« | Home | »

Enduring the Grind – Crunching as Company Policy

posted on April 20th, 2009 by bruce

Mike Capps, president of Epic Games, has come under fire recently for remarks about quality of life issues in the videogame industry made during the “Studio Heads on the Hotseat” panel at the IGDA Leadership Forum ’08 (video here).

The remarks which brought about the controversy were made at about 21 minutes into the panel. Taken in the context of everything he says, his remarks aren’t as inflammatory as they’ve been made out to be, but I still have experiences which would contradict some aspects of what he maintains is a good way to make video games. I’ll get back to him after a little background on crunching to make a video game.

For anyone who hasn’t studied the game development industry, the dreaded “crunch” refers to the period during which game developers work well beyond eight hours a day and sometimes through weekends in order to meet an important deadline, typically the shipping date of the finished game. It is, as you could imagine, generally dreaded. The video game industry is rampant with stories of employees forced to work marathon weeks of twelve hour days plus weekends for months at a time so that a project can ship when planned. Things came to a head with the infamous ea_spouse posting, revealing what were practically sweatshop conditions within a major American company. Any employee understands the difference between their passion being self motivated and being exploited, and it was obvious to everyone involved that what was happening was not healthy for the employees.

I have worked on projects which required extensive periods of heavy crunch. It is actually possible for you to, in a way, quit having a life outside your office during these crunch times. My first job with a major publisher required moderate to heavy hours for the better part of a year to finish Turok 2. So heavy was our crunch for the final months that, to this day, I can distinctly remember how new and novel a trip to a local shopping mall with my family felt on an evening after I had been freed from the bonds of crunch. It literally felt as if I had never been out shopping with my family before. I believe I can say I know how people freed from jail must feel.

How did this happen? What could have made so many employees let the needs of a company or a schedule dictate so much of the shape of their lives? In the early days of our industry, there was no industry to speak of; videogames were made by young people pouring their ideas for new creative experiences into personal computers on desks in their bedrooms. There were no working hours to speak of; the employees were either students who had taught themselves to program and wanted to make something like the games they played in arcades, or someone in an established industry who worked on a videogame idea in their off hours.

The passion for creating a videogame, as for any creative endeavor, can consume a person. It can become all that the developer thinks about, and all that he or she wants to do. Being young and passionate and working on something that was so entertaining, these early developers were willing to pour every spare waking hour they had into their creations. These people, propelled by a shared vision and a passion for what they did, eventually came together and formed companies. Naturally, the culture of these companies reflected the interests of its owners, and there was not so much an expectation of crunch as a desire for it. It was less a job and more a hobby, one you could devote endless hours to while getting paid. Friction naturally arose when the industry became large enough to need the talents people who did not want to devote the lion’s share of their life to it, and this has been the state of affairs for some time.

Is this “proper”? Because some people in an industry are willing to devote sixty hours of their life a week to their passion, does that mean everyone around them should as well? Is it better to turn away someone who does forty hours of high quality work in a week and replace him with someone who does more, but lower quality work? Even within the industry, the more dedicated people usually don’t expect everyone to share their passion to the same depth and extent that they do. As the industry ages, even the most passionate gamers within it want to experience a more well rounded life. There are far more pictures of children hanging on the walls in any given game company today than there were when I was getting started.

I believe recent high profile abuses of crunch have come from two factors. The aforementioned built in mentality that since someone is doing their “dream job” they should be willing to devote most of their waking hours to it, coupled with managers naive exploitation of the fact that few in the industry have ever made an hourly wage. The result is studios (infamously some EA studios) where managers simply expected employees to work nights and weekends as a matter of course, whether the needs of the schedule dictated it was necessary or not. Crunch in these instances becomes known as a grind (also referred to as a death march), which has no end in sight. There’s no discernible schedule, and no date or milestone to meet save for finishing the game. “Grind” is a pretty accurate description, too. The employees on these teams are eventually ground into a fine dust. There’s the camaraderie of a shared ordeal, fond memories of the times later and a product on the shelf to remind them of the effort, but little else that is ultimately fun about living through the experience.

Note that I said these managers are being naive. The first and most obvious argument against the fact that endless extended hours is productive is the mountain of evidence which led to the creation of the five day work week in the first place. Numerous studies, many of them cited by the IGDA itself, have discovered that after working a certain number of hours within a week an employee’s productivity drops off sharply. Sustained extended hours eventually wear out an employees to the point that they are spending as much time fixing mistakes due to fatigue as they are moving forward. You can multiply this effect when you consider that game creation is an inherently creative endeavor. It thrives on the people doing the creating being sharp, alert, focused, and passionate about what they are doing.

These naive managers do not seem to grasp the fact that video games are not stamped out by a machine on an assembly line (despite the frequent reappearance of some franchises on the shelf). There is no lever which, when pulled more often or more rapidly, will magically cause a game to appear in a shorter amount of time. They foolishly believe that if they have employees working on weekends, they are getting more value out of what they pay that employee. But things in a creative endeavor are never as simple as they seem. I have spent many an hour fixing simple mistakes put in by someone working extra hours, too bleary eyed to clearly see what he was doing. I’ve spent a few mornings shaking my head at the code I wrote late the night before. I have done it so often now that I recognize the point at which I am making the mistakes and I now force myself to stop, go home and think about something else for a while.

When you have to pay an employee to fix a mistake made due to fatigue, you are not getting any extra value out of their salary. You are wasting money. In effect, you are now paying that employee more money to produce the same amount of work, getting the product of their laborlater thanks to their diminished productivity. And you are reducing the quality of the finished game, which will be reflected in lower review scores and quite possibly lower sales.

Also consider that tired unhappy employees are much more likely to look for work conditions more conducive to happiness somewhere else. Employee turnover is not your friend in a place where intimate knowledge of the innards of a proprietary engine and production pipeline are required to add new features or content, or to fix existing ones. Every time an employee with that kind of knowledge walks out of a studio, the money and time spent so that she could acquire all that knowledge of your engine and processes will then have to be spent on her replacement.

Not all crunch happens because a malicious manager thinks he can get something for nothing. Often, it’s merely a case of a schedule being far too ambitious for the amount of time allocated to it. This has generally been the reason I have found myself working extended hours. In the early days, there were no production departments keeping a close eye on the progress of a project, constantly comparing what was left to do to the time left to do it and raising flags when it looked as if slipping dates were inevitable. Better planning and scheduling techniques can help mitigate the effects of ambitious schedules. Smart companies and teams are acutely aware that everything that’s desired for a game simply can’t be done given a certain amount of manpower, and are careful to scope the size of a project accordingly. Managing the dreaded “feature creep”, the last minute addition of features to a game, is a much larger concern today than in the past.

Which all brings me back to Mike Capps. He made it clear in his remarks that his company is very upfront that Epic’s “company culture” considers a sixty hour work week de rigeur. When someone knows what they are getting into, it’s hard to sympathize when he finds himself in it. Another earlier remark indicated that Epic employs profit sharing and thus the employees are motivated to produce a product with a smaller headcount to increase the portion of the profits for any one person. Everyone, not just the managers, shares the motivation for the extended hours, and there is a payoff for everyone involved for participation in the company’s culture.

He eventually makes the remark that he doesn’t think the forty hour week makes sense for our industry, and this is the part that gives me serious pause.

Is it possible that a person who expects to live a normal life, with friends and leisure and experiences they can only have miles away from their desk, can contribute nothing to the videogame industry? Is the knowledge in my head useless to a company because I have a wife and children I like to see in the evenings?

I really have to wonder, what is the motivation for his statement? Is he implying that people who are not so driven, so devoted to their desire to make games that they will do nothing else are of no use to the industry?

So I wonder, is this true in all forms of creative endeavor?

Don’t the best painters and writers and musicians devote their lives to their art? Why would anyone want to listen to a rock band that considered playing music a “normal job”? Would we have Mozart’s Requiem if his love of music had not consumed him?

But I also think I would hate to live in a world where artists had no life outside of their art, because they would have no life to inspire their art.

What would Van Gogh have given us if he had remained happily holed up in his studio avoiding courting so that he could devote himself to the technical details of painting?

Life has to drive a devotion to art for without a life, you have nothing to make art about.

Thus I find it interesting that Epic’s latest great work centers around a muscled space marine fighting monsters from below. This is not to imply that great human stories cannot be told in such a setting, it’s just revealing to me that with all of the prowess on display they have managed to apply a great degree of technical sheen to a game I played over a decade ago.

It’s obviously not necessary to make games that aspire to art, we are free to strive to merely entertain. But if we are calling for people to work a week and a half within a normal week, why do we also ask them to make the same game again? And then again?

Can crunch be removed entirely? Probably not. As I’ve said, game creation is a creative endeavor and thus will always inspire those making them to improve them, to make them more fun and engaging. Artists will want what they create to look as good as it can before potentially millions of eyeballs come to rest upon it. It’s almost inevitable that newly imagined features will creep in as a project progresses that everyone would like to see in the final product. In recent years, scheduling and production techniques which take this into account and respond and adjust the schedule as the game is in progress are becoming more popular.

Smart companies are now taking more and more input from employees during the scheduling phase. This not only produces a more realistic schedule (provided the employees are familiar with their abilities), but it also means that the employees are more motivated to deliver what they promised when they promised, even if it means working extra hours. At least in such a situation the employee has learned something about what they are capable of, and thus will be less likely to agree to an overly ambitious schedule in the future.

I’m not arguing that crunch should never occur, or that it can be excised completely. Personally, if I’m treated well at a company I don’t mind working extra hours to ensure the final product is as good as it can be made to be in a reasonable amount of time. If I’m convinced that my extra effort is making a game better, more fun and interesting, it will be done without any grumbling. But I have to question the culture that maintains it’s better to work more hours instead of just doing better work to begin with. I’d rather view this as a sometimes necessary compromise instead of the norm. And managers have to understand there is no linear relationship between man hours and creativity, in fact there may even be an inverse relationship.

I’m not so sure I could ever tell someone that their job “doesn’t make sense” within the context of a normal work week, that they should work a week and a half when their friends and family work one. Nor would I ever assume that because someone does that sixtieth hour of work that it will be anywhere near the quality of work done in the forty-first hour. I’d rather be treated like it’s my creativity and skills that actually matter, rather than my willingness to sit at a desk long after the sun has set.

1 Trackbacks/Pingbacks

  1. Pingback: My Dream Job and Rant « Ramblings about programming, gaming and other random stuff… on May 20, 2009

11 Comments

  1. TheJudge said on April 21, 2009:

    This is no shock. I worked at EA Seattle at the end of its run and they worked us like dogs during crunch time. The worst part is that you have no idea how long the crunch is for and many times it felt unnecessary. Making your testers or devs work 60-70 hours just to prove to your bosses is weak and unethical. It also underscores very poor planning.

    This article makes a great point of getting young people to work ‘their dream job’. Here in seattle Nintendo is the minor leagues of testing and we were told that over and over. Hey, this is your dream job right? And you know we can hire any kid out of high school to replace you. Brutal.

    Well written article.

  2. Phased said on April 21, 2009:

    I worked with TheJudge at EA Seattle, too, and it was by far the most grueling place in the 11+ years I’ve been in QA. Never in my life before or after have I done a stretch like I did there when I was doing 2 hr commute each way, 15-hr days, seven days a week, for 6 or 7 weeks. I was lucky if I slept for more than 4 hrs a night and was snapping at people at work left & right [which is weird for me because i’m a quiet & non-confrontational person]. The point in the article about more time = less productivity is so true, having personally lived it at EA and Sierra, who was just as bad as EA at crunch time.

    Being in the games industry you just expect to put in that much time and give up any life outside of work. It sucks, it really does. It should come down to managers being realistic when scheduling a project and not rely on having people work 60-80+ hrs a week as the norm or at least give some buffer room in case the project slips instead of doubling up everyone’s hours to make up for lost time.

  3. El Kabong said on April 21, 2009:

    I’m agreeing with both of those two. That said, the money I got when I was working like a dog was nice. It would have sucked if I was salaried instead of hourly, but when you pull in 1,500 a paycheck, it kind of takes the sting out of a 82 hour work week.

  4. Stefan said on April 21, 2009:

    First off I wanted to say this is a fantastic article…despite never having worked in video games, I have been a programmer in some long-term crunch environments, and am currently struggling through 60-80 hour weeks as a first year grad student. The descriptions here struck me as spot on and were quite evocative of my own memories.

    When discussing it with Jay, however, one other thing occurred to me. What the video game industry did wrong (to the extent that they were structuring this system intentionally) is that they made the senior staff keep working the long hours too. This prevents the system from being as self-perpetuating as it is in medical residencies, legal internships, academic positions, and on salusa secundus in Frank Herbert’s Dune (to pull in the fictional example which first articulated the effect to me).

    If you grind people down for 2-3 years, then let up on them and tell them how superior they are for having made it through all that, they will pretty much universally justify the ordeal as being necessary. They will also tend to dislike the idea of others achieving their same position without having gone through the same pain. It’s the Sardaukar principle at work.

    In the past year of grad school I’ve heard the “this is your dream job” line more than once to justify putting first years through extremely long hours. The difference between now and when I used to work as a programmer is that it’s often combined with the carrot, “It gets easier after the first year.” It’s looked on as a rite of passage, and people want to keep it around because it’s hard for us to accept that we suffered needlessly. Even just thinking, “I’m good at what I do after it, so it must be what made me good” makes it all worthwhile, and solidifies the system in your mind.

    If, however, you don’t have that mobility from suffering to superiority…then you get what happened to EA. And EA was hardly producing the Sardaukar of the video game world, despite owning the rights to that name.

    The way it works in the industry at the moment requires external enforcement…people within the system aren’t going to want to keep it that way, since even the senior programmers and designers are ground to the bone. That creates conflict, backlash, and tension. I think this may be a good thing, though, since it will hopefully lead to a more productive (and healthier) work environment for everyone involved.

  5. Shota said on April 21, 2009:

    Very nice article. Well done. Interesting!

    In a generalized way I would like to say that all new industries lag behind when it comes to regulation. If we look back in history I think we’ll find that the infancy stages of most industries (and videogames are barely out of the uterus) involved hard labor and long hours. Harder and longer then they did after they have matured a bit as a field of organized human activity. The sad thing is that all that progress was almost always a product of some hard fighting revolutionary minded thinkers who fought for workers rights. The problem is, i’m not sure if this day an age that ‘unionizing’ mentality stands a chance in the US anymore. I’m Marxist enough to hate the purely profit driven capitalist guiltless fat cats, (I was Marxist before the recent discrediting events punched the “for-profiters” in the gut) but who else is like me? Most people dream of growing up to be a fat cat.

    Sorry for the tirade, but I think at least some of the aforementioned is related to the heart of this article.

  6. pat said on April 21, 2009:

    shota – i dont think video games are in the same place the cigar rolling industry was when samuel gompers started to organized it way back when. i think this is more a matter of certain industries having higher expectations when it comes to when it comes to time spent at the office. in investment banking (to the extent thats still a profession) the expectation is that you basically live at the office and many of the people who do this take a perverse pride in how hardcore they are about about. when it comes to epic specifically i wouldnt be surprised if they think they are cocks of the walk because they put in long hours that result in bald space marine games. as for a company like ea, there may be some capitalist-fatcat-ism at play, but i wouldnt hold my breath for a…haymarket square (maybe? i dont have that many 19th century labor organizing references at the ready)

  7. GJ said on April 22, 2009:

    I loved the article, very well written and brings to light a lot of the issues in management of the gaming industry, which in many ways mirrors the way the movie industry churns out (which is a bad thing). I will now disagree with a few points and be called a fascist and capitalist. Having interned in college enough to know I’d never want to work in the gaming industry for the reasons stated in the article, combined with some close and relatively competent friends who still work in the industry, there are two fundamental reasons these conditions will never change. First is that the gaming industry is a “cool” industry–like sports, movies and the music industry–and therefore has an unlimited supply of labor. There are always people who want to get into it because it is cool. As a result, managers have free reign to treat their employees like crap, because they are expendable. Second is that the industry is profitable, so does not need to change for an economic perspective.

    Getting to the comments of the development cycle being poorly managed–the biggest game studios at this point are public companies, which means they are subjected to Wall Street scrutiny, i.e. they have to make every deadline, over promise timetables, etc, in order to keep analysts and shareholders happy. But keep something in mind. As poorly managed as most medium/big public game companies are, they are also profitable. When you’re running a company, you have two routes to go if you want to be successful. You either stay small and be a high value, boutique business with low overhead and a ridiculous operating profit. Or, you get really really big, and you shoot for a consistent 5-15% profit and focus on high volumes. And despite all of the bad press, most of the big gaming companies go the latter route, and do so successfully. Focusing on 2007 income after taxes vs. revenue (every company I’m about to mention lost money in 2008, stupid economy), Activision Blizzard made 227M on 1.33B revenue, EA made 72M on 3.1B revenue, THQ made 91M on 1B in revenue. The numbers switch around for the previous three years, but all of those companies were profitable in that time period. Those were the easiest to find American public gaming companies that deal in software (Sega, Nintendo, Sony all have so many other interests and this is just a comment response–but they exemplify this concept).

    The point I’m getting at is that there is money to be made in mediocrity. Yes, it is brutal on employees–but just because the development cycle is brutal and middle managers are idiots does not mean the upper management isn’t aware of this and isn’t planning around it. With companies as big as these, you’re not going to have top notch talent, you’re going to have a lot of warm interchangeable bodies, and you need to manage accordingly. It is great that Epic offers profit sharing as an incentive–but really they don’t have to do that because there is plenty of mediocre talent who will work for a regular salary. Ultimately, if a big company has an endless supply of labor and is making a profit, you are not going to see change any time soon, because there’s no economic incentive to do so.

    Now my challenge to the rest of you who know so much more about developers than I do–are there any studios that consistently churn out top notch games? Are they private or public companies? Are they owned by someone bigger? Are they smaller or bigger in terms of employees? Do they get profit sharing or stock, or are they just better run?

  8. christian said on April 22, 2009:

    GJ – to answer your last paragraph, as a software guy with no game industry experience, your best example is Valve Software. I don’t know all their details, but they’re small, consistent, and make cash. They were started in the first place in large part because Gabe Newell and others left high paying jobs at Microsoft and were able to keep themselves afloat until Half Life started selling like gangbusters. While I can’t say what their salaries, benefits, etc are, they are a prime example of a studio that does what they want and answers to no one, and likely takes good care of employees. Unfortunately, they are a model many point to, but which is difficult to establish.

  9. Bruce said on April 23, 2009:

    Well, that was…not really quick at all.

    http://kotaku.com/5224347/epic-boss-responds-to-exploitative-working-condition-claims

    Whether he is “clarifying” or backpedaling I can’t say. But at least the way he describes it now is much more reasonable and close to reality than just saying a forty hour week “doesn’t make sense” for us. To be honest, the crunch he describes for one Epic project, six weeks, isn’t really a soul crushing death march. I’ve endured far worse to make games.

    Just as in any industry that is driven by deadlines, it’s expected that anyone is willing to put in extra hours when it’s called for, and those employees who would refuse aren’t necessarily heroes to the people who would have to pick up the slack for them. Smart companies avoid it when they can, and a dedicated employee works extended hours when it’s necessary without assuming that should be the norm (and managers violate this at their peril). I still found his original assertion that since there are other industries where giving your life to “the firm” is the norm that it’s automatically sensible for ours to do so to be a bit offensive. Just because that’s how lawyers or investment bankers do their job doesn’t mean it’s good for us to do our job that way and, once again, yields to the unproven assumption that good creative work gets done past forty hours a week.

    I’m just continually -amazed- at how MUCH attention is paid to hours and how little is paid true quality of work down in the guts of the game production pipeline. When your well paid experienced senior developer spends days and days maintaining and improving a system jammed into the code base by one of those beloved young idealistic guys late at night were you -really- getting any extended value out of that noobs’ extra hours? One of the best artists I’ve ever worked with at times spent a fourth of his time wrangling art assets made by other team members into something that fit into our polygon budget. If a programmer engineers a data driven solution in forty hours that lets the game’s designers externally tweak and tune a greater portion of the game, his forty hour week was worth more than several months of a young programmer fiddling with a tweaky, hardwired system that forces the designers to ask a programmer any time they need to change a game parameter. And to answer your question, yes, I am one cranky son of a bitch to work with. I’ve MADE all of those mistakes and I’ll be damned if some noob is going to get away with them on my watch.

    Thanks to all for the reads and responses.

  10. GJ said on April 24, 2009:

    Bruce, do you know anything about Valve’s working conditions and compensation situation? I’m curious if there’s any correlation between quality and conditions.

    I also know some folks at Bethesda and they have said the company is known for taking its time and not rushing products. I believe this is actually true at least in terms of the launching of Fallout 3 and the Oblivion before it.

  11. Bruce said on April 25, 2009:

    I haven’t read anything about Valve’s policies specifically.

    “Not rushing” doesn’t necessarily mean there are never any long days, it just means that the deadline isn’t so set in stone that if the creative team needs another few months to add in some kickass features, they’ll probably get it. And the end result will equal better reviews and more dollars. Mind you not many franchises have that kind of power, and it also takes serious financial backing to agree to throwing another few months at a title.

    I probably shouldn’t have come down on Gears so hard. “Hits” in this industry are driven more by market reality than creative vision. Consumers have a pretty distinct and largely BORING set of criteria for what they want in a game, and I can’t blame developers for giving them what they want.

Leave a Reply