GDC 2010: Please Finish Your Game

Oh, shit, this reminds me: I have to renew my indie game developer membership card!

I found myself agreeing and disagreeing with Chris Hecker here. As a concept artist I value interesting ideas and multiple explorations (and iterations) thereof quite high. I often… no, always, value a good concept higher than a polished turd, or even a polished mediocre idea.

I’ve grown to like games which are not very close to the typical local optimums of fun mechanics – i.e. stuff that isn’t easily prototyped. It’s difficult for me finish anything because of my tendency to want things which require a leap of faith or foresight, and certainly a lot of work before it starts getting fun.

When it comes to pleasing me and the minority who share my preferences, neither small or big devs are succeeding. For me, finishing is more about coming up with deep and engaging mechanics than about keeping mechanics simple in order to make the project ‘finishable’ on other levels. I can take a lack of polish if what’s underneath is really interesting.

That said, I believe that there are a massive amount of unexplored fun simple ideas (e.g. Tetris) out there still and right now indie developers are probably in a better position to explore these than big developers (or ‘finishers’) are.

Theoretically, indies are also in a better position now to explore ideas which are further out there in ‘difficult to make art assets for’ territory, because people are OK with indie games not looking very polished. DF comes to mind, as an extreme case.

“I dunno, I probably should have made my saving throw and not said anything. My November 2009 talk wasn’t remotely about “what it is to be indie”, you should read the indiegames summary again.”

@Chris: Definitely not, thought it was a very interesting talk, and sparked some rad discussion and questions that I’d not asked before. Thanks for dropping by anyways, I appreciate the answers.

Hey Paul, if your opinions are not based in any logic, then what use are your opinions to us logical human beings?

Towards the content of the rant itself:

Actually, I think one of the finest counter-arguments against “Focus on your game and Finish” is Andy Schatz.

It’s a great story if you haven’t heard it. He was developing Dinosauria (multiple years of dev time planned) and he decided to take a break and make this little game he always wanted to try out called Monaco.

While it wasn’t a weekend project (i think it was like 6-8 weeks) it was certainly a departure from The Game He Had To Finish. If he hadn’t done that, well, then I would be much sadder because I love playing Monaco.

Of course, it comes full circle when you set your sights on the fact that he _is now_ Finishing Monaco. That’s great and I couldn’t be happier he’s Finishing it.

I guess I’ve always seen jam games and prototypes as a healthy thing in my career as a game developer, and it would seem the same would be true for Andy and Monaco.

One of my favorite quotes from an OG indie no less; Thoreau: “Foolish consistency is the hobgoblin of feeble minds.”

While I think you’re right, Chris, that more Finished games would be a great thing for the medium. The difficulty of that (for those of us below the poverty line, which I’m guessing is a vast majority) has much more to do with the many impasses we face, than it does the lack of a will to Finish.

@greg – i’m not sure i follow your question. are you saying that statements which aren’t logical arguments should never be said? how about such statements as increpare’s statement that he went into indie games after depression, and that it relieves it somewhat. that isn’t a logical argument. he’s just stating something true about himself. and something that’s interesting to other human beings, to boot. similarly, let’s say you tell someone that you love them. you aren’t making a logical argument, you’re just stating a status report or condition within yourself which is relevant to them, and may make them feel good. should people not tell others they love them because it’s not a logical argument, just a statement of internal feelings?

“Any particular definition or feeling of finishedness, or any particular definition of anything, when it imposes itself on or intervenes in a creative process, is legitimate to question, examine critically, and ultimately deny. At least I think so.”

Well I’m not saying that you can’t use roughness as a textural quality or anything. Or a deliberate open-endedness. But if “unfinished” as a word means anything at all (and if we can’t agree on the definition of words, then we can’t really communicate) it surely implies “not enough”, so in that sense it’s a judgement of the work. And if the creator believes that something is “unfinished” then that implies that they have given up before reaching a successful resolution.

That’s not to say that there is no value in unfinished work, I mean Picasso considered Les Demoiselles D’Avignon to be unfinished, but it’s still one of the most influential artworks of the 20th century. But I think that every creator who wants to create good work (as opposed to somebody who’s just interested in going through the process), is hoping to arrive at something that satisfies their own sense of “finishedness”, and to give up before you get there is always a failure of sorts.

