Zarfian Forgiveness in Choice of Games

I’m finally doing the legwork to put together IFDB pages for the games available from Choice of Games: Choice of the Dragon, Choice of Broadsides, Choice of Romance, and Choice of the Vampire.

I find myself a bit stumped at the question about Zarfian Forgiveness.

Let’s focus on Broadsides. It is possible to die in Broadsides, which IMO automatically means that it’s not “Merciful,” though it is impossible to get “stuck,” in the sense that you’re unable to make the plot progress. Furthermore, the game doesn’t have an UNDO feature, which maybe disqualifies it from being Polite; all actions are irrevocable.

We went to some effort to guarantee that it’s impossible to die early on in the game; no single action can cause the main character’s death (or cashiering or other bad endings). To find yourself in a no-good-ending state in the last chapter, you’d have to have failed at (almost?) every earlier chapter earlier in the game.

But no one of those failures are damning, so it’s never “obvious” that any one of them is a fatal mistake. So, on a fairly literal reading of Zarfian Forgiveness, Broadsides is Cruel. But that just doesn’t ring true to me; our automated tests show that you can get a non-death ending most of the time just by playing at random.

We also tried to guarantee that even the death endings are pretty cool, so even the “bad” endings feel fun.

So what are we, anyway? IMO, our game feels Polite, in spirit.

Well, wait, re-reading again, since every action is obviously irreversible, I think Broadsides is just Tough, not Cruel. I guess I can live with that, but it still doesn’t sit right with me.

Our games are not hard to win; we tell you when you’ve lost points. We just don’t tell you whether you’ve lost so many points that the game is unwinnable.

Then again, perhaps I’m wrongly conflating Forgiveness with a “difficulty rating.” Many text adventures are difficult to win because they are unforgiving; some are difficult to win even when they guarantee that the game is always winnable.

Does a game automatically upgrade to Cruel if there’s just one rare action that renders the game secretly unwinnable, even if the game is guaranteed to be winnable if you don’t perform that rare action?

What if the game is supposed to be guaranteed to be winnable, but the author introduces a bug that can get you stuck? Can you be Cruel by accident?

I think you may be, a a bit. As I understand it, Zarfian “cruelty” is deliberately very separate from difficulty.

Dunno. I’ve asked similar questions in the past (on Usenet) and the grey areas were left feeling pretty grey. I suppose they’re meant to be.

Prior discussions on this subject have been far clearer: Cruelty isn’t about bugs, at all.

In games, the difference between a death ending and a non-death bad ending is mostly cosmetic. Either way, the game is over and you lost (or didn’t win). So, if it’s not possible to die but still possible to get stuck in a bad ending (eg, prison), then that raises it a notch or two on the Zarfian cruelty scale.

If the bad ending prematurely ends the game (before the storyline has finished unfolding, that is), then I’d say it’s completely equivalent to death in game terms, even if your character doesn’t die. If the bad ending comes at the completion of the story, then it’s not exactly death, but not exactly fair if it’s easy to wind up at that ending.

I would rate those “choice of…” games as fully Cruel, because there is no ability to undo actions, and the negative consequences of your actions can show up many turns after you’ve made a choice, with little warning. To “do over” a bad choice, you’d have to play the entire game over again from the start, possibly several times, before you figure out what went wrong. That’s exactly the definition of a “Cruel” game. (As I recall, courting a mate in “Choice of the Dragon” was extremely difficult, and I had to play the game about 10 times to work that one out. I probably played “Vampire” 20 times before I concluded that both of the player’s potential lovers were doomed by the plot, and no choice I made could save either of them.)

But then again, Zarf’s scale was mainly designed for IF games, which are much more “open” in their approach, and it’s not like an official rating system anyway. So… grain of salt, I guess?

As I understand it, the cruelty system was created in the days before UNDO was very much use, so it isn’t really accounted for.

No, it’s not. If you make the game unwinnable and don’t realise that you’ve done so, that’s not automatically Cruel; you have to not realise that you’ve done anything irrevocable. All the Choice of Games games make it pretty clear that you won’t get to lawnmower their choices until you get what you want; you know you won’t be able to go back and have a do-over on the mutiny. So that would make them Tough.

It’s a semi-official rating system; it’s used on both IFDB and IFWiki, and it’s the closest thing to a difficulty rating we have. Personally, I wouldn’t enter a Zarfian cruelty rating for Choice of Games pieces, because the system really wasn’t designed for that kind of game.

