Save Systems and the Art of Game Flow

In Features by Chris Durston


The History of Saving

In the beginning, games were short. There were limited narratives, and games were not only capable of being played through from beginning to end in a single sitting but designed to be experienced in that way.

As games became lengthier, telling bigger stories and requiring more of a time investment, it became apparent that it would no longer be possible for someone to sit down, start playing a game, and finish the thing in one go. No longer could a player see all there was to see in a single session; games needed to come up with a method of allowing them to put it on hold, to leave and come back to pick up from the point they’d left it. There were limitations, though: early devices lacked storage that could be written to and read from, and any tiny little allocations of free space would have been too small to hold all the required variables. Passwords were an early solution: once a player reaches a certain point, give them a code which can be entered later to return to that point.

Image result for bomberman password

External storage eventually allowed home computers to store information about a player’s progress for loading in next time; consoles like the PlayStation also introduced memory cards, while other consoles had no storage on the system itself but opted to include battery-backed RAM on game cartridges, allowing save files to be kept on the actual game media. These days, of course, we have computers and consoles with internal and external hard drives, or even cloud-based storage.

We also have much lengthier and more complex games these days, which can take upwards of 40 or 50 hours of continuous play to complete and could include millions of potential combinations of variables, information about everything the player has done to that point. Fortunately, the storage capacity of today’s machines is such that we can store save states of enormous complexity. Without this facility, we’d be very limited in the size and depth of the games we could create and play, unless we were willing to play all our games in a single session each.

Types of Save System

The ability to save is something that dramatically affects the player’s experience, then, to the point of even making long games possible to play where they otherwise wouldn’t be at all practical. The save media, or save technology, is the important development which allows us to save, but there’s another concept that is probably more interesting: the save system. The distinction between the two concepts, at least as I’ll be using them here, is that the former is to do with the method by which the data is stored while the latter relates to the player’s experience of storing that data and the way this facility is represented within the game.

Image result for pokemon save game

I think there are three main methods of saving in modern games — there are most definitely examples of games that use more than one, or combine or overlap multiple systems, and there are probably even games that do something entirely novel that I’ve not thought of, but these are the primary three I can come up with:

  1. Save points are perhaps the classic implementation of a save system. These allow the player to save at predetermined times and locations: after leaving one save point, the player has to make it through a certain amount of the game before they’re able to save again. Save points generally use save slots, allowing the player to have multiple saved games.
  2. Quicksave allows the player to save at any time. This requires the player to manually choose to save, but isn’t tied to a specific point. There could be a single save slot, overwriting each time the player saves, or there could be multiple slots.
  3. Autosave is basically the game doing an automatic and continuous overwrite of the player’s progress. It’s kind of like an evolution of the idea of checkpoints, wherein the game sets a flag at particular points so the player can restart a level partway should they die. These days, games can continually keep track of progress so the player can quit at any time.

As I mentioned, there are of course examples of hybrid systems; perhaps the most obvious would be something like the approach employed by a lot of JRPGs (I’m thinking specifically of FFVII) wherein the player can save anywhere on the overworld but requires specific save points the rest of the time. This is a sort of quicksave-save point mashup, not quite allowing anytime/anywhere access to saving capabilities but not being as restricted as a save-points-only method.

Kingdom Hearts 3 and Negative Save Flow

What’s caused me to think about all this is playing through KH3, which — as has the rest of the KH series to date — uses save points, but also has an autosave function. I’m not used to KH having autosave, and I also think (from both my own experience and from having read a few other players’ accounts) that KH3‘s autosave function is a little spotty as to its frequency and comprehensiveness — which is to say that you could still lose quite a bit of progress if relying solely on autosave, I think — so I’ve been using the save points.

Image result for kh3 save point

Of course, save points as the only method of saving is what I’m used to in Kingdom Hearts; that’s what it’s always had, up to now. I’m really noticing with KH3, though, that the flow of my gameplay is really heavily dictated by the save points. Perhaps it’s because I’m now an adult, sadly: I don’t have the ability to play for hours on end nowadays, much as I wish that weren’t the case, so I need games to give me regular natural break points to allow me to leave and come back. In KH3, there are lengthy periods where you’ll go through a big, important chunk of gameplay — things you don’t really want to have to invest all that time and effort into only to have to repeat — and then get upwards of 15 minutes of cutscenes and other little transitional thingies before you’re able to save again.

Even if you’re using autosave, it won’t kick in until after each big gameplay-cutscene-cycle is finished. Your only options if you want to leave early are therefore to skip the cutscenes (which I can’t recommend doing, at least not on a first playthrough) or to replay a hefty chunk of stuff. There can be a really long time between opportunities to save, consistently so, and I don’t think that’s good or accessible design.

This isn’t to say that KH3 isn’t a good game overall — I love it — but it can veer a little bit close to overstaying its welcome in a single session for me, simply because I’m frequently getting impatient waiting to be allowed to save again so I can turn the PlayStation off and go do Important Life Things (much as I’d rather just keep playing). That real-life issue, which shouldn’t be part of my experience of playing the game, has affected my enjoyment of the game experience, which sucks.

(But I still really, really like KH3. Just to be clear.)

Image result for kh3 selfie

Using Save Systems to Create Positive Flow

It’s been an interesting experience going through this frustration at KH3‘s approach to saving. Autosave is often held up as a staple feature these days, something that modern games should include as a basic Good Design Thing, and perhaps if KH3‘s autosave were more frequent I’d feel more comfortable relying on it, and therefore less annoyed with having to get to the next save point. That said, I still think there’s something to be said for games continuing to implement a restricted save system. Sure, I’m frustrated by this particular restricted save system, but I think smart design can make use of this tool to add to the experience rather than subtract from it.

One of the most restricted save systems is the one employed by the Resident Evil series in a few of its earlier entries: save points are presented as typewriters, making them both location-specific and diegetic (part of the game’s internal universe rather than, say, a glowy thing that doesn’t seem to really exist in the game’s world). These typewriters aren’t free to use, though; players must make use of Ink Ribbons, an item in limited supply, to save their game.

Resi games are designed to put the player on edge, to be tense and frightening, and so this limiting of the player’s ability to save their game — to be free to leave the game, in some sense — is a great example of a game employing a functional element of its design to add to the tone and feel that it’s aiming to communicate. The intended reaction from the player is that they ought to be feeling nervous about the experience, so Resi makes use of every tool at its disposal in order to ensure that that’s the reaction it gets.

Interestingly, though, it’s not just tension and nervousness that this system ratchets up. It can also provide an opportunity for those feelings to be released, for moments of rest before resuming the onslaught, and that’s important too. If we think about narratives in any medium (movie, book, game, even music), no good story stays at one pitch throughout; there are always ebbs and flows of building up the stress, piling on problems, and then allowing resolution. The ups and downs are what create a feeling of emotional movement, and a Resi player’s rising fear as they go ever further from their last save point is allowed to calm when they reach the next (if only to immediately begin rising again).

Image result for resident evil save game

More broadly, and speaking about games in general now rather than any specific examples, I think clever and well-thought-through implementation of save points can provide the player with a feeling of rise and fall, of tension and resolution, of flow.

Speculating on Saving

I think an ideal save system for any game is one that helps to delineate the structure, building up and then taking a breath. It also needs to facilitate accessibility, creating natural break points at times and places that are sensible for players to leave from and return to.

I think of good save placement as being a little like well-structured chapters in a novel. Chapter breaks serve a few purposes in service to narrative, structure, and format: a well-placed chapter split not only provides a good break point, a cue to the reader that this is a viable place to put the book down and return later, but also outlines borders between the rise and fall of tension and resolution. Of course, chapters should also end in such a way that the reader should feel compelled to continue.

Image result for do you want to save your progress

Conceiving of save points similarly could build a kind of reflexive emotional response in the player: condition the player to think of save points as markers that this is a breather moment in the narrative, a chance to regroup from previous adventures in preparation for the next ones, and space them in such a way that the player not only has the real-life facility to turn off the game at regular and satisfying points but also the structural ability to divide the narrative into easily digestible chunks (again a point both for accessibility and for creating satisfying, not-overly-convoluted structure in the narrative and gameplay flow).

As I consider this, it hits me that in fact the earliest games to implement save points were probably doing this all along. Early Final Fantasy games paced their saving opportunities really well, both ensuring that players could take a break at regular and sensible times and providing much-needed relief while the tension was otherwise rising towards a climax. Where did things take a left turn, then?

Perhaps the assumption of autosave as a modern convenience has led games to depend too heavily on it. When the game is constantly saving and players can drop out at any time they like, how is the game to create a structure of build and release, of rise and fall? It has to assume that the player could quit at any moment, and make each possible next moment one at which the thread could conceivably be picked up. How, then, is a consistent and cohesive experience to be designed, with this in mind?

A novel with no chapters — or, to go one further, no paragraphs, just continuous prose — would be a tricky, inaccessible read not just because it would demand to be read in one sitting but because it would almost certainly find it difficult to structure its narrative in a way that made sense and felt sensible, which would also hamstring its ability to communicate cohesive themes (unless the theme was of madness, or the nature of language, or some other literary didacticism). Readers would have a difficult time putting it down and coming back to it; resuming the story would be intellectually challenging, and the emotional experience developed by the previous parts would have faded by the time the reader came back. You’d have to try to maintain the same themes and emotional resonance throughout the entire thing, rather than being able to build up and then release to a place where the cycle could be restarted. Deprived of a strong narrative and of any kind of thematic power, I can’t see that we could consider that a good novel.

A Vision of Flow

I hope that games won’t abandon the idea of having the player actively have to save, rather than simply sitting back and assuming autosave will do all the work for them. I think the kind of structure that well-placed save opportunities can lend to a game is something that probably gets overlooked when considering the quality of the assembly of a narrative — or a game as a whole — but I think it would be a real loss if it were to become a forgotten art.

It’s not that I have any belief that developers (or players) are getting lazy, or that autosave is a Bad Thing All The Time; far from it. I have a bit of a concern that dependency on autosave, which may be a legitimate quality-of-life improvement in a lot of places, might have the unintended side-effect of causing the structure of games to flatten, to be deprived of the ups and downs that make for a powerful emotional experience. Those things contribute to flow, a word I’ve been using without defining but that I basically think boils down to how closely-aligned the actual experience of each moment of the game is with what the player wants to be experiencing at that moment, whether that be a rise or a resolution of tension (that being the base of all narrative structures, arguably all artistic structures). It’s frustration at not being able to quit on one side, complacency at being able to quit any time on the other, and satisfaction as the golden mean sitting happily in the middle.

Do I really think that this horrible future, in which games are consigned to flatness and reduced emotional resonance, is likely to come to pass?

Nah. People are too creative, too determined to make something beautiful and worth experiencing. These kinds of cautionary notes are just the sort of thing that some talented person could spot and work out how to put a spin on, to turn into an even better experience.

I think I’ll say that I have faith and leave it on that positive note.


Before I go, credit is due to my partner Hannah for giving me the idea for this one.

Chris Durston on Twitter
Chris Durston
Contributor
Aspiring non-procrastinator; amateur writer of odd fiction and Thoughts About Games.