The year that the Hunt won
For various reasons, this year's hunt was not the best year my team's had. A few things were bad luck, but I think the big issue were endemic to this Hunt overall. Although the team whose name was the full text of Atlas Shrugged found the coin, I think it's equally fair to say this was the year that the Hunt won.
Warning: Puzzle spoilers ahead.
I frequently have insomnia before major events I'm excited about, and Mystery Hunt is no exception. Additionally, I'm in the middle of trying to deal with a home plumbing semi-emergency, and abruptly switching gears from work to play quickly is hard to do normally anyway. Between all these factors, I got maybe 3 hours of sleep on Tuesday.
Flying from the west coast to the east coasts sucks because it eats up a day of travel. Well, unless you take a redeye. I hadn't taken one before but was willing to give it a try this time. I lugged my newly-packed stuff to work in the morning and was luckily able to hitch a ride to the airport that evening, as opposed to having to take public transit or drive. Wednesday night was spent on the plane, where I uneasily caught about another 3 hours of (airplane-quality) sleep.
I arrived a day early to try to dayshift and recover from travel. I ended up spending the day poking around some IAP classes (none of which were ultimately interesting enough to have been worth dropping in), setting up my laptop for Hunt, and napping for maybe 2 hours. I didn't want to sleep too much and completely throw my schedule off; I was trying to get just enough so I'd crash early and get a full night's sleep.
Complicating matters, the friend who I was staying with had someone renting his guest room, so I ended up on the couch. This wouldn't have been bad, except the renter is in a startup or something and was having meetings at my friends house. They kept working until maybe 3 or 4 AM. Coupled with my nervous energy, I only ended up with maybe another 3 hours.
As I mentioned before, I was a little worried about this year's writing team, Manic Sages, as it is their first time organizing. Still, Codex was also new and they had done a great job, so I was hopeful this year would also be good.
While I was wandering around Thursday, I picked up the school newspaper. There was an article about Hunt, which was unusual. Maybe I wasn't paying attention in the past, but I think the article is usually run after the fact, recapping the Hunt. This made sense, though - by having the article run beforehand, it might get more students interested, and maybe the writing team could embed data for a puzzle in the article.
I was a little worried by a quote from one of the organizing leaders, which was that he wanted to "come up with something awesome that’s never been done before and make the hunt even more spectacular". The problem is that running Hunt is harder than it looks, and changing too many things at once (including learning to run Hunt) increases the likelihood of something going wrong. The advice we had given to Codex was basically that the first time you run Hunt, you should try to learn the ropes and just do a well executed standard Hunt. Save the crazy stuff for the second time you win.
Instead of being in Lobby 7, kickoff this year was in one of the school gyms. This wasn't the writing team's fault, but rather because the safety office had finally noticed that hundreds of people were packing Lobby 7 every year and Something Had To Be Done.
Nothing much happened at kickoff, basically just revealing that (as was roughly guessable from the pre-Hunt material) a big bank had gained control of the coin, and our task was to steal it back. The highlight of kickoff actually came from Codex, who was sitting with cardboard protest signs saying clever topical things like "Kickoff in Lobby 7" and "1% of the population controls 85% of the puzzles". Puzzles would be released at 1:30 (a bit later than usual), so we headed back to our HQ, with some people detouring for lunch. As an aside, some people on our team liked the extra time for lunch, though personally, it was a waste - I think food should be gotten before kickoff; I'm here for puzzles.
As 1:30 approached, we anxiously watched the decrementing timer on the puzzle website, chanting down the last 10 seconds before puzzles were released.
Nothing.
The timer reset to 28 minutes.
Was this a puzzle? No, the page looked the same, and a check of the source code didn't reveal anything interesting. Maybe technical difficulties?
28 minutes later, we were again counting down the last 10 seconds, and this time our patience was rewarded with puzzles.
The first round fell quickly. We completed it in about an hour and were actually left sitting around with no puzzles open, waiting for something to happen. This wasn't completely surprising, as the trend in recent years has been for the opening round to be a warmup (or for small teams, something they can complete and get a taste of the full experience). I think this is a great thing, and was glad the Sages had continued this.
At this point, I should note that we were somewhat underpowered this year. Some people who had hunted with us in the past were with other teams or just not present, and although we had 4 rooms (and in the past used them all), this year we basically only used one room (and occasionally spilled over to a second). To be fair, some of this is also our team learning to all solve in the same room, but it definitely felt like we had significantly fewer people.
