There’s been a blog debate brewing the past week about Mirror’s Edge, and exactly what’s wrong with it – or what’s wrong with our critiques of it. Did we shortsell its innovation? Are we missing its true design flaws? On Twitter, Brinstar and Mitch Krpata have both called the game “frustrating” – but it sounds like Brinstar really likes the game, and while I find it frustrating, I keep coming back to it as well. If I had to sum it up in one pithy phrase, I’d say that its core problem is that it looks like Rock Band 2 but plays like Mega Man 9; you want to settle in and enjoy the thrill, but imagine if Rock Band stopped the song every single time you hit a bum note.
But the other day, I said something less pithy: that this first version of the game was merely okay, but the sequel would be better. Keith Stuart anticipated this very argument, by asking if we would ever say something like that to, say, a film director. More than just a cop-out, it’s an argument against games as art, and we all want to say that games are art.
But you know what? Games are also software.
I come from a software background, as well as an artsy-fartsy one. I want to see games as art, but they’re also supposed to work as logically-constructed bodies of code. And in a lot of cases, reviewers need to see them as software rather than as art. Here’s why:
Sequels are not bad. In movies, sequels are usually a sign that the bankers are taking advantage of the artists. But movies don’t have to invent a new studio, lighting system, camera, and style of acting every time they’re made. Games often benefit from sequels. Noone remembers the first GTA, the first Ultima, or the first Burnout as fondly as the later ones.
Games can be patched. Critics have to start considering more and more that games will be fixed after release and ultimately, upgraded or expanded upon. To an arts reviewer, and even to a lot of games reviewers, this seems very fishy. Should you cut some slack to a buggy, short or unsatsifying game on the off-chance that it’ll get better with the next content release? If a good game like Far Cry 2 or Fable II has DLC on the radar, does that bump it even higher? My hunch is that while you shouldn’t err too positively, you should see the game as something of a platform that may or may not be worth building upon. I might cut Fallout 3 slack for a couple of its brazenly stupid but easily fixed bugs, and at the same time, feel eager to see how they expand it, whereas Fracture could get all the DLC in the world – and I’ll still hate the game.
Playtesting is crucial. I don’t know how much playtesting Mirror’s Edge received, or how much time they were given to act on what they learned. But I suspect it got a lot less than Portal. That game’s learning curve was paced pretty perfectly, and there were no cheap shots or headbangers that I can remember – whereas Mirror’s Edge has a number of them. Now, in movies, focus groups still carry some stigma (even though they’ve helped more of our favorite movies than we care to admit). But in games, as with software, we know that usability testing is crucial, because a team of developers cannot anticipate all the ways the players will behave. And to take it a step farther: think of all the data that Valve collects on its players after their games ship. What does that data inform? The sequels. And the patches …
(And btw, musicians effectively “playtest” their music in concert again and again. I feel like this is turning into an argument of why games are not like movies – and I knew that already.)
Not everything in the game is worth evaluating. This seems obvious yet it’s the one thing critics grapple with the most, because everyone draws a different line in the sand. We often hear that every element of a work of art should be integral to the whole. For example, in a poem or a short story, every single word should matter. We cut novels, movies and pieces of music a little more slack. But in a game, whole elements of the work could be considered features that are optional to the player.
A piece of software should have a vision – not visions, but a single vision that unifies the feature set and defines the audience it’s targeting and the user experiences it hopes to create. But once you’ve mapped that out (if you have), not every feature is equally important. Microsoft Word has a vision of giving users a powerful way to create text documents. You can argue that Word fulfills that vision even if you hate the word count tool, or even the spellchecker.
Game by game, we have trouble deciding what matters. Should bad sound design really cripple the grade? Did the graphics matter that much? Were the three redundant chapters at the end so boring that we should dock it some points for wasting our time? That’s all for us to wrestle with – just as the critics who liked Mirror’s Edge for its innovation, but hated its combat, might have their own opinions on how that turns into a final score.
But I definitely believe that games deserve more slack on this front than any work of art. They are not a unified experience. They are pieces of software with rich feature sets.
Games don’t pose arguments, they present systems with which to interact. See Ian Bogost on procedural rhetoric. Again, we know that games usually don’t have one single meaning that is transmitted equally to every player. But every time we tackle a game with strong and prominent themes, like Fable II‘s exploration of morality, or Fallout 3‘s portrayal of the many ways that people reorganize themselves after a nation-ending cataclysm, we risk looking for linear arguments when the game is offering a loaded scenario. See also Jonathan Blow’s discussion of the thought behind Braid, and what differentiates it from, say, a piece of writing.
