Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[theory]

A great logic puzzle that took me a bit of time to work through:

http://www.sporcle.com/games/dm21uk/codebreaker_logic

[liopoil]

that was tough! thanks After each new tile I uncovered I thought it would all quickly fall into place, but it didn't.

[Thisisnotasmile]

Took me 2 large post-it notes and a few refreshes.

[Kuildeous]

Yeah, that was a lot of fun. I want to see more like that.

I did have to guess a couple of times unfortunately.

I really couldn't quite get the prime-number clue to yield anything useful for me. I'm sure it presented some unique situations that really let pieces fall into place, but I just couldn't see them. Knowing the count of columns with exactly 1 prime didn't help me as much since I didn't know how many primes the other two could have. The upper bound of primes didn't help me so much, since by the time I got that clue, I already knew that the number of primes couldn't exceed that.

[theory]

Yeah, that was a lot of fun. I want to see more like that.

I did have to guess a couple of times unfortunately.

I really couldn't quite get the prime-number clue to yield anything useful for me. I'm sure it presented some unique situations that really let pieces fall into place, but I just couldn't see them. Knowing the count of columns with exactly 1 prime didn't help me as much since I didn't know how many primes the other two could have. The upper bound of primes didn't help me so much, since by the time I got that clue, I already knew that the number of primes couldn't exceed that.

There are only 9 primes.  That means that the columns have to be 1/1/1/3/3.  That in and of itself is not huge, but it's really powerful when combined with the Fibonacci clues.  Once your top row gets set to 2/3/13, it rules out a lot of locations for the primes.

[Kuildeous]

Ooh, yeah, I didn't actually lay out the number of primes. That clue does lay out that you could only have 1/1/1/3/3. If I had considered that, then I wouldn't have looked at a column thinking, "Well, that could have 2 or 3 primes."

The Fibonacci clue was quite helpful.

I may have to redo this puzzle with the mindset that a square is solved only when I can rule out all other possibilities. It'd be interesting to see the progression of clues.

[Watno]

How is one supposed to do that in 20 minutes? I think i took like an hour and misstyped quite often (no logical errors, though)

[Kuildeous]

I could see someone doing it in less than 20 minutes, but his logic would have to be impeccable. Now that I know the solution, I have no doubt that I can follow along in under 20 minutes.

And I'm definitely not the type of person who can figure this out from scratch in that short a time. That's for someone with a sharper mind than mine.

[eHalcyon]

The bottom row bothered me because M is the 13th letter which is still in the first half of the alphabet. But I guess they meant out of their subset without Z.

[Thisisnotasmile]

The bottom row bothered me because M is the 13th letter which is still in the first half of the alphabet. But I guess they meant out of their subset without Z.

I ran into the same issue when solving it, but afterwards I checked the google docs solution and there is a way to rule out the M before filling in that row.

[eHalcyon]

The bottom row bothered me because M is the 13th letter which is still in the first half of the alphabet. But I guess they meant out of their subset without Z.

I ran into the same issue when solving it, but afterwards I checked the google docs solution and there is a way to rule out the M before filling in that row.

Whaaat.

I will check that later then.

[eHalcyon]

Ahhhh.

I did the bottom row before working out the letters of the row with M.

[theory]

Here's a question I have that isn't related to this puzzle in particular.

Has anyone designed a logic puzzle whose central deduction is a meta-deduction?  That is to say, a logic puzzle which is logically deducible only because you are foretold that is logically deducible, and therefore at a certain otherwise undecidable point, you are guided by such meta-knowledge.

[Kuildeous]

The bottom row bothered me because M is the 13th letter which is still in the first half of the alphabet. But I guess they meant out of their subset without Z.

I'm not sure I see where the problem is. By that clue alone, M could be in that row, but M gets ruled out later.

The clue is technically correct, even when acknowledging that M is the 13th letter.

[GendoIkari]

Done. Never made any mistakes or entered the wrong thing, but it definitely took me longer than 20 minutes. Can't say how long because I did it off and on over the last several hours.

Definitely the hardest part for me was the prime numbers bit... figuring out where exactly that narrowed it down took quite a bit of thinking.

[GendoIkari]

Ooh, yeah, I didn't actually lay out the number of primes. That clue does lay out that you could only have 1/1/1/3/3. If I had considered that, then I wouldn't have looked at a column thinking, "Well, that could have 2 or 3 primes."

The Fibonacci clue was quite helpful.

I may have to redo this puzzle with the mindset that a square is solved only when I can rule out all other possibilities. It'd be interesting to see the progression of clues.

Wow, I also totally missed that no column could have exactly 2 Primes! That would have made that step quicker.

[Kirian]

The bottom row bothered me because M is the 13th letter which is still in the first half of the alphabet. But I guess they meant out of their subset without Z.

I'm not sure I see where the problem is. By that clue alone, M could be in that row, but M gets ruled out later.

The clue is technically correct, even when acknowledging that M is the 13th letter.

I ended up solving it in the same order as the Google doc, so that clue didn't bother me since you already know the fourth row is L-M-N-O by that point.

[Titandrake]

Here's a question I have that isn't related to this puzzle in particular.

Has anyone designed a logic puzzle whose central deduction is a meta-deduction?  That is to say, a logic puzzle which is logically deducible only because you are foretold that is logically deducible, and therefore at a certain otherwise undecidable point, you are guided by such meta-knowledge.

I think there are Mystery Hunt logic puzzles that are kind of like this. Where for instance, the puzzle has a unique solution only because you know that a certain block of squares must be filled in such that it forms a valid Braille letter.

I'm not sure if using meta-deduction is even possible; the puzzle has to only have one valid solution, so at that undecidable point you could try all possibilities and see which ones lead to a dead end. Doing that is guaranteed to solve the puzzle, but never uses the meta-deduction.

[Galzria]

I ran out of time typing in my final box. >.>

However, maybe it's just the way I work, I started out before I clicked "play" by writing out 1-25 in a 5x5 grid, as well as A-Y. I then scratched them off my list as I was able to place them in the grid.

[Galzria]

I ran out of time typing in my final box. >.>

However, maybe it's just the way I work, I started out before I clicked "play" by writing out 1-25 in a 5x5 grid, as well as A-Y. I then scratched them off my list as I was able to place them in the grid.

((My last 3 Cells were 6, 3 & 4. While I had the numbers on 3 & 4 and the letter on 6 early, pairing them up with their match as the clock was ticking down was scrambling my brain. Had I not paid attention to the clock I might've finished in time.... but I was feeling the pressure as it dipped below 00:30))

[liopoil]

Took me about 45 minutes. What stumped me a while was the clue from 8H, which was that the row with consonants only in the first half doesn't have a K in it. I immediately realized that had to be the bottom row, because that had the only vowel on the far right. From there I was just like, "Oh, so there isn't a K in there, whatever.", without realizing that it also meant that there can't be any consonants after M in it. I figured out that there must be L-M-N-O in the fourth row before that, so the M didn't confuse me. I also didn't start writing down stuff/making the grid until halfway through the puzzle, so that was a mistake.  The primes puzzled me for a little while. I wrote them all down, but I didn't get that there can't be any columns with 2 primes for a while. The Fibonacci clue and the alphabetical order left to right clue were by far the most helpful for me.

[Axxle]

I eventually got it, I should have written things down better since getting the answer to 14 was a chore.

[ConMan]

Here's a question I have that isn't related to this puzzle in particular.

Has anyone designed a logic puzzle whose central deduction is a meta-deduction?  That is to say, a logic puzzle which is logically deducible only because you are foretold that is logically deducible, and therefore at a certain otherwise undecidable point, you are guided by such meta-knowledge.

Sort-of. The self-referential aptitude test can be solved without assuming a unique solution, but doing so makes it easier to do so (I think you shave about 3 steps off the logic).

[gman314]

Here's a question I have that isn't related to this puzzle in particular.

Has anyone designed a logic puzzle whose central deduction is a meta-deduction?  That is to say, a logic puzzle which is logically deducible only because you are foretold that is logically deducible, and therefore at a certain otherwise undecidable point, you are guided by such meta-knowledge.

This puzzle is somewhat similar to the meta-deduction you asked about.

[DG]

Sort-of. The self-referential aptitude test can be solved without assuming a unique solution, but doing so makes it easier to do so (I think you shave about 3 steps off the logic).

This can also be applied to some Suduko puzzles. I don't know if it is definite rule of Suduko puzzles that they must have a unique solution but it makes sense to publish solutions that are unique.