Gamasutra: The Art & Business of Making Gamesspacer
View All     RSS
November 1, 2014
arrowPress Releases
November 1, 2014
PR Newswire
View All
View All     Submit Event

If you enjoy reading this site, you might also want to check out these UBM Tech sites:

The Digital Demise of Wonder
by Tom Battey on 05/04/12 11:57:00 am   Featured Blogs

The following blog post, unless otherwise noted, was written by a member of Gamasutra’s community.
The thoughts and opinions expressed are those of the writer and not Gamasutra or its parent company.


Has the internet robbed games of any sense of mystery?

I recently played through Polytron's Fez, a wonderfully absorbing little platform adventure game from the Xbox LIVE Arcade. The standout thing about Fez is its sense of mystery, the way it creates an ethereal feeling of wonder with a range of often obtuse, mostly logical, but at times seemingly impossible puzzles. It's the first game I've played in years with a notepad next to my control pad, and this made me pleased.

Except the whole time I played, every time I scrawled down a potential solution to a stubborn question-mark labelled room, part of me knew I needn't have bothered. I puzzled away for the joy of it, but I knew that at any point I could just turn to the internet and solve the whole game in minutes.

And eventually I did. I don't have the patience or the mental ability to decipher entire languages and counting systems, and I always knew that eventually I'd turn to the net to get those last few pesky cubes. And as soon as I did, the game wasn't fun anymore. That sense of wonder was lost, and the game boiled down to a mechanical process of visiting rooms and punching in codes until I'd hit 100%.

You could argue a lack of fortitude on my part, and certainly there will be a few stalwart individuals who will refuse to look online until they have solved every puzzle for themselves, but most people operate on a sliding scale of patience. If a solution eludes them to the point of true frustration, they'll hit the net for an instant solution.

In our interconnected, always-online world, is there any way to counter this? And is it even something we should try to counter? After all, genuine, point-of-anger frustration is something that should generally be avoided in videogames. No one wants controller-shaped dents in their walls. But I'd argue that games like Fez, which place such an emphasis on puzzles and mystery to engender a sense of wonder in the player, are somewhat undermined by the easy access of instant solutions the internet offers.

So how can designers overcome this? One way is to offer an in-game hint system. Leaving a game to check the internet, though quick and easy, is both an effort and an admission of defeat on a player's part, and most players will want to exhaust the options offered by the game itself before going online for a solution. An in-game hint system can offer a stop-gap between frustration and a cheap cheat-sheet, and a smart developer can weave such a system into the game's fiction in a way that doesn't cheapen the experience.

This was demonstrated rather well in Amanita Design's Machinarium. Machinarium is a point-and-click adventure title with a heavy emphasis on puzzles that, in the tradition of the genre, often initially seem obtuse and nonsensical. The game offers an in game hint-book with one caveat: to access a hint for any screen, the player must first complete a mini-game involving steering a key around hostile spiders.

It's a great system. First, the spider mini-game is deliberately somewhat unpleasant to play, which stops players from immediately accessing the hint-book as soon as they enter a screen, without at least trying to ascertain the solution for themselves. Second, the hints provided are clues rather than outright solutions, which, while making the actual solution to the given puzzle much more obvious, means the player still feels a sense of achievement upon finding the solution.

An important point is that the spider mini-game is never so difficult or frustrating that a player won't at least try it before turning to the internet. Certainly there will be some impatient completionists who won't waste their time with it and will go online immediately, but this system offers most players an acceptable middle-ground between bewilderment and frustration.

It's a system notably absent from Amanita's latest game Botanicula, which, incidentally, is a lovely little adventure title highly worth a look. There was one point, however, where I found myself stuck in a reasonably large environment with absolutely no idea where to go next. It turned out I just hadn't clicked on the right things in the right order in a previously explored room, but without any clues from the game, I found myself turning to net, which resulted in me kicking myself for not trying the solution earlier and enjoying the rest of the game slightly less from that point.

A well thought out in-game hint system in one way to stop players turning to the net. Another option would be to offer non-binary solutions to puzzles. That is, to make the solution to the puzzle dependent on the players actions within the game, rather than a single pre-defined method.

Most of Fez's puzzles have one solution, usually a code, which means an aspiring cheater can just download a list of rooms and their corresponding codes and have most of the game solved in minutes. There's one puzzle, however, where the solution is dependent on the time (in real hours) that the player started playing Fez. This means that while a player can look online and find out how to solve this puzzle, the solution cannot be given to them directly. They still have to go back to the game and figure out how the solution relates to their particular instance of the game world.

There are many ways of tying a game's solution in the player's previous actions; maybe a puzzle depends on how much of the map they have explored, or which order they visited certain rooms in, or how many of a certain object they have collected. This relegates internet cheat-sheets to a similar level as the Machinarium spider-book hint; they can shape the solution of the puzzle for you, but the player still has to go and actually solve it for themselves, and as such they retain some of the sense of achievement and participation from the puzzle.

