I’ve just started reading Tynan Sylvester’s (he of RimWorld fame) book “Designing Games: A Guide to Engineering Experiences”. I’ve only read one chapter so far but I’m enjoying his point of view.
The principle is that well balanced game mechanics help players (a) enter a flow state and lose their sense of self, and (b) go on to create arousal in the player, then (c) the fiction helps the player to label this arousal (e.g. a fiction around zombies allows the player to label their arousal as terror), where (d) the combination creates a pleasurable and memorable experience.
flow + arousal + labelling -> experience
He makes the point that arousal is undifferentiated, it is the job of the fiction to convert it into something meaningful that creates an experience.
In addition, if the mechanics fail (because they are too easy or too difficult) to help the player immerse themselves then the experience is likely to be compromised.
It’s funny, I’ve looked at RimWorld numerous times feel very drawn to the gameplay even as the art style has put me off. But I think I might need to try and get over that.
This chapter is essentially about developing taste and not something you can teach. He gives a few examples of elegance in terms of how mechanics interact but nothing prescriptive.
Essentially “design lots of stuff, test it out, develop your nose.”
Also, since RimWorld is in the Steam sale I decided to pick it up. I’ve only spent about 20 minutes with it playing the tutorial but I’m glad I got over my dislike of the art style as there is much to love here (yeah, yeah, I am way late to this party!)
RimWorld definitely appears to be one of those games where initial mechanics overwhelm is a potential problem. There are so many systems and they seem a little disparate at first.
Nah, there are so many different forms of interactivity – you have to figure out which ones you personally are most interested in (and it may be different for you as a player vs. as an author, too).
It may also be useful to think about the different layers of what’s going on in your game: they don’t all have to be linear or non-linear together. For instance most parser IF has linear stories, but moment-to-moment the PC can do anything that the “physics” of the world allows – much more non-linear. Or you could have non-linearity in the sense of say, litrouke’s January where it’s a static set of vignettes and you’re going to see them all (I think?) but you have quite a bit of freedom and some notable constraints on what order you experience the pieces in.
Or you can imagine a game where the overall story is “the same” every time but you are only going to see a fraction of the pieces on each run, so you might focus on a different aspect of the PC’s story each time, and only piece together the full picture of everything that’s going on through a bunch of playthroughs. Is that “non-linear”? The terminology is messy.
Then too, by non-linear are you thinking branching-dialogue, or something more simulationist (Social Democracy: An Alternate History, etc.), or something in between?
Huh. I wonder if you’d like Stacey Mason’s Responsiveness in Narrative Systems – it’s a 300-page PhD thesis but I think it’s not hugely unapproachable (beyond its size) and it’s a handy framework for thinking about when and how the game/system changes in response to the player, so you’re putting that effort where it’s more noticeable. Last year I attempted writing an outrageously abbreviated 3K-word overview of mostly the first half, which might give you a better idea of whether you want to dive into the whole thing…
Thanks for the link & esp. the summary. As you say, a 300 page thesis is quite an ask so I am grateful for that
You raise an interesting question about where non-linearity is important and I am not quite sure. I thought I knew until I considered some of your examples!
Update: I think my lack of certainty comes when I consider story and experience. What threw the cat among the pidgeons was the January example. My initial instict was to say that this was not meaningfully linear because although you can choose a route the end-state is the same “I have experienced every vignette”.
But although it’s not required, the order could matter very much in terms of what experience you have. If half the vignettes were happy and half sad, you’d likely have a very different experience if you got all-sad then all-happy vs all-happy then all-sad vs happy-sad-happy-sad-…
But there is no authorial narrative hear linear or otherwise.
An amusing illustration of non-linearity and branching leading to a single end state can be seen in Taskmaster Series 9 Episode 2. The two teams experienced radically different “narratives” en route to the same conclusion. (Rather off-topic to IF perhaps, but no mention of Taskmaster can ever be wrong.)
I think this is a pretty idiosyncratic definition of “nonlinear.” In a game design context, I have always seen “linearity” used to talk about whether things always happen/are seen/need to be done in the same order. You seem to be talking more about what’s usually referred to as narrative branching, which is not unrelated to linearity but is distinct from it. A game where you always complete the same tasks in the same order but get slightly different text at the end depending on how you completed them would, I think, be generally considered more linear than a game where you can do things in whatever order you want but the ending scene is always the same, even though the former game has slight narrative branching and the latter game has none.
I’m not really talking about my preferences, but about what I generally understand the term to mean in discussions of game design. It seemed like you and others in this thread might be talking past each other a bit because you are using a definition of “linearity/nonlinearity” that isn’t the most common one in this context.
I think “linear” tends to imply that a game doesn’t have flexibility of ordering and doesn’t have branching (or at least has very minimal branching), but if a game is specifically described as “nonlinear” then I would expect flexibility of ordering. I’ve never seen the term used to describe number of end states specifically.
It seems like maybe you’re envisioning, like, a flowchart of scenes that are in the game, and thinking, well, if that flowchart forms a straight line then it’s linear and if it branches then it’s not? But I think the concept as generally understood in this context is more about player experience. A player being shepherded from scene A to scene B to scene C and a player being shepherded from scene A to scene B to either scene C or scene D (depending on their previous actions) are both having pretty linear gameplay experiences. A player who can choose which of A, B, or C to explore first is having a nonlinear experience.
Ohhh, that’s a good point. I hadn’t picked up on that and I was mostly assuming Matt to be using “non-linear” to mean “anything other than a static non-branching story” but you’re absolutely right that nonlinear does have a specific meaning more like “storytelling that plays with ordering” and that’s confusing. Or at least the wrong connotations for the kind of stuff that I suspect he’s doing.
It’s the understanding I’ve formed from discussing and reading about game design here and elsewhere over the years. I can’t cite some kind of authoritative text that dictates that this is the definitive meaning, or anything, but for example here’s an article that defines it in an ordering-focused way. A lot of sources seem to consider narrative branching/multiple endings as a sort of subtype of nonlinearity, but I have not commonly seen that treated as a requirement for nonlinearity or the sole definition of nonlinearity, and the ordering aspect seems to be the primary focus mostly. (The article also touches on linearity being more of a continuum than a binary yes/no thing, and on the fact that a single game can be linear in some respects and nonlinear in others.)
There’s also the fact that in the realm of static fiction, “nonlinear narrative” generally means non-chronological, and I think those connotations tend to carry over to some degree.
Again, I wouldn’t describe a heavily branching story as linear, either, but at the same time it would never occur to me to dispute that January is nonlinear, because it is nonlinear in both the gameplay sense (you can choose to do things in different orders) and the narrative sense (its narrative is non-chronological), and I guess my argument is that this is not a “wow, if you squint I guess you really could see it as nonlinear!” thing, but a perfectly conventional usage of the term.
