A few months back, Clint Hocking published a blog post about his time crunching on Splinter Cell: Chaos Theory that spurred a lot of discussion about overtime in the industry. In the piece Hocking felt his own long hours on the project had resulted in something he called "brain damage" - he had been focused so intently on game development and done it so many hours on Splinter Cell, that all he could remember from that time was the game's development. It was as if the rest of his life during that period hadn't even happened.
Hocking heard a variety of responses to his piece. Some blamed his employer for making him work all those hours. Others decried the bad management practices that made the overwork necessary. Finally some thanked him for his efforts, saying the final game was worth his sacrifice. In a follow up post Hocking clarified that he didn't see this as the fault of anyone but himself - he had wanted to do it. He also said the overtime wasn't the result of overly bad management, it was just what happened given that particular situation and what was necessary to ship the game he wanted to ship. He added that for his subsequent project (Far Cry 2) he avoided repeating the same mistakes by distributing the workload. Most interesting though was his response to the people who thanked him and said "it was worth it." To these people he said "But you actually don't, and can't know the cost." He posited that the only person who can decide if the overwork was worth it is him.
What I like about Hocking's two pieces is that he neither says the overwork was justified nor does he say it was a mistake. He acknowledges the complexity of the problem of overwork in an elaborate creative endeavor like game development. And most importantly, I like that he wasn't bragging about all the overwork he did, and he didn't suggest it was the only way to make a good game. Of the many problems with overwork in the game industry, the fact that we so often fetishize it and see it as the "one true way" is one of our biggest problems.
Stories of artists who make crazy sacrifices for their work abound in our culture. For an example from outside of games, we can look to last year’s excellent film, Whiplash.
You may have seen Whiplash, written and directed by Damien Chazelle and starring J.K. Simmons as Fletcher, an obsessed, sadistic jazz teacher and Miles Teller as Andrew, his drum student. To talk about the film, I’m going to include some light spoilers. If you haven’t seen the movie and don’t want anything given away, now would be a good time to stop reading.
Whiplash is a film about obsession and what lengths people will go to become “great.” In the film, Fletcher pushes his students beyond any reasonable limit, all for the opportunity to “find his Charlie Parker.” Charlie Parker of course was one of the most brilliant and renowned jazz saxophonists of the 20th century, and early on in the film Fletcher tells a story about how a young Parker had a “cymbal thrown at his head” by his own mentor when Parker had screwed up a part. Fletcher maintains the physical assault motivated Parker to go home and practice harder until he was finally good enough. Fletcher uses this story as justification for his own sadistic teaching methods.
By the end of the film, Fletcher has done a number on Andrew to the point where Andrew risks his physical health and at one point even his life. Andrew pre-emptively breaks up with his girlfriend - he wouldn't have time for her because all he cares about is being a jazz great. Fletcher and Andrew are student and teacher but also bitter enemies, with the duo pushing each other to the edge in their mutual obsession with perfection. The film implies that Fletcher’s methods have pushed Andrew to finally become “his Charlie Parker.” Both Fletcher and Andrew may have miserable lives, but now they're making great music together, the two locked in a spiral of mutual hatred mixed with grudging respect that results in great art.
Except that it turns out the Charlie Parker story didn’t go down like that. Charlie Parker wasn’t actually attacked - he was “gonged off the stage” because he was deviating from established jazz tradition and experimenting in new modes, not because he wasn’t good. I'm not accusing the film of lying - it fits Fletcher's character that he would take the Parker story and twist it to his own sadistic ends. But it’s also true that the religious devotion to hyper-fast playing and ultra-precise tempos that the film says are so important are far from the most important thing in jazz music (perhaps Chazelle should have made a film about a technical death metal drummer instead). But Whiplash distorts the facts a bit and makes them into a hell of a story. As someone who used to play in jazz groups, I recognized the movie's distortions but still loved the film and what I took away from it: certain types of obsessive and destructive behavior can sometimes lead to greatness, but it always leads to misery.
The Myth of What It Takes
And so too goes the narrative for the game industry and our own problems with obsessive overwork. For example, a recent article about overtime on Kotaku feels conflicted in exploring the subject. The piece is well written and very thorough, presenting many different perspectives on the reality of overtime work, when it has been good and when it has been bad, and goes into attempts to avoid it. Yet at the beginning it can't help but say “just about every video game in history is the net result of countless overtime hours, extra weekends, and free time sacrificed for the almighty deadline.” This is a popular chestnut that developers may say all the time but isn't actually true. Not all good games are the result of massive crunch. Indeed, it is disproven later in the same article when several developers talk about the successful games they shipped without crushing overtime efforts.
I've worked with at least twenty different development teams at different points in my career, and I can assure you the approach and standards for overtime differs for each one. Some had hard crunch times, at times to dangerous extremes, while for other teams, overtime work hardly existed at all. And there was not a direct correlation between the quality of the games produced by these teams and how much overtime they did. My examples are anecdotal of course, but show there's a wider variety of approaches to overtime than many people often admit. One mistake we make as game developers is overconfidence that our method is the "One True Way" to make good games. And it's simply not the case.
Does that mean that overtime is always bad, or never works? No, sometimes it does work. If entered into in the right spirit, and with the right respect for one's team-mates and knowledge of one's own limits, it can result in good work. But in many situations it's not the only way to make a good game, and we should stop pretending it is.
So why does crunching keep getting touted as more ubiquitous than it even is? Veteran designer Zak McClendon put it this way after reading the Kotaku article, “Crunch reporting is in danger of becoming disaster tourism for the game industry.” There's something weirdly romantic about reading tales of teams working insane overtime to make the games we love, so these are the stories we hear most. We hear fewer stories about when things went well and everyone (mostly) went home to dinner with their families.
