Well, thinking back to platformers, there are several artier platformers with no death that don’t suffer for it, IMO; Small Worlds, Nevermore 3, Beulah and the Hundred Birds off the top of my head. (Small Worlds is short but a masterpiece, Nevermore 3 is a leisurely game that might take half an hour or an hour, Beulah is just a sketch but many other of Jake Elliott’s non-platformer games are interesting from an IF standpoint). In these games, if you miss a jump, the punishment is just that you fall down a bit and have to repeat a part over again. These games are mostly about exploration but I can imagine a pretty hard platformer that worked that way. The penalty for not making a jump would just be that you stayed stuck.
Those are obviously the exceptions in platformers, but I think that’s because platforming usually involves honing a physical skill. If you jump and fail, that makes you practice the jump again until (with luck) you can do it. The challenge tends to be avoiding death. In puzzle IF, the default state is to be stuck; the goal is not to avoid obstacles but to find the way forward. So it makes sense that failure just leaves you stuck; and it would be simply obnoxious for failure to force you to repeat something you’d done successfully, when there isn’t even any question of improving at it.
Now of course auto-restarting from an auto-checkpoint would take care of that problem, since the checkpoint is presumably set every time you succeed at something – if possible. But then are we really making irrevocable decisions? If we make the wrong decision, the game resets it for us. In a sense, that isn’t much different from zarf-merciful games; instead of saying “That wouldn’t be a good idea; the bugblatter beast would eat you,” the game says “The bugblatter beast eats you. you have died Press any key. There is a bugblatter beast here!” But then, all peril in IF is a matter of suspending disbelief. Maybe it’s easier to suspend disbelief about the existence of peril if the game actually kills and auto-revives you. (I don’t see how the auto-checkpoint would work with you getting stuck in a dead end, though.)
Yeah. It’s a way to do a tight retry loop without technically having a ‘dead end’ although the falling down functions as a ‘falling back’, which is what closes the loop. The existence of games like this is why I qualified ‘death’ as ‘otherwise arriving at a dead-end requiring a replay of whatever duration’ and also why I qualified by specifying ‘narrative’. Storywise, the part where you fall down a hundred times before making the jump, is omitted, just like any reset after a dead-end. It seems unlikely that your character is portrayed as a klutz who only gets it right once in a while. These are narrative dead ends, only cloaked, by situating them such that you can reset the situation merely by stepping to the left or whatever. It’s simple and it makes a lot of sense. But it’s a rare situation where a plausible version of this technique can be identified; sticking to it in every kind of narrative situation feels too much like throwing away a cardinal rule of good storytelling: “Raise the stakes.”
It depends on how much natural/randomised variation there is on every run; repeating something you’ve already nailed could be a way of getting more story out of it. If there is a possibility that a fresh combination of events could randomly occur, then it could remain fun, but absent that, I basically agree with you here.
Good points and I agree with your conclusion — it’s about the robustness with which you convey the idea that there are high stakes involved for the character. It’s not really about punishing the players themselves for screwing up; as I mentioned, I don’t really believe in that (except insofar as a ‘punishment’ interlude might make for good storytelling – a la Spider and Web).
I wouldn’t choose when to checkpoint based solely on the player’s last success. I am almost certainly going to write a specific ending for every way of failing in the story – so it’s not that much extra work to hardcode a more precise checkpoint at the entrance to every puzzle area/event, setting the player up for the available coming failures. (In my WIP however I solve this in an entirely different way. The game restarts you from the beginning of a level, every time. The only reason this hopefully isn’t terrible is that every playthrough of a level is highly randomised and variable, and also very short. And playing the same journey over and over is, idiosyncratically, required by the narrative even if you play it perfectly and never die; it’s hard to explain further in abstract, non-spoilery terms — anyway, I wouldn’t recommend it as a general solution. I’m not even sure it’s working for me. Experimentation is in progress.)
That game sounds very interesting! I should probably have said that my comments were assuming the more or less standard IF design – fixed world, the possible solutions have all been deliberately programmed in by the author, no or little story branching. Randomization does give you the chance to learn a skill and show it off. And if you can conceptualize your game in terms of discrete levels, then that makes it simple to set the checkpoints. A bushy sort of structure might make it harder.
Another advantage that I didn’t mention is that if you kill the player, you can give enlightening/amusing death messages, which it sounds like you’re doing.
The retry loop in most IF is something like one turn – if you succeed, you move forward, if you fail, you’re in exactly the same place. Which can lead to stasis. For another indie-game example, most of Machinarium has that kind of structure (though there are a couple places where I think you can make it unwinnable, one of which I’m pretty sure is a bug), but there’s one place where you have to
dismantle a ticking bomb. If you don’t solve that puzzle in time, well. (Video spoils a previous puzzle a little.) This switches up the kind of retry loop there is, and it does raise the stakes; most of the game has a leisurely sort of atmosphere of exploring a ramshackle city on a lazy weekend, but giving you as long as you want to defuse the bomb just wouldn’t work. Even though you see the bomb planted very early in the game. (It helps that you don’t lose an appreciable amount of progress when the puzzle resets.)
Thanks! It didn’t playtest very well. I believe the problem was too many story threads too early in the game; the effect was not to add intrigue but rather to obscure the signal with a lot of noise: kind of analagous to (without actually resembling) an RPG that shoves too many unrelated tasks into the quest log off the top.
I tweeted a while ago, that I want solving my game to be not just a process of deducing a correct sequential pattern, but an actual variegated skill: a set of ways of responding to various types of unpredictable stimuli. Also, you don’t always have to do the levels in the same sequence, although there are only 4 levels so you’re right that the level structure itself is not too bushy.
Or trying as hard as I can, at least. 8)
Thanks for the ref. That’s a tough loop to do well. The time limit is of course a mere formality, but to give it narrative weight they put you through a restart animation just long enough to be annoying. The redeeming feature perhaps is that maybe it’s the final puzzle of [a major section of] the game, so the story reward is so close you can taste it; then that loop delay could work – more like a suspenseful tease. If I had to live through a restart animation that elaborate throughout the entire game, however, I believe I would find it very offputting and would groan internally everytime I sense I have entered a difficult loop. I might not get through the game – seeing this response in myself and wishing to avoid it in others is the source of all my thinking here.
You have a playtestable version? I know you said it didn’t test very well, but can you send it to me anyway?
Yeah, that’s the only time that happens in the game, and it’s the fourth-act climax (and something that’s been set up since the beginning of act two or so – I’m approximating the number of acts). That could actually work in IF too – you can’t get stuck while gathering the materials for the climactic fight – but the big boss can kill you, and if it does you get a message like “That didn’t go well. Let’s try it again” and you find yourself outside the cave once more.
Thanks so much for offering – I would love for you to playtest a build: just not this build. The first playtest round (which involved only a handful of RL friends) fared so poorly storywise that I am just embarrassed to release it any further. A full plot rewrite is in progress, replacing all story text, and that should be ready for a second playtest round in perhaps a month. That’s when I am planning to open it up to playtesters who are actually involved in the IF and indie games communities. I will put you on that list! 8)