I spotted You Should Be Listening, an anime puzzle in the second round and jumped on it. Back in 2011, when we ran the Hunt, I had considered writing a puzzle about different (Japanese vs. English) voice actors for the same characters, and was pleased to see that this puzzle basically worked like the one I had been thinking about writing. It felt good that someone had also thought of it as a puzzle and actually written it. There were some added difficulty, like the images being hand drawn and thus being harder to identify. Although we never ended up solving half the puzzle (we only figured out some of the blanks on the right side of the slash), it was enough for us to guess at some reasonable answers. However, we got no response to our second answer attempt and kept staring at the puzzle, trying to figure out what to do on the left. Finally, Sages called us back and said our second attempt was correct. It would have been good to know earlier.
We later learned that Sages was having technical difficulties contacting us, both by phone and by email. Their email for us was wrong and they hadn't been noticing the bounces, and something about our phone setup and the way they called made things very sporadic. For example, they had sent an email explanation that there were in fact technical difficulties with the website, had been trying to call us about the completed first round, and were not intentionally creating big delays between our calling in answers and responding. So although we had gotten off on a somewhat rocky start, I was willing to chalk things up to bad luck.
There was The Obligatory Wordsearch, but it was already well underway and had enough people, so I reluctantly moved on. Definitely check it out though, the part I saw was cool.
I then helped solve Lost in Translation, a neat tangram puzzle. We were able to get the answer in a pretty reasonable amount of time. Looking good! The only worry was that the first round had been slightly on the large side at 17 puzzles; the second round also had a large number of puzzles and showed no signs of stopping. Being a smaller team than we normally were, large numbers of open puzzles would be a problem, as we would not be able to keep up with them. That's always the case, but I normally don't start feeling like there's a flood until maybe Saturday morning. This was still Friday afternoon.
I helped work on Color Sudoku next, but we got stuck on extraction and abandoned it. I considered going to my friend's place and sleeping, but there was a midnight event, and I volunteered to go to it as I wasn't in the middle of a puzzle and one of the other volunteers would be better suited for a puzzle we had just unlocked. The event was decent, but involved eating some desserts, so I was on a bit of a sugar high when I got back.
I considered working on a Jigsaw Puzzle, but concluded the effort vs. reward just wasn't worth it. It was getting late and I wanted to go sleep, but a second consequence of being a small team surfaced. People normally leave their stuff in the room, but that means someone has to watch it. On the other hand, everyone wanted to go home and sleep. I volunteered to stay and keep an eye on things provided I could doze until the other night owl got too tired to continue. I got maybe 2-3 hours of nap.
Incidentally, this was the first time I can remember that I've actually been disoriented when waking up. Very unsettling.
There was an odd call from HQ about a wrong answer we had called in - I didn't think we had a puzzle called Time Conundrum, nor did we have anyone working on it. Then again, based on the name, it was entirely possible we actually didn't have the puzzle. Or maybe I was hallucinating from sleep deprivation.
I'm the only one in our HQ for a while, and I passed the time mapping out a Text Adventure, eventually discovering the player was wandering around inside Feynmann diagrams, which was pretty neat. It seemed obvious that some quantum mechanics knowledge would be needed to finish the puzzle, so I set it aside.
Around this time, people started tricking back in, and I was hopeful that I could finally sleep. But we had also just learned that there would be an event in about half an hour that needed three people, and we only had one free person. I stuck around on the off chance we didn't have enough people, and only one other volunteer showed up, so I did end up going. This event wasn't particularly good IMHO, as there was no actual puzzle involved and no social interaction - it was a bunch of games of games of telephone played in a mesh with some malicious nodes. The event felt shoehorned into the Hunt (perhaps as filler or perhaps as a wildcard puzzle) rather than actually belonging, and was bottlenecked on the number of organizers available.
By the time we get back to our HQ, it's maybe 11 or noon, and I'm no longer feeling sleepy. While I could sleep, there are now people around, and I decide to work on the Time Conundrum with MM and J off in another room. As expected, it's a duck conundrum with the twist of time travel elements, and the goal is to figure out the stable, self-consistent flow of events. Like all conundrums, it requires a lot of attention to detail, and we get lost several times (mainly involving the definitions of age and time jumps).
Meanwhile, the list of unlocked puzzles has expanded, and so has my unease - rounds are looking like they're regularly consisting of maybe 20-30 puzzles, which is absurdly large. Normally, rounds are in the 10-20 range. Worse, talking to people suggests that everyone is finding puzzles to be running longer than normal - too many steps or too much gruntwork, even after you know what you're doing. To cap things off, the metas are looking like they're not backsolvable, as each answer only gives you a small bit of information relative to the answer.
But at the same time, it's still early in the Hunt. Maybe later rounds are more sane, or maybe there's a shortcut we haven't found yet, or maybe we're just not at our best.
Meanwhile, J has been pulled on to help finish Text Adventure. MM and I keep working on the conundrum, but run into a contradiction. We stared at it for a while, trying to find and fix errors, but I eventually decide that the giant whiteboard diagram we have is too untrustworthy and start fresh on paper. MM keeps trying to patch the existing timeline. I'm careful to make no assumptions this time, proving every step, and catch a few more errors. Meanwhile, MM is able to refine the data we have and we eventually converge on the answer at almost exactly the same time.
I again consider going back to sleep, as it's now around 8 PM. Just as I finish bringing my stuff back into the main room, though, L stops me to ask if I want to join a Mary's mob. Sure!
Finally after dinner, I've grabbed my stuff and am about to leave when I hear there's a Starcraft puzzle. Naturally, I really want to work on it, but I'm saved from having to make another painful decision of stay vs go when I hear that Snow Day is actually a Blizzard puzzle and the Starcraft part is mostly solved. I descramble the final clue phrase for the Starcraft section and am finally able to make it out the door. I make it back to my friend's place at 11 PM Saturday, about 36 hours after I left it.
...and the renter is on the couch, watching a movie with his girlfriend. grar.
Fortunately, my friend said I can use his room, so I do manage to get some sleep. This time I get a reasonable amount, about 10-12 hours.
Sunday doesn't start great either, as our QM isn't picking up the phone and I'm stuck outside for a while as I try to get in contact. I eventually get through and settle in to solve more puzzles.
Things aren't going well. We only really have enough solved to even approach 2 of the 6 real metas, and haven't solved either one. Our team normally gets a meta by Saturday afternoon or evening; 0 of 6 metas solved on Sunday afternoon is a bad sign. While we aren't having the greatest hunt ever, I also don't think we're that bad. Then again, I hope it is just us, as Sages isn't giving out hints and the alternative is that this will be the new record Hunt for time.
I check back in on Snow Day and see it's gotten further and is stuck on extraction. After familiarizing myself with the current status (which admittedly has too much data), I spot the solution.
As I'm continuing to bounce around puzzles trying to find something to work on, I hear that someone finally got the second round meta. This was another case of "we know what we probably want to do, but it takes an inordinate amount of time to do it". Solving the meta unlocked an obstacle training event on campus, and I head over with some others to tackle it. This is our first glimpse at the larger structure of the Hunt!
Earlier, I remember someone on Sages mentioning that although they were a large team, they needed everyone to run Hunt and had roles for people to play. Up until now, we hadn't seen much sign of this; with the exception of maybe the telephone event, it didn't seem like anything required more manpower than normal. The obstacle events must be where people were needed.
We were ushered into the training center, a room with Sages seated behind counters (most of which were closed), humorous notes jotted on the blackboard, lines for "amateurs" and "professionals", and a waiting area. We received a short briefing, then were shown into another room with just a laptop and a Griffin PowerMate. The screen showed a picture of a safe's dial, and turning the PowerMate was reflected onscreen. It took us 17 minutes to figure out the combination (mostly because we had forgotten about the safety briefing, which contained a vital clue), and chatted with the Sages representative. She said from the perspective of the training center, things were going at the expected rate. This was a little weird, because from our perspective, thing were going very slowly. I mentioned our frustration with puzzles taking longer than we were used to, concern that Hunt would end late, and how I was surprised that they weren't giving out hints (based on the solve rate and our projected progress so far, I would have expected hints to have started flowing at least by Sunday morning, maybe earlier).
My team is getting older, and there's not a lot of influx of new blood. We have day jobs, and (aside from me) people aren't willing to sty up as late. It's also good courtesy to leave things clean, which takes time. As such, we had already decided to stop puzzling at 4 PM on Sunday regardless of our progress in the Hunt. We ended up staring at the first meta for a while, as we had all but one answer, but were unable to come up with a working extraction mechanism. Because of that, we didn't actually end up stopping until maybe 5, after which we hung around and chatted. At the point we stopped, we had 1 meta and obstacle solved, all the data for another meta (but no idea how to extract), maybe 1/3 of one more round solved, and negligible amounts of the remaining 3 rounds. For our team, that's a pretty poor showing.
From the Sages' emails, it was clear things were unraveling. They were offering more and more hints and had first increased the rate of free answers to 0.3 per hour, then again to a whopping 1 per hour. For context, free answers are pretty powerful, as you can use them to get past that one puzzle that's holding you back from solving that one meta which is keeping you from endgame. In previous years teams went to events to earn maybe 2-3 free answers over the course of the entire Hunt, and there was no automatic granting. A free answer per hour is faster than you could solve most normal puzzles (much less these special extended edition puzzles), and was a sign of how badly things must be going. Sages HQ was apologetic and said they had originally intended to be done by Sunday afternoon and really wanted to be done by midnight.
Regardless, we didn't care - we were never interested in winning, and weren't even solving puzzles at this point. We headed out to dinner at 8, and then we split up for the night.
