Hacker News new | past | comments | ask | show | jobs | submit login
Nearly Half of Game Developers Want to Unionize (engadget.com)
483 points by toufiqbarhamov 88 days ago | hide | past | web | favorite | 526 comments

As someone who's been in the game industry a while, I can see where this is coming from. This is not about software devs, it's a special challenge specific to game devs where the lower pay and higher overtime expectations really affect people's lives. This is then made worse as studios keep closing down, because people will put up with more bs than usual for job stability.

No idea if unions are the solution, but this is definitely an issue specific to the game industry right now.

The biggest reason why gamedev jobs tend to have lower pay and more overtime compared to many other swdev jobs is really simple: there are lots of sw devs who are willing to put up with those conditions because they want to work on games.

Also, it's easy to claim that one "wants" to unionize on a survey. But that's quite a bit different from actually doing it. So I would take this survey with a large grain of salt. It's not like most game developers couldn't get other sw dev jobs if they wanted to, after all.

If you want to know what people really value, ignore what they say and focus on what they actually do.

Disclaimer: have worked for game companies before

> there are lots of sw devs who are willing to put up with those conditions because they want to work on games.

This is exactly the reason why I don't want to pursue a career in game development. I find games exciting but that doesn't make it feel worth to me to make compromises in pay and working conditions. Instead it can just be a hobby.

I used to be in the game industry. It's now just a hobby for me. If you're willing to work on small games in your spare time, you don't have to kill yourself to work on those things, and can still make things. There are times where I'm tempted to get a full time job in the industry again, but then I read articles like these and I remember why I'm still out. I have nowhere near as much passion about any other industries though, and it can be really hard to keep going to work every day at "corporate job X".

I’m trying to talk my kid into doing FX instead of games, so I get to see him in his twenties. It’s not working so far but he has figured out that EA and Activision are both bad actors.

Presuming you mean TV and movie FX then the working conditions aren’t great there either.

Also the huge multinationals are...huge - it varies a lot between different offices and teams.

There is no one EA or Activision experience, they are very heterogeneous entities.

While this is true its not the biggest reason. The game dev cycle is just different than the tech cycle and it opens a window to exploitation.

Games releases are seasonal so a common practice is to hire for an xmas release and gut the teams after that.

Shady bonus practices are common. You often have very little recourse into how yearly profit is calculated and paid out.

The separation of studio managing employees and publishers holding on to IP means that studios can close up shop without the rights holders being liable for any back pay or bonus.

And this is well known, yet people go into game dev anyway because they want to build games.

Sounds like a close parallel to "Hollywood accounting", where nested corporate structures are used -- and abused.

This is the correct answer. Management promises a unicorn, works you like a dog, and then pulls the rug out from under you, just before you deliver.

Alternatively, if you dig in, and refuse to deal with that bullshit, quitting, or being dismissed for not being a team player, half-way through crunch on a shitty project is a great way to burn bridges in that industry.

> Alternatively, if you dig in, and refuse to deal with that bullshit, quitting, or being dismissed for not being a team player, half-way through crunch on a shitty project is a great way to burn bridges in that industry.

I don't doubt it, and your comment supports my original point: that there is a surplus of devs willing to tolerate such conditions. If that were not so, quitting would not be near as much of a problem for one's future employment prospects, because employers would not have as many candidates to choose from in the first place.

I won't even play games anymore because I find those conditions to be deeply disturbing on many levels. I can't in good conscience support these workers grinding themselves to the bone just so I can receive slightly more motor cortex stimulation than I would spending the night on HN.

The film industry figured out long ago that they need to treat their employees with respect by giving them predictable hours, fair compensation for overtime, and guarantees that they won't lose health/retirement benefits across projects and studio closures. I do believe the game industry will get there eventually, but it seems it will take a lot of kicking and screaming.

You can't support these workers grinding themselves to the bone? What are you talking about?

There's no slavery here. I've been there. Unlike the people making your shoes in Bangladesh, everyone making your games is there because they love it. These are talented individuals who could easily go work in other industries.

People in the industry want better conditions, and more power to them. But this is not slave labor and no one is stuck in the game industry.

What you're saying is tantamount to not buying a painter's painting because he worked too hard on it. Trust me... the people you think you care about want you to buy the games.

> everyone making your games is there because they love it

That's simply not true. I'm sure most of them entered the industry for the love of it, but that doesn't mean they love their current job or being expected to work long hours.

> These are talented individuals who could easily go work in other industries.

There's a lot of shovelware in the games industry and I doubt many people love working on it. Do you really think the people working at studios churning out hanah montana games love their jobs? At best their hoping to work on something better in future, at worst the probably can't move elsewhere.

It can be a lot harder to break out than you think, the type of skills a game developer would have aren't in high demand in a lot of places and they could be missing some important skills used in the the more general industry, like SQL. This is an industry that loves to type cast, even something like getting a c# job as a java developer can be tough.

And then you've got the non-developer jobs. Graphic artists, writers and music composers don't have nearly as many options available.

Based on first hand experience, no it's not that hard to get out of games industry, not for a sw dev anyway. It's true that (for example) intimate knowledge of real-time 3d rendering is a rather niche skill, but OTOH knowing how to write fast code has much wider applicability, and most game developers will tend to be much better at that then your average sw dev.

If anything, it's harder to get into the games industry than to get out because there is relatively more competition for gamedev jobs than for other sw dev jobs. So I would argue that if you can get in, you can get out, and when getting out you have the added advantage of being more experienced than when you got in.

Late to the party but wanted to point out that in other industries like application development you are probably more likely to accept slower code if it means more readable and maintainable code whereas in games you often (opinion) need to really crank out the performance at any cost.

But I mean.. You know those interview questions people complain about? The ones with the binary trees and inverting? There is a certain group of people I'm thinking of who would have no problem with those questions... Besides High Schoolers! 'Cause let's be honest here, graph and tree traversal is High School level comp sci...

The reason there is so much shovel-ware is because there are so many people interested in making games. So many individuals involved in shovel-ware are talented. Perhaps not the best artists, or game designers, but certainly have enough technical chops to do other work.

That's a capitalistic outlook though. I don't particularly subscribe to "choice" or "free will" in general though to be honest..

You could just play indie games.

Maybe people are working themselves super-hard to make indie games, but it's self-directed and totally by individual choice. Tons of great indie games out there.

> totally by individual choice

So are people working for regular employers making video games. There's just more people who want to make games than work on insurance calculation (etc) software. That's why wages are lower.

I really enjoy paying small content creators too! It’s a strength of the internet.

As the film industry has demonstrated it is perfectly okay for independent productions to be non-union. There is little need or use for organized negotiations in that situation.

I agree, but I'm not sure any potential union would.

E.g. want to hire a union programmer for your 3-person team? Whole studio has to now follow union rules and bring all other developers into the union as well. Forever.

I don't think that VFX teams have a significantly better time than game devs. They compete for contracts with razor thin margins.

'Work like a slave on something you enjoy or get another job' is a scammy proposition. Games are highly profitable, no reason developers shouldn't negotiate for what their services are worth. Individual negotiation is difficult, collective negotiation is easier.

SOME games are highly profitable, and many are borderline and probably wouldn't be profitable without exploiting the devs. I don't like it either, but that's how it is.

If the issue is that someone will always be willing to accept a lower salary to break into the market, well then your fucked.

And it's not a case where you have the ability to strike, because if you did, someone will just come along and take your job.

The market is trying to tell you something, you just need to listen to it.

> Games are highly profitable

The gaming industry notoriously wildly swings from boom to bust.

"But that's quite a bit different from actually doing it."

I think if most of them could 'tick a box and join the union' they would - whereas actually establishing a union would be hard, yes.

There could be some terrible fallout as well - companies shutting down studios and opening up another one elsewhere etc..

Don’t forget the employees have to pay for the union.

Right now, they pay for not having a union, in dollars, stress, their health, divorce, etc.

Some do. Others may lead totally balanced lives, with a fair employer, earning good wages -- a union may serve them only detrimentally.

It's not as clear-cut as "unions are always good for all workers."

Those for whom a union would not be a benefit are so exceedingly few, but this is the standard anti-union message that has been given out for the last 40 years.

Game devs are hardest hit by bad conditions, but every worker in every company would benefit from collective negotiation -- use your imagination, you could have a much higher salary, more flexibility in your workplace, more autonomy, you name it. That's what unions do, they let you bargain for the things you can't as an individual.