… So that’s a starting list. It doesn’t even get into the obvious stuff, like the fact that not everybody has a PC that can run Crysis, or that people with regular TVs have a hard time reading type on PS 3 games (like me).
So here’s one case where I wish I had personally followed these recommendations: Spore. I reviewed it and also blogged about it, at length, trying to work through all the things that I thought worked and failed in the game. But after 10-15 hours with it, I didn’t get a full assessment of what was right or wrong with it. I would need to approach it the way I would evaluate a piece of software – and try to end up with a 10-15 page document that tries to assess the vision the game was trying to fulfill, the critical success factors that would execute that vision, and all the ways that it did or didn’t live up to them. If I had the time, I would love to do this kind of mega-critique with Spore, or even with Mirror’s Edge, which would be much simpler.
After all, it’s easy to write a favorable review – really, an appreciation – of a game that fires on all cylinders. But if we could dissect exactly what works and what doesn’t work about games like those two, it would probably be more edifying.
I would say that it’s not simply a case of video games being software. The long history of games is one where rules are modified by each generation, with each revision attempting to bring the game closer to its potential. The game of Basketball is still recognizable as the game that Dr. Naismith designed and played in 1891. The modern game however has dozens of small rule changes that have been made over the years, some to make the game more competitive, some to make it more entertaining to watch. You’re probably right that the constant revision of video games through sequels and patches is a legacy from software design, but it’s still not an anomaly for games in general.
I don’t think I agree that those points distinguish software from other art. Books can be revised, and movies get Director’s Cuts. Movies are focus-tested, and stories are read by editors and spouses and friends, with their comments incorporated into the finished product. And I argue that every piece of a game is just as essential as every piece of a painting. Whenever a developer puts something into a game or leaves something out, they are making a choice. If they make the wrong choices for a game, then that’s something that deserves criticism. A feature may not be used by a certain player, but its availability still reflects on the game. I would certainly be reluctant to criticize a game just for having Feature A, but if A means that the creators dropped the ball on making the controls right or making the voice acting not abysmal, then I think it’s valid to criticize the game for that.
A terrific and timely post, and I agree wholeheartedly with your thesis. I think part of the reason we don’t think of games as software has to do with the wide gap between our knowledge of games as experiences and their real existence as millions of lines of code. We toss around terms like “game engines” and “shaders,” but really we non-programmers have very little idea what all this stuff really means or how it works. It’s easier for us to limit our consideration of games to the standard gameplay, graphics, feature-list stuff.
I also think games have become incredibly deft at hiding the fact that they’re still software. Those of us of a certain age recall booting games via a command line and watching the program execute before our eyes. Today there is virtually no difference between inserting a disc to watch a movie or inserting a disc to play a game. It’s all magic now. Except when it crashes.
It’s a point well taken. Mirror’s Edge has now crashed twice in a row for me in exactly the same spot. Hard to appreciate anything good about it when that happens.
“But the other day, I said something less pithy: that this first version of the game was merely okay, but the sequel would be better. Keith Stuart anticipated this very argument, by asking if we would ever say something like that to, say, a film director. More than just a cop-out, it’s an argument against games as art….”
It’s actually just an implicit argument against games as movies, which I think you realized halfway through your post. Stuart’s remark depends on an assumption that film is a standard visual medium by which other screen-based art should be judged. This is a pretty arbitrary premise.
For instance, why not make television the benchmark? OK, so you critique a game and say that its sequel will be better. Might you also review the first season of a television show and say, “It’s fine, but it looks like the second season will be better”? Sure. TV series have to establish characters, find their voice, etc. In fact, many great TV shows hit their peak after a couple years on the air. In this context, looking forward to a game’s sequel is normal.
Games are not movies (or TV shows), so naturally their criticism responds to that context. It doesn’t have anything to do with something’s “art”-ness or not, it has to do with the fact that different media operate differently. Not exactly a revelation.
That said, I do wish that game reviews were more like film reviews and less like software reviews. This despite the fact that I agree with you on the shifting, relative importance of bugs and features from game to game.
You seem to argue that if reviewers would consider games as software, they might be more willing to consider the game as whole, in context. I believe the opposite is true. Many critics do treat games primarily as software, and as such overemphasize technical aspects that have little to do with the artistry or enjoyment of a game.
For instance, I just turned in a review of “The Last Remnant,” a game that has a number of graphical issues, like choppy framerate and texture pop-in. After finishing my review, I went to Metacritic to see what others had said. Sure enough, TLR lost a lot of points for the graphical stuff. What bothers me is not that such issues are mentioned — they’re noteworthy — but that they are treated as unforgivable sins. I mean, texture pop-in, c’mon, this has to be pretty far down on the list of things that interfere with one’s enjoyment of a game. Many critics act like technical missteps demand some untold number of obligatory demerits, like a missed landing in gymnastics. If you’ve got bugs, you just can’t score above a 7, sorry.
To me, this is a software mindset. If we were talking about Microsoft Word, bugs would be a big deal. A bug in Word could destroy your data or screw up your workflow. In software, bugs are central by definition. In a narrative medium, less so. Game criticism too often focuses on technical acuity at the expense of the narrative elements that can really make or break a gaming experience. Nobody cares about engines or shaders when they’re playing a game.
This isn’t to say that bugs can’t ruin a game, or technical brilliance can’t be the defining flourish of a transcendent game. My point is that, in general, the technical nits are overpicked at the expense of more interesting elements of gaming, and it’s because critics treat games TOO MUCH like software.
I agree with much of the post; I just feel that your thesis doesn’t match what you are actually arguing for.
P.S. I love what Charles had to say about the evolution of games in the very big-picture sense. Spot on.
That’s a very interesting observation, and one that I’ve been mulling over for some time now. I disagree that games are software, however. I believe instead that games are descendants of software, but have now evolved into something else (which nonetheless has many traits of traditional software). I’ve been long interested in trying to do a ‘pre-history’ of games, a genealogy, even. I think in that, you’d have to include software prominently along with film, novels, sports, architecture, theme parks etc. However, games are not now just software. They have elements of many other media, and therefore we can’t just review them as software. I think an outlet like an IGN or a Gamespot have actually been reviewing them as software for years, by breaking down each individual feature and processing them with good-to-bad ratios.
So as well as realising that playtesting is important, that sequels aren’t bad, and that games can be patched (points I couldn’t agree more with), we need to be able to pull back from the software perspective and look at them holistically, like we would do with a film or other media. Because unlike software, games are about experience, and experience cannot be evaluated through minutiae.
Well said, Chris. Those of us who make games for a living are acutely aware that we work on software, but it’s not so obvious to the players.
Thanks for all these responses and insights. I don’t really disagree with any of these points. John, you especially call me out on a good point – by pulling us back from a certain perspective on games as art, I risk strangling in the crib what we’ve been working for at places like the Onion, and legitimizing some of the “OMG there’s texture pop” nonsense that affects scores in the enthusiast press.
If anything, I really just want to broaden the idea of what makes a game. Certainly I think that comparing them to movies is less and less useful. But I also think we spend too little time thinking of them as a platform, rather than an object.
Either way, at the end of the day, I’ll always cut them all kinds of slack as long as they rock.
Nice read, Chris. There’s a very unhealthy focus on review scores and technical breakdowns these days to the point where it’s becoming a neurotic obsession. The enthusiast press is in a weird spot these days where it wants to cover games as art but they are also stuck in their traditional rut of treating games as consumer products too, with gaming zines acting as glorified consumer purchase guides.
I for one appreciate any critic who dares to review a video game without assigning a numeric score.
I guess the difference is that in a book, there’s no excuse to include anything that’s not necessary for the vision (the binding maybe?). In games, there are lots of elements you need no matter what, so we should focus on what’s important and cut slack in other areas.
And that’s exactly the job of the critic, isn’t it? To identify the elements that are most important in a game and base an argument based on those. Any review that doesn’t make an argument for what the vision or purpose of a game is and identify the elements important to the success of that vision/purpose has failed from the outset.
That being said, it’s important to note that the best games do exactly what you’re saying is so difficult: integrate every single element into a unified vision. That’s exactly why “Left 4 Dead,” for instance, is one of the best games of the year. And why “Grand Theft Auto 4,” which is great in so many ways, suffers: because the open world and the narrative are contradictory.
One of the ways “Left 4 Dead” succeeds is by recognizing that not every element has to be original and awesome. The plot and characters, after all, are derivative cliches and purposefully so. But it fits this co-op gameplay experience perfectly and that’s what matters. I much prefer that to, say, “Gears of War 2,” which has a godawful story and characters that significantly detract from the overall experience. A cohesive experience is better than trying to make every element equally awesome.
Perhaps that’s an example of how some developers need to realize they’re not writing a book or making a film.