Request for advice on A Killer Headache

Competitions IFComp
1

Hi everyone,

I’ve been trying to come up with ways to eliminate unwinnable states in A Killer Headache, and I think I just got an idea for one. Let me lay it out for you:

[spoiler]The biggest and most frustrating unwinnable state that remains is getting stuck behind the diner when the children are chasing you. The “official” solution was to go in the front entrance first, then sneak out the booth by breaking the window, so they never follow you out back. Once they see you go from the road to the parking lot, the game is unwinnable.

I was hanging onto this for a number of reasons, one of which was the lawnmower death scene, which would never even happen if it weren’t for this puzzle. But I doubt that anyone else loves this scene as much as I do, given that it’s very hard to figure out what went wrong.

But I just realized that a solution I came up with earlier for another too-easy state (don’t hate me!) might work for this one too. I didn’t have the pack of dogs enter the diner until a pretty late stage, when I realized you could win the game by outrunning the children, and never use the jukebox at all. Having the dogs means you must be in the dining room with the children before you can win the game. Well, it just occurred to me that this might solve the back-of-the-diner problem too. If I allowed the player to go in the employee entrance instead of stopping you with that dumb “shaking” excuse, the dogs could find you in the kitchen, forcing you to lure the children into the dining room and turn on the jukebox. You wouldn’t need to break the window in the booth, but I’d be okay with demoting that to an alternate solution.

There would still be a time pressure issue: you’d have to get the generator on and slip into the employee entrance before the kids arrived at the Shed. That might mean another unwinnable state, in the sense that “undo” wouldn’t give you enough time to wake up and escape. Maybe I could come up with a range of actions during the lawnmower sequence - attacking the horde gets you killed, but running away wakes you up, or something.[/spoiler]

Please, forgive me for the wretchedness of this puzzle, and help me come up with something better. I’m not sure if there are unanticipated loopholes or pitfalls to this proposal, so I really need someone to pick it apart. I would be deeply in your debt!

2

My personal solution was:

Fill up the gas can and start the generator first, and THEN go back to talk to the nun. It seems that the children don’t start following you until after you’ve eaten her brain, and doing things in this order meant that I never had to worry about the children following me into the back. I did have to wait a while in the diner for the children to come and deal with the dogs, though.

This is a “plan your moves” puzzle. As such, I feel that an unwinnable state is absolutely within the realms of acceptability. If it worries you, then perhaps what you want is a warning to save your game, right when the horde begins to follow the player.

3

That’s encouraging. I’ve been changing a lot lately, but it’s nice to know that if I don’t change this, it will still seem a little bit reasonable… [emote]:D[/emote]

As for a warning, I’d like to do that if I can do it in a diagetic way. But I’m not sure how.

4

Have a flashback to a pre-zombie memory of the guy playing a computer game. He dies in his game and says ‘Darn, good thing I saved a moment ago.’ And there’ll be another incidental character in the room who says, ‘Yeah, saving all the time is a really good idea. As a matter of fact I’d save right now if I could, but I’m not playing a computer game. Oh well.’

  • Wade
5

Or some graffiti. “Jesus Saves. He might need to restore later.”

6

I need to work on making the game funnier, clearly.

Wade, I am not going to make another game without major input from you. I was really disappointed when you didn’t like AKH, and I think if you’d been a tester I could have fixed a lot of that.

7

Hahaha… certainly, a way to push a game towards being one I would like is by letting me push the game around at the testing stage [emote]:)[/emote] But I don’t endorse such an enormously me-pleasing perspective.

More seriously, I can see that upon review, the kinds of things I said in my review of A Killer Headache were said by a lot of people writing their reviews independently of each other. So it’s likely one or many of us may have spoken about those observations if we’d been involved in testing. Though I’m guessing I may also have been slightly more lenient on some of the game’s toughness as a tester. General toughness and playability difficulties are interrelated but still separate. I know that in the context of IFComp, I’m probably not expecting or used to AKH-level toughness, and then there were playability issues on top of that. So I suspect I would probably have been a better player of the game had it been presented to me in a different context, though I’d still likely have been troubled by a lot of the same things. And when testing, that also makes me put on a different hat.

