Mystery Hunt 2017: The Puzzles

Jan 16, 2017 01:40

Here is a spoily summary of the puzzles I worked on and selected other puzzles, in order by round, rather than the order I looked at them. You might need to be logged into a team account to see the puzzles using the links.

Bigram Battleships (Fighter): This was a variant on the standard Battleships puzzle type, in which all the ships were one square longer than usual and instead of the regular clues, each cell contains two letters and each ship contained a valid word by reading its bigrams across or down. By the time I looked at it, people had already identified all possible 8 and 10 letter words and most of the 6s and 4s, but we were trying to find combinations that actually worked for each grid. Solved at 3:43 PM, but I don't think I was still working on it at that point.

The Fighter Meta:  When we had only two answers half an hour into the hunt, Charles suggested that this meta might have to do with the double letters both those answers had. This was remarkably on target, as each answer contained a sequence of four letters in an ABAA pattern, and the answers were all different lengths for ordering. The "B" letters in this order spelled out the answer. Solved at 4:12 PM.

False Prophets (Cleric): There were a bunch of pictures to identify. The correct words all take the form A AND B or A OR B where A and B are any single letters, or similar forms with three letters and two operators. The puzzle discusses two rival "orders" or clerics who thought some of the images were good, others should be forbidden, and they could not agree on the third. With the operators we quickly figured out that the orders disagreed on how to pricess expressions containing multiple operators, with one order evaluating the operator on the left first and the other evaluating the operator on the right first. We also figured that each letter of the alphabet was assigned a true or false value consistently by the two orders, and they liked the things that evaluated to true. I helped fill in a bunch of these, and then spent a little while trying to locate the spot marked on the map in California in one picture. Shortly before I determined that was Fort Ord, a teammate suggested that the answer might simply be an anagram of all the true letters, which was spelling CLAMP based on the ones we had figured out. Solved at 2:17 PM.

CHINA KILNS (Cleric): This clever puzzle had a list of clues and blanks for each, with one letter boxed and sometimes marked +1, +2, -1, or -2. I filled in ANTI-POSITRONS = ELECTRONS, and we quickly figured out that the answers were anagrams of other clues, and that we would chain them together in this way. For a moment we had people filling in the other clues in the blanks, ignoring the given spacing, but I pointed out the spacing and apostrophes in some blanks which matched the clue answers. Once about 2/3 of the clues were solved, counting letters helped us make most of the remaining matches without the answers, but by the time the chain was assembled we had most of those other answers as well. Solved at 2:44 PM.

The Cleric Meta: I did not actually work on this one. Charles suggested at 2:19 PM after we had both
Four-Part Harmony and False Prophets solved that he wanted to make WRITER'S CRAMP out of their answers WAITERS and CLAMP. At 2:26 he got BRAIN FREEZE and he wanted to backsolve the second  puzzle as BREEZE but it wasn't unlocked yet. At 2:53 we got that this was not just ailments by specifically the body parts from Operation. From 5 answers and their letters which were changed from the Operation body part this was solved at 3:18, and the other 5 puzzles in this round were solved immediately after they were unlocked, even the two in the row where we did not have any answers yet, by figuring out which remaining Operation part could have the necessary letter changes.

Net Work (Wizard): This was actually the first puzzle I was working on after the hunt opened. We quickly figured out that the answers were two-word phrases or compound words that needed to be entered in the spaces in the chart so that the directed edges connected the parts of the answers. Once we filled in all the parts, I and a teammate worked together to do the bookkeeping to figure out which edge was not clued, which (due to cluing provided in the puzzle) we knew would bethe final answer. Solved at 1:30 PM.

Getting Out of Line (Wizard): I worked on this one only a little at the start when we were trying to solve the clues, and transcribing the letters from the across ones into a sequence to spell the words actually needed. I wasn't still working on it when they actually made a crossword puzzle using the downs in no particular order except that some of them had a letter changed and these changed letters in the order the down clues were given spelled the answer. Solved at 5:46 PM.

The Wizard Meta: I didn't work on this one at all, but it looks like people pretty quickly figured out that the answers in this round were all things Harry Potter spells could do, and the initial letters of the spells in the order the puzzle was presented spelled the answer. Once we got the scavenger hunt answer at 4:27 PM, from the 4 answers we had the meta answer was guessed.

