Crunch: game development hell

[Thraxamer]

I admit, your mileage may vary. Maybe it’s not just industry experience that drove me to think there could be another way. It could be the fact that I’ve worked in other fields. Knowing that there’s more than one way to solve process problems can help, and it’s probably been an advantage.

But to your point, most of the producers that I known over the majority of my career have zero professional experience developing. Their development experience – if they have any – is confined to school projects and the occasional indie game jam. That’s not to say that’s not valuable, but it isn’t representative of what actual sustained development is or should be.

Ugh. In the process of writing this, I’ve reminded myself of an executive producer who insisted to our lead engineer that a massive, impractical change to our live product should be easy because “coding is just a bunch of if-then statements, with the occasional else.” Anyhow, that’s my baseline.

[TheirFeldspars]

I often see inappropriate videos accompanying the articles here, but this video of keyboards being crunched is perfect for this one.

[tlwest]

As a 30 year veteran in the industry, I’ve come to accept occasional crunches as facts of life, and regular crunches as a clear indicator that it’s time to shift to a different job.

The reality is that in a lot of industries, being late on a deliverable can cost the customer hundreds of thousands of dollars. Nothing like idling a team of 200 because the software they’d been planning for isn’t ready… So, no, they don’t always forget that the software was late…

As well, software management is a black art, and even the best can only give a probability curve of effort. If they bid assuming the 99th percentile of likely effort, they’ll be undercut by someone who assumes the 90th percentile, etc. As a consequence, perhaps 1 in 4 or 6 jobs aren’t going to make the ship date at 9-5 hours.

The result, sometimes a crunch is necessary.

On the other hand, my crunches might involve a few 12 hour days, a Saturday, and perhaps an all-nighter and might come once a year or so. I consider it part of the job.

But weeks of 18 hour days? That seems pretty much apocryphal to me and I can’t see how one wouldn’t burn out after a few days, let alone weeks.

[gracchus]

Agree with all of that. The fact remains that eternal crunch time is a sign of bad management – the games vertical just takes that dysfunction beyond one company to an entire industry.

@oldtaku

Of course now you’ve burned out all your people and the next deadline starts looming… good thing there’s an endless supply of optimistic young kids to run through the thresher!

Which brings up the issue of wages. Game developers aren’t paid by the hour, and the younger ones seem to have a blind spot for that aspect of math. It might be enticing for someone just out of school to start at $55k/annum, but if he’s working 60 hours a week for 50 weeks a year that translates into $18/hr. for an intense and high-skilled job that required a pricey STEM degree. And of course, tech bros have long convinced themselves they’re winners who don’t need unions like low-wage workers do.

[anon61221983]

[Mister44]

I know of the Captain, and his conspiracy to destroy the roofs of children’s’ mouths.

eta - oops - forgot the pic

[oldtaku]

@tlwest

But weeks of 18 hour days? That seems pretty much apocryphal to me and I can’t see how one wouldn’t burn out after a few days, let alone weeks.

1. Young guys - I’ve done it for up to a month (but not months, guess I’m not hardcore enough). No longer, because now I know 2) They’re only getting as much done (or less) as a good 6-8 hour day. You’re low gear for most of those 18 hours. And that’s the real kicker - you’re only there filling seats because management thinks you’re getting more done. Because it works in small bursts and they assume you can just extend the line upwards forever. If one guy can get 150% done by working one well rested 12 hour day, then obviously if you make him him work 16 hours a day for a month he will get 200% done that month, right? Sure, let’s commit to that schedule with EA.

Edit: What timing - according to The Collapse Of Visceral's Ambitious Star Wars Game Visceral crunched for weeks if not months on the leadup to Gate 3.5, after which EA cancelled the game.

[FGD135]

The solution, obvious it is:

This project management certification course could help you launch a new career

[Shuck]

@Thraxamer

I admit, your mileage may vary.

Yeah, I’ve worked with non-developer producers who had enough industry experience that they had good notions of how long things took, and experienced developers-turned managers who knew how long things took - but ignored that information for a variety of reasons. (As someone once explained to me, they felt pressure from publishers was to give unrealistic milestone dates, but also expected that they’d be able to get more time out of publishers when it came to it. Also developers constantly pushing to add as many features as they could cram in during the development period.)

@Thraxamer

I’ve reminded myself of an executive producer who insisted to our lead engineer that a massive, impractical change to our live product should be easy because “coding is just a bunch of if-then statements, with the occasional else.” Anyhow, that’s my baseline.

Ooooowwwch.

