author's notes for dahan konundrum

(minor spoilers for first you visit, major spoilers for dahan konundrum (but you're reading the author's notes, what would you expect))

let's start from the very beginning.

shape the self anew + boon of reimagining

back on february 10th, 2022.
patrick reached out to me asking if i wanted to think about a spirit island puzzle. of course i did.

that night, we decided on how the puzzle should work. we settled on one game, 3 players playing low complexity spirits, using basegame only. i spent a few more hours creating a game we could use. we spent the next few days verifying that the game was fully working.

on february 12th, i changed my minecraft username to include my current name. this may not have any clear ties to the puzzle, but it ends up making a pretty important change down the line, and it also motivated me to continue working on this puzzle at the end of summer. later that night, i ended up playing the best game with starlight that i have ever played. i hit all 9 of starlight's innate power thresholds. thinking back to that night, it gave me a really good euphoric feeling for the entire next week.

after checking over the game, patrick told me that he'll handle the script-writing part of this puzzle. i didn't really need to do anything for the rest of the month.

sometime around the beginning of march, we reconvened to work on the puzzle again. we've already somewhat forgotten what happened in each turn, though i had my diagrams of the island that i drew while creating the game, as well as the spreadsheet notes that i had. we decided on how the script should work, and patrick threw together the script. i went through it and made sure that everything important was covered, which was pretty difficult because a lot of details near the end were left out. patrick compared it to a 3-handed game, but i thought that solving the puzzle was much harder than playing a 3-handed game.

in the end, we were ready to share this puzzle at teammate's puzzle potluck event on march 5th.

patrick decided on the name “dahan konundrum” around that time, and the name hasn't changed since. i did consider renaming it to “space invaders” at some point, but i think dahan konundrum would be much better at catching the attention of all spirit island enjoyers doing the hunt.

nobody got anywhere on the puzzle. one team tried it, using tts, though it was apparently very difficult to use because tts's undo function sucked.

after the potluck, we decided that this was a good size to keep the puzzle at for it to be in line with a normal duck konundrum. we'll see how that went later, as we ended up not working on this puzzle for a couple months, because i became hosed.

on april 22, i spent a few hours creating a new minecraft skin. it was modeled after, and contained a lot of references to, starlight from spirit island, though it was also very heavy on pride flag colors. at this point starlight was pretty solidified as my favorite spirit.

lure of the unknown + enticing splendor

in june, we finally decided to reconvene. by this point, i had plans to change this puzzle into a puzzle with three games, each with increasing complexity, difficulty, and number of players. in particular, i wanted the first game to not have any expansion content and be a low complexity spirit. i also wanted to tie the games together somehow, possibly with each of them having the same things happening in the invader phase, such as fear cards and invader actions.

on june 9th, we finally got an editor for this puzzle. on that same day, i created game 1. it was with base shadows on board f, which was normally a terrible matchup. however, i didn't have to worry about that, because i was in full control of the rng and so could have an explore card miss at least one land whenever i wanted, as well as have shadows draw perfect-element cards in every draft. the game ended in the slow phase of turn 5, right when shadows achieved terror level 2. i drew the board state after every fast, invader, and slow phase, as well as made notes in the spreadsheet.

i tried to make this game compatible with russia 6. i think i got it mostly working, except for possibly the russia 5 fear bomb, as well as forgetting either russia 3 or russia 6 once. i thought about what events could happen on each of these turns, as well as whether i could make command beasts work, since then i wouldn't have to worry about events. i was pretty satisfied, and decided to write up the script later.

while i was writing the script, i found that the script worked best if i had a player playing the game and a spectator commenting on a few things that happened in the game, such as how lucky the player was with their card drafts. these characters didn't have names yet, so i named them “player” and “spectator”.