Lamentably, there's still a lot of folks in the software field who see themselves as the rock star, rather than the rank and file.

Also lamentably, there's still a lot of perception that a union means: you can't fire dead weight (you can, you just have to follow a process), pay is based on seniority (doesn't have to be, but what's wrong with that anyway? we don't have a good way to measure productivity or value, so what other metric should we use?), that they're corrupt (they're democratic institutions, so they're as corrupt as the members make them), etc.

All of it anti-union propaganda that's been working pretty well for the last 40 years.

I don't agree - gaming is notorious and specific for it's bad conditions.

People in most tech firms are not clamouring for a union.

When people are talking about it, it's for a reason.

If you make a movie, everyone involved is 'union'.

I don't see why gaming should be any different.

Unions virtually always benefit the worker, the biggest losers if there are some are the community/everyone else.

Yep. It's just frustrating seeing anti-union propaganda repeated, uncritically, unthinkingly, year after year, especially on a forum where people pride themselves on being thoughtful people.

Of course unions can be corrupt, but the only thing more corrupt than a union is management.

> totally balanced lives, with a fair employer, earning good wages

In that case you could view union membership as insurance, in case you ever lose that great job. Keep in mind that your local pipefitters union is not the only model out there - SAG-AFTRA seems to work for actors/actresses across a broad range of income, experience, and talent levels.

There is also a significant barrier to entry in SAG-AFTRA. Doesn't that mean union membership is just as probable as landing that great job?

No there isn't. It's very easy to qualify for membership in SAG as virtually every feature-length movie is SAG compliant. The only filter they use is checking there's a legit paystub so you didn't just shoot a short in your bedroom and try to pass it off as a screen credit.


Worked over a decade in indie films, never heard of any producer having a problem getting SAG to recognize paperwork.

The barrier to entry is a $3k initiation fee. That seems steep, but lots of people could probably put together that much money if they needed it.


Are you sure? To me that would be as critical as car insurance!

And are expected to participate in voting and walkouts.

most game developers couldn't get other sw dev jobs if they wanted to, after all.

The C++ guys working on the engine, sure. But they are a minority and the rest don’t have easily transferable skills.

I didn't go into game development, even though that's why I started to learn programming, precisely for this reason.

While I hope that the issues you mention above plus the cycle of game launching after a massive deadline crunch leading to development teams searching for new jobs (ie less stability), I don't know that it will happen. Why? People buying games.

Personally, I tend to buy the top tier of a game if for no other reason than they typically include the game/season pass (or whatever branding they go with) so that I don't have to manage it after the fact, of the games or expansions I bought recently, they've been the highest tier available: Assassin's Creed Odyssey, State of Decay 2, World of Warcraft Battle for Azeroth, Tekken 7, Destiny 2 Forsaken (which I really haven't played. Still in the Osiris area. Eh, oh well), Ghost Recon Wildlands, The Division. To me, the fidelity of the graphics and scope of the games (in that most everyone is chasing epic stories now and if it isn't 30+ hours, it is too short) means the price should bump up beyond $60 so while the tiers of games seems like a dirty move, it seems to me that it is so they can sell you $60 worth* of a $100 game instead of you looking at a $100 game and scoffing entirely.

So basically, until people purchasing games are willing to accept that a AAA game in 2019 takes more resources to create than a AAA game in 2009, I don't see this ever getting better.

* the $60 version is a full game and not 60% of one though so in reality at that price point you're getting way more value.

This is simply a marketing line, not reality. EA, Activision, and Rockstar are making record profits, not struggling to meet budgets. The base price of a game as you point out, gets you less than a whole game, for which you need to buy a season pass (sometimes as many as three of them) they have product tie-ins (Pizza rolls, Dew), they have DLC om-disc and off, and a world of microtransactions abusing gambling mechanics, and the disgusting “Live Services” model.


>The base price of a game as you point out, gets you less than a whole game...

This has become a meme at this point [1], but has anyone actually studied this in any type of academic or scientific manner? There are plenty of the games that have DLC, after release support, or any number of other revenue streams that would 100% be considered "complete games" by any reasonable standard. I know it is anecdotal, but compare Zelda: Breath of the Wild to Zelda: Ocarina of Time or FIFA 19 to FIFA 08. The release versions of the two newer games were probably more complete than the older versions even though they have DLC and micro-transactions respectively while the older were "done on release". In those two cases the extra revenue streams are helping to justify and pay for the continued support of the newer games.

[1] - https://www.ranker.com/list/dlc-memes-gaming/robert-carneval...

Those are outliers. Game industry have very small margins on average, it however tend to be hit or nothing. If you follow the news, studios are closing right and left.

But, the crunch does not happen because it would be effective. It is not effective way to create something. It happen because of mismanagement and culture that glorifies it - both among managers and workers. Crunch is more related to desorganization and lack of trust then something that would be smart to do.

beyond adding more story or visual add ins, what is a game that can't be completed without DLC? Never played Battlefront 2 (or 1 for that matter) and those tend to be tossed out there so I can't confirm or deny that one.

Diablo 3 base game - if you beat that, have you not beat the game? The DLC adds a new act to it, but for all intents and purposes it is an add on to a game.

What sometimes happens is that developers will create a game and then, very close to the end, some portions will be hacksawed out and turned into dlc. If done poorly, this can be very obvious and can ruin the game - the last Deus Ex game seemed to suffer from this.

This not standard practice. DLC plans will be specified at a high level early in production but will only solidify as the game gets closer to ship. There is a gap, which differs for different disciplines, between finishing work for the shipping build and its actual release. DLC is a great opportunity for the team to continue working at peak production on new content.

I've never seen content held back or cut out of a main release explicitly for the purpose of including in DLC. What sometimes happens is that features or half-finished content that the team decided to cut from the game is revisited for DLC. From my perspective as a game developer, modern games are never finished but at some point you have to start shipping things. DLC is an opportunity to explore ideas that would have otherwise been left on the cutting room floor, and provide some stability for large teams as they ramp down from full production.

I purchase most of my games on sale on steam for under 30 bucks. That's what they are worth to me, and clearly that's what they are worth to the publisher as I'm purchasing them legitimately.

Games made in 2019 have a larger audience than games made in 2009. I don't pay 600 dollars for a ticket to see a movie with a 300 million dollar budget. I pay the same as a movie with a 10 million dollar budget. Memento had a 9m dollar budget; I have not been 30x more entertained with a movie than Memento.

This has been a big issue in accounting, specifically public accounting. You are basically paid a salary and then you basically become a slave to the firm. You put up with lots of BS because you don't want to get fired or passed up for a promotion. The reason it is like that is because the partners of the firm get a cut of everything you produce...so the more you produce...the more they get. They quickly realize they need to figure out ways to keep you working more and more.

And I suspect that having sfx and vfx unionised which game dev is closer to might help.

I suppose this could happen to software engineering jobs outside of the gaming industry too once we end up with an equal number of over supplied software engineers.

    once we end up with an equal number of 
    over supplied software engineers.
I remember this being said 20 years ago & yet never seems to actually happen (so far).

I've been in the game industry going on a decade now and I have zero interest in unionizing. I do not want my future in any way controlled by yet another external entity that I have little control over and promises nebulous results.

As an employer I want to be able to fire toxic or under-performing employees at will without jumping through a hundred hoops to make sure I am not breaking any union rules.

As an employee, I want to be free to negotiate my own wages based on my merit.

If there is one thing I would want a union to do it would be to collectively enforce some kind of profit sharing plan. The interests of the employees and the employer align -- make a great game that sells well. Anything else, no thanks.

Politely -

You're management/executive at this point. It is obviously in your interest to be anti-union.

It is also the case that unions serve the interests of those who are not "rock star developers" more than those who are.

As someone on the employee side (not games), I too want my company to be able to fire bad employees at will. Having been at companies that do and don't, I very much prefer the ones that do lest the workplace become bloated with low quality.

> It is also the case that unions serve the interests of those who are not "rock star developers" more than those who are.

I'm not convinced in the current market that those interests need serving to the detriment of the better devs. Were the employers in a more favorable position I could see a different view. But right now any half-way capable dev (i.e. not "rock star") has plenty of current advantages sans union.

