

Adam - I believe you are solving the one after the change was made. It is now invalid down to row 8, not 9 now.

Tom - I was able to solve the previous version, but I'm not sure if I can solve this new version.

Funny picture, it popped out at the end.

I used lots of edge logic but still got stuck about 3/4 way through on the 2 in C10. Then I had to guess, and guessed right and finished it.

New version published by sensei69.

Gator... there are only 2 places c-17 could be,,, one leads to nada, the other r-5 does it ALL.

We have to wait for a higher authority, but me solved it.

There is NO way to know if the version that I just solved looks much like what the past commenters solved, but I solved completely no problem. Looking back at Gator's longest explanation that all looked vaguely correct as I look at my solved puzzle but I didn't see it until after the solve, so I don't know that it follows 100%. I do know that I did the "what if the 3-2 in C2 were both high" except I think my mental process was a plain "what if either started in R11" so that it felt more like regular (internal) edge logic.

Good challenge. Too bad such a complicated solve had its discussion further complicated by multiple versions.

Well, you all may be right but my son (who, being a biased mother, I think is extremely good at logic) and I did a lot of guessing that proved to be correct so we solved it in the end. I don't see any way out of guessing more than once on the top (main) part. The bottom 5 and 7 work with line logic. But, I like the result and did haver fun solving.

Re #28: Actually you can see all past versions of any puzzle. (Except for really old puzzles - we didn't always have versions).

The easiest way is to go to the "peek at solution" thing. If there are multiple versions then there will be a link on the page to the older versions.

You can actually play old versions too, if you open the puzzle in a new page (so there is no menu sidebar) and edit the URL to add "&restore=1&version=2" to the end of the URL.

thanks jan! i keep learning new things about the site. thank you for all of the thoughtful details and features!

solved

cool, Vaggelis

I would like to alleviate any confusion that may have come about from the comments and five existing versions. All the hints that were published are valid and reliable so long as you are solving version 1, 3, or 5. They are all the same. 2 and 4 are different from each other and from all odd numbered versions. So all the hints that sensei, Gator, and Jan had posted for version 1 will also help with version 5.

hmmm it might not have been a good puzzle but it sure made for a great discussion. :-D

I did have to guess a couple of times, but it was a fun solve and a fun image.

Sometimes in my solving I "guess", but it's not a random guess - I peer at the pic with different mental constructions of a line or angle, and I'm often (okay, occasionally!) correct. Does anyone out there understand what I'm trying to say?

Yes. It's prediction based on a perceived pattern. It's still considered guessing, because your not placing pixels where they must go as determined by logic. But it is an educated guess.

I'll often call that "face" logic or "arm is attached at the shoulder" logic or what-have-you.

But what if it's not? ;)

There's a gruesome puzzle idea!

Marz did that with her "in pieces" series. There wasn't too much blood...

Logically solvable, no guessing required.

Found to be solvable with moderate lookahead by gator.

Goto next topic

You must register and log in to be able to participate in this discussion.