Confusingness when playing

Mostly in parser, but also considering the thing this revolves around, I’d say choice based games as well.

What makes a game’s mechanics go over the edge from understandable to confusing? Where is the line drawn? Good examples just on either side of the line, maybe?

I’ve got a massive WIP which is very odd and super duper confusing, with many different levels to each bit. It is also ZIL, so I can’t do very special text effects (timed text is fine, so is hit any key and very limited input buttons are functional!): in particular, fade-in-out and/or multiple windows are off-limits. Well, at least for now. (The latter would be really helpful at some point in the future, and I know how to do it but it’s just very hassly.)

I can’t figure out techniques for introducing these mechanics in a way that doesn’t necessarily break the fourth wall. For context, the game not have NPCs per se, but there is a narrator.

Any tips?


This question runs into the problem of all puzzle design: the author is trying to be confusing, but not quite confusing enough to confuse the player. The audience demands a challenge but also wants to surpass it.

Therefore, every game is trying to do something which is rooted in familiar genre conventions, but which goes a little bit past them in a way which is surprising until you think of it and then seems obvious in retrospect. This is of course impossible. :) But you have to do it anyway!

The only possible answer is know your audience. Play a bunch of current games and see what current players think of as “understandable” and “familiar”. And then accept that there are a lot of kinds of player, with different levels of experience, and no single game will work for all of them.


One technique is to ramp up the complexity of the game mechanic. Here I think about Portal and Portal 2, where the gameplay started as just jumping, and moved on to using one portal, and then two portals, and then added enemies, and then in part 2 it added all kinds of new physical elements. Throwing the player into part 2 would have been overwhelming, but the gradual building of difficulty allows the player to overcome obstacles that would otherwise be insurmountable.

I would also recommend early “usability” testing if you’re worried about overwhelming the player. The tester should be instructed not to worry about rough edges, but just to comment on “confusingness” or accessibility of puzzles.


It seems like you’re asking about how to present a challenging, obscure puzzle in the most accessible way. That’s quite the conundrum.

You could take a page from the Zelda games that force the player to learn a couple of things early on to progress, and then the player takes their new found knowledge to greater heights later on. Maybe work out the challenging puzzle you want to have happen, then dissect a couple of key elements and make one or two mandatory (yet easier) puzzles out of those elements. Teach and prepare your players, essentially.

I would try not to present these learning puzzles as in game tutorials though. It should feel organic to the game. When someone learns without knowing they’re being taught, that’s the sweet spot.

Edit: I think I said what Phil said, but in a different way. :crazy_face:


I recently played a critically acclaimed game that was often talked about having surprises and secrets. Many of them are competently designed and require some observational skills that aren’t common in mainstream games. I can see why it’s so appealing.

But I actually think I would’ve liked the game more if it was more confusing and obscured. I play a lot of “jank” and hardcore puzzle games with strange and counterintuitive mechanics that lead to surprising discoveries. In some ways, I want games to go over the edge. Confusion is really a subjective state of mind: once you understand a puzzle, no matter how high concept it is, it is never going to be confusing to you.

So, it definitely depends on the audience you want to cultivate. You are going to playtest a lot, I imagine, to find that audience niche you want to get. Fairness is kinda a “language-game” (to use a Wittgenstein-ism): it entirely depends on the community’s implicit rule-set.

I’m always looking for hard puzzles, so most puzzle games are not going to be for me. You probably shouldn’t pander to me if you’re not going for puzzle aficionados. But if you are, I’m always happy to playtest :slight_smile:


I would be delighted to reach out to you when I’m ready!


Playtesting is really the only way to know for sure.

Here’s something I do:

Nobody I know IRL except my BFF will play parser IF, but they will read a transcript. So I print out a transcript of the beginning few scenes where the player is learning the mechanic and ask them to read it with a piece of paper covering the next player commands as they go. Even non-players can be really helpful with saying things like “That’s way too much information all at once.” You don’t have to get a bona fide parser-playing play tester to get feedback on how you’re introducing a learning curve.


Mate in 6 chess problems fit that just fine. No problem moving the pieces. Knowing where the pieces should go, almost impossible.

You just have to figure out the key move/idea that makes everything easy!


Nintendo uses a four-step structure for its mechanic levels. If you use something similar, it’ll probably be easier to figure out from a player’s POV.

  • safe environment
  • unsafe environment
  • twist
  • conclusion

GMTK video and transcript


I can think of several types of unknowns in a game - knowing what the state of the world is, what actions you can do and what you can do them on, what the immediate results of those actions will be, what the more distant consequences of those actions are. Even in an entirely non obfuscated game it will require some amount of exploration and experimentation to learn these things.

I don’t think confusion at the level of a single mechanic or object is desirable or something that needs to be balanced to make a fun challenge. If the player cannot tell what an action is doing, or cannot tell what state an object is in or how it got there, that is almost always a failure in design, though I’m sure there are exceptions.


I can’t really add much to what has been said about testing. I never seem to know what people think, so testing is my only recourse. Try to find new people for every round, even if that means initially turning some people away.

Phil’s suggestion (and example) re: incremental complexity is a great one that I want to second.

But I really want to extrapolate from Zarf’s suggestion regarding the audience. I think it’s important to figure out who your audience is. Not everyone will love every game equally. If possible, you want to figure out who will love your game. What kind of player would be attracted to your work, and what kind of information would they want to have? This sounds like kind of a weird, abstract question, but testing will probably bear it out. This gives you targets and a way to imagine your work as it might be experienced. Since you cannot make all games for all people, think about your audience and build toward it.

Very relatable. I really struggled to find a good voice for a “tutor,” and I am still struggling, since I can’t repeat the same trick in the next game!