At GDC this year, one of the sessions offered developers a platform to rant for five minutes about whatever pisses them off most. Anna Marsh of Lady Shotgun chose to rant about how overtime in the games industry is a ‘bad thing’. It was both funny and comfortably familiar, not least because Anna used to work here at CA.
It’s an argument I’ve heard many times: overtime ruins people’s quality of life and doesn’t get the work done better or faster. We all know that overtime can be bad, but is it always true? Is there such a thing as ‘good overtime’?
I’ve been doing development overtime for 30 years now. I remember all-nighters back in the 80s and 90s; a colleague going out for pizza, not coming back, and being found asleep in his car outside, door open and one leg out. I remember double all-nighters, finishing that last line of code and dashing to the duplicator with the master, with the game on shelves 48 hours later.
WHEN IT COMES TO THE CRUNCH
They’re fond memories. It was hard, but it was great fun too. Projects were short, results immediate, and the teams were tiny. We were doing it for us, and not for anyone else.
By the 21st century, we figured out how to accurately predict how long a project would take. Sadly, this didn’t mean the dates matched the publisher’s requirements, and we still found ourselves burning game quality and the wellbeing of the team to achieve near impossible deadlines.
Back in 2000, we ate a tonne of takeaway food finishing Shogun: Total War. It wasn’t good for us but we were doing great work, which founded the Total War franchise.
Nowadays, the Total War team does far less overtime than we used to. We’re better at planning, and judging the quality/cost/time trade-off. We’ve also learned to flip the team’s mode from ‘explore and innovate’ to ‘get the bastard finished’ without needing the pain of overtime.
If you’re trying to do something truly original it’s much, much harder, but even on a franchise sequel it only takes losing a key developer or two to throw your plans in to disarray. The pressure to work extra hours is impossible to completely escape.
WHEN LABOUR ISN’T WORKING
But does overtime work? There’s plenty of research that shows that extra hours worked don’t directly equate to more stuff getting done, and that there’s a point where working more is counterproductive. But it’s not the same for everyone. And it seems like the more creative and talented you are, the higher the limit.
At award ceremonies, the developer receiving the award always, always thanks the team for the exceptional effort they put in to making the game. This isn’t because overtime effort is necessary to win awards; it’s because award-winning developers are driven, and driven developers spend their time doing what they love.
I think that overtime can be divided into good and bad, and that good overtime benefits both the devs and the projects. With good overtime, the extra hours are not the thing that benefits the project. They’re a side effect; a symptom of people fully engaged in work they think is fascinating and worthwhile, and they stay later because it’s more fun than whatever else they would be doing.
The line between work and leisure blurs and can sometime fade. The team puts their soul in to the project and it shows in the end result. They achieve great things. Games development is art, not manufacturing.
Gladwell’s 10,000 hour rule states that 10,000 hours of practice will make you a genius at any particular activity. Plenty of us got those hours of practice early on in our careers, much of it through overtime, and we’ve built on it since then. I think we owe it to the next generations of talent to provide an environment which allows and encourages them to make themselves geniuses in their chosen field. If a thing is worth doing, it’s worth doing obsessively.
NO BAD THING?
So, good overtime improves the quality of a game, develops and exercises talent, gives space to innovate, lets you help struggling colleagues, or squeeze in one more feature, or polish off an annoying rough edge. It helps win awards, and deal with unforeseeable events.
Bad overtime is not completely voluntary, is enforced to meet an arbitrary deadline or cut costs, goes on too long, is too intense, is planned for from the start, isn’t recognised or rewarded, or it’s there to fix problems that are not of the team’s making.
Overtime is like cholesterol. The most delicious bits of game development are full of it. There is good overtime and bad overtime and too much of the bad kind will kill you. It can’t be avoided completely, but with careful dieting, the good compensates for a little bad.
[Interested in contributing your own article for Develop's readers? We're always on the lookout for industry-authored pieces on development-related topics. Email firstname.lastname@example.org for more details.]