When I went to sleep, I was unsettled that not only had the coin not been found, Sages was giving out even more hints. Not a good sign.
Early in the Hunt, our team privately joked that if Hunt went past midnight on Sunday, then the Hunt itself would win and everyone, including Sages, would lose; their punishment should be to write it again next year (and the next, and the next...) until they got it right.
Ive been doing Hunt for a while. The longest one I participated in was the Time Bandits hunt in 2004, which had it's own share of major issues. Time Bandits ended (I think?) at 9 AM Monday. The next longest was the Matrix hunt in 2003, which ended at 7 AM Monday. Hunts since then had gotten better about keeping time under control, with most ending Sunday evening.
I woke up at 10:30 AM Monday, intending to meet up with N for lunch at 11.
There was no email that the coin had been found. Hunt was still going on.
Despite our HQ shutting down, we were technically still in the game. We had accumulated 17 free answers (not counting ones we had already spent). This was enough for us to get all the remaining answers in the last round and still have one free answer left over. Craziness. And sometime overnight, someone on our team finally solved the first round meta, and we were eligible to run the corresponding obstacle training event. Most of the remaining team is off socializing at a teammember's house (it turned out to be further than I wanted to go). I want to check out the obstacle training event, but my phone's run out of power (necessary to setup an appointment) and I forgot to bring my charger. I wandered over to the training center, but as expected, they're pretty frazzled and don't have the bandwidth for me to run the event. Oh well.
Someone on our team is still solving puzzles, so I run one more data gathering task over in a nearby library. Side note: I think it's poor form to locate important puzzle information is located inside a building which is closed for part of the time. Anyway, as I'm sending the pictures, I see an email from Sages that the coin has finally been found. The winning team only had 5 of the 6 metas, which was a trick by Sages to reduce the time even further, as it's usually the last remaining meta which is the worst.
It's a new "record": 3 PM on Monday, with massive hinting (and even then the Hunt was incomplete). Sages certainly did do something that's never been done before, and I hope won't be done again. To their credit, it sounds like this wasn't intentional and that they tried to fix things when they noticed (too late) that theory and practice were, in practice, not the same.
I missed wrapup, as the length of the Hunt pushed it past my return flight. I'm really curious what the Sages and other team's experiences were like, and I'll hopefully be able to catch a recording.
Although I was pretty grumpy at the time, now that I've gotten some sleep the puzzle constructions themselves was pretty impressive. There was some errata - not as few as the previous year, but still only a handful. Things generally weren't broken, and I'm guessing Sages did a good job testsolving.
My major gripes were that each puzzle had a higher than normal effort:reward ratio. A lot of the time it felt like there was a bunch of work for the sake of work, and that the puzzle didn't benefit from the tacked-on elements. Even if the puzzles had been normal length, the metas didn't really seem backsolvable, so teams had to finish more puzzles (which in reality were also longer, compounding the problem). My guess is that finishing the Hunt without hints or free answers would have taken twice as long as it did. It's not surprising that the Sages' estimates were off, as it was their first time running things, but it was surprising how far they were off and how long it took them to react. I guess the former is somewhat understandable given how fast they finished last year's Hunt (Saturday at 10 PM), but I have no explanation for the latter.
My armchair psychologist take is that Sages wrote a Hunt they wanted to solve (which is sort of natural; your playtest feedback will reflect your team ethos) and applied their own experience to extrapolate other team's status. Unfortunately, this didn't mesh well with our style, or perhaps other teams. I bet most teams rely on backsolving to make up for being smaller (certainly ours does, and we're not even a small team - we're probably normally medium-large), and that didn't seem possible this year. Longer time to complete puzzles meant it's more likely you'd be caught up in the time release waves, and only a large team would be able to keep up.
Over dinner on Sunday, some of my teammates pointed out that the Time Bandits hunt was significantly more broken in terms of puzzle construction, and therefore was a worse hunt. I'm not sure I agree, as I'll tolerate some brokenness in favor of the satisfaction of solving more puzzles. In Time Bandits, we were able to get through to endgame. This year, we finished a third of the metas, with little hope of getting more.
This has nothing to do with the Sages or this hunt, but I'm a little disturbed that we left some low-hanging answers untouched. For example, there was a puzzle where the extracted instruction was to write a Cracked-style article, which we started on but never finished. Not much effort and we would have gotten another answer, as compared to spending lots of time on a different one and potentially not coming up with anything. It says something about the coordination level of our team; I'm worried that we might be no longer competitive (even if Hunt hadn't been broken). Don't get me wrong - I love seeing everyone, and I don't want to win or write, but I would like to be on a team that comes close to finishing. Hopefully next year we'll do better.
