Puzzle Description:

The silhouette of the space shuttle Endeavour in a very unique setting over Earth's horizon, from a photo taken by an Expedition 22 crew member aboard the International Space Station just prior to docking on Feb. 9th. The image was captured 183 miles over the South Pacific Ocean. The white dot is a reflection. The lower left corner is the Earth's Troposphere, with all of the clouds and weather. The diagonal line is where the Stratosphere meets the the Mesosphere.

#1: Teresa K (fasstar) on Feb 24, 2010 [SPOILER]

Okay, I know I did one of the Endeavor recently, but you have to admit that this one is different. The one before was from a graphic. This one is from a photo that my brother (scientist at NASA) sent me. I could not really do it justice in such a small grid and only B&W. Here is the photo, along with some others:

http://www.sacbee.com/static/weblogs/photos/2010/02/space-shuttle-endeavours-missi.html

You will need to use what I call zigzag logic. When you get stuck, you will notice you need to place a series of 2s, with a one on the left and a one on the right, which tells you this is a zigzag rather than many blocks of 4. There is only one way to place the zigzag. Then the rest is line logic.

A fun puzzle and an interesting image!

splendid image and enjoyable solve...you were right about the zigzag. :)

Well, I'm not convinced of the validity of "zig zag logic". I can certainly see when such a zig zag solution is possible, but I can't quite convince myself that it is the only possible solution. It seems like the proof of that would have to be fairly elaborate.

Teresa I think you have to make it solvable for us to accept this entry! Sorry!

Thanks, Telly and Sally.

Jota, I think I will just withdraw this one from the contest. I don't like to edit a puzzle I have already published unless it's absolutely necessary. In this case, I kind of like it the way it is. :-) There are so many possibilities for the contest; it won't take me long to come up with something else.

Jan, if you declare this puzzle not solvable without guessing, I will accept that decision. But you said your're not convinced, so let me try to convince you. Hehehe.

I would like to point out that this type of logic was acceptable in several other puzzles here. The solution doesn't seem elaborate to me. Although you have to look a few moves ahead, it's just a matter of visually following the diagonal line to determine which direction it will go.

Here is the point where I got stuck after doing as much line logic as possible:

http://come-over.to/Play/PBN/7910a.jpg

In the shaded area (R23 C20-21), those will either be both black or both dots. The former would interfere with the row below, so they will be dots. That helps to go forward, but not far. This is where I got stuck again:

http://come-over.to/Play/PBN/7910b.jpg

When you have 1,2,2,2,2,2 one way and 2,2,2,2,2,2 the other way, you know that has to be diagonal. The question is whether it starts in the upper left or upper right (shaded). If you visually follow the diagonal, it is easy to see which way it must go. Yes, that is looking several steps ahead, but even with my terrible short term memory, it is easy to "see" without guessing. To me, this seems much easier than some of Gator's logic, which I would definitely call elaborate.

Gator, I love your elaborate logic, even when it strains my brain. :-)

I agree with Teresa on this.

While the diagonal is not inherently obvious in many cases, in this one it is clear.


The step-by-step of a staircase:

square c1r6 - c8r13

There is only a 1 in r6 and all the vertical clues in that block start with 2. Therefore this (yet unplaced) block will extend into r7.

R7 has only a 2 so the black block (still unfixed) must either extend to the left or the right. (After this step, the direction is determined as there will be a white under the first block, barring that space.)

Now, *that* horizontal block is also in a vertical 2, and since the row above is now complete, this block can only extend downwards.

And the cycle repeats, forming a staircase (or diagonal) ((or zigzag)).


The square of c19r24 - c26r30 is the same, but rotated to the right.

____________________________________

Of course, even when it's not as simple as this one, it can sometimes be teased out. I just put forth some logic on this to claim that #163 is solvable. http://infrapinklizzard.deviantart.com/art/Logic-163-155347359

And the debate continues... will it ever end??? It basically comes down to this: with a diagonal or zig-zag line, is there ever a way there could be any other solution BUT the diagonal line, and/or other than the intended direction? If so, then is it truly logically solvable?

if you do some looking ahead in the bottom left region, specifically the bottom row, the two diagonal lines will solve themselves out anyways eliminating a need for 'diagonal logic'. however, you have to look 2-3 steps ahead which normally isn't logically acceptable on this site. cool puzzle still, Teresa.

Thanks, Ant.