A union is ultimately just labor doing collective bargaining, which is certainly something capital+management does. What is bargained for isn't set in stone; while lots of people are interested in making sure firing isn't arbitrary, labor doesn't have to bargain over making people unfire-able, or even on specific compensation. A well-run union bargains for what its membership wants. If a union of software developers wants competent productive co-workers, then it can leave at-will firings on the table and instead push for limits on hours worked per week, or bonuses based on revenue (with access to the books to double check), or employment agreements that vigorously protect individual IP rights to anything made on their own time, or sane office layouts instead of open plan, or a generous severance package with at-will firing, or provisions for no confidence votes for some management positions, or, well, any number of things.

What else can you imagine beyond compensation or job security that could be improved in the lot of a software developer? Maybe even some things could yield productivity benefits that management would like if they could see their way through it? And if some of these things are more likely with collective bargaining, why would you rule it out as a tool?

In the current market, I'd rather bargain for myself. We shouldn't pretend what a union wants is what every one of its members wants. Leaving at-will firings on the table for instance is unlikely to garner popular support. I personally don't want specific severance requirements for those fired, otherwise you end up with defacto unfire-able situations because you've artificially increased the cost to remove a bad employee. I don't want to force them to run their company a certain way by rule, I'll just voice displeasure and leave if necessary. Due to the vast differences in employer packages and the freedom of choice given to the employee in the current demand-heavy market, I find individual bargaining to currently be more beneficial than collective bargaining.

> if some of these things are more likely with collective bargaining, why would you rule it out as a tool?

Because of all the negative externalities that happen to some employees and employers. If it was all upside there would be no reason. It's not ruled out as a tool, just its use in the current dev (general, maybe not game specific) market.

> We shouldn't pretend what a union wants is what every one of its members wants.

We shouldn't pretend that what a union wants is arbitrarily disjointed from the individual interests of its members, either.

Every collective has the potential to depart from individual preferences in some way. How much that happens varies in practice.

And it's pretty likely that your employment agreements are not an exception, even if you've arrived at them without any collective work on your side.

> Leaving at-will firings on the table for instance is unlikely to garner popular support.

This assumes the only incentive for joining a union is ultimate job security. That's a popular anti-union conception, but that doesn't make it correct. Any popular benefit that is not commonly conferred could provide a cohesion point for collective bargaining.

> I personally don't want specific severance requirements for those fired, otherwise you end up with defacto unfire-able situations because you've artificially increased the cost to remove a bad employee.

Increasing the cost of something is very different from making it defacto impossible. Though it is one way of providing an incentive against doing something arbitrarily. And it's not clear a negotiated price for firing would be any more "artificial" than any other negotiated price, up to and including your salary.

> I don't want to force them to run their company a certain way by rule, I'll just voice displeasure and leave if necessary.

That's pretty much how unions work. It just turns out to have more leverage if you multiply it by the size of the participating workforce.

> Due to the vast differences in employer packages and the freedom of choice given to the employee in the current demand-heavy market

Again, the freedom of individually negotiated compensation can co-exist with collective bargaining. A sibling comment even points out examples.

Hollywood unions provide protections on the low-end and the high-end while letting those at the high-end negotiate better pay packages...

Doesn't make much sense. Collective bargaining is just that. If you trust the company to do what is best you should have no problem because the company will bargain for what it finds important. Also you seemingly care a lot about what would happen with a union, but essentially nothing about how you are treated by the company. That isn't really a very strong position to make an argument from. If your whole idea is that you don't mind taking other opportunities, presumably it would be worth it for you to move to a company that isn't unionized in the same way you would if you didn't agree with what he company was doing.

"A well-run union bargains for what its membership wants." how? if it's via democracy then no thanks, as that comes with so many issues it's not even funny.

I think making these arguments at this point shows just how effective the decades long campaign of anti-union propaganda has been.

How great did those rockstar employees do when google and other massive companies engaged in anti-poaching agreements to drive salaries down? How good do they do when they end up divorced and miserable after putting in yet another 2 months of 100 hour work weeks?

I think not recognizing the arguments as built instead of propaganda-fed shows how effective the anti-corporate narrative has become.

Unionization doesn't stop illegal collusion, prosecution stops illegal collusion. What is with all the appeal to emotion with divorced, miserable, 100 hour work weeks, etc? Most employees did just fine anyways. Every single job would get value out of a union if the justification is simply saying "but tell that to the employee that was divorced, overworked, blah blah emotional language". At a macro level there are practical concerns.

There are many different types of unions, ranging from the destructive (i.e., Teamsters) to the cooperative (Hollywood guild-unions).

The anti-union propaganda did an amazing job at making most people think that all unions were like the corrupt Teamsters unions of NYC and Chicago, when most of them are nothing at all like that.

Most employees did just fine anyways.

Categorically false. Most employees did fine because unions negotiated the labor rules that led to those "fine" working conditions. Those conditions have worsened as unions have lost their leverage.

> Unionization doesn't stop illegal collusion, prosecution stops illegal collusion.

You could make that same argument against the concept of defense lawyers. How would you feel if you had to defend yourself against criminal charges without a lawyer? Do you think justice would be served if you had to go it alone against a well-funded, experienced team of prosecutors who want to convict?

Yes, defense lawyers aren't perfect and can't stop all prosecutorial misconduct, but they still help a lot. Ditto with unions.

> Unionization doesn't stop illegal collusion, prosecution stops illegal collusion.

And unions have labor lawyers. Wouldn't it help the prosecution to add positions whose job would be to identify such issues, and which have visibility across employers?

Today, without a union, how would you as an individual ever know if companies are colluding against you and other workers? What would you do to cause this offense to be prosecuted?

> Every single job would get value out of a union if the justification is simply saying ...

I agree with this sentence, but I draw from it the opposite conclusion. I think most jobs would get value out of a union.

I don't understand your logic here. If something is desirable for everyone, then everyone should do it. We shouldn't assume it's not possible for everyone, and therefore conclude that it must not have been desirable in the first place.

Unionization helps because of the threat of well organized action against the company, including litigation.

Not all unions are like public employee unions.

If they unionize, game developers are most likely to copy the union/guild format of Hollywood unions, which provide salary and health protections but otherwise let the members determine employment terms on their own (i.e., contracts for days, weeks, full projects, long-term, or open-ended are all acceptable).

Protecting bad employees is not really what game industry unionization is all about. Its more about standardizing job titles and accreditation, like film industry unions.

Admittedly I assumed this thread had devolved to general development unionization. If game industry employees are valuable, not easily replaceable, limited to only a few companies, and can't easily repurpose themselves then unionization has value.

Unions in Hollywood place a useful role of insuring proper credit and arbitrating when there is a dispute. Some game publishers have tried to avoid credits, partly to make it more difficult to know which employees to poach. But I have seen credits sorely abused in some instances.

It would be great if we could fire bad bosses.

That's the beauty of the current over-abundance of demand for devs. You fire bad bosses by quitting. Simply put, if you want to pick your best company, you have to be willing to find it. Similarly, if an employer wants to pick their best devs, they have to be willing to do things like not have bad bosses. In markets where demand is constrained, mobility may not apply, but neither do the benefits of cross-company unions.

I've been an employee for 80% of my career and from a moral standpoint I believe in treating people fairly.

The current unionization efforts provide almost no details on their goals beyond "everything will be better for everyone". On either end of the employee/employer relationship I have a hard time seeing how anyone can support such a generic effort.

A hard stance on working conditions and compensation is a starting point for a real conversation. The majority of employees would have no problem supporting this, and since I already value work/life balance and fair compensation, my motives and interests would align.

Then this doesn't affect you. Even in industries that have unions, the smaller shops are frequently non-union, and if you treat employees well, they won't have any incentive to organize.

The community theatre down the street doesn't require every performer to be an cardholding AEA member, and it wouldn't make financial sense for anyone involved (management, workers, or audiences) for that to change. The mere existence of a union doesn't necessarily change anything unless workers want it to change.

For big game dev companies that have repeatedly shown that they don't believe in treating people fairly, this would bring professional representation to the workers' side. Even if the details are light at this point, I don't see how you can be against that in principle. The employer/employee relationship is inherently asymmetric, and this attempts to correct for that.

Let's use SAG as a blueprint for a game industry union.