Like last year, the winning team makes me apprehensive again. Fortunately, unlike Codex or Sages, they've written before. Unfortunately, the one they wrote was Time Bandits. Hopefully they've learned since 2004.
Warning: Puzzle spoilers ahead.
I frequently have insomnia before major events I'm excited about, and Mystery Hunt is no exception. Additionally, I'm in the middle of trying to deal with a home plumbing semi-emergency, and abruptly switching gears from work to play quickly is hard to do normally anyway. Between all these factors, I got maybe 3 hours of sleep on Tuesday.
Flying from the west coast to the east coasts sucks because it eats up a day of travel. Well, unless you take a redeye. I hadn't taken one before but was willing to give it a try this time. I lugged my newly-packed stuff to work in the morning and was luckily able to hitch a ride to the airport that evening, as opposed to having to take public transit or drive. Wednesday night was spent on the plane, where I uneasily caught about another 3 hours of (airplane-quality) sleep.
I arrived a day early to try to dayshift and recover from travel. I ended up spending the day poking around some IAP classes (none of which were ultimately interesting enough to have been worth dropping in), setting up my laptop for Hunt, and napping for maybe 2 hours. I didn't want to sleep too much and completely throw my schedule off; I was trying to get just enough so I'd crash early and get a full night's sleep.
Complicating matters, the friend who I was staying with had someone renting his guest room, so I ended up on the couch. This wouldn't have been bad, except the renter is in a startup or something and was having meetings at my friends house. They kept working until maybe 3 or 4 AM. Coupled with my nervous energy, I only ended up with maybe another 3 hours.
As I mentioned before, I was a little worried about this year's writing team, Manic Sages, as it is their first time organizing. Still, Codex was also new and they had done a great job, so I was hopeful this year would also be good.
While I was wandering around Thursday, I picked up the school newspaper. There was an article about Hunt, which was unusual. Maybe I wasn't paying attention in the past, but I think the article is usually run after the fact, recapping the Hunt. This made sense, though - by having the article run beforehand, it might get more students interested, and maybe the writing team could embed data for a puzzle in the article.
I was a little worried by a quote from one of the organizing leaders, which was that he wanted to "come up with something awesome that’s never been done before and make the hunt even more spectacular". The problem is that running Hunt is harder than it looks, and changing too many things at once (including learning to run Hunt) increases the likelihood of something going wrong. The advice we had given to Codex was basically that the first time you run Hunt, you should try to learn the ropes and just do a well executed standard Hunt. Save the crazy stuff for the second time you win.
Instead of being in Lobby 7, kickoff this year was in one of the school gyms. This wasn't the writing team's fault, but rather because the safety office had finally noticed that hundreds of people were packing Lobby 7 every year and Something Had To Be Done.
Nothing much happened at kickoff, basically just revealing that (as was roughly guessable from the pre-Hunt material) a big bank had gained control of the coin, and our task was to steal it back. The highlight of kickoff actually came from Codex, who was sitting with cardboard protest signs saying clever topical things like "Kickoff in Lobby 7" and "1% of the population controls 85% of the puzzles". Puzzles would be released at 1:30 (a bit later than usual), so we headed back to our HQ, with some people detouring for lunch. As an aside, some people on our team liked the extra time for lunch, though personally, it was a waste - I think food should be gotten before kickoff; I'm here for puzzles.
As 1:30 approached, we anxiously watched the decrementing timer on the puzzle website, chanting down the last 10 seconds before puzzles were released.
Nothing.
The timer reset to 28 minutes.
Was this a puzzle? No, the page looked the same, and a check of the source code didn't reveal anything interesting. Maybe technical difficulties?
28 minutes later, we were again counting down the last 10 seconds, and this time our patience was rewarded with puzzles.
The first round fell quickly. We completed it in about an hour and were actually left sitting around with no puzzles open, waiting for something to happen. This wasn't completely surprising, as the trend in recent years has been for the opening round to be a warmup (or for small teams, something they can complete and get a taste of the full experience). I think this is a great thing, and was glad the Sages had continued this.
At this point, I should note that we were somewhat underpowered this year. Some people who had hunted with us in the past were with other teams or just not present, and although we had 4 rooms (and in the past used them all), this year we basically only used one room (and occasionally spilled over to a second). To be fair, some of this is also our team learning to all solve in the same room, but it definitely felt like we had significantly fewer people.
I spotted You Should Be Listening, an anime puzzle in the second round and jumped on it. Back in 2011, when we ran the Hunt, I had considered writing a puzzle about different (Japanese vs. English) voice actors for the same characters, and was pleased to see that this puzzle basically worked like the one I had been thinking about writing. It felt good that someone had also thought of it as a puzzle and actually written it. There were some added difficulty, like the images being hand drawn and thus being harder to identify. Although we never ended up solving half the puzzle (we only figured out some of the blanks on the right side of the slash), it was enough for us to guess at some reasonable answers. However, we got no response to our second answer attempt and kept staring at the puzzle, trying to figure out what to do on the left. Finally, Sages called us back and said our second attempt was correct. It would have been good to know earlier.