A third, potentially more powerful (but definitely harder to design and manage) solution to the problem of internet cheating is to take a look at ARGs, or Alternate Reality Games. Extra Credits have done a great piece on ARGs for the uninitiated, but basically an ARG brings elements of a videogame into the real world, through the internet, presenting a puzzle that can only be solved by online cooperation, and relies on interactions, sometimes physical, sometimes digital, in the real world.

The previous two 'solutions' I outlined treat the internet as the enemy, as something to be avoided or circumvented in order to retain a game's sense of mystery. An ARG embraces the internet, and makes online communication central to the game experience.

There is a particular puzzle in Fez which, perhaps inadvertently, led to an ARG-level of cooperation across the internet to find the solution. The solution was found by an organised brute-force testing of all possible solution, and as far as I'm aware an 'actual', in-game solution has yet to be found.

Now, ignoring the fact that the final solution to this one puzzle is still just a code that anyone can now find and input in minutes, this puzzle is interesting in that it brought players outside of the game, onto the internet, to discuss possible solutions. Rumours of possible solutions have sprung up, been tested and declared false, all by people who clearly love the game. In this instance, using the internet has actually increased the sense of wonder and mystery about the game, and has created such a sense of connection that certain players are effectively playing the game even when they're not actually playing the game.

That's quite powerful stuff. Now, integrating ARG principles into a normal, non-ARG game would be an incredibly difficult task for any developer. And even then, they would have to try and ensure some kind of ongoing solution, otherwise all their efforts would achieve would be a prolonging of the usual internet cheat sheet approach (despite, or in fact thanks to, the efforts of the online community, the Fez Monolith puzzle can now be solved with a cursory glance at the top of a Google search.)

Still, these methods prove that there are ways to combat internet cheating, and potentially embrace it to enhance the game experience. If nothing else, it proves that the internet doesn't necessarily spell the demise of mystery and wonder in game; it just means that developers must employ some puzzle-solving thinking of their own if they want to avoid their players effectively spoiling games for themselves.

Related Jobs

Rensselaer Polytechnic Institute
Rensselaer Polytechnic Institute — Troy, New York, United States

Assistant Professor in Music and Media
The College of New Jersey
The College of New Jersey — Ewing, New Jersey, United States

Assistant Professor - Interactive Multi Media - Tenure Track
Next Games
Next Games — Helsinki, Finland

Senior Level Designer
Magic Leap, Inc.
Magic Leap, Inc. — Wellington, New Zealand

Level Designer


Bart Stewart
profile image
It's probably instructive to think about that word "cheating."

What's the point of playing a single-player puzzle game? If it's the pleasure of perceiving elegant (or at least correct) solutions to a series of puzzles, then looking up answers online is, in a way, cheating yourself of that pleasurable experience.

If on the other hand the point of playing a game is to win -- to conquer and get past challenges -- then looking up the answers to puzzles is not cheating at all; it's simply being efficient in reaching the "win" state.

The point here is that different people are going to approach games in different ways, and it's to be expected that they'll respond to any particular game based on their preferred style of play. Someone who enjoys exploring a complex problem-space will spend hours tinkering with just one puzzle in SpaceChem, while someone who enjoys simple and pleasantly repetitive activities will turn to YouTube a few times before deciding that it's a terrible game, not even really a game at all. (Of course it is a game; it's just not *their* preferred kind of game.)

So the first thing in designing a puzzle game is to understand that even if you design it for people who enjoy thinking about how to solve puzzles, it's going to be played by people who just want to win. And those latter folks are going to share the answers with the world to prove they won. This is why things like "thottbot" came into existence.

This means the puzzle game designer has a choice: build the game knowing that people will put the answers to every puzzle online and hope that the people who enjoy solving puzzles won't be put off by that, or design the game's puzzles in such a way that the solution to each one is unique to that player at that moment.

People have been trying that first approach for a long time now. And as you note, it just doesn't feel like it works all that well in an age when it's easy to share solutions.

The second approach is harder. It cuts way back on the number of puzzle types that you can imagine (or borrow from Martin Gardner's "Mathematical Recreations" column in Scientific American). It also has the downside -- a significant one for a commercial game -- that it excludes the "I just want to win" gamer almost entirely. They're not into puzzle games to start with; thinking and imagining and perceiving are not their preferred forms of fun, and when those are the only way to win they won't play.

If you're OK with that, then how do you make a puzzle whose solution is unique to each player?

A starting point can be found in SpaceChem. Rather than defining a single perfect solution to each challenge, SpaceChem's "convert these inputs to those outputs" puzzle style allows for multiple ways to accomplish each goal. This means the player can either accept the first solution that meets the requirements, or keep thinking and looking for ways to solve the challenge using either the fewest number of symbols or the shortest number of cycles. In a way, this design creates two games in one. One game is about finding any solution (which works for the "just want to win" players), and the other is an efficiency game -- find one of two mutually exclusive optimal solutions -- which the "perception is fun" gamers can still enjoy even after the "win" gamer has moved on.

(In a way, this non-binary, "multiple solutions to challenges" design approach is also the theme behind the Looking Glass school of gameplay, as found in games like System Shock and Thief and Deus Ex. I believe this is why these games appeal to the thoughtful/imaginative player as well as the dextrous action gamer.)

SpaceChem's isn't quite the "unique solution for each gamer" kind of puzzle, though. "Win" gamers can still brute-force solutions, and they'll still upload to YouTube the most efficient solutions (in both cycles and symbols) for every challenge.

Something close enough to unique might be a puzzle design where the puzzles are procedurally generated, and where the number of possible puzzles is something like a hundred million or more. A game might consist of a set of 30 puzzles randomly selected from the full set of possible puzzles. Even working together, gamers will not figure out all the solutions and post them online. Of course, in this case you have to be sure that no one can figure out the procedural generation algorithm! (Good luck with that.)

Finally, there's the puzzle design where each puzzle really is unique to each player, so that the solutions can't be looked up but must be solved personally. The "elapsed time since the start of play" puzzle in Fez comes close to this form. Taken to its final form, it might be possible to design a game where the kind of puzzle you get in one round depends on some elements of the puzzles presented in previous rounds.

The solutions to the first few puzzles would still be posted online. (Actually, that might not be a bad thing as it would help new players ease into the game.) But the further into the game, the less likely it would be that someone else would have gotten the same puzzle, solved it, and posted the solution. In a way, this is really a variation on the "massive solution space" approach, but it does have the virtue of including the player's actions in the generation algorithm.

There are almost certainly other ways of designing a puzzle game so that each puzzle has more than one correct solution. The point is that this is probably the necessary approach if you really want puzzles that remain as much fun for the perceptive gamer as for the persistent player.

Tom Battey
profile image
I think you're right in that designing puzzles depends on your intended audience. If you're design a straight-up 'puzzle game', then there's not so much of an issue, as you'd hope the people playing would be the kind of people who actually enjoy puzzling the way to each solution by themselves.

I'm not big on my puzzle games. I lack both patience and a suitably logical mind, so it doesn't take much to get me stumped, and when I am, my persistence diminishes rapidly. I'd never take on something like Riven, and while I love the idea of SpaceChem, I'm pretty sure actually playing it would quickly drive me mad. And that's fine, I totally get the appeal of these games, they're just not for me.

I do however have a soft spot for the old adventure games, though usually more for the story and the ride than the actual puzzles. The puzzles in these games, however, do add an air of mystery and immersion that they'd lack without them, and games like Fez and Machinarium have inherited this allure. So while I may well suck at the puzzle aspects, I'm more willing to persevere and try to crack it for the joy of being able to progress further in the game. I like it when a game makes me feel smart, and rewards me with more narrative or world to explore.

I don't believe that every game should be designed with every player in mind, but I think the internet cheat-sheet is a particular issue/temptation for this kind of player, who inhabits a sort of middle ground between the perceptive and the persistent. We relish the feeling of solving a puzzle for ourselves, but not so much that we're willing to give up seeing the rest of a game's world in the face of a particularly stubborn puzzle. Knowing the solution is just a mouse-click away makes it all the more tempting to just cheat our way through to the end and forego a sense of real achievement just to see what else a game has to offer.

The solutions you've outlined will help with that, and I do think designers are aware of this - recent Zelda titles have given quite exhaustive (but largely optional) hint system for struggling players, where older titles would have just expected the player to deal with it.

I like the idea of puzzle solutions that rely on the solutions to previous puzzles - that way a game is sort of 'training' a player, and getting assistance in earlier puzzles isn't such a sign of defeat when the choice to do so actually affects what comes later in the game. It would be interesting to see a game designed with this as a design principle.

Nathan Mates
profile image
The problem is that not all puzzle games are exactly logical. OldManMurray's takedown of Gabriel Knight 3 at shows games being deliberately obtuse for the sake of being challenging. When games defenestrated any sense of logic or fairness, gamers went straight for walkthrus. It'll take quite a lot of work for games to earn back a sense of "don't run away" -- especially when some gamers are complaining that games are too easy. Those asking for harder games led straight to the GK3 debacle.

Tom Battey
profile image
Just read the Gabriel Knight piece, and it was a laugh, but it does raise a good point - the point that game puzzles stop being logical is the point I usually lose interest in them.

When you solve a puzzle using logic - even it is whacked-out adventure game logic - you feel smart, you feel like you've learned something, and it creates a positive feedback loop with the game. You feel smart, so you want to progress, to persevere and prove that you are, in fact, smart.

When you solve a puzzle like the one outlined in that article - and I've no idea how anyone could solve that beyond random trial and error - all you feel is a sense of 'what the hell?' If you get stuck on that puzzle, and have to turn to a walkthrough, only to discover how obtuse the solution is...well that's just infuriating.

A games puzzles should always be rooted in logic - even if it is a logic invented for the game. It has to make some sort of sense, otherwise it's just not fun to solve.