Thanks for your interesting response, Anythony – I hope you don’t think I’m splitting hairs.

*'(and if we can’t agree on the definition of words, then we can’t really communicate)’*

That you know that you can establish that you don’t agree on something presupposes some communication has taken place (though maybe it’s not what you would consider real communication?). Borges (I think – I can’t find a reference off-hand, though) didn’t like dictionaries at all – does that mean he didn’t care for meaning or communication? I agree it can be very frustrating to talk to someone who has very different ideas of what words mean, though : /

*’But I think that every creator who wants to create good work (as opposed to somebody who’s just interested in going through the process), is hoping to arrive at something that satisfies their own sense of “finishedness”, and to give up before you get there is always a failure of sorts.’*

I find your mentioning of process interesting. I had never really though of things in that way before with regards to games – I don’t think I’m focused on process as a thing in and of itself (beyond its practical aspects), but I do think that the process by which a work was created is very frequently present in the experience of the end work. When I encounter an obviously big-budget game, I might think “how many tens, or hundreds of people spent years of their lives working on this?”. That’s obviously not what the game is about, and people with personal experience are going generally to be more sensitive to it than others, but I think in principle it might be something that people can make use of.

A number of arts do without the notion of ‘finishedness’ just fine – one can take improvisation, say. An improvisation does have a beginning, a middle, and an end, but notions of ‘finishedness’ are a little vaguer. Say one was to make a game as one might improvise a piece of music (A lot of this might be said to be present in the time-constrained jam games, but maybe better still the live-coding environments that one sees springing up in some places nowadays) – would working in this style mean that one was foresaking the creation of ‘good work’? I see it as a different process to the one you talk about, but not necessarily any worse nor any less focussed on the work being so-created. Which is more suited to the production of games? I think they’d both be pretty good.

Say someone makes a game, and they deliberately leave out a large section – because they want the people experiencing it to feel the sensation of unfinishedness. Is the game finished? You can say “to the author, yes, to the others, maybe it’s not so clear”. I guess I’d be okay with that. As to whether the work should be regarded as, ultimately, finished, I personally think it’s going to be a more than usually arbitrary labelling either way.

The term finishedness can be applied to a lot of things in a lot of different ways by people. You might say that, on average, there’s some core meaning that’s still present in most uses.

*’But I think that every creator who wants to create good work (as opposed to somebody who’s just interested in going through the process), is hoping to arrive at something that satisfies their own sense of “finishedness”, and to give up before you get there is always a failure of sorts.’*

Sometimes, I try to banish ideas like ‘success’ and ‘failure’ from my mind entirely. Both the words and the ideas. One might say “well then if you manage to successfully foreclose such concepts in making a game then it can be said that you’ve succeeded in your intentions”. That doesn’t sound much more expressive to me (there might, though, be a more logically straight-forward way still of wording it). Though I should probably wonder more strongly to myself what the hell I mean when I say what I said in the first sentence of this paragraph.

I think it’s possible to say that finishedness is more about process than its absence. But maybe the gesture of denying it is, if it is just a gesture, possibly more about procedure.

requoting

*”is hoping to arrive at something that satisfies their own sense of “finishedness”, and to give up before you get there is always a failure of sorts.”*

What if you cultivate a distrust for your own gut values? Criteria like this can loose some of their applicability.

If one has an artist who successively denies what comes to mind to him as constituting finishedness of his work, this maybe does assume some core meaning of finishedness (or unfinishedness), one that they’re really centering the whole work around.

For me, these concerns come from a critical examination of the creative process. Ultimately, I want to make art. I do think there’re artistically interesting things to be gained from formally manipulating the artistic process, but given that it’s not really a rational process, nothing that can be pointed out as a ‘part’ of it can really justify its presence. So. Actually maybe that’s a key difference: I’m not viewing artistic creation as being as rational a process as you and maybe some others are?