Crunch is discussed so much that I think it has become fetishized. Developers see it as “proving themselves,” “displaying their passion” and it contributes to a misplaced macho attitude about “what it takes” to make great games. Developers end up seeing crunch as a "badge of honor" and compete with each other on who has the most atrocious overtime war stories. Part of this comes out of competitiveness in the culture of software development, but also the competitive nature of games themselves. By this twisted perception, whoever works the hardest wins.
I think another part of the problem is that creativity itself is a hard thing to understand. Much as we like to write post-mortems, give lectures and write textbooks about game design (and I'm guilty of all of those), at the end of the day there's some undefinable magic involved in making a good game that we will never be able to completely understand through analysis. Working from the same best practices and principles, someone can make a great game and someone else can make a game that doesn't catch on with anyone. There is no magic answer to this problem, and that can be very scary. In our quest to make great games we desperately seek anything we can latch onto to increase our chances of success. So we make "life-destroying crunch" one of the prerequisites for making great games. But we don't really know that overwork is the answer more than anything else. It's just something we know how to do.
And finally, we accept overtime as an excuse for our own bad decisions. When we're there late at night working on a game, instead of wondering about what mistakes kept us at the office until midnight yet again, we create the myth that it's simply impossible to do it any other way. We brag about our overwork to justify its existence.
Truth and Consequences
In the worst cases our tendency to fetishize and brag about overwork allows teams to be exploited by predatory management practices, like unscoped feature creep or substantial changes in direction without adding time or budget to the project. Obviously overwork to make up for bad planning should (and often is) seen as a failure. But that overwork is partly made possible by our industry's acceptance of overtime as "what it takes." In Whiplash, no one forces Andrew to break up with his girlfriend or put up with abuse from Fletcher: it's just what he thinks he has to do to be great. Once you start thinking that way, people will take advantage of it.
And I should say that what I'm talking about here is overwork that we will on ourselves, as is the case for Hocking or many times myself, or for Andrew in Whiplash. We all know there are many instances in the industry where overtime was not optional, whether ordered by management or strong peer pressure. I personally haven't experienced this, or rather, haven't experienced it where I didn't also want to do it, but of course it exists, and that's a different situation than what I am talking about here. In a large team environment, it can be hard to differentiate someone wanting to redo something because it's beneficial and they honestly want to versus peer pressure situations where suddenly everyone's doing it whether they want and need to or not. The complexities of those problematic situations are not something I'm delving into here.
Self-inflicted overtime definitely happens for a lot of indies. One might think indie development would be an opportunity to break the cycle of overwork, but often it doesn't work out that way. Startups in general are known for their grueling work loads, hence the expression "People are ready to work 80 hours a week so they don't have to work 40." It's harder to brand indie crunch time as "exploitation" because typically they have a financial stake in what they are making and are really only "doing it to themselves." But when people across the industry brag about working long hours, it can make desperate indies resort to crunch time all the quicker. And it's not hard to find cautionary tales about the downsides of that sort of thinking.
Now working on an independent game myself, at several points on my project I have set lofty goals and made promises to people about what I would show them when, and this has resulted in many late nights. But I've been making a conscious effort to look at each instance of excessive crunch time and wondering what I did wrong. And I think that's the key - when you do end up working overtime, be cognizant that mistakes were made and that it shouldn't be like this. With our industry's misplaced macho bravado about crunch, the stories of teams that avoid long hours don't get nearly as much play on the Internet, which in turn skews developer's perception of what "everyone" is doing. But I'm heartened by certain indies who have been smarter than me about managing their time, making great games while deliberately not killing themselves. And I'm happy to say I've at least drawn a hard line at making it to family dinners, something I was worse at during my years in AAA. One step at a time.
The Cycle Repeats
In the end, I don't think self-imposed crunch time is always bad. Things don't always go as planned, and putting in extra hours is sometimes the best solution. But we need to be honest about it: extended overtime periods are often not that useful and not worth the long-term cost. Instead of complain-bragging to your friends about your overtime, ask yourself "Could this have been avoided? What mistakes led us to this point?" We need to stop bragging about crunch as a way to paper over our failures.
Halfway through writing this I remembered I'd written about overtime in a piece before, years ago. In a Game Developer magazine post-mortem for Drakan: The Ancients' Gates, I talked quite a bit about the overtime that was taken on for that game. Re-reading it, I realize that I was somewhat guilty of the same type of overwork fetishization that I now recognize as part of the problem. That article posed the question "is it worth it?" and in the end I basically concluded that it was worth it in many cases, and that's why we keep doing it to ourselves.
But now I see that as letting ourselves off the hook way too easily. I think "is it worth it?" is a question we need to keep asking ourselves, for every project, for every deadline. As soon as we accept long hours as just "how you make games" we end up doing more of it than we should and we risk putting ourselves into a cycle of misery that will drive many people out of the industry and make those of us who stick around less successful creatively not to mention miserable.
In the end, "Is it worth it?" is a question only you can answer.
Thanks to Zak McClendon, Dean Tate, Katie Chironis, Clint Hocking, and Margaret Rogers for their help in putting together this piece.
Post Script: This blog post was delayed numerous times due to some hard deadlines I set for myself on my game, and the overtime that resulted in me having literally no free time to work on it. The irony of this is not lost on me. Beyond the situations that led to overtime on my main project, I am also guilty of overscoping this post - for the 2000 words you read above, I've got another 2000 and zillions of examples I ultimately cut out. This post script is here not to brag about this, but to publicly promise myself I will do better next time.