Re #10: Adam, in many cases, there are just too many possibilities to eliminate to say that it is solvable by (human) logic, but *in this case* the 1 in row six gives a definite starting point and completely excludes the possibility of 2x2 blocks.

OK, I think I have a way forward on this one not involving "diagonal logic". It is elaborate though.

Look at row 24 on the right. R24C20 may or may not be filled in depending on whether or not R24C19 is the 1 clue or part of one of the 2 clues. Let's assume that R24C19 is the 1 clue, and let's look how this would affect row 25. No matter how you try to place the 2 2 clues in row 24, this will cause row 25 to be invalid. So we know the R24C20 has to be black, as R24C19 cannot be the 1 clue. Of course this doesn't get us very far.

Now focus on the 4 clue in row 29. Notice how there are only two places that it can go: columns 1-4 or columns 6-9. Notice that if any one of the cells (R29C1, R29C2, R29C3, R29C4) is black, then they are all black. Also notice that the same logic applies to the 4 clue in column 2. If R27C2, R28C2, R29C2, or R30C2 is black, then they are all black. But look at what happens here - the 4 clues cross each other. So we can conclude that if R27C2, R28C2, R29C2, R30C2, R29C1, R29C3, or R29C4 are black then every single one of these cells is black.

Now look at the clue 5 in row 30. If the 5 clue extends to the left even one cell (R30C4), this causes R29C4 to be filled in and consequently all of those other cells as previously discussed. So R29C1-R29C4 (and R27C2-R30C2) will be filled in, unless the 5 clue in row 30 extends all the way to the right (with R30C1-R30C4 being dots). The problem is when this happens R29C6-R29C8 are black and R29C9 is a dot. This is impossible, so the 5 clue in row 30 cannot extend all the way to the right and must extend at least one cell to the left (R30C4). This allows for a lot more line logic.

Looking at row 30 again, the remaining 3 clue will either have to go in columns 11-13 or 21-23. If it goes in columns 21-23, then the 2 clue would have to go in columns 25-26. Using edge logic to see how this affects row 29, you can see it will cause row 29 to be invalid. So the 3 clue must go in columns 11-13. A lot more line logic.

Now only the the upper left box, lower right box, and R25C6 are left to fill in. Look at column 6. Either R12C6 is the 1 clue or it is part of the 2 clue. You can use the summing technique to KNOW that it is part of the 2 clue. Columns 1-5 require 10 cells to be filled in. Rows 6-11 require 11 cells to be filled in. This leaves 1 cell for R6C6-R11C6. If R12C6 was the 2 clue, then you would have to have 2 cells filled in R6C6-R11C6, but this cannot happen. So R11C6 gets filled in as R12C6 has to be part of the 2 clue. Line logic finishes it out.

Thanks for taking the time to look at this Gator. You are right, that is elaborate! And very well thought out. But that is really beyond my working memory capacity to figure out in my head without marking any squares. On the other hand, I CAN see the look-ahead moves I described in #8 very easily. My question to you is this: Is my way correct or incorrect? I've tried it twice, and it works out nicely with simple line logic once the left side diagonal is started.

I'm also going to mark this one as solvable with logic, but it definitely fits in the "unique solution, much lookahead required" category that Jan has proposed.
http://webpbn.com/read.cgi?type=T&id=215&what=allrep

Found to be logically solvable by Gator.

I was trying really hard to figure this out logically, (because I hate to guess), but I was getting so stuck that I finally looked in the comments, and wow! I was gamely trying to follow along until I got to Gator's explanation, and then I just gave in and assumed that the diagonals were where they were supposed to be. :) Definitely not typical difficulty for one of Teresa's puzzles, but challenging, and made me think of the "South Pacific" musical. I thought it was going to be something about Bali Hai. :)

Oh I love the music from South Pacific. I remember the original movie, I was only 11 or 12, and very impressionable. Our high school chorus performed all the songs from South Pacific one year.

http://www.youtube.com/watch?v=81NROmUb7o0

i enjoyed solving this :-) very nice image, also.

Fine puzzle. No rating from me on solvability, when a puzzle gets very complicated (for me, that's pretty much anything beyond simple line logic) I almost always guess. Have to admit I guessed quite a bit in this one, but I did see the diagonals and followed them.

As I was solving this (terrific!) puzzle, basically from top right to bottom left, I saw what looked a lot like a WWII fighter plane, as used in the Pacific(!).

I like what it really is better.

Goto next topic

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