I don’t know if I’m responding in a satisfactory manner to what you’re saying. I suspect not.

zaphos:
*So perhaps an alternative goal to ‘finish your game’ could be to refine the art of the ‘unfinished’ game – to elevate the short form of mechanically-interesting games to do things not possible in the long form. I think “rom check fail” may actually be a step in that direction …*

Would such works not count as finished per Hecker’s criteria, though? (Given that Mr Hecker decided such a work had these qualities – any such evaluation will naturally be super-highly subjective).

in that sense i don’t think even braid was finished, because it didn’t really exhaust the possibilities of rewinding time in a puzzle platformer; there is still a lot that could have been done with the mechanics that he didn’t do. i can’t think of very many games that exhausted all or even most of the possibilities of a particular game mechanic or set of them.

that’s one of the other issues i had with this talk, it was elevating one type of game above another type (saying that finished, longer games are better than shorter experimental games). i don’t make short experimental games myself anymore, but i still like to play them, and think they’re just as important as the kind of games i make (the ones that take years to make and have dozens of hours of game time), and i think they’re the core element of the indie game community (even if they don’t get on the tigsource frontpage very often). they’re the bread and butter, whereas the polished stuff is like desert.

to extend that analogy, his position strikes me as a ‘let them eat cake’ thing, so out of touch with how most indie game developers live and work (their problems with productivity and lack of time) that he wonders why people don’t spend thousands of hours baking delightful cakes instead of quickly preparing bread.

I actually substantially agree with Paul on this, even though I don’t think he’s expressed himself in the best way.

Creative games development has been supported by hobbyists and part-timers since the 1970s or earlier, and that hasn’t gone away just because the commercial end is now dominated by blockbusters.

Most of these developers don’t have the money or freedom to take a year or two full-time, and run things as a business. Few can afford to hire an artist (as blow did). Few can decide to be poor for three years while they work on their dream game. As for me, I just want to contribute for the love of it, without turning it into my day job.

Add that to the fact that there is no infrastructure to support the lives of people who, like Stephen, want to make art in the medium rather than entertainment or saleable product.

So I think Paul is right to find it annoying that these lectures tend to devalue that bedrock of hobbyists and artists. And Paul is being very even-handed here, given that he himself is involved in developing the kind of full-time, commercial indie game that hecker and blow have championed.

I don’t agree with Paul that Phil and Jon and Chris shouldn’t be lecturing on these topics because they don’t have experience, or because they’re not entitled to be famous. But I do think that all three of them can sometimes come off as though they think everyone should do things their way – that is, in a kind of quasi-commercial way. This is like suggesting that every painter ought to get a big grant from the Canadian government, or a loan from an investment fund, and start a graphic design firm.

I think if anything there’s a tendency for people to spend too much time developing failed ideas in game development. It makes no sense to ‘finish’ your game if you can express the full merit of your idea in a half-finished sketch! But by the time you’ve secured your funding, and assembled your team, and sold your buttons and t-shirts on your website, how the hell can you back out? You can’t, you’ve got to see the damned thing through to its ill-fated release day.

One thing that puzzles me about this whole argument is that people who make these ‘Finished’ games are constantly complaining that it destroyed their souls! Phil recently wrote that Fez has left him as an empty shell of himself. Obviously a lot of this is about financial stress, and presumably Hecker and Blow don’t have to worry about that anymore, but why go forcing that on other people if they don’t want it?

Apart from being bad advice as far as improving the state of the art, it seems to be horrible personal life advice for the vast majority of indie developers.

Chris: if that’s the case, then I think that it’s just poor strategy to be worried to deeply about Finishing a game. If you’re working 1 hour a week after your day job, or if finances are a stress, then to commit to some monolithic project like Aquaria, for example, is to take a colossal risk, financially, personally, and creatively. Hopefully it paid off creatively and financially in Aquaria’s case, but Derek and Alec have both said it was a massive burden personally. And that’s the best case scenario, where you get mainstream press and win the IGF grand prize! For every Braid, Aquaria, and World of Goo, there are ten games that can’t find an audience (like Eufloria) or that get mired in development hell (like Fez).

I don’t find your talk insulting, in any way, but I would give you the exact opposite advice, if you’re not one of the ‘idle rich’. I’d say: don’t commit so soon after leaving EA to completing a giant project. Get some small things out there so you can establish a) cashflow, b) creative momentum and c) a fanbase. Feel your way, rather than taking a massive leap into the dark. Manage your risk exposure.

The problem with your point is that if it takes (say) one month to make a 1-hour game, it will probably take around 5 months to make five 1-hour games. But it *won’t* take 5 months to make one five-hour game. It might take two years. Game development doesn’t scale neatly like you suggest it does, and the last 10% of polish can easily take up 90% of the time.

Your options are to either budget 9 times as much time for your projects, or to just write off the last 10% of polish. And if money, time or energy are in short supply, I think the latter plan can often be the wisest one – since losing 10% of polish usually won’t reduce the quality, audience or impact of the title by anything like 90%.

Lots of interesting responses, so little time… I’d just like to take up this one little point here:

“Game development doesn’t scale neatly like you suggest it does, and the last 10% of polish can easily take up 90% of the time.”

I think that’s the crux of it – that last 10% is hard. It can be demoralising and take a real toll on the developer. And it doesn’t really help you to grow your creativity. But that last 10% is VITAL to the quality of the end product, as a user experience. It’s potentially what elevates an “interesting” game to a “fucking masterpiece” to a discerning audience.

By all means, experiment. Rough sketches and half-realised ideas are cool with me. But I have played thousands of pretty okay games, and comparatively few fucking masterpieces – there’s always room for one more.

Chris: if you were just railing against too many 1-hour games, I guess I would agree with you. Maybe you should have called the rant ‘make sure your game is playable by other people!’

I just want to say one more thing, which is this: I understand that Spore is frequently taken to be a prime example of a game which wasn’t Finished, and didn’t live up to its full potential.

I see it differently. For me, Spore is a prime example of overinvestment. It seemed to me that in Spore there were, early on, incredible systems for creature editing and procedural planets and multiplayer sharing, but nobody had bothered to prototype the rest of the game. But once the editor had already been shown at E3, and everyone had grown excited about the exciting idea, it was too late to cancel it.

Now it’s left to the development team to flesh out the gaps in the design: how should we build a game out of these systems? In game development, it very frequently is the case that an exciting idea doesn’t translate to a compelling reality. These are the games that *shouldn’t* be Finished. In fact, trying to Finish them creates a kind of black hole, which years of work can disappear down for no benefit.

How would an extra five years of development have fixed Spore’s problems? Would it have solved the tension between evolutionary realism and creative player input? Doubtful. Would it have made the process of colonizing thousands of planets less repetitive? I put it to you that the problem with Spore is that the fundamental concept makes for an interesting simulator but a bad game.

I bought Spore for $50 or whatever it was and I have to admit I felt very disappointed. What if the team had spent much less time on it, employed fewer people, and sold it for $20? What if they just made an evolution simulator and left out the rest of the goal-oriented gamey stuff? I’d say I would have been much more satisfied with that product.

I would counter that if they’re masterpieces, then the last 10% wasn’t left out – it’s not necessarily a measure of detail or flashiness, but of completeness. Opera Omnia probably doesn’t need to be presented like Cave Story to feel complete.

Dwarf Fortress is obviously still a work in progress, so its deficiencies can be forgiven for now – it’s clearly a work of enormous scope, and the last game that anyone would criticise for lack of ambition.

But Narbacular Drop is an interesting choice… a great idea with fugly presentation that was immensely improved (and subsequently hailed as a masterpiece) when it was remade as Portal?

“I’m not saying jams should be banned, hopefully that’s clear at this point.”

I didn’t imply that you were, nor is that related at all to my question which you evaded.

Bennett#75 – This is probably not the right place for general Spore critiques, but for me, the core part of that game that deserved more depth, in the sense I’m talking about in this rant, was “editor consequence”, where how you made your creature/vehicle/building in the editor impacts its abilities and behavior in the world. There is something really deep there, that we didn’t find and explore, I think.

Chris

“Like I said, try to remember when Portal hadn’t been announced. Narbacular Drop actually communicates its hook very well, despite the fact that nobody would call it Finished.”

That’s kind of what I’m getting at though.

Narbacular Drop: great idea, “interesting” result. Interesting enough for Valve to buy it in fact. But nobody would call it “Finished”. Despite its brilliance, there were still areas which were obviously letting it down.

Portal: same great idea, but this time refined to the point where every aspect of it is quite flawlessly realised: people call it a masterpiece and one of the greatest games of the decade.

Thanks for sharing this with me and time card calculator!