for the rest of the summer, i accidentally lost most of my puzzle-writing motivation for two reasons. the first one was because i was living at et, and i was worried about other people living at et seeing my notes and getting spoiled. the other reason was because i accidentally got addicted to vault hunters, and then later enigmatica 2: expert skyblock. (these are both minecraft modpacks)

on july 19th, my plans for this puzzle changed again, to instead having 4 games, with the first and last having one player and the middle two games having two players. this way, the 4 games could also have spirits increasing in complexity, and each game would be adding one expansion. my initial plans were to have the spirit in game 4 be starlight, since it's the least complex out of the three very high complexity spirits we had. in summary, my plans were for game 1 to be shadows with no expansions or adversary, game 2 to be keeper and fangs with branch & claw and no adversary, game 3 to be two high complexity spirits from feather & flame against a low-level russia, and game 4 to be starlight against russia 6. i still wanted the games to be combinable together.

i was given a rude break from my minecraft addiction at the end of july by an event called “my parents picking me up and taking me on vacation”. several places that we were traveling to didn't have very good internet, so i decided to continue working on this. however, i discovered that syncing fear between games 1 and 2 was really difficult, and i also didn't remember which cards were in jagged earth, which was important because i couldn't use those cards in game 2. i ended up getting a few things done such as the general turn structure of game 2, but it wasn't very good progress.

during that vacation, i had the idea of giving each player a name that kinda reflected what spirit they would play. i hoped that this would motivate me to work more on the puzzle, because i could name the starlight player lumia, after my minecraft persona. i named the shadows player paul. finally, since game 1 was a dialogue between two people, i made the non-player in game 1 the same person as the player in game 4.

the names of the spirits i had set were paul (shadows), bjorn (fangs), spike (keeper), fern (wildfire), riga (downpour), and lumia (starlight). all their names meant something related to the spirit they were playing. paul was a pun on pall, bjorn was bear in some scandinavian language, spike was partially a reference to brawl stars, fern was a substring of inferno, riga was a substring of irrigate, and lumia came from the latin root of lumen, meaning “light”.

most of these names persisted until after mystery hunt, but only two of them would persist until the puzzle was actually released in abcdefg. luna has had that name since december, but rae, dalia, and undine's names weren't changed to their most recent names until february 2023, which is after mystery hunt. i'll be referring to the characters by their names in the version that is released, since these authors notes will end up being read by people who saw the puzzle, and not just myself.

a bit later, i started trying out the name lumia for myself. its first public use was in the credits of gph 2022, though this didn't really proliferate into real life until mid-september. and now it's my public-facing name.

gift of living energy + spur on with words of fire + unrelenting growth

i finished enigmatica 2: expert skyblock on september 14th. if not for schoolwork, i would become much more free in the next month, until i decided to carry someone else through the end of the modpack around the border between october and november. however, i also foresaw myself having plenty of schoolwork, so i decided to try to cut down on the amount of work needed for this puzzle by changing game 4 to an even more complex spirit. (this is supposed to not make sense btw)

justification: unlike starlight, fractured can easily cast down a board on turn 3 in the fast phase (with my free manipulation of rng), and for the two turns before, fractured only needs to play 1 card.

i kept the name of the fractured player as lumia for now, primarily to bamboozle people who knew me into thinking that game 4 would still be with starlight. though i did have luna as a backup name for the fractured player. i would have a pretty good excuse for keeping this name, since i had the idea of naming this character lumia before i had the idea of taking the name for myself. i was thinking about asking for opinions on whether this was a good idea, but i eventually decided to change it, which was probably for the better because it would be pretty confusing for testsolvers when they wanted to talk to me during the testsolve.

on september 29th, i finally figured out how i would control the fear cards. i would just not try to sync between games 1-3, and instead just have game 4 count all the fear generated in all games combined.
this ended up being my final plan for this puzzle; to have games 1-3 visible from the very start, and then once the partial was gotten from games 1-3, game 4 would be unlocked, which would use the scripts of games 1-3 along with another script just for game 4.
games 1-3 would each be only affecting their own island of 1-2 boards, but game 4 would use all 6 boards split into 4 parts, where each part is on the islet corresponding to a game from before. (i'm honestly not sure how best to word this.)

