Friday 19 July 2013

The Swapper Postmortem - What Went Right

I feel like this is going to entail a lot of brain-work on my part when I should be writing my MA dissertation on sustainable contractual justice, so if you don't mind I'm just going to launch straight into it.

Needless to say, spoilers throughout.

What Went Right
Balance of Subtlety in Exposition
It is so hard to find the sweet spot between obvious and obscure. That we seem to have hit upon it is probably a matter of luck more than anything else. The challenge you have with the sorts of narrative mechanics we're using is that:
  • it's hard to keep the player's attention for long, owing to lack of interactivity in the narrative itself and lack of on-screen visual resources (no fancy animations, no facial expressions)
  • when you do have the player's attention you generally have to cover a lot of ground very quickly; a kitchen-sink approach to sci-fi tropes convolutes matters
  • different players pay different attention to both optional and forced narrative delivery; you need to give everyone the bare bones to keep moving forward, but give those that want it that bit more
  • it's super hard to playtest objectively, and - if I can say it without being lynched - informed opinions about games narrative are harder to find than bugs are
My fear in advance of release was that the game would be too difficult to fathom. This, to some degree, has been borne out. An awful lot of reviews refer to the protagonist as a man; in fact that little detail is a great barometer for whether or not someone's been paying attention / I've made it too obscure. 

Incidentally, I read some interesting discussion about whether or not the protagonist's gender is a spoiler. Part of me thinks we should have asked reviewers not to mention it (in fact I rather assumed most would choose not to for players' sake) because it does give you some clue as to her identity and to what's going on. The point made on the comments thread was that since the gender is never actually specified to you there's no reason to assume she's a boy or a girl, and the fact that she turns out to be a girl ought no more be a spoiler than if she turned out to have red hair, or be called Clive.

It's true that we don't do a lot of work in The Swapper to guarantee players' following of the story. In Penumbra we had load screen texts where the player directly reminds you of the salient stuff that's happened up to now precisely to solve these problems; indeed it's something games like Driver: San Francisco and Remember Me do to mind-numbing excess. In The Swapper there are three cinematic scenes which progress the plot, and for the most part they're the only time that we really underline the important points. In fact, it was only a week or two before release that Olli - I think with good insight - insisted we re-record lines for the final Theseus scene which more clearly emphasised the complicated existential situation Scavenger/Chalmers/Dennett finds hirself in. 

On another sidenote, I had male voices in mind for all the characters when I wrote them, but the possibility of a female cast was on the agenda. Olli decided on the switch, and I was all in favour, the week before we recorded. What we left open - because, really, it didn't matter - was the original gender of Chalmers and Dennett. I think maybe we left in some clue in one direction or another, but in the end that character sounds female because it's a female body, and I don't know that assumptions can be made beyond that (unless I've forgotten some further reference in the final script, which frankly is possible).

Stimulating Thought, Not Providing Answers
I think there is an answer to the mind/body problem. I think there is no such thing as a soul or a non-material realm, and I think consciousness will turn out to be identified with some kind of physical function or arrangement, with its experiential quality essentially explained away. Nothing would give me more pleasure than to construct a game like a logical argument for that conclusion, and there are some stubborn explanatory gaps in The Swapper's backstory which might suggest a conclusion of that sort (more on which later), but in the end what accounted for a lot of what people liked about the story and its philosophical content was precisely that it didn't push them firmly in one direction or another (more on which here).

The goal of challenging people's preconceptions regardless of what they were provided plenty of substance for the story. We knew we needed a way of representing these different points of view without bias, and we knew that however the Swapper device itself was involved in the story its use would have to reflect the deep-seated epistemological and metaphysical problems that face those two points of view. You can almost see how the whole story logically arises from those few initial premises.

In the end a lot of bad video games try to be power-fantasies; and presenting a thoroughly materialist manifesto in The Swapper would have been just that for me and, I should think, some large part of the audience. There's nothing unsettling about having your deeply held beliefs reinforced to you. The literature arguing against materialism is vast owing not to some huge percentage of philosophers who believe in the soul, but to the fact that no materialist theory seems to be quite satisfactory to anyone yet, and we need to keep working on it. That observation is at the very core of the game, and of what makes the mind so fascinating in the first place.

The Ending
I've said it before, and hopefully I'll have a chance to say it again: when an ending that I've worked on goes down well it's the greatest complement. It's probably a by-product of my abnormal fixation with logic - just as a correct answer in a maths equation is more important than showing your working, somehow if an audience approves of an ending then they implicitly approve of the whole thing. Certainly in my mind the ending is always the logical conclusion of what's come before (or just as frequently what comes before are the logical steps to reach a particular conclusion).

On The Swapper, as on Penumbra, the ending was stubborn until the last. We went through countless alternatives - some were concepted out, others were ideas that survived mere days. One decision we took early on was to have the game begin and end on the planet, Chori V (so-called during development because one of Carlo's first choir tracks played during that sequence), and that it should obviously involve the use of the Swapper in some meaningful way. This limited us vastly. There's only so many things the Swapper can do, and only so many actors in play at the end of the game. We could let you swap with the Head Watcher, or starve to death, or... about 20 different variations on those ideas. We didn't even consider allowing the player to get off the planet, or be rescued; that didn't seem to deliver on any of our themes, nor did it seem cohesive with the isolationist feel or the substance of the story.

In fairness to him, Olli wasn't satisfied with any of my ideas until the one that made the final game - he didn't let me sit on my laurels - and I think what we got in the end is better than anything we had before. We'd spent days on Skype to-ing and fro-ing over what was basically the same set of ideas pitched slightly differently when Olli suggested that once on the planet we needed some kind of goal or gameplay, and that maybe there was some kind of emergency broadcast antenna which the player could go switch on/off, for some reason or other. None of us seemed to particularly go for it, including Olli, but it did take us on that extra leap towards the idea of a distress signal and a rescue ship. We had to bodge it a bit - the characters have to be unaware the ship is on its way initially else their motivations don't fit - but the idea basically made it through unscathed. 

What worked, I think, was that it put the central problem in stark relief. Previously the alternatives we were working with were swapping with the Watchers, swapping with the antagonist, or simply waiting to die. What made me uncomfortable about those was how hard it is to map them onto particular player motivations. I don't know if the player's chosen to swap with the Watchers because he believes he has an eternal soul, or because he's a reckless adventurer. By bringing in the rescue ship we're able to give the player the binary choice between swapping, or dying, and that's something we can much better map. Consequently if you choose to do what you have to to go on you get a poignant / slapstick scene in which your survival seems assured, but no one seems particularly to know or care whether it's your survival in particular. Provided the moment of that decision conjures to mind the issues we were trying to present throughout the game, all at once, then it's done its job. If the bit that comes after that can prod you just a little bit more then all the better for it.

A small, focused team. The Swapper was worked on by a game designer / artist, a level designer, a musician and a writer. Each of us had a high degree of ownership over our respective roles, with Olli (our team lead) keeping an eye on the overall product. Going into reviews I think everyone had the same fear I did: that each of our areas would turn out to be the weakest link. That most of the reviews commented on the coherency of the complete package I think is testament to what a small, dedicated team can do that is a far tougher demand of larger developers. It also, of course, reflects what you can do when there isn't a marketing department breathing down your neck, or a concern that your game won't sell unless the protagonist is a buff 30 year old male.

Next time: what went wrong.

EDIT: It occurred to me about three minutes after putting this post live that I said I'd run it by Olli before posting, and I forgot to. Having now done so it's apparent that Olli is concerned that a discussion of the game like this so soon after release will influence players' experiences of it. I think that if people choose to read this and that changes, for better or worse, their opinion of the game, then that's fine and good; but in the interests of me keeping my job on the next game from Facepalm I'll be delaying the 'What Went Wrong' until we're somewhat further out the gate (okay, Olli didn't actually threaten me, and I hope he would hire me again whether I did what I was told or not, but best to play it safe, eh?)


  1. Have you played the game? Are you going to? Most interestingly for me, where do you think the reviews went wrong? What could we have done better?

  2. First up, thanks for this amazing game. I played The Swapper close to its release, and I especially admired the atmosphere and (as you mention above) the philosophical depth brought about by only sketching out the story (and of course the fantastic visual and sound design). But since you asked, I'll focus on what I thought could have been handled better: the gameplay itself.

    Don't get me wrong: I thought the puzzles themselves were great, with perfectly pitched difficulty, but I just thought that they felt increasingly like artificial barriers to my progress. In the beginning of the game, the relationship between the gameplay mechanics and the game's themes is perfect, but as you get more familiar with the main conceit of cloning and swapping, the puzzles start to feel like work.

    If we look at the best-in-class example, Braid, we see how this problem can be avoided. In every new world, Braid introduces a new quirk that doesn't just force you into a new gameplay approach, but actually makes you consider its central theme in a new way. This ensures that the link between gameplay and narrative/themes stays intact and the game feels like a whole. This does happen in The Swapper – the first time you have to progress by swapping between airborne clones, thus dooming them to fall to their destruction, comes to mind – but not nearly often enough. Having to deal with limitations like the lights and a limited number of clones doesn't make me consider the existential problem at the game's core any differently – it's just the plot developments that do that.

    I hope this is of some use to you in your further endeavours. Once again, enormous respect for the game, and I'll be sure to follow your work in the future.