If I were a small non-union indie shop and I was lucky enough that some highly talented AAA developer wanted to work with me, I would be unable to hire him if he's union unless I converted my entire shop over to be union. If I didn't, he would be in direct violation of his union contract and could be fined or lose his membership.

I understand that not every union functions the same, but it is not hard to imagine that wherever there is a lot of money at stake the union will do everything in it's power, to well, seize power.

Considering the movie industry is dwarfed by the video game industry from a revenue point of view, I can very easily imagine this scenario repeating itself.

> If I were a small non-union indie shop and I was lucky enough that some highly talented AAA developer wanted to work with me, I would be unable to hire him if he's union unless I converted my entire shop over to be union. If I didn't, he would be in direct violation of his union contract and could be fined or lose his membership.

I've never heard of unions working like this, this may be a US-specific thing. Unions are just workers right groups, if the company is big, the group is inside the company, if it's too small, you just apply to the local union. They help to negotiate better laws and benefits for the whole industry & also directly at a company level in big companies.

Union works this way in the US? Wow, this is not a clever way to achieve anything. I understand why some people are anti-union then.

Not all unions function this way, but this is one possibility.

I'm not a fan of unions either. That said, I won't work more than 40 hours in a week more than a handful of weeks out of the year. I'm 44yo now, and when I took my current job, I stated I only have about 1 crunch week a quarter in me. If I have to do more than one 40-50+ hour week a quarter, then it's not me that is the problem.

I don't work in gaming though. I would love to, but the pay for business development is frankly better with a lot less stress. I wouldn't mind seeing a push legislating that any salaried worker in a week must be granted an extra paid day off at 45 hours and each 5 hours above in a single week. That would stop, or at least compensate, the abuse.

I'm with you on the crunch limit, few weeks a year at most and even then anything beyond 50 hours is pointless from a productivity point of view.

Addressing overwork at the legislative level would get my support if there were a way to trade pain now for benefit later (e.g. your extra time off suggestion).

and how do you get such things addressed at a legislative level?

Maybe unions can be useful after all!

Are you suggesting that a significant portion of legislation is because of unions? Maybe a part of it, and probably so at the state level. As it stands, I shouldn't have to be in support of a union in order to support legislation.

Working with toxic or under performing employees is just as bad as having them work for you.

Absolutely. And it's morale poison for the rest of the team. I speak from experience.

Then don't let them into the union? Or, alternatively, fire them from the union. A modern union is not restricted to follow the same practices as ones for different times and industries.

Then the rules for firing toxic and under performing employees can be written into the union contract and how the union conducts its business... This isn't an unreasonable ask, no one is going to want to be saddled with a toxic person they can't get rid of.

These are all very old problems brought up when workers want protection from abusive employers by forming a union and is not unique to the video games industry at all.

Even railroads and textile mills didn’t ask their employees to work 80 hour work weeks.

There are some serious bad actors in the game industry, Rock Star being one of the worst.

Where I've worked we valued work/life balance as much as possible. Did we have a bit of crunch close to a release? Yep, but it was short and a few weeks at max. Most of us enjoyed it due to the extra camaraderie.

If there were a unionization effort that focused on 1) a hard 40-50 hour cap on work hours (or with paid overtime) and 2) enforced profit sharing plans, I could support it.

The current main unionization effort basically says "things will be better for everyone" on their FAQ. Total bullshit.

Saying that 'most of us enjoyed crunch time' seems like a bit of projection, especially for people that might have families or other social obligations outside of work that crunch time eats into.

We were sane about it, family obligations always took priority. There is a difference between a sprint to the end and a death march like Rock Star.

Just so you know, this is the knife edge that leads you down to crunch.

Literally every time I had a 3-4mo 80-100hour crunch it always started out with "we're going to push for a week or two" which just transitioned into "just one more week" for the next 3 months.

Also, shame for celebrating a crunch. You may have enjoyed it but there's a high likelyhood someone else went along just to not "rock the boat". I don't care who you are putting in more hours puts stress on other parts of your life.

I just don't understand how that industry has their head up their ass from top to bottom for such a long time. There's some interesting problems in that space but after my 6 years I'd never go back.

Shame for celebrating a crunch? I'm talking 50 hours a week instead of 40 for a period of a few weeks. It is possible for a company and project to need that extra work and not turn into a sweatshop afterwards. I am 100% against the crunch that is par for the course in so many studios, but I am also disappointed by an attitude that work is always just work and god forbid it ever intrude on life outside.

I worked for five years at early and late stage startups before transitioning to games, and I cannot recall there ever being a time when extra work was not needed to get over the finish line.

If a union ends up being the only way chronic overwork can be addressed in the industry at large, so be it. Perhaps the real solution though is a legislative one, as suggested below.

Did you compensate your people for the extra work? Because otherwise it had certainly best not be intruding on life outside of work.

> I worked for five years at early and late stage startups before transitioning to games, and I cannot recall there ever being a time when extra work was not needed to get over the finish line.

I worked at such company too. It had zero to do with need for overtime and and a lot to do with people wanting to work this way. E.g. unwillingness to prioritize, unwillingness to negotiate, wish to be seen like the one who stays late and thus finding work to stay late when not needed. Overtime is not seen as failure of organization, so the organization does not learn how to do it. People staying late are seen as heros and people managing projects so that overtime is not needed are not rewarded, so latter leave and former create culture.

To large extend, people who stay in such companies don't believe it is possible to make deadline without overtime, so they are not even trying.

It's trivial to never have overtime if you either 1) have no deadlines or 2) have a project so unoriginal you can plan its schedule to the hour.

Show me any project with dynamic requirements that has some sort of deadline, be it time or money running out, and I guarantee there has been some extra work at some point.

I am not saying it is trivial. I am saying it is not rewarded. It takes skill and willingness and some companies are not motivated to do it. Nevertheless, it is quite possible to not cruck when you decide you don't do that, period.

Prioritization, negotiation and saying no. Estimates large enough that they have buffers. You don't need on hour predictability. It is precisely when you have dynamic requirements when you are supposed to use tools like that.

After crunch, there are typically many bugs and convoluted code. It just adds to overall time in long term. Pretty much all studies found crunch to not be effective. It is not about achieving more.

> but I am also disappointed by an attitude that work is always just work and god forbid it ever intrude on life outside.

Fuck yeah it's "just work".

I want my employees and coworkers to be happy, healthy and well actualized so that when they're at work they're focused and giving it 100%. Plenty of hard business data there shows that when you make sure someone's time outside of work is well respected they'll perform much better on the job.

Burnout and turnover can have a brutal impact on an company/organization's ability to execute, the list goes on. You're also self-selecting for a workforce that has the flexibility, which means your viewpoint is a lot less diverse.

This is the same crunch bullshit I got fed before in that industry.

I've got a kid, wife, friends, and hobbies that I place very high value on. Can I do a few weeks of extra hours? No problem.

Yup, but that's a place of privilege that lets you do that.

What if you were a single dad?

What if you had an elder family member to take care of?

What if you were living paycheck to paycheck and had a second job to be able to make rent?

I've found some of the best teams I was on had a diverse set of people who bring their unique perspectives into the fold. Otherwise you could end up with a racist soap dispenser[1].

[1] https://www.youtube.com/watch?v=8PlUf30rvyA

You can imagine me to be as privileged as you wish if it helps your argument.

> What if you were a single dad? > What if you had an elder family member to take care of?

As I mentioned above, family takes priority. If your team is so immature that they feel resentful when another member can't work those two extra hours a day because their Dad is sick, you've got another set of problems.

> What if you were living paycheck to paycheck and had a second job to be able to make rent?

This is a bit of a stretch. We are limiting the scope of our discussion to workers in the game industry. If you need two jobs to make rent in this scenario, find a way to cut expenses.

Yeah, but how many people do you work with that have those types of commitments? How many people didn't participate in that crunch at all?

I'll hazard that the answer is low, because the industry self-selects for people who are willing to take abuse for 'prestige'. They take pride in it, just the same way you did.

> This is a bit of a stretch. We are limiting the scope of our discussion to workers in the game industry.

Yes, yes we are. My first job in the industry was $35k/hr as a dev and I've seen worse salaries for design or art.

Let me pose a question to you. How would you live in Seattle, on $35k/yr in a way that one accident/major expense wouldn't put you in a situation where you're living paycheck to paycheck? Keep in mind an average 1br apartment is ~$1800/mo[1] so you're going to be well over 50% income : housing ratio.