this was also the point when the editors started running through the script for part 1 and looking for glaring errors or really difficult areas in the gameplay or the script. the first question to come up was how to best play the game out. in their experience from the testsolve of the older version, tts didn't work well, so we ended up trying this new thing called excalidraw. it seemed pretty useful while i was watching, so after that run through we made a note to officially recommend using excalidraw, or any other app that could draw and save past board states easily.

patrick and cat found a few glaring issues, in particular, i don't say whether this game is actually against russia 6 or not, and it isn't supposed to be, at least not by itself.

i finished drafting the script for part 2 on october 3rd, and again, i discovered by watching patrick and cat run through it that i did not describe how setup worked at all. in particular, the board layout being opposite shores instead of standard was pretty important.

games 3 and 4 were pretty tied together. except for the first half of game 4, the games were planned and the scripts were written together. in fact, game 4 interferes with game 3 directly, and the removal of all buildings of game 3 in game 4 is discussed in game 3's script. i finished game 3 and it was run through before the end of the first week of october. the editors liked the sudden disappearance of the buildings from game 3, since this would likely occur at a point where a solver is wanting to take a break from this puzzle.

in october, i made a bunch of aesthetic changes to this puzzle. first of all, luna had all of her script lines lowercased, but i had the spirit island terms still start with a capital letter, mostly to show “importance”. also, i color coded all of the lines in the dialogue depending on who said them. rae's lines were purple, luna's lines were magenta, bjorn's lines were red, dalia's lines were dark green, fern's lines were orange, and undine's lines were blue. you can see the vestiges of these colors from the solution excalidraws.
i also changed the background color of the google docs to be light gray to make fern's lines easier to see, as well as converting to a pageless setup so page breaks wouldn't just be random inconveniences when testsolving.
everything was given its own style. extractions were in italics, and checksums were underlined. the text that marked transitions from one turn to the next had completely no formatting. all lines that weren't dialogue was black, except for the text marking turn 5 of game 3, which was magenta, hinting that luna may have gotten involved here.

on october 14th, i finally got an answer assigned to this puzzle. it was 10 letters long, but it wasn't too bad to split this between the four parts. after re-reading through my notes and my diagrams to make sure everything was accounted for, i started looking for places to put extractions.

on october 27th, i settled on the current extraction mechanic for the puzzle. naturally, i wanted to gate the answer behind part 4, and gate part 4 behind a partial. i thought about what i could use as a partial, and eventually i settled on the phrase “russia level six”, which could be shortened to “rus lv six”. i then spent a while thinking about places where you could get a letter of “rus lv six” in the standalone parts, and the corresponding letter of the answer in the parts that were influenced by part 4.

starting october 18th, the backerkit for nature incarnate was opened, and they revealed a lot of things. naturally, i wanted to include a reference to it somewhere in part 4. the only reveal that would be relevant on the puzzle was the reveal on october 28th of growth through sacrifice being replaced by a different card, roiling bog and snagging thorn. fortunately, i did actually use growth through sacrifice in game 3 (though the card was completely unnamed), so i was actually able to mention and use roiling bog in game 4's script. i also included a cheeky reference to the newly announced fangs aspect unconstrained in part 2 a bit later, since bjorn was already complaining about b4 being blighted, so it would be pretty easy to add.

the puzzle finally became actually testsolvable on november 27th. patrick added a main puzzle page that linked to the 3 parts that should be visible, as well as a few suggestions for how to do this puzzle, including the resources such as the card catalog and the faq (though we didn't find links to the rulebooks until partially into the first testsolve). it also had the enumerations at the end. i also finally changed the fractured player's name from lumia to luna, since it just worked better for the spirit (i.e. time and light are much less directly connected than time and the moon).