@tlwest

But weeks of 18 hour days? That seems pretty much apocryphal to me

I’ve regularly read/heard from people about crunch like this (though more often 16 hour days, 7 days a week for months and months at a time). And yeah, they destroyed people. The 18+ hour work-days seem more like expectations/work culture than actual specific demands from management. Though having said that, working with Asian companies, I’ve come across situations where people were working stretches of what may have been close to - or actually - 24-hour days (and then ending up in the hospital) during crunch periods.

[Thraxamer]

I’ve done it, although you don’t need to take my word for it. I can point you to a team that I’ve worked with that can confirm over 6 months of 18 hour days, 7 days a week. You just have to mention “Survivor: Outback - The Interactive Game” to them.

I can point you to a few more who can also confirm a similar level of crunch, only the magic phrase is “Charlie and the Chocolate Factory.”

Note that both games were garbage. The teams knew they were cranking out garbage.

Each one of those people who committed to those crunches did so because they were following a dream of working in game development. What was the option?

To quit before you had a finished title to your name? A credit on a finished title means a whole lot more to a potential employer than zero credit for something you bailed on.

To burn all of your bridges by leaving an otherwise untenable situation? Game dev is a tight knit group. Sure, you can burn those bridges… if you want to make bank software for the rest of your life.

I mean, that was the mindset, and that was the fear. Fear and desire… those can be very powerful motivators.

I think your low-grade shovelware houses, third-party port houses, and mid-grade producers of TV and movie tie-ins probably experience more of those nightmare scenarios, because they’re chasing a fast buck. Build a franchise? Who has time for that, with another Christmas just around the corner? But that’s another story.

Now that there’s enough data to support the thesis that death marches are bad for business, things are improving… even in those lower-tier studios. Well, the ones that are still around, anyway.

[tlwest]

@Thraxamer

I’ve done it, although you don’t need to take my word for it. I can point you to a team that I’ve worked with that can confirm over 6 months of 18 hour days, 7 days a week.

Okay, I believe you, but how did you physically survive?

By the time you add in commutes, preparing meals, eating meals, shower, laundry, handling minimal aspects of life, I can’t imagine you’re looking at more than 2-3 hours sleep a night, which is pretty much a gateway to madness in a week or two, let alone the month or two.

Simply dying on the commute becomes a near certainty as you’re well into hallucination and likely malnutrition territory.

The crunches that I’ve experienced that left exactly zero time for anything beyond eat & sleep (and not much of that) were maybe 14 hours a day, and in reality, closer to 12 (we’d break for lunch, coffee, and dinner, if it was going to be a 9-11 sort of day.).

They felt like 18, but I was not working 9am-3am straight, no way. (And I’ll say, more than 7 days straight of 12 hours days, and all they were getting was a technically adept zombie.)

Anyway, I’ll just assume that you guys were simply made of sterner stuff, because even the craziest big-law consultants that I knew (and the factory workers who were committing suicide because working conditions were so bad) were not pulling hours like that for weeks at a time.

[Edit: I just realized my perspective might be a little different. I’ve almost always been a contractor when doing crazy hours, so when I have a 14 hour day, that’s billable hours, which means I’m at my desk and concentrating on work the whole time. Breaks for meals, coffee, etc, aren’t billable (by me, anyway)]

[Thraxamer]

Well, the first time we did that, sometimes we’d end up sleeping at our desks. Sometimes, we just didn’t get more than a couple of hours of sleep. When I was younger, I could push myself through a lot, as could a lot of other similarly motivated young people. Coffee also helped. Lots of coffee. And lots of OTC stimulants from truck stops (most of which are now illegal).

The first place – long dead, now – had a mandated 10am-4am day, 7 days a week. It lasted for just shy of 6 months. Lunch and dinner were catered and were always pizza. Magic Lantern – maker of shovelware – was in the literal middle of nowhere. We all lived in town, no more than a couple of miles from work. We had another office in Dallas (part of an ill-considered business move). They had the same hours, so many of them slept at work. One guy’s marriage ended during that crunch. Like 3D Realms, we had a model where we could potentially get a substantial bonus payout based on project completion within a certain window. We weren’t quite as competent as 3D Realms, however (to put it mildly).

The threat of being fired loomed over all of us, which none of us wanted. It was a common threat, though.

For our efforts, we got a week off at the very end. Sleeping a full night’s worth felt weird for a while.

The game was crap, in large part because we were all in such poor shape, mentally and physically. I must have gained 20 lbs. during that time, from the pizza and the lack of motion. Many of us got sick for a while near the end. We came to work anyway, though, because that was the mandate.

I went to work for another company as soon as I could after that. This company was in the Chicago suburbs. Again, the prices weren’t too bad in the 'burbs, so you could live close to work. And, once the project started to go south, crunch became a mandated thing. We didn’t go quite as long as my worst crunch. Maybe 4 months. We had a mandatory 10am-midnight schedule, 7 days a week. We were expected to put in as many extra hours as we could on top of that. Social pressure, it turns out, is an effective thing. So most of us stayed until 2am, at a minimum. We’d get asked by management if other people on the team were pulling their weight, every now and again. Dinners were bought, but not lunch. They had a vending machine, though. Most of us wanted to get a gig somewhere else, but… again… we were afraid of just bailing. The industry seemed way too interconnected to risk that sort of a blemish on our resumes.

I malingered one day, though, and looked up labor law at the federal and state level. Illinois had a 1-day-in-7 law, which meant what this company was doing – the whole “7 days a week” thing – was illegal. I casually mentioned it to our HR manager one day. Suddenly, we were back down to 6 days a week. We were told to go home at midnight. It felt almost like it wasn’t crunch anymore.

A few months after that crunch, a fellow team member died. His heart gave out. I have no idea if our schedule had anything to do with it, but I wouldn’t be surprised.

I ended up taking a job in Eugene, OR not too long after that. My second project in Eugene, we did some extra hours, but nothing approaching the crunches I’d endured. Maybe 70 hours a week. Lunch at the desk, dinner catered… you know the drill. Mandatory Saturdays. Still, the first significant crunch in Eugene almost did a number on my marriage. Everyone has their limit, and my wife was near hers. So I made my choice. I drew a line in the sand with my bosses. I was ready to walk. They relented, and I ended up moving into project management. I was grateful for that.

It’s better now. It really is. My former co-workers and I are a bit wiser for it all. A bit less willing to let work become an all-consuming thing. A bit more willing to walk, if needed. A bit less afraid of the consequences. Most of them aren’t in game development anymore. Some went to film effects, some went to government contractors, and some went into other fields unrelated to software development.

On the other side of it all, none of the companies I’ve worked for since then have been quite so demanding. They seem to have learned that it’s better to have employees who are awake, aware, and happy to be there. Who knew? Maybe it’s getting better for everyone. I hope that trend continues.

[tlwest]

I am truly aghast. You also have my condolences. I’m horrified that 18 hours/day for 6 months isn’t apocryphal.

Amazing to think how much money the company lost to fund the destruction of so many people’s lives. it’s just lose/lose everywhere.

Also, I wonder if rural tech companies are more prone to abusing their employees simply because there is less competition to jump to. All the places with long crunches that I know about tend to bleed employees like crazy (or used contractors paid by the hour).

[Thraxamer]

Well, I learned some valuable lessons. It wasn’t a complete wash. If I’d been a little smarter and a little less eager, I might’ve gotten wise a little earlier in the process. One of the biggest lessons I’ve learned is that I really don’t have to ensure something like that. Walking away from a bad situation doesn’t make one a failure.

I absolutely agree that the lack of competition was a factor. The only response was to move, and moves are incredibly expensive, not to mention physically and emotionally draining. If you don’t make much to begin with (there’s a lower cost of living in the rural areas, after all… which means you can get paid less), then a move can be nigh-impossible without some sort of external assistance.

Although, I should note that the company in the Chicago 'burbs had a “no compete” contract they made every new employee sign. If you wanted to work there, you had to agree to not work for another company in the same field in a 50 mile radius. If you refused to sign, you didn’t get the job. I only found out about that contract on my first day, so… it felt like a no-win situation. Some people tried to get out of that, but it was a legal nightmare for them.

The company in rural Illinois with its sister studio in Dallas was interesting, just because we got to see another perspective on the whole crunch issue. John, the head of the Dallas studio, was very vocal about the fact that he planned to get rid of the Dallas devs at the end of the project, because “that’s just the way the industry worked.” He figured they’d just sign on again, when the next project came around. I guess that was an older model for game development? It was the first I’d heard of it.

The Dallas devs weren’t too keen to stay on after that crunch. Go figure. And yet, because that game was reviewed so poorly, we had several return in the hopes that the next game wouldn’t be such a blight on their resume. That taught me something about the sunk cost fallacy, to say the least!

I’m happy that you haven’t had to ensure any nonsense like that. No one should have to do anything like that, in order to work in a field they enjoy.