Are there any resources on adaptive puzzle pacing? I’m thinking here of programming constructs similar to the discovery of the ruins in Photopia, or the plane ticket in Shade.
I’m writing an adventure that is heavily reliant on tension and suspense, and I’d like to have dehydration/oxygen level pressure without actually killing the player immediately. I’m well on my way to implementing it, but I always try to learn from others’ mistakes, and I’m having trouble finding similar constructions.
To clarify -
I’m writing a game where the player has to survive on a space ship not meant for habitation past a particular point in the journey. To do so, they have to deal with heat, cold, thirst, and the ship’s automated repair routines. I’m constructing them in such a way that new players will not suffer for not knowing IF tropes - so looking would not decrease their oxygen level in a EVA, for example, nor would most things except actions which lead to the solution of the puzzle. So this is a concept more based around narrative pacing rather than puzzle design, i suppose. It’s a bit of a nebulous issue, and I’m probably exposing how new I am as a result.
A comprehensive illusion of scarcity works as effectively as scarcity itself.
The most interesting part of any time limit, in narrative terms, is the last few desperate seconds. The preceding countdown is just background detail. If players are wasting turns, let them know in increasingly strenuous language, but there’s no reason to actually penalize them until they’re almost ready to solve the last critical puzzles… at which point, oh my, the air is getting thin.
With any real time limit there’s a chance players will waste resources early and discover, at a point past their more recent saves, that the game is unwinnable. Instead, let them run out of a resource dramatically – as or slightly before they use the last bit they need – regardless of how much they wasted experimenting. So long as the writing drops hints that such waste is real waste, tension remains. Analog descriptions (“the needle is well into the red”) work better than digital ones (“10.5%”) for obvious reasons.
Those last few seconds, however? As brutally time-sensitive as you care to make them. So long as it’s crystal clear that the clock is ticking, the shift from illusion to reality can be both visceral and fair.
Noooooooo! Mercy! On behalf of puzzle-challenged players everywhere, I promise to suspend disbelief when I survive my eighth bumbling attempt if I don’t have to reload the game eight times.
I agree that ratcheting up the tension towards the end is great, but I think there are ways to do that that don’t involve brutal timing puzzles. The threat of death ratchets up tension - actual death relieves it, because the game is over. At least for me. Maybe that’s because I tend to die a lot in those situations?
It’s a really good thing to be considering. One thing you might think about is whether you’re willing to force the plot along when the player is clearly struggling. Let’s say it’s been 20 or 30 turns since the player has done anything productive - is there some way you can clue the player in with an updated message (ie “You glance frantically around the cabin, looking for something, anything that will let you get at the Snuggly Bunny Deep Space Warmth Suit . . . and then your eye falls on the bolt cutters, gleaming dully in the emergency lighting.”) I don’t think it’s too far fetched that panicky people get ideas, and some of those ideas could be failures (that point in a productive direction), but it would keep things moving along, which for a new or stuck player, would probably excuse some heavy handedness.
I agree, but for me this is partly a result of the awkward UI experience that these puzzles engender. Either I have to remind myself to save when the action picks up, or I have to use UNDO a bunch of times. The first essentially kills the tension and the second destroys immersion by making me guess exactly where I am in the unfolding sequence of events.
TADS has a nice autosave extension that automatically creates a saved checkpoint to roll back to, and extends the usual RESTORE / RESTART / UNDO / QUIT prompt with a RETRY option. It’s used to pretty good effect in All Hope Abandon. As a player, it feels about the same as the checkpoint system in open-world games, where some missions require you to succeed to advance the plot. I’m thinking specifically of Assassin’s Creed 2 and Red Dead Redemption. RDR had an additional refinement where you could skip a mission altogether after failing it more than three times, advancing to a “success” outcome without penalty.
The Duel That Spanned the Ages has a few timed sequences that are well-handled. One puzzle during the midgame has you open an airlock while low on air. You do eventually die but there’s a generous amount of time to explore the area in the meantime. As an example of what not to do, I would cite Nevermore and the need to periodically refresh your alertness.
Also Oxygen may be worth a look; the major puzzle is oxygen-related and the tension arises from danger to others rather than danger to the player. This is an avenue that strikes me as very promising to explore; placing NPCs in jeopardy lets the player formulate a goal (“stop X from dying”) rather than reacting to a mandatory, author-imposed one (“avoid dying for the Nth time”).
I would encourage you not to rely on atmospheric messages alone. One of the first things I do now when confronted with such messages is to type >Z for a few dozen turns, just to see if I will eventually die and to establish a sense of the turn limit involved. Especially do not make them progressively more urgent if there is no real urgency in game terms. (Divis Mortis was a slight offender in this regard.)
In my WIP, I have a situation where there’s a threat. I made a simple “every turn” message that keeps repeating until you start to deal with the threat. Originally I intended to have this escalate until you die, but I was surprised how effective it was at directing my attention to the situation - when playing seriously, I never felt compelled to let it drag on. I think I will eventually change it so it doesn’t get boring if you try to mess with it like bcressey, but I haven’t decided how yet.
Just sitting with the game and adding little bits of encouragement to push the player towards a goal has resulted in a lot of drama. I’ve really been surprised how well this has worked out. My approach has been to define the beginning and the end pretty clearly, and then fill in the middle bit by bit. It’s been a great way to control how long the game will be, and how much interesting stuff happens along the way.
Hmyeah, I think like capnmikee I pretty much agree with this. I still haven’t caught up to Divis Mortis, but Kevin Mintmeier’s “Containment” from the JayIsGames comp did something similar; it started with a bunch of atmospheric messages about how the nuclear power plant was about to melt down, and my reaction was “Am I about to suffer an annoying timed death?” It was a relief when I realized they weren’t. A timer would have been especially problematic in that game, because it involved a lot of backtracking and experimentation to see what was going on.
I suppose you could argue that, if you’re trying to establish tension, you won’t want the player to feel the relief of knowing that there isn’t really a timer. But you have to balance that against the sheer annoyingness of timed deaths, which scales with how much of the game the player will have to replay. Especially if it’s not really clear that you have to save before the game becomes unwinnable. (And if the puzzle requires a long enough sequence of moves that the player may save after the game becomes unwinnable, that’s – well, hello there, sludge puzzle from Losing Your Grip!) For myself, I suspend disbelief enough that I’m happy to treat a situation as urgent even if I really have unlimited turns. But (like gravel) I’m really bad at puzzles.
(Now that I’ve read Ben Cressey’s approach, I’d be tempted to add a counter that counts the number of times the player waits after the atmospheric messages start ratcheting up. Eventually it would say “You’re just messing around to see if there’s really a timer, aren’t you? A giant foot comes from the sky and squashes you.”)
Oh, and one thing you can do is hand-roll the sort of checkpoint system Ben talked about – in Everybody Dies there’s a timed puzzle at the end that automatically rolls back to the beginning if you run out of time, and Earl Grey does that throughout as does I Expect You To Die. Might be more or less technically difficult depending on what your game is like – all three of those games are very linear, I think, so it’s easy for them to roll the state back. It’s also kind of frame-breaking (though there’s some justification in “Everybody Dies”) – but ragequits are even more frame-breaking.
Same here. I also start "z.z.z.z.z.z.z.z.z.z.z.z"ing if I get a light source - I want to see if it’s got a limit, and if it does, how long I’ve got.
If a game wants me NOT to do it, it had better have proved itself to be engaging, immersive and fair (it may be difficult, as long as it’s fair). I’ve approached too much crap with too much goodwill, and that’s made me set a few hard limits on what I’m prepared to endure.
