Jason Schrier's Anthem piece for kotaku is done and out, it's a doozy


#21

It’s a hell of a piece. I really hope Bioware can find a way out of this mess and a way to make games without so much human misery resulting from it.

About the only senior Bioware/EA person mentioned in here that actually comes away looking good is Mark Darrah, purely for being willing to basically go “no, we need decisions and we need to make this game into a game.” If that’s indicative of where the studio’s willing to go on future projects, hopefully it can help avoid this happening in the future. If not… I don’t know how willing I’ll be to support Bioware’s future games.


#22

That article took me a LONG time to read so for anyone who wants the cliff notes, here’s the major points I took from it.

  1. Bioware now has an epidemic of “stress casualties,” in which people are so depressed and anxious from work they are put on doctor-mandated leave for weeks or months, and many of them do not return. While we hear about Big Names leaving Bioware, there are also many devs without that clout who are similarly burning out and leaving, in absolute droves.

  2. Like many EA studios, Bioware was forced into moving to the DICE-developed Frostbite engine, which excels at high fidelity graphics and big arenas, but fails on every other metric. EA didn’t want to pay out licensing fees for another engine. The problem is that as an in-house engine, it’s horribly documented, so any studio that has trouble with Frostbite is told to contact a central Frostbite team that can assist them. This is further messed up because requests for assistance are filtered based on who EA deems will make the most money. So big shooters and FIFA get priority while Bioware has to sit on their hands waiting for assistance or until they can just code around the glitch/bug/issue to make things work. It’s horrific. It’s just horrible.

  3. Bioware is now two studios, Edmonton and Austin. Edmonton took the lead on Anthem, but Austin is the studio that actually has live service experience, given their work with SWTOR. They saw many of the issues that Edmonton’s design decisions were putting into place, but could not get anyone to listen. Now, the game has been essentially handed off to them and they have at least some optimism on fixing it.

  4. Leadership changes. So many leadership changes and a massive, massive issue of the team needing things nailed down about mechanics and lore and the story, et al, only for the panel of people leading things just hemming and hawing and never handing down decisions.

  5. EA worried about meme-able crap like the bad bad faces in ME: Andromeda, so they shot extremely precise and high quality mo-capped NPC interactions. But doing so is very expensive and essentially locked them into the decisions they’d made, so further changed became impossible without throwing out a ton of expensive work.

The entire article is extremely worth reading but it is also very long.

My… not “favorite” bit but the most compelling thing about all this is I can feel Schrierer’s anxiousness and concern for Bioware in the writing. It’s completely palpable, how worried he is about the studio.

Excellent work.


#23

For anyone else in this thread who was unfamiliar with the “big bang” model of software development, I found a summary.

Unlike nearly all other popular [software development life cycle] models, the Big Bang model is unique in that it requires virtually no planning, organization, best practices, or typical procedures. Instead, the Big Bang model is fundamentally about simply starting the project right now, at this instant, with no formal development structure or organization. It is typical that very few if any team members, let alone the customer, are completely familiar with what the requirements are for the project, nor what is necessary to meet those goals, and thus every single aspect of the project is developed and implemented on the fly.

distant sound of eternal screaming


#24

As a project manager, this is the shit I have nightmares about.


#25

I can’t get past the line of ‘Bioware Magic’, that it will come together in the end like with…
Mass Effect? That was the death knell of that franchise, the thing that burned an entire audience, and made me stop playing Bioware games, was that nobody had any plan or vision in the end and so everything done before the three colours in a room was meaningless.


#26

In the mechanical/chemical process world, you know, the world where making mistakes actually matters, we call that kind of project management “Big Bang” because it will cause one - either an explosion that destroys your building, or a financial explosion that destroys your company. Or both!

I cannot SMH hard enough at the idea that that is called “project management” because it’s literally the opposite.

Some of my friends wonder why I have absolutely no faith in any of the Silicon Valley Tech companies that have started to branch out into making Actual Physical Stuff and this shit is part of why. Software developers have entirely the wrong mindset to do anything including, apparently, making software.


#27

I came in here to say essentially this, but I’m also kind of curious to see how this piece will be received compared to his Red Dead 2 story which basically told a similar tale but ended with a game that people liked.

Not that there isn’t an immense value to people who actually care about labor being able to get a look behind the scenes at the way these games are made, but it always kind of bums me out that a lot of people who would write the RDR2 article off as heresy at worst or just the cost of doing business at best will come away from this with little more of an opinion than “EA bad.”


#28

There’s a lot of stuff in here that kinda makes me wonder if there are some egos that need to get checked.

Don’t call it Destiny. Bioware magic. Project code names Dylan, Joplin, and Morrison. Dev criticism/feedback being dismissed by higher ups. Austin studio criticism/feedback being dismissed by a studio that seems to consider themselves the “A Team” (which is something we also heard about back around Andromeda, if memory serves.)


#29

Man fuck “Bioware Magic” that’s some bullshit like your boss saying all your coworkers are family.


#30

Your comment caused me to mull over the Frostbite sections a bit more, and I think this puts a gigantic neon sign out there for reporters like Schreier and others to go and find the Frostbite shared services team and hear their stories. If I had to guess, this is a group that EA considers to be a cost center, not an area for investment, which means there aren’t enough people working on updating the engine and helping the internal teams responsible for working with it. (This is wild speculation, and I may be very wrong, but it seems consistent with what we’ve seen come out of EA/Frostbite.)

