Web Paint-by-Number Forum
Comments on Puzzle #9357: Another one of my signs.
By Tom O'Connell (sensei69)

peek at solution       solve puzzle
  quality:   difficulty:   solvability: moderate lookahead  

Puzzle Description:

#1: Billie Patterson (bpat) on Jun 24, 2010

Is there a way to solve it without guessing? I had to guess at the step pattern in the bottom left.
#2: Tom O'Connell (sensei69) on Jun 24, 2010 [HINT]
you have to do edge logic on c-20
#3: Susan Duncan (medic25733) on Jun 24, 2010
I solved it using edge logic. Nice image Tom
#4: MarioMax (MarioMax) on Jun 24, 2010
What is it?
#5: Gator (Gator) on Jun 24, 2010 [HINT]
I would call this more like affirmation logic than edge logic. Edge logic only let's you make R20C20 a dot. This let's you fill in R14C20. Now if you look at every possible place the 6 clue can go and you see how this will affect column 19, in every case R9C19-R11C19 will be dots and R12C19-R13C19 will be filled in. You can now do a little more line logic.

This still doesn't get us to the end though. I'll work on this some more tonight.
#6: Cynthia Lynn McDaniel (mcaardva) on Jun 24, 2010
I should put this sign on my computer at work. Except that I work at a church...
#7: Tom O'Connell (sensei69) on Jun 24, 2010
lol Cynthia
thx Gator
#8: Teresa K (fasstar) on Jun 24, 2010 [HINT]
Gator, I got stuck after filling in R14C20, with a dot in R20C20. If I look at all the possible placements of the 6 in C20, there are six different possibilities. I don't think that is looking ahead one or two moves, but it requires holding a lot more info in short term memory than I am capable of. I had to use trial and error just to follow your hints.
#9: Merili (merilinnuke) on Jun 24, 2010 [HINT]
I also used edge logic with the 7 & 5 at the bottom.
#10: Jay Hughes (BFavreHOF74) on Jun 25, 2010 [SPOILER]
I wish everyone would "Stop screwing around" and thank Tom for a good puzzle.
#11: Trish (tryingmysoul) on Jun 25, 2010
Oh, I finally got it. It's a rebus.
#12: Tom O'Connell (sensei69) on Jun 25, 2010
thx jay lol
#13: Gator (Gator) on Jun 25, 2010 [HINT]
OK, here is a different way to tackle this one.

After the initial line solving, look at the 7 clue in row 19. It can be in columns 9-20. Starting at column 9, it would have to be in R19C9-R19C15. If you look how this affects row 18, it would cause R18C11-R18C13 to be black which makes row 18 invalid. So R19C9 cannot be part of the 7 clue (but it's not a dot as it could be part of the 3 clue). Moving right, the same logic will make R19C10 and R19C11 not part of the 7 clue. This lets us make R19C16-R19C18 black. More line solving.

Again looking at row 19, we have two 3 clues to place. If they are to exist all the way right as possible, they would have to go in columns 6-8 and 10-12. Looking at column 18, we can see that the two blacks already there have to each be part of a 3 clue. Now look at placing the 3 clue in row 19 in columns 10-12. Notice how this would make R18C11 and R18C12 be black. But this makes row 18 invalid as those two cells could not be part of the 3 clue and they are not the 1 clue. So we know that both 3 clues of row 19 cannot be on the right and one must be on the left. So R19C2 and R19C3 are black. The rest solves normally.

Very challenging puzzle Tom!
#14: Web Paint-By-Number Robot (webpbn) on Jun 25, 2010
Found to be logically solvable by Gator.
#15: Tom O'Connell (sensei69) on Jun 25, 2010
thx for your patience Gator, i worked long & hard on it :)
#16: JoDeen Mozena (ozymoe) on Jun 25, 2010
Liiiiiiiiiked this one verrah much! Thank you, Tom!
#17: Tom O'Connell (sensei69) on Jun 25, 2010
thx JoDeen
#18: Joe (infrapinklizzard) on Jun 25, 2010 [HINT]
Another way to solve it:
After line logic, edge logic on the 6 in c20 makes r20c20 white and r14c20 black.

(As per Gator) the 7 in r19 is limited by the 3s in r18. Straight line logic says the 7 could be anywhere from c9 to c20. However, if it is in r9, 10, or 11, a block of four will be forced in r18. This cannot be, so the seven must start no earlier than c12. This makes r19c16-18 black.

Now I did edge logic on the 5 in r20. Again, straight line logic says it could be in c8 to c19. However, if it starts in c8, 9, 10, or 11 it will cause a conflict in r19. Thus limited to c12-19, r20c15-16 must be black.

The rest is line logic.
#19: bugaboo (bugaboo) on Jun 26, 2010
great solve
#20: Tom O'Connell (sensei69) on Jun 26, 2010
thx bugaboo
#21: Teresa K (fasstar) on Jun 27, 2010
I went back and started over. I think I had a mistake the first time, because the second time, I was able to do it with not much difficulty.

Really cool puzzle, Tom. (Thanks, Jay, for explaining the meaning.)
#22: Jota (jota) on Jun 27, 2010
Nice!
#23: Jane Doe (telly) on Jul 6, 2010
I had to read Jay's comment to get it...thanks. What's the little h at the bottom?
#24: Tom O'Connell (sensei69) on Jul 7, 2010
its a lil n
#25: Edith Clark (eclark) on Jul 19, 2011
Cute, very cute. Not sure if this is what you guys do for edge logic, but after I got the top done, I was stuck. So I started putting the 6 black pixels in column 20 in each possible place until I could use color logic and not get an error. Do you mean when I use edge logic I'm supposed to remember what will happen where before I start filling in? I'm sunk.
#26: Teresa K (fasstar) on Jul 19, 2011
Edith, read the FAQ on the left, then click on the link for Advanced Puzzle Solving Techniques at the top of that article. Those two pages will be really helpful.
#27: Edith Clark (eclark) on Jul 21, 2011
Thanks, Tom. I think that is more or less what I do. :)

Goto next topic

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