For that reason you have people who work second jobs or have other commitments to be able to make their rent.

[1] https://www.rentcafe.com/average-rent-market-trends/us/wa/se...

Let me reiterate, I am not advocating crunch. I am saying that it is a reasonable expectation for a high performing salaried position that a few weeks of extra hours over the course of a year is completely reasonable.

And $35k for a dev job? When was this, and if it was anytime even remotely recent why on earth did you accept? A new QA hire at my last job would've made more than that.

You don't live in Seattle proper if you are making $35k. Live in a suburb on the line. Don't have a car payment if you can avoid it. And to pull that income : housing ratio down get roommates. I have done this.

And frankly this is a pretty silly argument to make in a discussion around why game developers should unionize.

Shame for you telling other people how they should feel.

Family obligations are quite clearly not taking priority when you are crunching. By necessity, the partner is doing all household and kids related work at that time. Obviously missed deadlines happen outside game development too and the partner is likely to have period where she/he has more work to do too and the other one takes it.

But crunch is normally defined as sustained overtime - articles I seen it require it to be over 6 weeks. So it is quite a lot of time.

Yea exactly. The difference between crunch and extra work to hit a deadline is entirely one of frequency and duration. My wife couldn't work in a more different industry than I do, but she is also subject to the occasional bout of extra work. We support each other and it's no issue.

If I wanted a predictable job that always has upfront hours, I suppose I could go back to making hourly minimum wage with zero benefits.

>If there were a unionization effort that focused on 1) a hard 40-50 hour cap on work hours (or with paid overtime) and 2) enforced profit sharing plans, I could support it.

>The current main unionization effort basically says "things will be better for everyone" on their FAQ. Total bullshit.

What is your familiarity with how a union works? An industry-wide unionization effort cannot effectively state what the future will hold.

When employees decide to try to organize a union within their company, they decide what the issues are. Their task is to convince their coworkers to join the union, so they will be focusing on the pain points unique to their own employment situation. Maybe in one company, it's crunch time, while in another, it's keeping a more consistent labor force despite project sizes changing wildly over time.

The new union will then negotiate a contract on behalf of its members. Each new union will be focusing on different things.

An industry-wide organizational effort that focuses on work hours and profit sharing would not catch the interest employees at a shop with good hours, good pay, but zero job protections and inadequate health coverage (for example).

Well my brother worked construction in the northeast where it is unionized, and it was an absolute nightmare for a variety of reasons. He was unable to work without a union card, getting that union card was very difficult without knowing someone because union jobs are highly coveted and thus kept artificially scarce, and once a member of the union he was forced to work with many lazy workers hiding behind the job protections afforded by the union.

What's to stop the same thing repeating itself in a game developers union?

I also worked for a construction union in the northeast, and our employer sent the worst employee's pay check to the job site via same-day courier to ensure that, once duly compensated, he could be immediately dismissed. The rest of the crew was thrilled to see him go. Construction is dangerous work, and none of us wanted him around.

At the same job, another employee, young apprentice, was found napping. He popped awake and was apologetic. Partied too hard the night before. The foreman sent him on a coffee run to get his head together. He wasn't fired, presumably because the foreman never reported him, and (at least for the rest of that job) it never happened again. Almost 20 years later, he still probably gets shit for it.

Bad employees can coast by in any job, especially hourly ones where the upper management is rarely onsite with the workforce. First dude got fired in short order despite the job protections. Second dude didn't, but (to my knowledge) not because of any union regs.

Anyway, I digress. There is, of course, nothing preventing some union horror story from taking place. There is also nothing preventing even worse labor exploitation from taking place (and we're already at 80+ hour weeks in some cases).

The biggest difference, though, is the labor market itself. The laborers in construction unions are temporary hires for a specific job. At the same time, most construction unions are craft unions with apprenticeships, where each new employee represents a risk/investment to training that person for many years. Nepotism abounds, because if you train a new hire for, say, two years, then they wash out of the program or leave to study engineering, they could have spent that time and energy training someone else.

By contrast, I would expect most gaming unions to look more like a public employees' union. It would represent permanent employees of a company, likely salaried. Not much would change in the hiring process outside contact negotiation. You get the job? Congrats, you can join the union.

The same thing that stops it from happening in Hollywood. Even the best actors and directors choose to join SAG and the DGA even though they don't need to.

Some? Hah.

In my ~6 years in the industry I don't know of a single coworker in the entire Seattle area who didn't run into a project that underwent serious crunch. Sample size of ~400 people or so.

I've also got longtime friends who broke in separately from me(and worked on some high profile stuff) in the SF/Bay area that say the exact same thing.

Pedantic note: In the 19th century textile mills and factories in general asked for 14-16 hours a day, 6 days a week [0]. In the UK, in 1831, for minors under the age of 18, daily hours were LIMITED to 12 hours a day[1].

[0] http://www.striking-women.org/module/workplace-issues-past-a...

[1] https://www.parliament.uk/about/living-heritage/transforming...

Haha fair enough. The gaming industry is at the level of textile mill exploitation of workers.

To spell it out: 14 to 16 hours/day x 6 days/week = 84 to 96 hours a week.

Textile mills used to and still do in some low cost locales


> If employees don't like it, they're free to leave.

They're also free to bargain collectively. Why should 'freedom' mean capital consolidates and organizes as much as it likes, while labour is reduced to isolated workers making individual choices?

Its 80 years of propaganda that makes people think like this. It is to be expected, since many of these beliefs are from emotion, and not from cold business logic.

The cold business logic says to consolidate and make partnerships/business unions.

Zarath 88 days ago [flagged]

Shh, stop ruining the capitalist narrative.

In many cases, that also means

> I don't like doing this, so nobody else is allowed to do this for you either.

Unions very often prevent non-union labor and, as such, enforce their own desires on others. Don't get me wrong, sometimes that's necessary. But there are negatives involved in unions, just like there are positives.

Ok? And what about game companies who prevent people who want work-life balance from being game developers? 2 sides of the coin.

Positives and negatives. 2 sides to the coin. The tone of your post makes it sound like you disagree with me. The argument of your post agrees with me.

Programmers are not the only ones being mistreated by gaming companies. QA testers, graphic designers, etc... cannot simply "change jobs".

It's also not as simple as "you're free to leave". Most people working in the games industry are young. Many of them do live paycheck to paycheck. In many cases they may have also moved across the country for the job. Quitting a job is in and of itself a substantial risk for many people.

There is a reason we have worker protection laws. That reason is that companies can and will abuse workers without them. This has been proven time and time again.

>Most people working in the games industry are young. Many of them do live paycheck to paycheck. In many cases they may have also moved across the country for the job. Quitting a job is in and of itself a substantial risk for many people.

They knew this industry was like this before they accepted the job. Why do we need to protect people from their own poor decisions? These aren't people who lacked opportunity and got stuck in a bad place in life and need a helping hand, these are highly privileged people who willingly signed up for mistreatment.

> These people are only being mistreated because they allow it.

Yep. And they are allowing it by not unionizing.

Yep, that's a good point. How many years have they had now to unionize? The video game industry isn't exactly new.

Skills in game development do not necessarily transfer to skills in software development

Excuse me? Video games are software by any definition I'm aware of and highly complex software at that. More than most, software engineers on AAA video games need to have an understanding of the full stack including the underlying hardware to be effective contributors.

What is the skill gap that a software engineer working in games would have moving elsewhere? The only one I think of would be domain experience in TDD and unit testing. Neither is widely practiced in game development, but they aren't universal for software development more generally either.

This is missing the fact that the vast majority of game developers are not "software engineers" While what you say is true for a large amount of software engineers in games. The vast majority of game developers are in Art, Design, Audio, QA, etc, where this does not apply.

I can't speak for all game development degrees, but at least for the one my university offers they don't have much focus on strong CS fundamentals, and offer more of a high level overview of programming, writing, animation, etc, especially focusing on combining these in existing game engines. People who graduate with these kinds of degrees would be employable as game developers, but wouldn't be strong software engineers. They could transition I'm sure, but it would take time to learn what they missed in university, and given the work hours expected of them in game development it seems free time for learning would be at a premium.

> As an employee, I want to be free to negotiate my own wages based on my merit.

Two things wrong with this.