I’d also wager that the prioritization for more profitable projects may not be as explicit as the article implies, and that there are likely more factors than that going into which tickets get solved. Anthem’s inconsistent direction and pre-production status likely didn’t do the team any favors when it was trying to get help from the Frostbite team. After all, that team is going to be judged on how well it supports the big games coming out in the near future, and it would make sense for it to prioritize the higher value ones for institutional survival.

“Sorry, we couldn’t get to it because this game that has been in development hell was eating up our time” isn’t a good look for that team.

This also speaks to a broader issue — because Frostbite is a proprietary engine, EA can’t hire people who have existing expertise with it, unless those folks have already worked there. Given their current labor practices burn people out, brain drain is an even bigger problem for them, compared to those development shops that work with third party engines. If the people who know how to use your system quit, you have to spend time getting other people up to speed.


#31

Speaking as a software developer, this is 100% true. The only reason stuff like this gets any traction, is that the state of software design and management is so very bad that things like this sound good. The other problem (which you can find numerous examples of in the article) is that even software people say “No, that can’t be done in less than a year”, you’ll find yourself with 3 months to do it anyway, because the consequences of failure (unless you are Boeing, nobody dies) are generally so weak, that forcing features into untenable timelines tends to work out at the business level, even if the quality is awful.


#32

“The game was codenamed Dylan after Bob Dylan”

Huh, that’s pretty stu-

“Project Joplin”

Project Morrison”

Oh fuck off

Edit: Also, Bioware Magic is the most toxic, blind, misplaced brand loyalty ass bullshit, I’ve heard in a hot minute. It speaks to a staggering level of hubris and absolutely naked disregard for your workforce that no matter what their working conditions are things will be fine because of the name on buiding.

Miss me with all of that shit, it is disgusting.


#33

“We can patch it later” is an approach that Silicon Valley thinks is true everywhere and there’s going to be a lot more imploded companies in the next decade or so when they discover that this isn’t the case.


#34

I wish we had a Japanese Jason Schrier who could tell me what happened in the ten years I waited for FFXV… I don’t even want more DLC for that, I just need to know how it ended up like it did. It’s painful, actually.

Every game good or bad needs post-mortems like this. Since game companies keep their secrets so tight we never get to know what happened or why. BioWare PR can complain about the reporting but they’re obviously not reflecting themselves since this same thing seems to keep happening to them.


#35

This tangentially apropos, but I just listened to a podcast episode about Elizabeth Holmes and Theranos and it touches on exactly this part. If you’re making an app and it doesn’t work, you change it until it does and it isn’t a Big deal. Now, Apply that philosophy to the medical field and suddenly the stakes are waaaaaaaay higher.

The patch it away solution model is steeped in hubris and callousness and it almost always ends up hurting people in the end.


#36

I suggest this video, but the answers contained are not surprising in the least.

tl;dw: Nomura is and has always been a terrible director and he kept changing things and juggling way too many projects during Versus XIII’s development. It eventually got to the point that in order for this hellbeast to finally get finished and for Nomura to wrap up other projects, they had a co-director take charge towards the end and get it ready to ship.


#37

While their is plenty of jaw-dropping bits in this article, it also feels like anyone who has read one of these post mortems recently could’ve written the bullet point version of this off the top of their heads.

Couple things -

  • The Frostbite parts make me wonder about Apex Legends. Did Respawn get grandfathered out of that requirement, or has EA started to recognize that it’s not the one size fits all solution they wanted it to be?

  • The culture of a project gone bad where everything is treated negatively is so recognizable. Long meetings with nothing decided. Long waits for changes that are outdated before they’re implemented. “We need someone with a vision.” “Here’s someone with a vision.” “No, sorry, not that one.” Sometimes it’s nice to realize that every workplace is as fucked up as your own workplace.


#38

Okay I’ve been thinking a bit more about not really this piece but its general purpose, because frankly I don’t really know what it’s for.

Sure the working conditions in the Bioware offices were awful, but the same goes for almost any other game studio out there. Where’s does this story fit in the larger context of working conditions in the games industry and efforts to help improve them? How does this help games workers?

I’m also wondering if we would’ve gotten the same piece, if Anthem’s Metacritic rating wasn’t 55, but 85 instead? Maybe we would’ve gotten a short paragraph in a review how hard it was for folks to make this thing, but considering that the loot is cool and all that, it was all kind of worth it maybe?

Having this split between “getting a short, meaningless paragraph in a review, when the game was good, but getting a long detailed report about what all went wrong, when it was bad”, inadvertently supports this shitty attitude that suffering for art is ultimatively worth it, if the results are good.

I don’t know, it feels kind of icky to now see dozens of people starting to armchair analyse the internal workings of a game studio, when they probably wouldn’t waste a single thought on them, if the game’s quest design wasn’t that boring.


#39

Respawn wasn’t acquired until December 2017, by which point Apex Legends would have been late into development. I’m sure EA’s mandate of Frostbite doesn’t include complete scrapping of half-finished games. Fallen Order is probably not going to be in Frostbite, either.


#40

Draws awareness to the problem and pressures management to improve (which several developers said was why they wanted this to go public). If they don’t, then the developers there know then action is absolutely necessary.