Not Everyone Feels The Crunch

[Following Gamasutra's recent, much discussed article on 'quality of life', we follow up with EA_Spouse colleagues and other industry execs with a clear message - crunch doesn't have to happen.]

Game developers know certain things are inevitable -- software bugs that need exterminating, for example. And schedules that invariably lead to excruciating crunch times.

But wait! Crunch - the bane of developers hoping for a superior quality of life - needn't rear its ugly head if the proper steps are taken, say some.

For instance, many studios that create serious games for military, medical, and governmental purposes have found a way to skirt crunch. And so have some companies that build more traditional consumer games.

What's their secret?

Ken Yeast, the director of engineering at LA-based 7 Studios, recalls vividly what crunch was like at Electronic Arts where he managed the engineering staff for Lord Of The Rings: Battle For Middle Earth.

In fact, one of his teammates was Leander Hasty, the man who put the "spouse" in "EA_Spouse," and whose experiences became legend throughout the industry via an open letter from his fiancée.

"Some people romanticize crunch and say it pumps you up and makes you feel exhilarated," Yeast says. "And that may be true, but only for short periods. It's akin to running. If you've been running at a pretty high speed, you can then sprint, but only briefly. In game development, if you're crunching for any period of time longer than a week or two, your productivity drops very quickly, especially in the quality of your work."

Indeed, nearing the end of the Rings project, Yeast and his team were doing the equivalent of long-distance running.

"Ironically, we had Labor Day off," he remembers, "which was something we had to fight for. And then we didn't take another day off until the product shipped 60-something days later. For two months we worked seven days a week, 12-14 hours a day. The attitude was 'just do whatever it takes to meet the deadline and don't give any excuses about not having the resources. We'll just keep piling resources on.'"

"But, of course, you can't just keep piling resources on. And the increased workload leads to mistakes that resulted in even longer work hours. That's just the nature of crunch. You're better off dialing it back and then you get a better product."

Fast-forward to 7 Studios today where, given Yeast's painful experience at EA, eliminating crunch is an extremely high priority. But, he says, you can't just say no to crunch and everything works according to plan. There are times when Yeast admits that he's not as successful at eradicating crunch as he'd like to be.

"We put a lot of effort into planning, into being dynamic, and being able to adapt to changes," Yeast explains. "But most important is to watch for the consequences of your decisions and understand them."

"At EA, higher-up management wanted a very high level of delivery without seeing the consequences of some of the decisions they were making. If your people are suddenly working extra hours, find out why. What is the mechanism that's broken?"

Yeast is a firm believer in post-mortems. "After you finish a milestone or the delivery of a game," he says, "analyze what went right and what went wrong. That's a great strategy for catching the problems that cause crunch."

Andrew Eades is another developer who'd been crunched in an earlier life and decided "never again." The studio at which he and his partner David Amor had been toiling - frequently for "ridiculously silly hours" - went bust, and, at the end of 2003, the two founded UK-based Relentless Software. Its name stems from the duo's solemn oath to never, ever work overtime again.

"Some people find crunch invigorating but, frankly, after the first few days, it loses its appeal," says Eades, Relentless' owner and director. "We take the extreme view that you should never do it. We define 'crunch' as anything over a 40-hour week."

In fact, developers at Relentless work 9-5 with an hour for lunch, five days a week, for a total of 35 hours. The key word, says Eades, is "work".

"We'd noticed that at many studios there was a big difference between the amount of time people spend in the office and their productive time," he explains.

"Because many studios are set up to resemble playgrounds, people develop a slacking-off culture. They play games, drink coffee, and chat with their mates, which are very nice things to do but not exactly useful in the production of video games."

Eades says his policy of maximizing the 35-hour workweek is what enables his team to avoid crunch. "We don't have crunch because we have an unbreakable rule that there can be no crunch. It's not allowed," he adds.

That's in addition to Relentless' "very strong project management skills that give us a very measurable week," Eades notes.

"Every week is the same. We know how long it takes to do stuff and we can predict it much better than if every week was completely random. That's why there's no mass panic at the end of the project. We just finish, exactly as planned."

But instilling such discipline into a company that's not used to it may be fruitless, according to Eades, who admits that his no-crunch policy works because Relentless was built that way from the ground up.

"I think altering a studio's policy to one like ours would be really difficult," he says. "Our policy is so fundamental and so strange that, unless you reinforce all the policies that back it up, it would be very easy to lose one's way and go back to old, easier habits."

"You would have to have an absolute emphasis on delivery of milestones; that's number one. And then you'd have to be really firm when your publisher says, 'Can you work the weekend?'

"Your answer, without exception, would have to be 'No!' That's not an easy thing to do. So, in the long run, the way we did it - from scratch - works best."

But not everyone who has experienced crunch - and is hoping to escape it - has the option of launching a new studio. Others look around and see that the grass may be greener outside the commercial games sector, particularly where developers are making serious games.

Ben Sawyer, co-director of the Serious Games Initiative, believes that his environment is no panacea for developers.

"Serious games creators do get hit by crunches... just different kinds," he says. "Clients often take a long time to get back to you and they gum up your process, but then they still want to see deadlines get hit. So you're often crunching at the end."

"But, at the same time, many customers don't necessarily have large projects and they are often more understanding of your schedules, so you can get some leeway. However, no one should think that serious games development is the escape that it sounds like it could be."

On the other hand, Joe Straitiff has made the switch and doesn't regret it for a minute.

Straitiff worked at Midway, Infogrames, and Stormfront Studios, and was part of the class-action suit against EA (that also involved Hasty as the lead plaintiff) where, as a software engineer 3, his last project was The Urbz: Sims in the City.

"They were trying to get us to work at least 10 hours a day, seven days a week," he recalls. "I didn't quite give them that but I do remember 14-15-hour days a few days in a row every now and then, and I know I did some 80-hour weeks."

Today, Straitiff is senior software engineer at San Mateo, CA-based Forterra Systems which produces private and secure MMO virtual worlds for corporate, government, defense, medical, and educational clients, not gamers.

"I've been here over three years now, and it's been regular hours 99% of the time," he says. "And the other 1% is never a lot, just support for one of the major shows. Last year there was no extra time necessary at all."

While Straitiff can't vouch for the fact that Forterra is typical of all serious games studios, he does believe that the serious side has a much higher quality of life.

"On this side of the fence, our deliveries have to be solid while still being maintainable," he explains. "The customers are fewer - paying a lot more for the product - and they expect regular updates. We are selling them a platform, so we spend a lot of time estimating and tracking progress, adjusting everything along the way."

In comparison, at commercial games studios, Straitiff recalls schedules that everyone knew were impossible from the start, "but the pressure was still there to just get it done. Everyone wanted to ship a game every Christmas, which amounted to about six months of dev time for a title, clearly not enough for a decent game. So everyone crunched and hacked away."

He believes that if the games industry could just get past the "we've got to ship for Christmas or we're dead" attitude, conditions would improve considerably.

"At Forterra, I'm active in the Process Workgroup where we're constantly striving to make engineering process sane, predictable, and useful," he says. "We're shipping a platform with periodic releases that we expect to maintain and improve for years to come, so the ‘hack it and ship the pig' attitude just doesn't work for us."

Would Straitiff ever go back?

"Well, interesting opportunities are always a draw, especially since the memories of the bad times tend to fade over time," he says. "But I wouldn't budge unless I found a company that has all its ducks in a row and could convince me of that. Besides, working on cutting-edge game technology here is fun; distributed simulation is complicated and working with that type of platform is an interesting challenge."

"The biggest missing piece is seeing your game on the shelf at the local game store," he admits. "That never goes away."