game 1 got testsolved once on december 8th. unfortunately, since i didn't find links to the rulebooks yet, the start was very slow. though a lot of the beginning was slow, since it was actually somewhat underclued with which growth option gets taken, until everything adds up in turn 3. in the end, game 1 ended up taking 2 hours, and the group only finished the first turn of game 2 before leaving. this wasn't a good sign, since our expectations were that game 1 would only take around 1 hour, though we reasoned that since these parts are parallelizable, it would be shorter.

powerstorm + lure of the unknown + exaltation of the incandescent sky + gift of power

now for the large testsolve of the entire puzzle. this would end up taking 6 days, from december 22nd to 28th, and it took over 10 hours. the testsolvers said that unlocking part 4 is very cool, which is what i was hoping for, but they also knew that this puzzle was way too long, and i didn't disagree with them. they compared it to first you visit burkina faso, which was infamous for being extra long. however, patrick helped me reason with them that not very many people will see the puzzle, and that it was fine to leave this puzzle in for the select few teams that it was written for. we all agreed that we need to at least smooth out each game individually, since if games 2 and 3 are this long by themselves, game 4 will not be solved by any teams during hunt.

i guess an important detail of this testsolve was that it took as many shortcuts as it could. in particular, it basically didn't look at part 3 at all, and after unlocking part 4, the testsolvers basically wheel of fortuned the answer from just a couple letters from parts 1, 2, and 4. i did accidentally make the extractions in part 4 much easier, and that was helpful in making the puzzle more manageable, but i did also want teams to actually do part 4 entirely.

as cat pointed out, i had a chronic problem where a player would say that they wanted to do something, but they ended up not doing that thing, which would very easily throw solvers off. this was a pretty common occurrence in game 3, and i even felt its effects when reading through the scripts, since i didn't realize that the plans that were being talked about were not gonna get carried out.. so obviously, that was something i needed to change. there was also a general lack of cluing within parts 2 and 3, such as me not naming many cards that are relevant, particularly sky stretches.

the testsolvers had two major requests. the first one was to make setup faster by adding excalidraw links to the puzzle of the game already set up for them. this wasn't too bad, though these files were definitely too big to hold all the information for all 5 turns of the game, which i would find out when i write the solutions later.

the other major request that the testsolvers had was adding more checksums, which was also a change i agreed with. unfortunately, i wanted to add them into the script, but i couldn't really find a good place to add them, so i procrastinated on that for several days. i was eventually convinced that the checksums didn't need to be said by anyone and could just be added as an extra line at the end of each turn. i was fine with that, so i just wrote a bunch of random observations i made that were true for both the original game and the relevant part of game 4. the checksums for game 4 i made to be just the fear counts in the game, partially to reward teams who kept track of that, partially because i didn't feel like thinking of any better checksums.

this puzzle definitely needed to get all of its parts checked before hunt. the main testsolving group agreed to look through game 3 and all 4 parts of game 4 to make sure everything worked.

on january 4th, i had the idea of asking officials from gtg whether i could tease an unrevealed major power from nature incarnate in this puzzle. to my surprise, they said yes, and gave me exaltation of the incandescent sky.
my first thought was “wow, those sure are some numbers”. my next thought was “where in game 4 would i put this?” i first tried at the very beginning, when luna was looking at her days that never were. however, it felt a bit weird to talk about it in the beginning and then forget about it in later turns.
my next attempt was to still have it in luna's days that never were, but mostly ignore it until the growth phase of turn 4. however, anything i could really say about it ran into the chronic problem i had earlier of a player saying something but changing their mind later.
so i ended up having luna see it while gaining a major on turn 4, saying that it looks useful but it wasn't for this turn, and just toss it into her days. this was where it stayed up through mystery hunt.