1) Professional unions, such as Hollywood and sports unions, do not establish rigid payscales like non-professional unions. They set a salary floor and some basic guidelines for non-salary pay (e.g. residuals are typically mandated). If game developers unionize, the union will likely be a professional union.

2) Salary negotiation is intensely problematic, and many companies are doing away with it, union or no. For example, Reddit a few years ago eliminated negotiation in favor of delivering their best offer from the start. Salary negotiation has sexist biases and rewards people for skills that A) have nothing to do with the skills that are required for the job and B) not everyone qualified for the job possesses. Salary negotiation is about enabling employers to pay employees lower salaries; it benefits the employer entirely and not the employee. And that goes back to pnathan's comment: the whole point of unions is to act in the interest of the employees against the employer, so the whole "as an employer" section is ultimately irrelevant.

> I do not want my future in any way controlled by yet another external entity that I have little control over and promises nebulous results.

You mean like.... an employer?

> As an employer


Emphasis on the "As an employer":

> I run a very small indie game studio in Boulder, CO.

I've been an employee for 80% of my game development career.

Doesn't change the current "conflict-of-interests" situation that I wanted to highlight.

Hmm, if being in management makes him conflicted from making anti-union statements, does being an employee also conflict someone from making pro-uniom statements?

High-performing, in-demand, highly capable workers have always felt like this. Unions don't benefit you, the way mandatory healthcare doesn't benefit young healthy people.

Why would any of those things happen when none of the potential union members want them either? A union is democratic after all and does not automatically have legal agreements with the company or requirements from other employees.

No one want's to protect under-performers for whom they'd have to pick up the slack. And no one wants their compensation set arbitrarily. Especially since in software anyone can jump ship and find performance based compensation somewhere else.

They just want to work the hours specified in the employment contract rather than being singled out in isolation from each other and required to work far beyond the amount contractually agreed on under threat of termination.

So you'd just have a group that discuses abusive uncompensated overtime and request changes in a united way. Because a single person complaining or quitting over mistreatment has zero impact.

What is your role in your organization? Some roles fit better for union support than others, imho.

Predominantly engineer, but I've been a first employee at a hugely successful game startup and now run my own very small indie game company. I'm intimately familiar with the pros and cons of being on both ends of the equation.

Is it truly merit, or simply what the market will bear?


Not to be pedantic, but this headline is not accurate. There’s nothing that makes me confident that respondents to the game dev conference survey are representative enough to say “nearly half of game developers.”

This is a really important point as trying to represent a bunch of people’s opinion based on a single, limited survey is not productive.

I would think that conference attendees would be fairly representative of commercial game developers.

I don't know enough about statistics to say how big the error bars would be here, but I'm constantly astonished by how well statistical methods can work with even seemingly small samples.

The issue isn't with sample size. It is with sample method.

The population of gamedevs who go to a conference is probably not a random selection of gamedevs in general.

I would expect smaller game companies to be less well represented.

Its a sample size of 4000 and electoral polls are smaller than that 1k or 3k

sample size is irrelevant if the selection criteria are poor.

If the sample size is large enough the selection criteria doesn't matter.

Take a sample of 5000 random people (an enormous sample size) attending the Republican National Convention and ask them which party they will vote for.

I guess that's technically true, but only if your sample size is 100%. :-)

There were examples of very large polls that failed spectacularly because of the badly chosen samples


The issue isn't with sample size.

The sample is almost certainly biased, we just don't necessarily know how much and in what direction.

> The issue isn't with sample size.

But considering how many developers don’t live where the conference takes place and/or can’t get there I have to agree with the top here and say this survey is severely limited.

it would indicate the better paid ones who can afford to g to industry conferences

Well, you can do a lot with a small sample provided that it's a random sample.

It seems like you'd have to know how conference attendees differ from the population of game developers you're interested in.

Statistically speaking 4000 is a good enough number. I mean, election polls in the US usually use less than that for a population of 300 million and they're usually close to the real number.

Except in 2016. And that 4000 is a biased population of only game devs that attended a certain conference. Maybe if it was a totally random sample of 4000 of all game devs it would be statistically significant.

>Except in 2016

In 2016 the national polls were very close to the final result--within a couple points. Remember that Clinton won the popular vote.

State polls were less accurate, but the general consensus is they were wrong due to undecided voters breaking more for Trump than is normal in the final few days--after the final state polls were conducted.

That being said, your point stands. Sample method is generally more important than sample size, and 4,000 game devs who self selected to attend a conference is very unlikely to be a representative sample.

When they do these election polls, do they stand in the street randomly choosing people coming out of the conference centre where one of the major parties is having their AGM?

>> Statistically speaking 4000 is a good enough number.

It seems to me the GP comment meant that because it was a survey among people who attended GDC it's not representative. If that is why they questioned it, I'd say it's an even more relevant survey, as those are the people who care more about the industry and have more influence. No mater how you slice it, I'd say it's an indicator not to be ignored.

It’s an indicator, but just unlikely to be an indicator for the population of game developers.

They are taking a very good care to select the right 4000 people.

Unions would very likely increase game prices. I will happily pay more for games in exchange for better conditions for game developers.

Game prices have already increased via less consumer friendly practices. Predatory monetization schemes like lootboxes (i.e. gambling), DLC (mixed bag), cosmetics, always online DRM, company specific storefronts, digital sales, "early access" and releasing unfinished games have all reduced costs and increased profits while sticker prices remain static. Sticker prices have even gotten better sometimes since many games go on deep discounts quickly.

Looking at the bigger picture, as a whole, adjusted for inflation, looks like games have actually gotten a bit cheaper. An NES game was $30-$40 in 1988[1], that's $65-$85 in 2018 dollars.

You also have a whole category of "basic" games (like phone games or indy games) that are nominally (unadjusted) cheaper than a retail NES game for the full version. I'm not an expert on this category by any means but I bought a few games from the Switch store that are very comparable to an NES game for $5-20. That's only a max of $9 in 1988 dollars!!

I'd agree that "consumer [un]friendly practices" are still a problem though. (could be a case of a few "whales" making large purchases on stuff like loot boxes subsidizing everyone else)

[1] http://videogamecomicads.blogspot.com/2012/12/sears-nintendo...

DISCLAIMER: I only play single player console games.

Don’t buy the industry bullshit. They’re making record profits through gambling and micro transactions.



Perhaps you're replying to the wrong comment? I certainly didn't say, nor imply, they weren't. Nor am I "buy[ing] the industry bullshit" (whatever that means), I'm speaking as an end consumer.

Have to adjust for amount of buyers too though, and the cost of different kinds of games.

When someone says "X has gotten cheaper" they mean when you walk into a store and buy X you pay less than you did in the past, either in real (adjusted for inflation) or nominal (unadjusted for inflation) dollars.

They've also decreased, a $60 game in 2006 should cost ~$75 today. Prices haven't kept up with inflation.

Most games have $80-$120 editions that include all of the actual content. $60 is usually the bare bones experience. That is on top of monetization in game.

But prices don't exist in a vacuum either. The gaming market has exploded and is vastly larger than it was back in 2006.

Steam found games were typically much more profitable when heavily discounted than when full price in general.

This is not typical of "most" games. Some games, usually AAA titles from major publishers do this, but the content is usually unimportant to the game as a whole.

The discussion is about unionization. I'd assume that means we are primarily talking about AAA here. But I could of been more explicit.

How much of content offered is sort of irrelevant. A better metric would be how many gamers buy these things. I'd guess quite a few since they keep going all in on the model (it's getting worse).

I always got the impression it was to give gaming journalists something to write about. I never met anybody actually buying these special editions.

Except now they have an additional "season pass" for nearly the same cost of the game for what used to be included for free.

This is wrong because you assume that content from the season pass is already ready when the game releases. Season pass is a different term for expansions that you used to pay in the past.

I've seen that on Reddit as well, you think companies have a year worth of content and holding onto it ... The content is just not ready / created.

What? No where did I suggest that content was ready to go - I'm simply stating that a finished game now has less quality content for the same price than it used to.

Again do you have proof for that? Because I'm pretty sure game have more content nowdays than they had 20 years ago,

I really can't agree with that. A small minority of games are like this yes, Battlefront 2 being the shining example of bad monetary practices. But something like RDR2 or just about any RPG is packed with content.

How so? BF2 has a ton of content for $60 and you don't have to pay for it.