The Leaning Tower of Sheshach (Dynast): I was working on this one early on, helping to come up with the translations for the words and then brainstorming how we were going to assemble a tower out of them. It was pointed out early on that Sheshach is Babel encoded with the Atbash cipher (in Hebrew), but it was after I stopped working on this puzzle that somebody figured out a translation of each word was Atbash of a translation of another word, using some pair of the 6 languages used to encode this puzzle. Solved at 3:49 PM.

On Numbers and Games (Dynast): I started the work on this puzzle, identifying several of the games and pointing out that the years in the posts corresponded to the years the games were released, and the games were presented in alphabetical order while the number of likes presented a different ordering. And I pointed out the flavor suggsted nim was involved. A teammate figured out that the initials of the games in the new ordering spelled NIM SUM BGG IDs. Unfortunately this meant that we had to get all the games to solve the puzzle, and I moved on before that was done. Solved at 6:07 PM.

The Despondent Dynast Meta: Another meta I did not work on. It was at 3:26 PM that it was suggested this was a Mah Jongg meta, and very quickly those working on it found that each answer could be assembled into a winning Mah Jongg hand of a pair of like letters and three triples each consisting of three of the same letter or an alphabetic run of three, with one letter missing. Starting at 3:46 PM we started guessing answers with the ending FOURTH given ?????FO?R??  assuming the answer was Mah Jongg-related. FOURTH was confirmed just after this when Sheshach (see above) was solved to match the H. The 4th guess we submitted, at 3:56 PM, was correct.

Corporate Chains (Linguist): I started working on this one when we unlocked the linguist around 2:50 PM. This was a simple puzzle with several product slogans each with a word replaced. The slogans were given in alphabetical order of their products. We got most of the products and changed words quickly. We also figured out that the wrong words were synonyms of the right words in other slogans. Using this to form a chain and keeping the first one in place spelled another slogan, this time unmodified, for SKOAL which was the answer, solved at 3:02 PM.

Dot Matrix (Linguist): This puzzle was on 3 pages of old-school computer printouts given to us when we rescued the Linguist. Once photos of these pages were posted in our spreadsheet I helped transcribe data and solve clues. It was clear we were going to do something with Braille, but I had moved on before teammates figured out that the puzzle answers needed to be entered in the rows in Morse to make them fit. When the blocks of three answers were read off as Braille, they spelled the answers to the clues in the boxes at the top. The marked Braille letters in each block spelled one word per page, and these were entered in the rows of the last set of boxes as Morse to spell the final answer in Braille. Solved at 4:46 PM

Sooner or Later (Linguist): The last linguist puzzle to unlock, after midnight. By the time I jumped in here, most of the clues were answered but nobody had made sense of the sooner and later and red and blue boxes. I figured that out at 1:07 AM - each answer is made of a group of letters from the first half of the alphabet before or after a group of letters from the second half of the alphabet, and the colored boxes tell which half comes first in the word. At 1:12 AM one of our UK team found the message in the last letters of the first halves and at 1:16 PM it was solved.

The Linguist Meta: Charles was trying to convince me this was an IPA meta early on. I looked at it but didn't see where that was going. Hours later, looking at the set of answers we had, I recognized HOAX as a transdeletion of the language XHOSA. Some of my teammates were staring at me like "what is that mess of letters" but I explained for their sake it was a language and then solved ANTI as LATIN. Very quickly the rest of the answers we had at that point were extracted and we filled in the gaps to get the answer at 7:58 PM.

Marvels of the Ancient World (Dungeon): I didn't work on this puzzle, but I knew a group was working on this game reconstruction puzzle for the game 7 Wonders. After spending almost 4 hours doing this at 7:50 PM they found that the answer to the puzzle was SEVEN WONDERS. Really? Well, I guess if you make it this obvious nobody actually guesses that until they have solved the puzzle. :-)

The Dismal Dungeon Meta: The only thing I contributed to this puzzle was pointing out there were lots of Es and Ns, but teammates extended this to include Ss as well and to take them as compass directions. These were used to trace paths through the dungeon, with any non-compass letters in the answers matching letters at the intersections of the given dungeon map (with one column of letters filling in for each solved puzzle). We figured it was something to do with the starting and ending spaces, but with needing the answers to get the directions and also to get columns of letters, it took many answers. Fortunately we solved a lot of the puzzles from this round quickly, though, so we had 8 of 11 answers by the time we solved it at 7:24 PM.

Ye Olde Seventhe Duck Konundrum: The Cones of Duckshire (Economist): I sat down with three other players to play out this game around 5 PM. Since I knew it was a Dan Katz hunt, I had packed Team Luck's old DK solving kit, including a bag of multi-colored mini poker chips, two decks of playing cards, a bag of about 700 letter tiles, a 24x24 board for playing letter tiles on, and a duck (a yellow rubber one that you might take into the bath, if you are into that kind of thing, like Ernie). After looking at this one I left the cards and board behind, and initially the letter tiles, though I soon went back and picked out letter tiles matching the initials of the colors to use as the cones. We sat down and played straight through it with never needing to backtrack more than one turn, having written down (among other data) who used their powers on who so when we got to the clue that needed that at the end of the game we could look that up. Solved at 6:09 PM.

The Economist Meta: I did not work on this meta at all, but when somebody figured out the answers were all things found on the backs of the US state quarter series from a while back, it was solved quickly at 8:31 PM.

Rulers (Thespians): About 6:45 PM I found some of my teammates had started on this one, cutting out the rulers and trying to match them between dots, while also solving some of the little rebus puzzles. I first solved most of the remaining rebuses they didn't already have. After a while, when we matched one rule to the dots for LION and HEART, I suggsted this was RICHARD, which matched the seven inch length of this ruler. We didn't recognize the other epithets, but we were able to look them up. After solving most of them and extracting the letters from the monarchs' given names, we thought we were looking at UNREADY INSIGHT and while UNREADY was ETHELRED, we weren't sure what to do with the other half. Finally I solved the unsolved rebus (BIG) and we confirmed that name to make the clue UNREADY IN EIGHT, telling us to use the spelling ETHELRED and not AETHELRED. Solved at 7:23 PM.

Changing Rooms (Thespians): This was one of the last puzzles solved before we found the coin. I missed the first part, which was a cryptic crossword variant with two un-enumerated clues per row and a set of "room" clues whose answers snaked around the grid without overlapping. Also, each clue's wordplay gave an extra letter. The extra letters from the across clues spelled STAR BATTLE so we then proceeded to solve that puzzle type using the room clue placements as the regions. This was the part that was happening when I looked at the puzzle, and I tried to solve it but I didn't finish it before going to watch our final battle. Eventually, when that was solved by remote members, the clue from the extra down letters, EVERY FIFTH IS NEXT TYPE, applied to the letters on spaces matching the stars in the star battle, gave LITS, so now the same regions were used to solve a LITS puzzle. The spaces used in the LITS solution which had stars in the star battle solution extracted letters from the cryptic spelling the answer, solved at 3:37 AM.

The Thrilling Thespians Meta: When I started looking at this one, teammates had already formed HAMLET and MACBETH by placing a letter between left side answers and right side answers, and guessed that the left side answers started with A through E for ordering. They also had guessed LLOYD GEORGE formed OTHELLO with an unsolved left side answer, but I pointed out that the right side answers ended with A through E and they should order them that way, so LLOYD GEORGE had to go with EMAIL SCAM. Some research turned up CAMILLO from The Winter's Tale which and we came up with CALIBAN to go as the first one, which let us guess it was spelling IAMBI (the plural of IAMBUS, a variation of IAMB, the metrical foot, as well as a sentence I AM BI which answered the question posed in the flavor text) at 7:40 PM. Note that we solved the Dismal Dungeon 16 minutes earlier, so we had a bunch of puzzles open at this point.

The Most Interesting Puzzle in the World (Chemist): After finishing Thespians, sin_vraal and I were looking at this puzzle together just after it opened. The title and picture in the meme used in this puzzle referred to the "most interesting man in the world" ad campaign used by Dos Equis beer, and he guessed the answers would all contain double X. That was half right; one set of clues gave such answers, while the other set gave answers in alpha order which matched those answers but with other letters in place of the Xs. This led to an instruction to bring them root beer, and when another root beer procurer on the team failed, I was sent out to locate some. Since it was near our Stata headquarters, I looked at the vending machines at the corner of 26 & 56, none of which had any root beer as an option. I remembered the two sets of vending machines in the basement of the infinite corridor from the solution to last year's dog show runaround and went there but they were not there any more. So I went to Laverde's and bought the root beer. The resulting answer was finally called in at 8:12 PM.

The Chemist Meta: We had a bunch of answers already when somebody figured out all the answers were made of two atomic symbols with an O between them. The first hypothesis was that the difference in atomic numbers of the answers was going to turn into another atomic symbol to spell an answer, and Charles had me code up a list of possible answers. But the extraction wasn't looking good, and they figured out it was the average instead, at which point we submitted the answer at 10:00 PM with only the second-last and third-last puzzles unsolved. We backsolved Spontaneous Reactions as Kona from this, but we failed to backsolve 12-Step Program even knowing that the answer had to use two elements that averaged to Einsteinium because nobody recognized Bhopa as a thing. (I was going for T lots, i.e. the parking lots at T stations).

At this point we had all 6 character metas solved and we unlocked The Big Bad Battle, a cool interaction where six of us got to play a board game, moving our tokens around a large hex board to reveal parts of a message which was written in very distorted letters made with lines between the centers of some of the hexes. In between moves we got to play a trivia game where each set of answers related to one of our characters' special abiltiies (which were their metapuzzle answers). In each round of trivia we had to give 11-N answers in 60 seconds, where N was the number we rolled on a d10. Everybody could help answer questions, but only answers given/repeated by the player in charge of that round counted. The first of these rounds we rolled a 1 so we had to answer 10 questions. We managed to get the 10! So when I was up as the chemist and rolled at 8 I was feeling really, really good. We never failed a round of trivia and after four rounds I figured out the answer, submitted at 10:56 PM.

The Broken Bridge Meta: Somehow I didn't look at any of the puzzles for this round, but teammates solved the meta, figuring out that each answer clues a four-letter word and those words, when properly arranged, form a long word ladder with four gaps which could be filled in with the words HUGE DUCT TAPE ROLL, the hilarious answer submitted at 12:08 AM.

Oh, You! (Criminal): My teammates had already decoded the Caesar-shifted clues for this word search and solved them to get a bunch of words starting with OVER or UNDER. And they found the ends of each answer in the grid, Caesar-shifted the same as their clues, and took the letters above or below these answers in the grid. I was there when somebody started Caesar-shifting these letters back by the same shift the word above or below was shifted and helped fix an error in the message, to solve the puzzle at 9:59 PM.

The Crafty Criminal Meta: Early on I made the observation that all of this meta's answers (that we had to that point) consisted of a 6-letter word and a 3-letter word. (Ultimately, some of the second words were of different lengths) Somebody else later figured out that the letters n the second words were from a restricted set, and then identified the set as the letters in BLACK or WHITE which are disjoint sets. These provided the black and white pegs for a game of Mastermind, which took some time to solve with the non-dictionary answer PEGMAN (apparently a nickname some people use for the Google Street View icon you use to position yourself n the map).

Stackuro (Minstrels): I proposed at 8:48 PM that this was a Siamese twins Kakuro in which the given clue numbers were products of the two sums for each entry in the respective solutions, and this proved correct. I came back later to help finish this one. Solved at 11:16 PM.

Chromesthesia (Minstrels): I transcribed some of the colors from the letters and I also found the message hidden in the black space. One of my teammates had the tools to analyze the 16-bits-per-channel PNG, which had additional data in the low bits, so I moved on from this one before they managed to feed the file to some sort of speech codec to get audio out of it, and tried to solve a puzzle from that, but ultimately backsolved the answer as the only sound-related word from a list of possibilities matching what was needed for the meta, at 1:57 AM.

The Modest Minstrels Meta: I looked atthis, but did not really contribute. Teammates had already figured out that the songs were in different keys or "modes" and the order of those modes. It took a while to figure out because of an error in the puzzle we managed to work our way around (which was later fixed with an erratum on the round page), but it was solved at 1:49 AM, triggering the private versions of events as we had all the quest metas solved and enough points to open the endgame once the events were done.

The Puzzle at the End of This Book (Cube): I went out to retrieve the copy of the puzzle, which was a small hardcover book which was Setec's own version of The Monster at the End of This Book. I noticed some of the stuff in the book but handed it over to teammates to cataloged the Morse Code messages on each page, the large green words, the hidden Braille letters, the fonts that the red word Puzzle was written in, and the semaphore Grover was doing with his arms on each page. They also identified the characters at the end of the book, and knew they needed one more hidden message and gave me a shot. I found it with hidden Pokemon names in the text on each page and they proceeded to extract the answer at 12:58 AM.

The Curious Cube Meta: Charles and I both thought that we were meant to dissect the grid into cube nets, but it seemed underconstrained. Charles eventually figured out that since we were meant to enter the answers from the starred squares, that each star should be on a separate path and only cube nets which are path hexominoes should be used. Now that we could cut out, fold, and label 8 small cubes, I suggested the colored squares with 2 per color were meant to be matched face to face (inspired by a similar clue in the SUMS 2016 puzzle Psychosomatic) and teammates figured out this would form a 2x2x2 cube. When assembled, this led to a cube with UP UP on one face and DOWN on another. The answer read around the other faces, solved at 1:14 AM.

Open and Shut Case (Warlord): I looked at this one, recognizing that these were snippets of the particular variety of javascript where the type coercion rules are abused to allow you to write programs with only punctuation marks. In this case, a long program of this sort had been broken up into 26 separate files named with letters of the alphabet. We were told certain ones went before other certain ones but otherwise we were on our own to find the order. I wrote some spreadsheet formulas to count the excess or deficit of opening/closing brackets and parentheses in each file to attempt to limit the possibilities. However, I also summed those over all the files and ended up with the totals not balancing, which confused us. Ultimately, this was backsolved as needing to be of the form --XX and programming related, and I guessed REXX which was correct, at 1:48 AM.

The Woeful Warlord Meta: I did not work on this one, but Charles determined that the four-letter answers with a high frequency of Xs and Os were giving four separate tic tac toe games with two Xs and an O already played, the way the puzzles were laid out on the screen. Using only those letters, each game had a forced move by O to avoid letting X win, and another such move by X after that. Extracting the letters that fell on these spaces spelled the answer, solved at 1:34 AM.

The Maniacal Merchants: The penultimate map section opened up a puzzle directly on the page. This was a meta using the bits of equipment that many of the other puzzles had given us. I didn't work on it, but it was solved at 12:51 AM, replacing our inventory with 6 enchanted items, one of which clearly corresponded to each character.

Events: After we finished all the metas, Setec called us to start arranging private sessions of the three Saturday events for us so we could get unstuck and move on to the endgame. The big-head-racing event reminded me of those math puzzles that ask you for the minimum number of races needed to determine the ranking of N racehorses if each one always runs at the same constant speed (no two horses at the same speed) and K of them can participate in each race. In this case, the "horses" were people wearing big head costumes like are seen in a break between innings at several major league ballparks, though the heads were ones of famous MIT alums, and the heads in each rase were selected for us, but we needed to be able to predict the correct order for all four competitors in one race to win. In our private version we got the announcer telling us what happened but no actual head-racers. After the first two races, we had three possible orders for the 6, and since one of the uncertain ones was not present in the 3rd race, we had a 2/3 chance of our pick being right in that race, and it was. The hungry hungry hippos played differently with just one team present. They gave us a time limit to collect a certain number of balls and then to spell words with them. The pub trivia worked the same as in the real event except with nobody else there, and they still made us wait the full 2 minutes after each round, and we got our 9 correct answers by round 4 which was the soonest practically possible.

The Final Foreboding Fortress: We went off to a room and selected 6 members to play the roles of the characters. The sorcerer quizzed players in turn with a question that could have many possible answers, but we needed to find the one which matched the quirk for that character's answers in The Big Bad Battle. Just one of these for each character, and we could keep trying until we got them all. We got 5 of the 6 on the first try. Each time we got one, the hexagonal pad (think board game hex blown up to human pawn size) that character was standing on was flipped over revealing one or two digits. It them took us a little while to figure out how to use the digits. Clearly it was an ordering, and we wanted to use the + or - enchantments on the weapons here, but what took time was figuring out what we were shifting. Finally we realized we should be shifting the letters from The Big Bad Battle's final answer and this was solved. We got our runaround instructions, which were straightforward-ish but with a fantasy bent on things, so a line of trees was a line of ents, etc. I did realize that they were taking us way off to the east side of campus, leading me to say more than once, "they hid the coin way the $#%^ over here?" but indeed it was there, taped to the side of one of the large diamond-like art pieces near the Sloan building, E62, found at 4:23 AM.

mystery hunt, puzzle hunts, mit mystery hunt, puzzles

Previous post Next post
Up