Hey, when are we getting Choice of Crackhead Psychopath? I’m hoping it’s cruel in the zarfian scale.

everything is pretty cool and fun and shallow in trash culture.

I always thought Zarfian Death Cruelty was about simulator death, and other such simulator-y dead-ends that a player may blunder into, and not death as a legitimate story ending. No?

C.

It’s probably no fair for me to butt into this discussion…

Mm, no. What’s not accounted for is multi-turn undo. (At the time some Z-machine interpreters only supported one turn of undo, and the Inform library was built to enforce this limitation.) So it makes a distinction between “you screwed up but know it” (and will undo immediately), and “you screwed up and will find out later” (and will have to restore).

What I had in mind was “the best story ending” versus “other endings”. (Infidel has always been a prominent case of a best story ending where you die.) I don’t think I put any thought into games with several true endings. I probably would have said, well, it becomes a matter of which ending the player wants to reach.

The Choice-Of game model (though not the choice-game model in general!) seems to presume that every ending of the game is equally valid – there are no “you screwed up, ha ha, you’ll have to start over” ends. I guess it could still depend which ending the player is aiming for, but I don’t know how much your players think about that sort of thing.

Yeah, that’s what I meant; I should have been more specific. (I thought about saying ‘multi-turn undo’ and then didn’t, for no very good reason.)

This talk of multi-turn undo puts a question in my mind: does anyone know of any games where “undo” as the first move produces an interesting or useful effect? (Perhaps in a situation where most alternate first-moves bring a precarious situation to a premature end?) With multiple-undos you could get a regular “Memento” going, beginning at the end.

First I thought “I’ve hit on a stroke of genius!”, quickly followed by a deflationary “but with the proportions of genius here, I’m likely re-inventing an awesome wheel.” In any case it’s taken me 10 years (and counting!) to implement a relatively simple game design, so perhaps I shouldn’t invest myself too much in bringing anything more complicated to fruition. If you do decide to run with the gimmick, however, should it prove to be original, do please feel free to credit me for coming up with it! 8)

I’ve seen it suggested, and I wouldn’t be surprised if there’s a game or two that uses it as a one-off cool effect, but I don’t think there’s anything in IF that uses it as a core mechanic. There are a number of games that are centrally about working backwards, but not by messing with UNDO. (Outside IF, Braid is pretty close to it.)

(Though games that work by messing with RESTART are well-established.)

Pick Up The Phone Book And Aisle does it. It’s a hack, of course; ordinarily trying “undo” when the turn count hasn’t advanced yet will get you “You can’t ‘undo’ what hasn’t been done!” This can even produce problems when you enter a command that doesn’t advance the turn count.

I thought of Braid too.

I can’t personally say I’d ever run with it, but as gimmicks go, it’s certainly worth implementing a funny response to :slight_smile:

That’s my interpretation, too. And, as I see it, the game would continue to be Tough even if it had infinite undo, as long as the negative effects happen multiple turns later.

For example, CYOA #1 Cave of Time is Tough by that definition. According to this chart, as soon as you reach page 22, you’re cut off from all of the good endings, but you won’t know it until you explore its entire subtree.

I do think the definition of “irrevocable” could stand to be clarified, not just to clarify infinite UNDO, but also to games (like ours) which don’t even allow you to freely save/restore. If you want to go back and change your mind, you have to restart the game, from the beginning.

I think right now the background assumption is that unlimited save/restore is possible; the question is whether the player will know when to save/restore. In Polite games, you never have to restore, because you can just type UNDO once. In Tough games, it’s obvious when you need to save. In Nasty games, it’s obvious right after you needed to save, which lets you UNDO once and then save. In Cruel games, you have no idea when you need to save; if you don’t know the game is Cruel, you might not even know whether you need to save.

Of course, there’s the separate problem of knowing when to restore. In games where you can get stuck, even if you saved before an irrevocable action, you may not be sure whether you’re in an unwinnable state. It makes the game much harder, because you need to not just explore the game world as it is, but also in other possible branches.

What would you make of a game that said this? “You are dead. You should never have eaten the cake 57 turns ago. Click here or type OOPS to go back 58 turns, to the moment before you ate the cake.”

For another turn 1 undo, see the most recent Inside ADRIFT.