vigor of the breaking dawn + skies herald the season of return + voracious growth + entwined power

the day i got back onto campus, which was the day before the start of iap, i started working on writing the solution for this puzzle. my much improved sleep schedule and my improved mood from being away from my parents let me start working pretty early in the morning, and do work intermittently all day. of course, i was gonna use excalidraw for my solution. however, after i finished turn 1, excalidraw started lagging pretty badly, so i had to export everything as a link and then clear the drawing before moving on to turn 2. i believe this was mostly due to the pictures taking up a massive amount of storage and increasing the file size of the drawing.

i had two different links for every turn, one being for games 1-3 and the other being game 4. so in the end i had 9 links in total, since i combined turns 4 and 5 for games 1-3. this worked well, until it was time for the solution to get postprodded. one of the postprodders asked me if i could just use images directly instead of links. i told them that i'd try, but when i tried, the image i'd get from each link ended up being 130 mb in size. i also tried converting into svg instead of png, but that didn't shrink the file sizes enough. i needed to shrink the drawings down by a factor of 10 in both dimensions, and this got the file size to about 25 mb for each turn, which was manageable.

when i looked at the postprodded solution later that week, the solution didn't actually include the images. so maybe i just didn't need to download the images, and that would have saved a lot of hassle.

also, i had to structure the solution with all the deductions in an appendix, since i was really unsure what deductions i would need to write out, so i just listed a bunch of deductions i thought of. i also had pretty long author's notes roughly detailing the most important parts of the journey of this puzzle, and it was in all lowercase, like this blog post. at that point, the author's notes were still shorter than the appendix.

this puzzle was one of the last few puzzles to be completed. the teammate writing discord had categories to mark the status of puzzles. this puzzle was in "done-3" which is the fourth done category, which means that there were at least 150 puzzles done before it. i looked at the postprodded puzzle in the bootes round. it looked amazing, with all the vibrant colors, but on a black background this time.

i have to thank everybody on teammate who worked on this puzzle with me. my coauthor (patrick), our editor (cat), the main testsolvers (eggy and pei), and the other testsolvers (ariel and agotsis), the postprodders(who i think were andrew and evan but i didn't quite pay attention), and the "factchecker" (harrison). this puzzle was too big and specialized to actually be factchecked before mystery hunt, so our factchecks were the main testsolvers going through each part once to verify correctness, and then the one that puzzup recognized a factcheck, which was checking whether the postprod was correctly done.

in the last few days before mystery hunt, i was incredibly excited for this puzzle (as well as my other puzzle, set off fireworks) to get seen and potentially solved. it was pretty far back though, so i wasn't sure whether et phone in answer (one team whose members i often play spirit island with) would actually get to the puzzle. cj even asked me whether there'd be any puzzles that he'd recognize as one that i wrote, i told him “you can maybe tell which ones i touched”.

birds cry warning + portents of disaster + fragments of yesteryear

then came mystery hunt itself. of course, you know that hunt ran a lot longer than we planned. what you may have learned more recently is that the round that was most impacted by this was bootes. it was originally 3 subrounds: ship, modules, and place. initially, we tried rearranging the puzzles in this round to move harder puzzles later, so dahan konundrum would be moved from near the middle of the round to near the end. then later, ship and place were removed entirely. dahan konundrum was under the place subround, so nobody ended up seeing it during hunt.
in fact, when bootes round was rearranged, i told et that i was pretty sure they wouldn't see my harder puzzle during hunt, though i wasn't quite expecting the reason to be it getting removed from hunt entirely.

i was honestly not too hurt by the removal. sure, there was a teaser, but i would be fine as long as it is released any time before nature incarnate becomes available. and i could also share this puzzle with the public anyways, even without the hunt surrounding it. though i still thought that it would be better better with the hunt surrounding it, because the answer is still an answer in the bootes subround, which is pretty hard to change.

on sunday night, the authors of the removed puzzles had a meeting to discuss what we wanted to do with the removed puzzles. there was pretty widespread agreement that we wanted the puzzles to be released as an actual hunt later in the year. abhijit suggested may 2nd, because that is when the bootes culmination at midnight is. we agreed that we would reconvene to discuss concrete plans in march.

during that meeting, i suggested that there was enough time to make edits to our puzzles before they get released. i thought about adding a part 5, but that would make creating the game and making everything consistent very tedious. however, i still wanted to make changes, so i made a few changes to the last two turns.

unlock the gates of deepest power + sky stretches to shore + blood draws predators + pent-up calamity

first of all, i wanted to make more use out of exaltation of the incandescent sky. right then, it was just getting tossed into luna's days. the main reason i wanted to add a part 5 was so this would get more use, but i decided to take advantage of the free space of dalia not doing anything in their last turn, and get them to play exaltation. this was made much easier since keeper gains a very large amount of energy through growth, so it wasn't particularly difficult to get them to play this. additionally, keeper thresholded powerstorm earlier in the game, so i could basically play the same cards plus exaltation, and the thresholds would still be hit.
however, this changed up a lot of turn 5, affecting the actions of every single player. first of all, bjorn reclaimed and played an additional card in their growth. next, fern gained gift of living energy instead of fire in the sky, because dalia was just barely short on energy to use everything. the effects of exaltation made a huge difference on rae's board, since now rae would have 4 card plays and could threshold terrifying nightmares and max the innate. finally, undine and luna took advantage of the threshold of exaltation to repeat blur the arc to create more dahan and generate fear at the same time.

secondly, i knew that the extractions in the later turns of part 2 were really bad. so i made them easier. now, solvers didn't have to keep track of fear earned by just this islet for the entire game, and they also didn't have to realize that they needed to look at what was happening on other boards for any of the extractions. on the other hand, a couple of extractions in the other games had to change, since the number of presence on the islet of part 3 of game 4 and the number of explorers at the end of part 1 of game 4 changed a lot with these revisions.

thirdly, i made even more cosmetic changes to the puzzle, changing a few names, colors, and even genders. rae, dalia, and undine finally got the names that you're currently seeing. some of these name changes also came along with gender changes, meaning that this puzzle has trans characters.
also, i changed a couple colors, specifically luna to yellow and rae to white, which were the colors of the spirits' primary elements. this freed up purple and magenta, so i changed the extractions to magenta and the checksums to purple, and i also realized that i haven't used light gray, so i made all the extra information such as the turn separators light gray. (i know moon is technically light gray, but it's also the closest thing to white, and i wanted rae's lines to be white)

sometime in february, teammate had a debriefing of hunt in its entirety. after the official stuff ended, we started shitposting in voice-chat-text. one thing that was mentioned was that the previous scavenger hunt answer was sitting there unused, so i thought of a mechanic that i could use to make that happen. i figured out one mechanic i could use that made it rely on this puzzle, which is also why i made these name and color changes. it was harder than i initially thought though, because the dataset i was using combined with the mechanic that i had in mind was pretty restricted. anyways, i successfully found an extraction that worked, and wrote a testsolvable draft of the puzzle. however, i haven't yet sought anybody to testsolve it, and it was decided that the scavenger hunt would not be replaced anyways.

as may rolled around, the rest of the bootes round was announced as a full hunt. this announcement happened during microsoft puzzle hunt, so it might have been at an imperfect time. however, it means that i finally was able to release my post-hunt write-up.

one day, i was in a conversation that cj was also in, when something about the amount of writing we did was brought up. i wouldn't really have said anything in the conversation, but then i realized that i had this puzzle which was quite long, so i decided to check. the five documents that were just the puzzle documents took up 30 pages in total, which cj said was around 10k words. i added up the word counts for each game, and the total was around 10.8k words. this also made me curious about what other puzzles were really long. it turns out that the longest puzzle that was actually in hunt was moral of the story at 4.2k words, though the quiz bowl puzzle was also above 3k words.

this realization, along with the large number of blog posts written about hunt, made me a lot more willing to write words. i almost definitely wouldn't be writing this blog post as its own blog post if not for the realization. at this point, it's more than half as long as the puzzle itself.

i had a friend testsolve (which ended up doubling as a factcheck) all these changes in late may. at this point, there was still some lines that were possibly misleading, and there were even some gameplay mistakes. however, he went through everything very thoroughly, and managed to finish the first 3 parts with the gameplay all correct (though he did have some mistakes while extracting). he did notice some suspicious things, like all the invader cards being the same, and a lot of actions being taken suboptimally. when he unlocked part 4, he had lots of strong sentimental emotions while reading it for the first time, including almost crying at the replacement of gts by roiling bog and the use of the threshold effect of exaltation. the testsolves spent 2, 4, 6, and 10 hours on the 4 parts in order, for a total of 22 hours. by then, it was too late to make any changes to the script, so i just left the puzzle at that difficulty and postprodded the revisions.

these revisions, along with some small color changes for accessibility/contrast purposes, were the content of the first 5 pull requests i've ever done. i mentioned that this would have been fewer if i was more competent, then someone mentioned this would also have been fewer if the reviewers were more competent, but then i told them it would have been zero if they were more competent.

in one of teammate's spreadsheets, we can see all the puzzles in abcdefg along with their fun and difficulty ratings and hours spent, both ratings starting from 1 as the least fun/easiest to 7 as the most fun/hardest. most puzzles in abcdefg were between 4 and 5 in difficulty with at most 7 hours spent (with sensorium at the highest of both ranges), while dahan konundrum was difficulty 6 with 10 hours spent. when other teammates looked at the spreadsheet, they pointed this out and said that dahan konundrum sounds like it would be scary from just this.

the happenings in the last few paragraphs made me deeply consider how much effort dahan konundrum takes to solve. at the beginning, you only have games 1-3 unlocked, but those already add up to 7k words. once game 4 is unlocked, the word count goes up to 11k, over a 50% increase. also, the base game of spirit island is above 4 in bgg complexity, and this puzzle uses all the expansions, including the currently unreleased nature incarnate, which have even higher complexities going up to 4.5. these factors, in addition to this puzzle being a duck konundrum, means that this puzzle was probably the hardest one in abcdefg. in addition, hiding the fact that you would need to do each part twice makes this puzzle by far the most deceptive in abcdefg.

i don't know how i should feel about having written a puzzle that's too hard for a puzzlehunt that was advertised to be really hard puzzles.

these characteristics have shown their fear generating capabilities during abcdefg. ue described unlocking part 4 as "the most terrifying thing possible", and cj screamed at me (i believe this was a combined scream of anger and fear, where the anger is at the answer itself) when galactic submitted the guess that unlocked part 4.

there were some factors in the space place meta that made backsolving dahan konundrum much harder too. two of the most notable had to do with the second-hardest feeder, first you visit. for one, these two were the only two that clued their arrows with cardinal directions instead of physical directions. also, the length of the answer to first you visit was 12 characters long, which fits the (6 5 1) enumeration of the partial that unlocks part 4.

while looking around at teammate's testing site, i tried submitting the actual answer first. it turns out that it didn't let you submit the partial after the puzzle was already solved, so backsolving this puzzle would have locked a team out of part 4. someone on teammate offered to make part 4 unlock after submitting the answer to the puzzle, but i dispreferred that to locking teams out of part 4. maybe we could have changed it to keep the answer submissions working, but this would be the only puzzle where it actually matters and i didn't think this was worth the effort. i did get asked why this puzzle's partial worked differently from sensorium's partial, and the main reason was really that the partial of this puzzle could actually be submitted in the answer checker, while the partial for sensorium could not.

i must admit, i don't expect others to find this very interesting, and some may even find this post hard to read, either due to its length or the lack of capital letters. but at least i enjoyed writing about myself and something i spent over a year working on. and now that it's been released to the world, i can finally rest, and actually use that mystery card i added to et's copy of spirit island back in february.

and write dahan konundrum 2 to put in a hunt later :hohoho: