Over in the Vintage Game Club forum, Michael posted an interesting question, “Do modern gamers still want puzzles?”:

I’m working on a blog post focused on this, but I wonder if puzzles in narrative games are a relic of a bygone era of gaming – or are they a necessary ludic element? Has shooting replaced puzzle-solving as the “gameplay” aspect in narrative games? Is there still an audience for puzzle-solving narrative games? Why not just play a game like Professor Layton and the Curious Village if puzzles are your thing?

(He’s also posted a related blog post.)

Actually, to be honest, my first reaction to this was more that it was a ludicrous question rather than an interesting question, but I started to come around after thinking about it a bit more. Of course, shooting is far too restrictive as a gameplay concept, but say we expand that to include other methods of directly manipulating your character/environment in ways that don’t depend too much on the details of the environment at that particular moment: jumping, pressing colored buttons on a plastic guitar simulacrum while strumming, rolling a ball around the world to suck up items. (Obviously the last two don’t show up so often in narrative games, though now that I think about it, I rather like the idea of a narrative game that you can play with a guitar controller! Like Gitaroo Man, only better.)

If we call the above “direct gameplay” (a lousy name, somebody please give me a better one), then there are real differences between direct gameplay and puzzles. In the “direct gameplay” side, it’s a lot easier to judge the difficulty of any particular situation: if you have two rooms in a shooter, one with twice the number of enemies or three-quarters the available ammo or whatever, you can predict relatively easily how much harder one of those rooms is going to be compared to the other room. Also you can train players to do a better and better job, so that they can reliable successfully make it through situations after playing the game for five hours that would have them dropping the controller in frustration if they’d seen that after five minutes. And you can even tune the difficulty level, so that players of different skill can be presented with appropriate challenges while still getting the same narrative payoff. (Or, as I find myself doing increasingly, players who care more about skill than the gameplay mechanic can turn down the difficulty level.)

With puzzle-based mechanics, though, none of that works: person A may find puzzle 1 obvious while running to gamefaqs (or giving up in frustration) when being confronted with puzzle 2, while Person B may have exactly the opposite experience. I suppose “none of that works” is a bit of an exaggeration: we have intuitive ideas about what puzzles are easier and what are harder, and I pretty much agreed with the difficulty ratings for puzzles in both Zack & Wiki and Professor Layton. And it is possible to train people to some extent in puzzle-solving games: Zack & Wiki again provides an example of that. But it is true that your ability to predict whether or not people will be able to solve your puzzles is much lower than the corresponding prediction in direct gameplay, and that matters: if there are 20 puzzles and you can only predict correctly 90% of the time whether or not people will be able to solve them, then that means players are going to have to run to gamefaqs a couple of times during your game. And I suspect that most game players (and game designers) would rather avoid that.

Having said that, this argument doesn’t run entirely in the “direct gameplay is good” direction: the nice thing about running to gamefaqs is that it reliably works. I’d prefer to never have to go to gamefaqs, but it doesn’t bother me too much if I have to do it a couple of times over the course of playing a game, and at least I can be confident that I’ll be able to finish any puzzle-based game should I chose to do so. Whereas, in direct gameplay games, if something really is too hard for me, then I’ll just stop right there: I can think of a few games (Metroid Prime 2, Final Fantasy VI, Final Fantasy X) where I found the gameplay to be so annoying as to be insulting, and even if I could have finished the game (which I’m sure was the case in those three instances), I had no desire to go through the last few bits of gameplay if that’s the way the designers felt like treating me, despite the potential narrative payoff at the end.

Of course, the boundaries between these categories of gameplay elements aren’t sharp; unsurprisingly, the fuzzy categories can suffer from both problems. Boss battles in direct gameplay games are the best example of this: those are situations where you typically have to both do some puzzle thinking (to figure out the boss’s weakness(es)) and also to be quite good at the relevant direct gameplay techniques. If the game doesn’t balance both of those right, then you can have the double-whammy of the player running to gamefaqs to figure out what to do and also pushing the buttons in just the right way to carry out that plan. (And, indeed, boss battles were my downfall in the three examples I mentioned above where I gave up on the game.)

So, where does the above rambling leave us? I appreciate Michael’s question more than I did when I first saw it, but I still don’t like the suggestion at all: my gaming world would be much much less rich if designers avoided puzzles entirely. And, as he admits, it wasn’t posed completely seriously:

I’m obviously playing devil’s advocate a bit here, but Ben’s frustrated exit from Grim Fandango made me think about these questions and how modern gamers would respond to them.

But if we accept the (to me) obvious fact that game designers need to have puzzles in their gameplay toolbox, that ghettoizing puzzles only impoverishes us all (well, only is too strong a world, given the massive heap of awesomeness that is Professor Layton), what should they do about the above issues?

One obvious suggestion: hints. My first reaction is that that’s a lousy suggestion, that it’s really hard to have hints work well. But if I’m remembering correctly, I didn’t have to go to gamefaqs a single time when playing through Professor Layton to solve a puzzle. (I did have to do that to locate some of the puzzles, but that doesn’t bother me at all.) And, while I would generally prefer to solve a puzzle without hints, they did a good job of giving hints that pointed me in the right direction while still leaving me with a bit of creative thinking to do. So, done right, hints can do a lot to alleviate this issue.

Of course, they’re hardly a panacea: I’m sure I would have felt differently about Professor Layton if I’d needed to look at the hints for half the puzzles instead of a tenth of them, and I’m also sure that, for many of the puzzles in Grim Fandango, you wouldn’t be able to give hints without completely spoiling the puzzle. This is a real constraint on game design, it demands creative solutions, and if the conclusion of this line of reasoning is that the puzzles in Professor Layton are better-designed than those in Grim Fandango, that’s a conclusion that I can accept.

The other technique that this suggests is to use aggregation. This is a standard technique for reducing uncertainty: if you have multiple variables with a wide range of uncertainty, and if those variables are uncorrelated, then you can reduce the uncertainty by increasing the number of the variables. (This is why, e.g., long playoff series are good, at least if you want the better team to win.) In concrete gameplay terms: instead of making the player solve three specific puzzles to progress along the narrative, make them solve three out of five puzzles to progress.

And again (surprise, surprise), Professor Layton does that: they give you a zillion (well, not a zillion, but 150 or so?) puzzles to play; and many of the gameplay/narrative gates force you to solve a certain number of puzzles to pass without restricting exactly which puzzles to solve. (Which is hardly unique to puzzle games, of course, e.g. Super Mario 64.) That’s not the case for every gameplay gate in Professor Layton; is it the case that the mandatory puzzles in that game are easier or more predictably solvable (at least with hints) than the optional puzzles? I don’t know, I’ll have to go back and see. (That certainly is the case with the mandatory gameplay gates in Super Mario 64: they’re battles with Bowser, and those depend on relatively generic platforming mechanics.)

If nothing else, writing this has taught me one thing: not only is Professor Layton a massive heap of awesomeness because of its puzzles and because of its art style, but it’s also a massive heap of awesomeness because of the way in which it navigated the puzzle gameplay design shoals. (If it weren’t late at night, I would come up with a Scylla and Charybdis metaphor here – good thing I’m responding to The Brainy Gamer instead of Living Epic.) Hmm, come to think of it, what’s up with that apparent claim of his that Professor Layton isn’t a narrative game? I sure remember a plot in it…

It’s nice that the Vintage Game Club is giving rise to questions like this. Don’t get me wrong, talking about the game itself with people is a lot of fun, but I’m also happy that it’s giving rise to questions that go beyond the boundaries of that one game. And, reading back through this, I realize that I’ve taken the question in a fairly different direction than Michael has, because I’ve completely ignored the question of what modern gamers want. To that end, I’ll just suggest that a million modern gamers do seem to have some fondness for puzzles…

Post Revisions: