next »

[BorisTheBrave]

So I was playing La Mulana, and once again had to go for a walkthrough, when it struck me that the puzzles in that game are almost unique in style compared with other games. After thinking about it, I think there's three ways to design a puzzle game:


1) Inside the box
All the rules and interactions are stated in advance, but the content of the puzzle changes each time.
Examples: point and click adventures, chess puzzles

2) Edge of the box
The game introduces new rules at a pace, so the majority of the difficulty is understanding the new consequences of each rule.
Examples: zelda dungeons, braid, space chem

3) Outside the box
The puzzle solutions don't have preestablished rules or patterns. Often, you need to break some implicitly defined rule or trope, or apply real world knowledge.
Examples: La Mulana, ARGs, parts of Undertale and the Metal Gear series.


Obviously, games mix and match between the three, sometimes dabbling in outside the box for a single puzzle (often a Guide, Dang It moment), but it's usually pretty clear which one a game is going for.

What do you think? I'm really struggling to think of any games that lean really heavily of outside the box puzzles except La Mulana.

[s0]

Frog Fractions

[quantumpotato]

http://ifdb.tads.org/viewgame?id=aearuuxv83plclpl

Play this game where you set the letter A-Z on your alphabet gun, then point it at things and change one letter in them to another. Bam, new nown. Turn glass to grass? Try it. A Wall into a Hall? 

[J-Snake]

What do you think? I'm really struggling to think of any games that lean really heavily of outside the box puzzles except La Mulana.

Myst is the typical game for that.

1) and 2) is essentially the same point, there is no distinction as it is typical that most puzzles don't involve all existing mechanics. So 2) rather describes the aspects of pacing, not the puzzle system. 3) is what I consider "fake puzzles".

[TitoOliveira]

How do you think The Witness fit into those categories?
I feel like you have to approach it like an Outside the Box, because you start out not understanding the rules of a given "puzzle type" in the game. And then it transition to the other two.

[J-Snake]

According to my notion the Witness is a mix of 1) and 3). It is a collection of (mostly simple) puzzles and fake puzzles. And when there is running out of ideas a fake difficulty is added to some puzzles which adds nothing to the puzzles themselves. There is nothing more to it regarding the puzzle space. Activating lasers and wandering the landscape is only a means to control pacing and possibly show off artistic values.

You can characterize fake puzzles by their special case existence. They can pretend to have a "fake consistency", but they actually don't follow a set rule that lives within the consistent puzzle space. They only try to give a clue what their special case rules might be. And for that you can take any real world inspirations, like projecting geometry to compose flat pictures, as a typical example.

[quantumpotato]

According to my notion the Witness is a mix of 1) and 3). It is a collection of (mostly simple) puzzles and fake puzzles. And when there is running out of ideas a fake difficulty is added to some puzzles which adds nothing to the puzzles themselves. There is nothing more to it regarding the puzzle space. Activating lasers and wandering the landscape is only a means to control pacing and possibly show off artistic values.

You can characterize fake puzzles by their special case existence. They can pretend to have a "fake consistency", but they actually don't follow a set rule that lives within the consistent puzzle space. They only try to give a clue what their special case rules might be. And for that you can take any real world inspirations, like projecting geometry to compose flat pictures, as a typical example.

The bird tones audio puzzles in the Witness bugged me. Especially from JBlow's talk about "a puzzle should contain exactly the core concept and nothing else". They have you listening to audio and you have to drag to match the pitch - a cool puzzle IMO! Then they start introducing drum sounds and phone buzzes that make it suddenly more difficult to identify the bird pitch - frustrating

[BorisTheBrave]

I'd reject calling these puzzles "fake puzzles". (fake difficulty, sure)
They're a different style, but there is certainly something that needs to be puzzled out. I actually found it quite refreshing to be playing a game where you are expected to go "off the rails". You had to sleep on some of the hints, not to help think about them, but to gain a new perspective.

I never played much of the witness, but it does seem to blend all three. Each time a new rule is introduced without explanation, it's type 3 (as the rules themselves don't have some obvious meta-rule). After you get the rule, there's some exploration of that idea, type 2, and finally there's just some puzzles that are simply hard, even once you've got the key concepts and clear understand what needs doing, type 1.

Myst on the other hand, I see as entirely type 1. Sure, it throws you in with no explanation, like the witness, but almost all the puzzles behave in the same way. They are either lock and key puzzles, or explore/manipulate puzzles. The indiviual puzzles vary in presentation, but none of them subvert expectations set up in the early game, except maybe the "ending puzzle".

[afender7]

So 1 would be Stephen's Sausage Roll, 2 would be Braid and 3 would be the majority of text-adventure games.

1 and 2 are similar, in the sense that you can always take a 1 and turn it into a 2: for example, A Good Snomwan Is Hard To Build has a set of rules that never change and are all accessible from the very beginning of the game, which makes it a 1 (this is excluding the extra 'end-game'...), but one could easily imagine adding levels further into the game which contain some additional objects that interact with the previous ones in unique ways (e.g. fire that melts snowballs), thus adding new rules and turning it into a 2. In other words, all 2s necessarily contain 1, and all 1s can theoretically be turned into 2s.

3s, however, are a different type of puzzle entirely: they have no cosistent systems, or objects/elements that appear across multiple puzzles and behave exactly the same way in each one. I guess they are basically an evolution of text-adventure puzzles, as well as puzzles found in many point-and-click adventure games.

I tend to prefer 1>2>3 in roughly that order. 3s usually annoy me because they have a tendency to require you to follow the same line of thinking that the designer did, and it's very hard to design a puzzle in that style which will make logical sense to everyone who plays it. Some people simply won't think to try a certain sequence of actions, or combination of objects, and so they'll either give up, or solve it through blind trial-and-error and feel like they didn't actually accomplish or learn anything.

[Karin E. Skoog]

The bird tones audio puzzles in the Witness bugged me. Especially from JBlow's talk about "a puzzle should contain exactly the core concept and nothing else". They have you listening to audio and you have to drag to match the pitch - a cool puzzle IMO! Then they start introducing drum sounds and phone buzzes that make it suddenly more difficult to identify the bird pitch - frustrating

I quite liked this part of the game. I actually found it to be an interesting extension of the core principles of the game and while, yes, the additional noises were distracting, I thought it was clever in what it said about the design (especially in the simplicity of the design):


It's a personal interpretation, but I interpreted those puzzles as - there are these elements you try to focus on (in life), but there are also all of these different distractions that get in the way and muddle the message you're trying to focus on (the clarity of message).

When you strip away these distractions, you can gain moments of clarity...but there are also times where your instinct works to your detriment and what initially appears to be a distraction, instead brings you moments of clarity. (There was a specific puzzle in that section that brought me to that second conclusion.)

[afender7]

The sound area is probably the most misunderstood. It is simply the audio version of the perspective-based puzzles. People complain about the sounds being annoying, but that's the point! Just like how many of the trees and other obstacles in the shadow area are "annoying" -- the sounds are just more jarring because of our sensitivity to noise.

[FriendLight]

I find it curious that point and click adventures were categorized into group 1 with chess puzzles. After all, the most common criticism I hear about point and click puzzles is that many puzzle solutions or interactions feel arbitrary. If the player doesn't have the same frame of reference as the developer (for example some real-world knowledge) how the game world reacts to the player can feel illogical.