14



11

I have As shown here: Note: the position is easily seen to be legal, as the knight can make back and forth moves while the black pieces get into position.


11

Would a physical puzzle of some sort be appropriate? I'm thinking of something like a maze, with a lot of endpoints, but missing the middle section. Each user is given a piece of card with the centre of the maze on it, but each user's card is different, and leads them to a different endpoint. The endpoint could just be a password/phrase to be communicated ...


7

UPDATE #4: After some asking on Matplus a few weeks ago, I found that my 42 was already beaten. But I have an ego. It feels so good to have come so close! Anyhow, here’s the “new” records. Enjoy! Without Promotions-43 The 43 Mates: With Promotions-47 The 47 Mates: I always knew that three pawn promotions. I just couldn’t figure it out myself. The same ...


7

The answer to a puzzle can be reduced to either being correct or incorrect. Thus you can make almost any puzzle have two outcomes by simply giving a correct option A and an incorrect but persuasive option B. The real issue is if the puzzle can be undermined if the solver simply tests both options; essentially reverse engineering the puzzle (e.g. a typical ...


6

For both of the puzzles you link to, unless I am mistaken, the following is true: Define the "N-restricted" version of the problem to be what you get by replacing "a random positive integer" with "a random positive integer <= N" and assuming that different integers are chosen independently. Then the limit, as N tends to infinity, of the answer to the N-...


6

The answer is Moreover, First note that Step 1 Step 2 Step 3 Step 4


6

Not sure if you meant something like this, example puzzle: The King's Gambit: Chess opening that begins to be popular again: 1. e4 e5 2. f4 (White is weirdly offering a pawn) Now when you know how message works, you can try it yourself. Or I should be describing hide and seek? Sorry for format, hope it is enough. And solution for this is obvious: ...


6

Here's one: HAISU with a twist and another: https://patents.google.com/patent/US5324037A/en A completed, solved game puzzle 10 is a mobius strip made of multiple columns 24 and rows 22 of blocks 20, or block-like pieces. Each block 20 has opposing display surfaces and opposing connecting surfaces which are orthogonal to the display surfaces. The ...


5

The results for both puzzles are: and an incomplete one: The cipher is a variant of: But: A note for the alphabets:


5

Here is 36 (the 37 from the OP has flaws, see comments), contributions given by: Picture:


5

The interesting thing about anagrams is how their difficulty increases rapidly with the insertion of even a single new letter. For example, a short anagram like WROC can be solved within a glance. This is because of the number of ways of ordering the letters is the factorial of the number of letters: there are $4!=4*3*2*1=24$ possibilities to order the four ...


5

As far as I know, they are called Operator Search Puzzles. You can refer to this link: https://brilliant.org/wiki/arithmetic-puzzles-operator-search/ The "24_puzzle" is an extended puzzle of this category. PS: I am new to SE, so any constructive criticism is much appreciated :)


5

I think "error" can be an anagram indicator, but there are some caveats. Run Time Error (4) I would be fine with this. "Error" is applicable to the previous word in the same way that "news organization" might give SEWN; it's an error of the word "time", so a "time error". (The strictest of Ximenians typically don't allow nouns as anagram indicators at ...


5

Simply put, a hint should give you more information than you already have. As you say, though, a hint that simply gives the solution isn't a good one, but that doesn't mean a hint can't give too much information. It just means a hint can't give you the information you want. As an example, this is what I would probably consider to be the best hint on the ...


4

The issue with a cipher puzzle is generally not being "too broad", but being solvable at all. A cipher usually isn't too broad. As long as it's not a one-time pad, it probably has a single clear best decoding. The problem with ciphers is that they are intentionally made to be hard to solve. The point of most ciphers is that nobody can decode them easily. So ...


4

On the popular British TV gameshow, "Pointless", they often have a round that has a category and then a series of the words that fit that category in anagram form (presented as English words). In this case the anagrams are of varying difficulty, but with the presence of the given category, this gives clues to the players. The way I tend to attempt them is to ...


4

There are a lot of ways to make words from other words. PART is TRAP backwards. PLAIN is PLAN with an additional letter inserted in the middle. BARRING is BAR plus RING. MONDAY is DYNAMO with the letters rearranged. Many rebuses require the solver to first recognize the words in the picture, and then manipulate those words in some way to produce the answer. ...


4

You can get a bit more excitement from the letters if you assign each number to a range, albeit an irregular one. So 1=A, 2=B, 3=C,D,E, 4=F,G, 5=H, 6=I,J, 7=K, 8=L, 9=M,N,O,P,Q, and 0=R,S,T,U,V,W,X,Y,Z. So for example PUZZLING STACK EXCHANGE becomes 90008694 00137 30351943. To decode, reverse the process. With some of the letters fixed, guessing the ...


3

The cipher is decoded by The answer to the first puzzle is: The answer to the second puzzle is:


3

Here are three puzzles with $3$ visual images, each being $2\times 2$ Raven matrices: This last one is a little more unique, but quite easy like the previous ones: I cannot find such a puzzle with just $2$ visual designs. I guess it is unlikely because the more objects there are in the puzzle, the more specific the pattern. Like, I might have the sequence $...


3

I have 35, with the following: Picture: The secret to the moves here is:


3

$p_{n+1}$ is defined as the lowest prime factor of the number $2^{p_n}+1$. Let’s try something: The question is, Sample starting points: Hmm.... Let’s use It naturally follows that


3

Simplicity isn't a bad thing Don't get caught up in wanting to make your puzzle as complicated as possible while still being solvable. This puzzle is lent a great deal of uniqueness by the fact that they'll be putting it together over the course of 18 months. Don't go overboard in making it more than it needs to be. I don't know what level of puzzle your ...


3

How about a simple one and the hints are words instead of letters? (And these words hide the letters for the real words.) Here is one possible way. Suppose you want to hide a word AVOCADO. There are $7$ letters. You then need to create $7$ words that start with letter 'A', 'B', 'C', up to 'G'; and consecutively end with letter 'A', 'V', 'O', up to 'O'. ...


3

I'll give some info but I'm not sure how good it will be for making a puzzle. I know this site loves graph theory, so maybe there is something there. This picture is of a Youngs' Ladder and comes from my research into the Heawood conjecture. Here are the properties (In this case n=7): The numbers {1,..,3n-1; 3n+1} are present. The three edges on ...


2

I think because


2

Unsure... partial Edit:


2

If you're looking for a simple and pure visual puzzle, you can use: "Spot The Twin" a.k.a "Find Two Identical Image" Puzzle It ranges from an easy one like this... ... to an intermediate one... ... or even to a hard one... ... or if you want to make it in a single style, you can use something evil like this ;)


2

How about pre-constructed Minesweeper positions? As for the desired features: 1) + Pure logic 2) + Purely visual 3) + Arbitrarily scalable (consistency checking proven to be NP-complete) As for the disliked features: 1) + No words required 2) - Almost, but not quite, everyone already knows minesweeper


Only top voted, non community-wiki answers of a minimum length are eligible