I can say that as a tester I am quite accessibility/playability-oriented. I don’t think ASchultz will mind me saying that in testing Shuffling Around (mind you, it was very early days when I first saw it) for a long time I kept saying things like ‘I think this puzzle is too hard to guess, and this part needs more clues, and I don’t understand this part, and the player needs more help here’ etc. And assessing how the difficulty of a puzzle game like that is traveling is very difficult for all concerned. Andrew was understandably concerned that it could become too easy as he kept working on it. But I’ve got a broad feeling that if you’re on the inside, if you aim for a bit easier than you perceive it should be, that may broadly coincide with your actual desired difficulty as interpreted by a person who never was on the inside.

Anyway thanks, and let me know if you want me to join the test team on your next project.

  • Wade
8

o/t-ing to my game:

I agree with what Wade said and in fact worried about this. The original bit had a lot of extra puzzles etc. and no hints, and it started with a nasty 7-letter puzzle instead of the 4-letter thing.

I tried to consider that things would be tougher than I thought (and even with that I underestimated)–the problem I guess was that I said, well, there are only so many possibilities–and can’t you work through each one? The problem is, you need other people’s bad experiences to see how to avoid them, because you can only guess so much, & that’s potentially awkward even with people you trust.

Also, I had a problem with hints seeming potentially artificial–though I think it’s better to have artificial hints than none at all. Because you can always go in for a little fourth-wall humor if nothing else. The gadget/slider didn’t come along til later, and they required planning.

I also have to admit that I knew the game was probably too hard in the first draft, but I didn’t know how to make it easier. That said, I didn’t want to sit on my hands until something popped in my head (didn’t work in 2011,) and seeing what people tried and amalgamating that data helped a lot.

A few transcripts from different people often gave me enough notes to figure how to make things more streamlined & I found stuff that testers would just throw out could be really useful or lead down a good path. In any case, I think you need to be able to say “This should/could show the way later.”

And I say more streamlined as opposed to simpler here–I think a game can be a great way to get people to think and try more than they’d originally have thought, but they shouldn’t be forced into it.

My concerns SA might become too easy ultimately seem invalid–looking back, I think if the game is immersive enough, the people will ignore the logic-grid hints for complex items like, say, the drainage and just try and get through.

If it’s not, and they’re stuck, they’ll appreciate the way through. So I think that is a win-win.

Either way, nothing’s too obvious except for the fun introductory puzzles & even those frustrated some people.

As for helping the player, that’s so subjective that it’s often hard to tell when your idea is a big help or small help. But I think it’s valuable to keep notes on where people got stuck and try to empathize with them. And it’s important to try to give hints several ways in any case. It was also frustrating for me to tell a tester “Yes, you’re right, it’s too abstruse as-is. But I don’t know what to do, and I’ll take my lumps if I can’t think of anything in time.”

But I know that asking people to revisit certain puzzles–and their explaining what they thought of the new version–was tremendously helpful.

It wasn’t until post-release that I decided to

combine the slider and gadget, add alternate/extra clues for 7- and 8-letter puzzles, let the player know there are no words above 8 letters (except for the 2-word puzzle) and allow for a substance that will let you see 1) what can be changed and 2) the basic story ahead.

The alternate clues, of course, required time to seem valid and not too spoiler-y. I may still have to ask my testers which specifically are good/bad. And I’m still implementing the last bit. But I hope this shows that there are a lot of ways to make small and big strides for hints & I think they are all good.

Because I admit I don’t mind being semi-fooled by a puzzle, especially when I realize it was clued other ways. In my case, I want the subjective and objective descriptions to clue what an item should be, and I think combining “mathy” and “English-y” hints is quite a good idea.