Bury, Sep 2018
Vault 17 is a game with a reputation. A couple of months after it opened, word spread that it had a zero percent success rate – not a single team had beaten it. Of course, that’s not a good thing in an escape room. It’s easy to make an escape room unreasonably difficult, the trick is to make it challenging but also scrupulously fair. However, I’d also heard that it had subsequently been beaten, and more importantly that it had been overhauled to make it a more reasonable game, so was curious to give it a go.
The backstory is that you’ve spent many years living in an underground vault after nuclear war reduced the outside world to a wasteland; but now, with the vault about to fail, you’re doing your best to escape. Fans of the Fallout computer series will immediately recognise that premise, and there are several affectionate nods to those games in the room’s decor and design – including optional blue jumpsuits to wear while you play, if you wish. There are also very pointed references to current-day politics, which I suppose could be controversial for some players but which I found a very welcome touch of humour.
As well as giving the story, the briefing also explains that you’ll need to progress through a series of chambers, with a different requirement to open each successive door. This is therefore a very linear game, primarily consisting of a sequence of challenges to be beaten one at a time. It’s also one that deliberately avoids use of padlocks, instead using maglocks and other hidden mechanisms. Neither are there any walkie talkies or hint screens – if you need help, it’s delivered by an assistant named Lex, whose robotic voice emerges from a speaker in the first room.
Vault 17 has a great deal going for it, which makes it all the more unfortunate that I found it simply not very fun to play. That was despite having unusual puzzle ideas rooted in the story and setting, a clear narrative that throws in a twist or two; and also a very impressive set. They’ve done a lot with, I think, not a lot of budget; the result is dusty and cluttered with discarded industrial whatnots, but also manages to create a gritty and convincing atmosphere where the decorative highlights are properly impressive. Lights and locks often appear to be under the direct control of the host rather than triggered by sensors, but the design thereby manages effects that would otherwise be difficult to achieve. That all gave it the potential for a first-rate escape room, one with a genuinely cinematic feel. However, all that falls apart if it’s not also fun and well-designed.
Amongst a variety of gripes, inevitably I’m going to complain about the searching involved in the game, which we naturally struggled with. Searching is far from my favourite part of escape rooms anyhow, particularly in a dark and cluttered space, and if items are too unreasonably hard to spot then it can seem an unfair task – but even so, if we fail to spot something then ultimately that’s our fault for not looking carefully enough. However, the structure of the game made items hard to find in a different way. With such a linear game, we focused hard on whatever the next puzzle in the sequence was, and the host was dimming the lights in the areas we’d left, which discouraged us going back and checking for things we’d missed. Perhaps the intention is that players should go back and search harder only when an unsolvable puzzle makes it clear that there’s something else to find, but still: failing to find something in an area that the game is actively driving you out of feels unfair in a rather different way to simply overlooking something extremely well hidden.
A more acute frustration was a set of audio clues that were simply too quiet to be made out, even with repeated attempts. We eventually got past that by a combination of guesswork, gamemaster clues and luck. The audio balance seemed odd in other ways too, with the periodic ‘time remaining’ warnings coming through very loudly, but the hint system being too quiet to hear unless we hurried back to the start area to listen to them. That seems like a problem that’s very easy to fix, so it was surprising to find that it was a shared complaint for other groups who played Vault months earlier.
(I don’t mean to imply that the host was obstructive – he was responsive to requests to bring the lights back up, and repeated hints when asked to.)
While the puzzle content was superficially creative, that was undermined by unnecessary inconsistencies, such as where puzzles appeared to need a certain number of inputs and turned out to need one or two fewer; or where one of a set of puzzle components looked different to all the rest, making it unclear whether it should be included or not. Other puzzles relied on us having found search targets, which is fair enough, although each time that meant an interruption of flow as we went back and searched once again; and then a surprising number of puzzle ideas all shared a similar substitution mechanic in a way that became repetitive.
There’s certainly a danger of viewing everything through the opposite of rose-tinted glasses, with each frustration adding up to make the rest of the game appear worse than it was. I can imagine some teams could have a much better experience in Vault; the criticisms above are somewhat subjective and may not bother some teams, particularly ones who do well at the game. I think there was a subtler reason I disliked it though: a great deal of the game felt like we had to jump through arbitrary hoops. The puzzles were themed well for the environment, but rather than feeling that we were finding clever ways to master the environment with occasional help from the host, many tasks instead gave the impression that we were dependent on the gamemaster’s whim, and were allowed to progress when he decided we’d got close enough to the intended actions.
I’d expected Vault to be substantially reworked from its initial incarnation, but comparing notes with teams who played Vault at a much earlier point, it doesn’t appear to have changed all that much. I didn’t think it was impossibly hard, though it’s certainly very difficult (yes, we ran out of time, with maybe a puzzle and a half left), and there are various reasons for arguing that its difficulty level isn’t entirely fair. Still, the spread in the ratings between me and my teammate don’t reflect a difference of opinion, but rather genuine uncertainty about how other teams may find this. If all the above doesn’t put you off, then by all means give it a try: expect a tough game that’s not always very fair, take the rougher edges in your stride and underneath all the problems there’s what could be a distinctive and interesting game. 