We later learned that Sages was having technical difficulties contacting us, both by phone and by email. Their email for us was wrong and they hadn't been noticing the bounces, and something about our phone setup and the way they called made things very sporadic. For example, they had sent an email explanation that there were in fact technical difficulties with the website, had been trying to call us about the completed first round, and were not intentionally creating big delays between our calling in answers and responding. So although we had gotten off on a somewhat rocky start, I was willing to chalk things up to bad luck.
There was The Obligatory Wordsearch, but it was already well underway and had enough people, so I reluctantly moved on. Definitely check it out though, the part I saw was cool.
I then helped solve Lost in Translation, a neat tangram puzzle. We were able to get the answer in a pretty reasonable amount of time. Looking good! The only worry was that the first round had been slightly on the large side at 17 puzzles; the second round also had a large number of puzzles and showed no signs of stopping. Being a smaller team than we normally were, large numbers of open puzzles would be a problem, as we would not be able to keep up with them. That's always the case, but I normally don't start feeling like there's a flood until maybe Saturday morning. This was still Friday afternoon.
I helped work on Color Sudoku next, but we got stuck on extraction and abandoned it. I considered going to my friend's place and sleeping, but there was a midnight event, and I volunteered to go to it as I wasn't in the middle of a puzzle and one of the other volunteers would be better suited for a puzzle we had just unlocked. The event was decent, but involved eating some desserts, so I was on a bit of a sugar high when I got back.
I considered working on a Jigsaw Puzzle, but concluded the effort vs. reward just wasn't worth it. It was getting late and I wanted to go sleep, but a second consequence of being a small team surfaced. People normally leave their stuff in the room, but that means someone has to watch it. On the other hand, everyone wanted to go home and sleep. I volunteered to stay and keep an eye on things provided I could doze until the other night owl got too tired to continue. I got maybe 2-3 hours of nap.
Incidentally, this was the first time I can remember that I've actually been disoriented when waking up. Very unsettling.
There was an odd call from HQ about a wrong answer we had called in - I didn't think we had a puzzle called Time Conundrum, nor did we have anyone working on it. Then again, based on the name, it was entirely possible we actually didn't have the puzzle. Or maybe I was hallucinating from sleep deprivation.
I'm the only one in our HQ for a while, and I passed the time mapping out a Text Adventure, eventually discovering the player was wandering around inside Feynmann diagrams, which was pretty neat. It seemed obvious that some quantum mechanics knowledge would be needed to finish the puzzle, so I set it aside.
Around this time, people started tricking back in, and I was hopeful that I could finally sleep. But we had also just learned that there would be an event in about half an hour that needed three people, and we only had one free person. I stuck around on the off chance we didn't have enough people, and only one other volunteer showed up, so I did end up going. This event wasn't particularly good IMHO, as there was no actual puzzle involved and no social interaction - it was a bunch of games of games of telephone played in a mesh with some malicious nodes. The event felt shoehorned into the Hunt (perhaps as filler or perhaps as a wildcard puzzle) rather than actually belonging, and was bottlenecked on the number of organizers available.
By the time we get back to our HQ, it's maybe 11 or noon, and I'm no longer feeling sleepy. While I could sleep, there are now people around, and I decide to work on the Time Conundrum with MM and J off in another room. As expected, it's a duck conundrum with the twist of time travel elements, and the goal is to figure out the stable, self-consistent flow of events. Like all conundrums, it requires a lot of attention to detail, and we get lost several times (mainly involving the definitions of age and time jumps).
Meanwhile, the list of unlocked puzzles has expanded, and so has my unease - rounds are looking like they're regularly consisting of maybe 20-30 puzzles, which is absurdly large. Normally, rounds are in the 10-20 range. Worse, talking to people suggests that everyone is finding puzzles to be running longer than normal - too many steps or too much gruntwork, even after you know what you're doing. To cap things off, the metas are looking like they're not backsolvable, as each answer only gives you a small bit of information relative to the answer.
But at the same time, it's still early in the Hunt. Maybe later rounds are more sane, or maybe there's a shortcut we haven't found yet, or maybe we're just not at our best.
Meanwhile, J has been pulled on to help finish Text Adventure. MM and I keep working on the conundrum, but run into a contradiction. We stared at it for a while, trying to find and fix errors, but I eventually decide that the giant whiteboard diagram we have is too untrustworthy and start fresh on paper. MM keeps trying to patch the existing timeline. I'm careful to make no assumptions this time, proving every step, and catch a few more errors. Meanwhile, MM is able to refine the data we have and we eventually converge on the answer at almost exactly the same time.
I again consider going back to sleep, as it's now around 8 PM. Just as I finish bringing my stuff back into the main room, though, L stops me to ask if I want to join a Mary's mob. Sure!
Finally after dinner, I've grabbed my stuff and am about to leave when I hear there's a Starcraft puzzle. Naturally, I really want to work on it, but I'm saved from having to make another painful decision of stay vs go when I hear that Snow Day is actually a Blizzard puzzle and the Starcraft part is mostly solved. I descramble the final clue phrase for the Starcraft section and am finally able to make it out the door. I make it back to my friend's place at 11 PM Saturday, about 36 hours after I left it.
...and the renter is on the couch, watching a movie with his girlfriend. grar.
Fortunately, my friend said I can use his room, so I do manage to get some sleep. This time I get a reasonable amount, about 10-12 hours.
Sunday doesn't start great either, as our QM isn't picking up the phone and I'm stuck outside for a while as I try to get in contact. I eventually get through and settle in to solve more puzzles.
Things aren't going well. We only really have enough solved to even approach 2 of the 6 real metas, and haven't solved either one. Our team normally gets a meta by Saturday afternoon or evening; 0 of 6 metas solved on Sunday afternoon is a bad sign. While we aren't having the greatest hunt ever, I also don't think we're that bad. Then again, I hope it is just us, as Sages isn't giving out hints and the alternative is that this will be the new record Hunt for time.
I check back in on Snow Day and see it's gotten further and is stuck on extraction. After familiarizing myself with the current status (which admittedly has too much data), I spot the solution.
As I'm continuing to bounce around puzzles trying to find something to work on, I hear that someone finally got the second round meta. This was another case of "we know what we probably want to do, but it takes an inordinate amount of time to do it". Solving the meta unlocked an obstacle training event on campus, and I head over with some others to tackle it. This is our first glimpse at the larger structure of the Hunt!
Earlier, I remember someone on Sages mentioning that although they were a large team, they needed everyone to run Hunt and had roles for people to play. Up until now, we hadn't seen much sign of this; with the exception of maybe the telephone event, it didn't seem like anything required more manpower than normal. The obstacle events must be where people were needed.
We were ushered into the training center, a room with Sages seated behind counters (most of which were closed), humorous notes jotted on the blackboard, lines for "amateurs" and "professionals", and a waiting area. We received a short briefing, then were shown into another room with just a laptop and a Griffin PowerMate. The screen showed a picture of a safe's dial, and turning the PowerMate was reflected onscreen. It took us 17 minutes to figure out the combination (mostly because we had forgotten about the safety briefing, which contained a vital clue), and chatted with the Sages representative. She said from the perspective of the training center, things were going at the expected rate. This was a little weird, because from our perspective, thing were going very slowly. I mentioned our frustration with puzzles taking longer than we were used to, concern that Hunt would end late, and how I was surprised that they weren't giving out hints (based on the solve rate and our projected progress so far, I would have expected hints to have started flowing at least by Sunday morning, maybe earlier).
My team is getting older, and there's not a lot of influx of new blood. We have day jobs, and (aside from me) people aren't willing to sty up as late. It's also good courtesy to leave things clean, which takes time. As such, we had already decided to stop puzzling at 4 PM on Sunday regardless of our progress in the Hunt. We ended up staring at the first meta for a while, as we had all but one answer, but were unable to come up with a working extraction mechanism. Because of that, we didn't actually end up stopping until maybe 5, after which we hung around and chatted. At the point we stopped, we had 1 meta and obstacle solved, all the data for another meta (but no idea how to extract), maybe 1/3 of one more round solved, and negligible amounts of the remaining 3 rounds. For our team, that's a pretty poor showing.
From the Sages' emails, it was clear things were unraveling. They were offering more and more hints and had first increased the rate of free answers to 0.3 per hour, then again to a whopping 1 per hour. For context, free answers are pretty powerful, as you can use them to get past that one puzzle that's holding you back from solving that one meta which is keeping you from endgame. In previous years teams went to events to earn maybe 2-3 free answers over the course of the entire Hunt, and there was no automatic granting. A free answer per hour is faster than you could solve most normal puzzles (much less these special extended edition puzzles), and was a sign of how badly things must be going. Sages HQ was apologetic and said they had originally intended to be done by Sunday afternoon and really wanted to be done by midnight.
Regardless, we didn't care - we were never interested in winning, and weren't even solving puzzles at this point. We headed out to dinner at 8, and then we split up for the night.
When I went to sleep, I was unsettled that not only had the coin not been found, Sages was giving out even more hints. Not a good sign.
Early in the Hunt, our team privately joked that if Hunt went past midnight on Sunday, then the Hunt itself would win and everyone, including Sages, would lose; their punishment should be to write it again next year (and the next, and the next...) until they got it right.
Ive been doing Hunt for a while. The longest one I participated in was the Time Bandits hunt in 2004, which had it's own share of major issues. Time Bandits ended (I think?) at 9 AM Monday. The next longest was the Matrix hunt in 2003, which ended at 7 AM Monday. Hunts since then had gotten better about keeping time under control, with most ending Sunday evening.
I woke up at 10:30 AM Monday, intending to meet up with N for lunch at 11.
There was no email that the coin had been found. Hunt was still going on.
Despite our HQ shutting down, we were technically still in the game. We had accumulated 17 free answers (not counting ones we had already spent). This was enough for us to get all the remaining answers in the last round and still have one free answer left over. Craziness. And sometime overnight, someone on our team finally solved the first round meta, and we were eligible to run the corresponding obstacle training event. Most of the remaining team is off socializing at a teammember's house (it turned out to be further than I wanted to go). I want to check out the obstacle training event, but my phone's run out of power (necessary to setup an appointment) and I forgot to bring my charger. I wandered over to the training center, but as expected, they're pretty frazzled and don't have the bandwidth for me to run the event. Oh well.
Someone on our team is still solving puzzles, so I run one more data gathering task over in a nearby library. Side note: I think it's poor form to locate important puzzle information is located inside a building which is closed for part of the time. Anyway, as I'm sending the pictures, I see an email from Sages that the coin has finally been found. The winning team only had 5 of the 6 metas, which was a trick by Sages to reduce the time even further, as it's usually the last remaining meta which is the worst.
It's a new "record": 3 PM on Monday, with massive hinting (and even then the Hunt was incomplete). Sages certainly did do something that's never been done before, and I hope won't be done again. To their credit, it sounds like this wasn't intentional and that they tried to fix things when they noticed (too late) that theory and practice were, in practice, not the same.
I missed wrapup, as the length of the Hunt pushed it past my return flight. I'm really curious what the Sages and other team's experiences were like, and I'll hopefully be able to catch a recording.
Although I was pretty grumpy at the time, now that I've gotten some sleep the puzzle constructions themselves was pretty impressive. There was some errata - not as few as the previous year, but still only a handful. Things generally weren't broken, and I'm guessing Sages did a good job testsolving.
My major gripes were that each puzzle had a higher than normal effort:reward ratio. A lot of the time it felt like there was a bunch of work for the sake of work, and that the puzzle didn't benefit from the tacked-on elements. Even if the puzzles had been normal length, the metas didn't really seem backsolvable, so teams had to finish more puzzles (which in reality were also longer, compounding the problem). My guess is that finishing the Hunt without hints or free answers would have taken twice as long as it did. It's not surprising that the Sages' estimates were off, as it was their first time running things, but it was surprising how far they were off and how long it took them to react. I guess the former is somewhat understandable given how fast they finished last year's Hunt (Saturday at 10 PM), but I have no explanation for the latter.
My armchair psychologist take is that Sages wrote a Hunt they wanted to solve (which is sort of natural; your playtest feedback will reflect your team ethos) and applied their own experience to extrapolate other team's status. Unfortunately, this didn't mesh well with our style, or perhaps other teams. I bet most teams rely on backsolving to make up for being smaller (certainly ours does, and we're not even a small team - we're probably normally medium-large), and that didn't seem possible this year. Longer time to complete puzzles meant it's more likely you'd be caught up in the time release waves, and only a large team would be able to keep up.
Over dinner on Sunday, some of my teammates pointed out that the Time Bandits hunt was significantly more broken in terms of puzzle construction, and therefore was a worse hunt. I'm not sure I agree, as I'll tolerate some brokenness in favor of the satisfaction of solving more puzzles. In Time Bandits, we were able to get through to endgame. This year, we finished a third of the metas, with little hope of getting more.
This has nothing to do with the Sages or this hunt, but I'm a little disturbed that we left some low-hanging answers untouched. For example, there was a puzzle where the extracted instruction was to write a Cracked-style article, which we started on but never finished. Not much effort and we would have gotten another answer, as compared to spending lots of time on a different one and potentially not coming up with anything. It says something about the coordination level of our team; I'm worried that we might be no longer competitive (even if Hunt hadn't been broken). Don't get me wrong - I love seeing everyone, and I don't want to win or write, but I would like to be on a team that comes close to finishing. Hopefully next year we'll do better.
Like last year, the winning team makes me apprehensive again. Fortunately, unlike Codex or Sages, they've written before. Unfortunately, the one they wrote was Time Bandits. Hopefully they've learned since 2004.