I won't argue if you thought it was a lot.

I can't think of games in 2006 that offered additional, free content. They sold them individually as map packs or weird, one-off add-ons (Horse Armor).

Horse Armor was basically the start of the bullshit, I'm not referring to additional free content, - but to match older games in terms of quality play time you often need that season pass.

Play time is a terrible measure for cost of creating a game. Yes, maybe the average playtime of old NES games was shorter, but the amount of artists involved was minimal and they didn't do it by having more content, they did it by stealing coin-sucking GAME OVERs from arcades.

I care about quality play time over just about any other metric. How long is a game actually enjoyable - maximizing that seems like it should be the clear goal, no?

I can't think of a game that is shorter than its predecessors and requires a season pass purchase, or really any 'essential' DLC for a single player game at all. With regard to multiplayer games, they don't have a cap on playtime and traditionally sold expansions (for example Battlefield 1942 Road to Rome).

Multiplayer games are often particularly brutal about this though - if you want to be able to play with the largest population you need to buy the latest content expansion - this didn't used to be the case, compare Call of Duty 4 with the newer Black Ops series for example.

Example of a $60 season pass?

Battlefield games and call of duty games come to mind. You can buy the DLC a la carte when it’s released or spend on a season pass to get all DLC when it’s released.

Battlefield games no longer have a season pass.

Fallout 4 on release had a $60 season pass - last one I ever paid for. But many games are still doing it, I see black ops 4 has one of similar price, as does Civ 6.

Fallout 4 had a $49.99 season pass which was an increase from the previous $29.99[0].

Black ops 4 had one for $40[1]. Can't seem to find historical info about civ6, but I know a "large expansion" was not included in the season pass.

[0]: https://ew.com/article/2016/02/17/fallout-4-season-pass-pric...

[1]: https://www.polygon.com/2018/6/13/17447600/call-of-duty-blac...

Maybe it was after release but I seem to remember the prices matched of the title and the season pass, I'm not in the states though so perhaps an international pricing difference, you get the point though, $50 is still nuts. I've edited the parent appropriately. I suppose the ~$100 range is more correct than the $120 range for a full game these days.

Every single game does not have this.

I would be incredibly supportive of a game developer union in which developers can push back against being made to practice these unethical behaviors.

Unions traditionally are there to defend the interests of its members, not to address consumer related unethical behaviors.

> Unions traditionally are there to defend the interests of its members, not to address consumer related unethical behaviors.

Many unions and guilds have codes of professional ethics. A quick google finds this example:


I guess I was thinking more like a guild, then? I licensing boards? "Sorry I can't shove random RNG-for-money, boss, I'll lose my membership/license".

These union-skeptical comments like the GP rely upon the assumption that unions a hypothetical tech union would look identical to failed unions of the past, and are incapable of doing anything different.

Example of a modern non-American union where workers exert control to better the company over management decisions:


I've really appreciated Skill Up's commentary on this, particularly in the context of his AC: Odyssey and COD: Black Ops 4 reviews:



Both are thoughtfully argued and well worth watching.

> Game prices have already increased via less consumer friendly practices. Predatory monetization schemes like lootboxes

I suspect loot boxes decrease the price of a game to the median player. I suspect that in many cases, they've reduced the modal and often even median price to $0.

The probably also increase the mean price, though.

Labor cost does not determine a product's sale price. That is a disastrous myth. Demand changes selling price. If a company cannot price to sell to their demand because of capital or labor costs, the free market has stated that the business is not viable.

You are correct, but I would caveat that labor cost is just one of many factors in determining the minimum price a product can be profitably sold at.

For any particular company or even country to some extent (since the games industry is quite globalized), this is true. However, if there were widespread unionization among many of the biggest studios then that could conceivably lead to a market-wide price hike, which would be largely absorbed by consumers due to a lack of suitable substitutes. Yes, the supply/demand equilibrium point would shift and demand would likely drop somewhat, but I think it's quite likely that the market demand for top-tier games isn't perfectly elastic.

The optimal selling price for a game does to change becase developers add a fixed cost.

Suppose you have 1 Billion in fixed costs and 1$ in per unit cost and are given two options. The equation is just (number sold) * (unit price - unit costs) and that 1 Billion dollar fixed cost is irrelevant when maximizing profit (or minimizing loss).

What it would do is reduce the number of games created, as the industry becomes slightly less profitable. Which would then spread consumer spending over fewer titles and thus recuperate the burden of a union.

This is the current problem with the game market. There was a race to the bottom in game prices over the course of the last 8 years that's caused many titles to barely break even. Throw in an over-saturation of developers who are willing to price their products for less and you have an industry that is sustained through people chasing their dreams while eating ramen for dinner.

Labor / Employee costs is a lot of the times the most significant costs to a company. It's not the sole determination but it's definitely one of the top 3 factors Labor (forever and growing) > Infrastructure (large but incremental) and R&D (medium and ongoing)

Right, and if the costs of running your business are greater than the revenue you can generate from market demand, then you are not running a viable business.

Depends on industry film has a high labour cost because you pay actors a lot - I suspect the % of film labour costs due to Bectu/ ISATSU members is not that big.

It really doesn't matter what the market bears because software is so malleable that companies spend dozens of hours creating loot box systems, marketplace systems, data driven pricing models etc. to squeeze every last dollar out of the customer, regardless of whether the demand is high for these things or not. It doesn't matter if their product is half broken or is barely a product at all, as long as they can create 50,000 skins. Then again, I'm not sure what demand for these systems are, but considering how universally detested they are, I would assume not much.

> Labor cost does not determine a product's sale price. That is a disastrous myth

It's not a myth. It just depends on whether the pricing is demand driven or supply driven. Commodity products absolutely have their prices set by costs (labor among them).

Maybe. It’s also possible that being less able to offload the costs of bad project management onto your workers will just make them pursue more realistic project plans.

I see lots of AAA games loaded down with unnecessary cruft that adds little value to the overall experience and it makes me sad to think a bunch of poor devs probably had to grind through crunch to add it.

I expect it’ll be a little from both columns.

Genuinely curious as to what you consider "unnecessary cruft."

If you look at most AAA games from a structural perspective, they are generally composed of many systems and modes of play. Compare to indie games, which typically have a much smaller number of different systems and modes.

Think of a first-person shooter. You can add the ability to upgrade weapons and armor (a system) and add a fishing minigame to provide resources for upgrades (a new mode of play).

Specifically, look at Just Cause 4. Your basic systems core to the gameplay are shooting things, the grappling hook, the parachute, the wingsuit, and vehicles. The unnecessary cruft is stuff like your ability to upgrade the grappling hook, the deployment of squads to capture territory, and supply drops.

My impression is that AAA studios aren't stupid... they know that these extra systems and modes don't contribute much to the game. But they do contribute to customer's evaluation of whether the game is worth spending $60, and this contribute's to the bottom line. This is the same reason why manufacturers pack needless features into a product and skimp on core functionality. People are more likely to shop on the basis of feature lists than on the quality of core functionality.

This is why I'm skeptical... the unnecessary cruft is there because it drives purchasing decisions, so it's not going away unless you change the way people purchase games.

Others have mentioned RDR2 and the upgrade systems in Just Cause. They're both right and symptomatic of the larger problems.

I think Ubisoft games are the worst at this. Tons of experimental or different game elements that don't contribute to the core game. You can play chess or other board games in the middle of Assassin's Creed III. It does nothing. They had this whole city-building subgame that added nothing. They had an inchoate pirate game in there that was actually engaging enough that they made an entire sequel centered on it, but it added nothing to the actual ACIII game itself.

I can see evidence of all sort of "wouldn't it be cool if. . ." features that really don't need to be there and I can't help but think that if PMs actually had to think through what they're trying to do with their game they wouldn't do this as much. Bioware actually admitted to this after Mass Effect 3, claiming that they had never managed with so many resources at hand before and they couldn't really resist the impulse to bite off more than they can chew. This led to them adding in a bunch of extraneous side-quests and stuff that they didn't really get a chance to put enough time in to do it justice or make it worth anyone's while.

Well, RDR2 for sure. So much in that game is there to make the game world realistic and beautiful (and it succeeds greatly!), but not a lot of attention to lasting gameplay.

I'll happily pay the lowest price they'll sell the game to me for, and routinely wait for years after a game's initial release to get that low price. Nothing stops you from buying multiple copies of a game you like and want to support. Or going to https://www.mobygames.com/ and looking up the Credits for a game you like to find the devs to contribute to directly.

How do you contribute to a dev directly? Also, how do you decide who to contribute to?

When it comes to AAA games (which is who is really exploiting their employees rather than indies) it sounds like you're picking someone at random, (e.g. "Today's your lucky day Shading Engineer #4, please accept this venmo.")

We pay so little for games these days. Even the big titles still sell at $60. They inflate the price through DLC and other add-ons, but it does seem like something like a RDR2 should cost at least $80-100 these days.

The unfortunate side-effect is that smaller studios often charge much less for their titles than what they should in an effort to please customers who are conditioned to pay very little, and then can't make up for price in volume.

> We pay so little for games these days.

It's true. A lot of people complain about $60, but that's significantly cheaper than what NES games cost back in their heyday (adjusted for inflation).

And just look at not just team size and effort (probably 50-100x comparing modern AAA games to NES titles), but amount of content: most NES games used difficulty to pad out their length greatly; a gamer who's reasonably skilled at, say, Contra, can beat it in like 30-60 minutes. In comparison, a game these days with a 6 hour campaign is thought of as pretty short, and usually they have other content other than campaign: challenge modes, multiplayer, minigames, etc.

Its a weird industry. If you're there for appeasing people who enjoy art, story and gameplay, there's now a market for that. And those people don't mind either paying a little more for the title nor a 2-3 hour long game, as long as the experience is great.

But if you're there to chase teenagers for their parent's money, or even just the lowest common denominator of gamer, you'll probably have a bad time. Those customers want very long campaigns, AAA graphics and as low of a price point as possible.

This is pure speculation, but I'd imagine they would increase game quality too. Many AAA games are heavily marketed but lacking in depth and features; it would make sense that if game designers had more power in the organizations they work for, they might be able to push back against unrealistic publisher-driven deadlines which lessen game quality.

Game prices are going to increase anyway.

Last year, EA made over a billion dollars in profits. If they aren't using that money to improve the working conditions of developers, what makes you think the solution is to give them more money?

When you look at the prices of the games themselves, they're fairly cheap today. For the last couple console generations, the cost of a new AAA game has been consistently around $60. A NES game 30 years ago could cost $50, but when you take into account inflation that's about $100 in today's money. Companies like EA seem to make up for that by selling more copies of games, having deluxe editions, and jamming in microtransactions.

Only the most bare-bones version of a AAA quality title will cost you $60; the absolute minimum of content which could still be called a game. To get access to all of the content (frequently including story content), you're looking at a $100 starting point ($70-$90 for the "deluxe" edition, $30 for a "season pass"), going ever-upwards with the more in-game content and physical ephemera you want.

And that doesn't include any in-game transactions.

>Only the most bare-bones version of a AAA quality title will cost you $60; the absolute minimum of content which could still be called a game

That's fairly disingenuous - the last game I bought at full price was Spiderman. A quick Google says the the average user on the default difficulty clears the story in about 20 hours.

Sper metroid came out in 1994 (25 years ago), and the same quick search says that most people will clear it in 8 hours or so.

Single player games today are also regularly putting out free balance patches (nioh), large free content updates(Gran Turismo) and smaller chunks of paid content that rival the size of other games for very reasonable amounts (Witcher 3).

The production value behind those "bare bones" titles is still magnitudes larger than what you had back then, while the "play time" is at least comparable.

Statistics show that most people never finish the games anyway (I certainly don't) so it's only fair that those people who must really play everything pay extra.

Developers will probably just move development to Asia or Eastern Europe. At the end of the day, unionized devs will be at a substantial competitive disadvantage against ones that are willing to put in more work.

We all say we'll happily pay more for the same product. But those words often aren't backed up by actions.

"Unionized" is not the same as "not willing to put in more work". It's more about "not willing to put up with more bullshit".

Developers who actually sleep and spend weekends with their families produce a better product than those who just accept dead marches. If managers are aware that they can not get away with those shenanigans it's possible that we'll actually get better games.

> "Unionized" is not the same as "not willing to put in more work". It's more about "not willing to put up with more bullshit".

The point is, though, many people are willing to put up with that "bullshit" (which, judging by what you say later, mostly revolves around working hours contrary to what you write in this first paragraph). As much as we try to convince ourselves that crunch time and long hours don't work, plenty of studios demonstrate that is does. RDR2 is probably one of the best games of 2018 and it relied on crunch time. At the end of the day, a studio that can't demand crunch time is going to be at a disadvantage compared to those that can.

In an ideal world we'd all have well paying jobs in fields we love that offer good work life balance. But we don't live in the ideal world, we live in the real one. And many game developers are willing to make the sacrifice of long working hours.

It's not work if you are just looking at your monitor like a zombie. That's just heating the chair with your butt. So yes, bullshit. If a manager doesn't realize that, he should be fired.

You didn't address the more important point: Managers can get away with it by moving production to cheaper places. It's already happening anyway.

What a lot of people don't realize is that unions can also cause wage suppression when they decide that more members working is better than fewer members working at higher pay.

Managers can also get fired when they plan unrealistic goals or keep changing signposts instead of forcing the devs to solve their problems. Unions can make that happen.

Game developers already heavily use outsourcers and freelancers, many of whom based in South America, Eastern Europe, and Southeast Asia. The biggest games will have hundreds of contributors from internal support studios in places like Romania in addition to external partners.

I expect this trend to continue along the lines sketched out by the film industry: smaller core teams with specialized support studios/groups brought onto projects as needed and heavy use of contractors to scale staffing at different points of the project. This works in Hollywood in large part due to the union protections afforded to that temporary workforce, from actors to camera operators to screen writers.

As it stands now, video game development looks a lot like the (film) VFX industry and I think both are in an untenable place for long term sustainability. Something needs to change and I think organizations like SAG are a model for that.

If developers could cut costs by outsourcing, they'd do it whether the workers unionized or not.

Many already are. Unionizing would likely accelerate that shift.

Also, if developers could simply increase prices, they would, regardless of what their labor costs were.

They are doing it. Outsourcing has a cost unto itself though, just like unionizing has a cost.

And the game quality would suffer.

General software development gets moved overseas and comes back all the time.

Software development is hard. Pushing it to a cheaper third party usually increases costs a lot in the long term due to the increase is development complexity and generally a decrease in worker skill.

A strong union can also prevent companies from moving some jobs overseas.

> And the game quality would suffer

The Witcher 3 was a pretty good game (Polish developers) - many would call that an understatement. The Uncharted series since 2 or 3 relied heavily on outsourced development (especially for art assets). Arma 3 was developed by Czech devs. These countries probably produce lower budget games on average, and don't have lots of existing infrastructure to leverage. But that makes what they have accomplished all the more impressive given what they were working with.

What makes you so confident in the class that developers in countries with lower costs of living are worse than ones in wealthy countries? Do you think that Polish or Bulgarian developers are on average worse than American ones, and it so do you have something to back up that claim?

There's also, I dunno, all of the Japanese games that people love. Chinese companies are also starting to increase and market overseas, and some of those games are high quality. South Korea too. Asia is a powerhouse for entertainment software, even if we might argue they have a long way to go for business/enterprise/SaaS software.

Japan has been a powerhouse for games since the 1980s. Have people forgotten about the NES?

Witcher 3 or Arma were not outsourced. They were developed in house (in Poland and Czech Republic respectively).

Moving development doesn't necessarily entail outsourcing. Publishers can contract foreign developers for the whole game. Blizzard is already doing this with Diablo Immoral, handing development off to NetEase

> General software development gets moved overseas and comes back all the time.

General software development also isn't unionized. As soon is it is, moving overseas becomes vastly more attractive.

> Software development is hard. Pushing it to a cheaper third party usually increases costs a lot in the long term due to the increase is development complexity and generally a decrease in worker skill.

What's actually harder about outsourcing is management and quality control. The developers in cheaper countries aren't somehow inherently worse.

Many publishers have studios around the world, working on entire games by themselves. Why keep funding an expensive studio in the US that's getting unionized? That better be high caliber, which most studios aren't.

> A strong union can also prevent companies from moving some jobs overseas.

Maybe in a big monolithic company, but that's not how game publishers operate.

I would guess that maybe just less games would be made.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact