I was just exposed to the scale of outsourcing in AAA game development

There's a Reddit thread making the rounds at the moment suggesting it knows the reason for some of the flaws present in Starfield, the credits for the game are 45 minutes long and it's because 27 other studios were involved in development of the game.

I know outsourcing has been a part of AAA game development for a long time but i assumed i don't know... one or two other studios, maybe even five if you've got just an insane amount of art that needs to be created, but 27? Half the studios involved with Starfield belong to seemingly your one stop shop for video game development Keyword Studios, which owns 70+ other studios all around the world including... you know... places where labour is cheaper... Keyword itself does not require that its name be shown in the credits of your project so that it can stay, according to chief exec Andrew Day "under the radar", but it's not exactly staying hidden.

It seems like all AAA game development functions like this now, Elden Ring has over 30 other studios attached with what looks like the bulk of its 3D art being credited to a studio in India if i'm reading the credits right, i even checked out industry darling Bardur's Gate 3, if this is how the biggest studios work surely there's less of it happening at Larian? Well i don't know exactly how many of these are studios or just doing PR, marketing, localisation or playtesting, but there are 60 other companies that "contributed" to BG3.

I'm kind of blown away, i'm sure in most or all cases the primary studio is responsible for putting all the pieces together to shape their vision and it's probably not as big a deal as i'm making it out to be, maybe your studio ends up in the credits even if all you did was have a phone call, but the scale of the whole thing, the thousands of people involved, it's way more than i thought.

As one user on Reddit suggests "look at the credits for Fallout 3 it's like an indie game". So i did, guess how many external studios worked on Fallout 3? Two, one for the bulk of localisation and one for Russian.

I don't want to tie this in to the decline of quality in the AAA gaming space but... maybe it's a contributing factor? Is bug fixing and general coding being outsourced too? Code is famously a fragile and fickle thing what if you spend most of the games' development fixing things other studios from around the world have broken, what if the primary studio used to have 100 devs to put the pieces together and now it only has 60 because the rest are outsourced? That's ok... we've got patches now...

There's no question for you here it's just something i found out about, which is weird because i've been watching this industry since the mid 90's and i don't hear any journalists or industry insiders talking about it. I think about the recent closing of Volition and everyone finding someone to blame, but with scales like this and Volition probably being only one of the 20+ other studios involved in its projects it's better just to blame the entire system.

Is this news to you too, or have you always known?

i won't say i've always known- because that's clearly a lie.

however- i did get an inkling once i determined i have sufficient time to file my taxes while waiting for credit scrolls to conclude (that's not any shade against credits, though i'm not sure Bill who manages IT in Finance Unit Twelve is terribly relevant).

honestly some of the best insight i've gotten into the industry is the Play, Watch, Listen podcast. everyone's mileage is going to vary on the personalities- but one thing you can't deny is the breadth of experience between one of the world's most prolific voice actors, games journo-turned-presenter-turned-writer, indie dev darling and a name card composer.

Mike Bithell in particular is constantly dropping interesting pearls of game dev knowledge- and one of the things you understand quickly is open world games with a need for 4k assets more or less necessitates armies of artists and professionals to build the parts which are then in theory assembled at HQ.

i'm very much the opposite of an expert- but it's my layman understanding that scale and the outside work it requires- is not necessarily the aspect that brings down modern games (though i'm sure it can exacerbate issues). i hear it's frequently the internal processes and management at the 'home' studio that makes the difference (or so it would seem).

it's my understanding that the work done by the massive asset houses is pretty mercenary- if you need miles and miles of 4k city building textures, it seems pretty transactional.

but making sure your design doc is solid, having really robust testing, a publisher that is supportive, decent cash, and good project management- that reads like a wishlist more than a list of requirements, so you can see where things can start to come apart.

it's not my original thought, but i've heard modern AAA games dev described as building digital cathedrals- and as such they evoke all the mastery, splendor, and abject misery required to build the real thing.

I'm not surprised at all given how much is involved in game development and how many games have multiple big studios that used to make AAA games on their own working as "support studios" so that the games can come out. Beyond the basics like generating art assets there are a lot of very specific roles that you don't necessarily need throughout development (like sound mixing or VA recording or other things) that it makes sense to outsource now that these things have to be done well and can't just be some random studio employee recording Jen from accounts receivable at the publisher doing voice for an NPC like in the old days.

I don't think it is really the cause of games having bugs or other issues unless the tasks are split unwisely or leadership is bad. Code is fragile but game engines are much better these days and things like art assets are more plug and play than they once were, especially if using a common engine like Unity where the artists probably understand the specs required. I don't think paying someone in Bangalore to make you some really nice flooring textures for your buildings is going to compromise the game quality.

I do think that it's probably a massive logistics headache that makes hitting your dates harder, though, and that might contribute to quality issues and especially crunch.

I only played a couple hours of Starfield so far and it's failed to grab my interest, although I will keep going and see it it ramps up. But from what I can tell, the flaws of the game are more related to its core design and not any particular issues in development. Like, at some point very early on Todd Howard or others in the core leadership team had to make some decisions and it *seems* like those are the decisions that everyone is complaining about. Namely, the lack of free-form exploration and the somewhat-bland setting. Those issues would be there whether this was a tiny indie dev or the monster that is Bethesda. It doesn't look like a case of "great concept & design but poor execution" that I might expect would be the case if mediocrity was due to the decentralized development. That might be more the case in something like Cyberpunk, where the core idea is solid but the execution was crappy at launch due to bugs and performance issues. (at least from what I've heard. Haven't played it myself.)

@wollywoo: I would add to this that if we want to look at Xbox 360 era Bethesda as "the good old days before outsourcing led to all these buggy games" then, umm...we might want to check our memory banks on what those games were like. Gamebryo. Truly the worst.

Cyberpunk was, in my view, mostly a victim of its own hype. I played the game at launch on Xbox Series X and while it definitely had problems (bad driving, horrible implementation of police, some pretty nasty bugs though not nearly as many as we saw in videos) it was my third favorite game of the year.

The problem was that they promised a lot more choices and interactivity than they delivered. And the first mission actually plays out this way, with a lot of interesting choices and different approaches. Then the rest of the game does not. At all. That plus the shallowness of the open world that was promised to feel like a living breathing place led people to feel like they'd been lied to and sour on the game. Also the PS4/Xbox One versions were hot garbage running on the base versions of those consoles.

I always think about an interview with a game director years ago which i unfortunately can't trace, who said something along the lines of art not being the problem with games but the coding, testing and bug fixing, the artists can crank out tonnes of art but the rest of production is what takes so much time, also something i think Brad said in the Bombcast about a Final Fantasy artist who spent ages working on rocks because there wasn't anything else to do.

Those are just random memories of mine from 2 games but they always gave me the impression that internal art generation wasn't a problem, at least... 10+ years ago.

But I also don't know what the takeaway from this is actually meant to be because like... Fallout 3 wasn't some rock solid release either it was all sorts of fucked up at launch because Bethesda releases games with a ton of jank baked in. It is arguably one of the things they've always been known for. Hell people began to see it as an endearing quirk and largely excused the way it was broken a lot of the times because well if it was TOO bad then modders would fix it.

I think the larger issue with some peoples response to Starfield is that game dev moves considerably in the time it takes Bethesda to make these behemoths and this time the fact that they barely did anything at all to keep up has not been overlooked the way it might have been in the past.

@wollywoo: I played less of the game than you likely, and saw the quick look. Maybe im just not in the know of tons of space operas but i don't get the take online that the game is so bland and boring and uncreative or that the exploration isn't no mans sky like so it's bad. Does it just become unplayablely bad soon. with fallout 76 i instantly new that was baren and poorly thrown out with little passion before the fixes. This IMO feels better than fallout 3, but im not a big fall out person and people love that. People seem to like skyrim less each year and i loved it so i cant figure where ill stand though. But I actually think just a no mans sky but sim game isn't what a mainstream audience was looking for and there are plenty of those around. (Though they should have done maybe a planet to space thing where clouds or freezing over fog covers your screen for a load and got the same point across, and flying to a planet when near enough ask to beam and a video loads, i assume a mod will fix that.)

TLDR: I guess I'm not seeing all the issues online discourse lead me to believe, but again that could be due to being on that side of the fence GB's quick look was. Heck even jeff was positive on it and i expected him to eat into it for everything and say he played like 2 hours and is quitting. I heard it gets great 20 hours in from the ign review. Fair enough that people think that's too long, but it also seems this isn't supposed to be a game you just lay into in a weekend. Skyrim is a game you got lost in and you can't get that same experience obviously as you're space traveling, space for all its mystery is kinda dull, unless you happen upon stuff a lot but a lot of it is empty by design. I guess again I'm just also playing it via streaming on gamepass, so my issues aren't gonna be as evident, I'm not super sensitive to fps, and I'm super impressed by the physics and graphics environment wise (good on you outsourcers lol). Anyway don't see a issue with outsourcing at this point gamers can either let them do that or they do it all in-house and take 2 decades + per game to get it to the point it looks good for gamers who think this game is ps3 graphics in their mind (reddit takes). Our bar for a "good game" is going too high after all the great games this year, especially given I did a poll and a ton of people think this year of games isn't good at all/is average.

@topcyclist: Eh, to be clear, I'm not judging the game yet really. It could be something great. It's just that it's failed to grab me much so far. My comments were more based on reviews I've seen and general discourse.

@efesell: Regarding Fallout 3 my point was less about it being a finely polished gemstone and more that it only took the team within the studio to make it and not 30 other companies around the world, it had the "expected Bethesda jank" of the time but that didn't prevent it from getting 9/10's because the rest of the experience was so phenomenal. I'm not saying outsourced studios do bad work or don't care i have absolutely zero evidence of that, but i think there's something to be said for having a team in one building believing in and realising a cohesive vision, and i'll use classic and indie games as my evidence for that thought.

Shortly after i made this thread i saw a video about why modern games take so long to make, the video... didn't really answer the question i don't think, but it did bring up an interesting fact, Starfield took the same amount of time to make as Oblivion, Fallout 3 and Skyrim combined, and that's despite having 27 other studios working on it, or maybe it was because of it? Is there another reason why game development is so crazy right now? I don't think it's graphics, we're always making new tools and engines to create and implement everything faster and faster, there's a complication somewhere beyond the usual suspects.

@cikame: Games are bigger and more complicated than ever. Engines and tools have not kept up with the work demands for high fidelity assets. But everything else is more complicated too. Animation, scripting, level design, systems, etc... And games cost a lot more to make so there is more of a commitment to trying to make them as polished as possible, which certainly doesn't always happen but still gets more time taken.

Long gone are the days when you could just whip something up in a year with 20 people and make it work. PlayStation games were all 700MB or under. Current games often break 100GB. It's just a whole different scale than it once was.

era and genre can really be a non-linear force multiplier.

people frequently note that something like Arkham Asylum was built by a team of 40 that at max was 60- which on it's face seems miraculous.

but then again that game fits on a DVD and is actually a very tight metroidvania with clever asset use and a series of combat arenas. so it makes more sense when you compare it to the behemoth systems-driven open worlds that target 4k and are fully performance -capture.

I work for an energy company with big projects which often involve many third parties creating deliverables. Everything they create for the project is reviewed at each revision. I imagine game developers have a similar system for the individual review of assets, rather than a more hands-off approach where they just hand it all over at the end for the main developer to fix. There's probably more communication and cohesion than it would initially appear.

The fact that Bethesda shackled themselves to a system that was barely held together with string, spit, and happy thoughts by the time Oblivion rolled around, seemingly for the main purpose of being able to pick up and move almost every object in the game in 'realtime' (IE: when you pick up an object and just sort of move it around the world as opposed to just adding it to your inventory) because 'the memes' and the videos of 1000 cheesewheels rolling down a hill really helped Oblivion and Skyrim reach a wider audience set a dangerous precedent. Because everyone just dismissed it at the time as "charming jank," so there was little incentive to fix any of it and it's pretty clear those at the top took the wrong lesson and just decided to NEVER fix it to "make the games go viral." Plus the community showed that they'd happily fix shit for free.

A lot of this is theoretical, a lot of this is based on actual knowledge, but you look at games like Outer Worlds or Dishonered, that have a very similar "feel" but you can't move shit randomly around, and it's 1000X more stable out of the box and you start to wonder just how much the ability to move shit in realtime is worth VS its cost. And that's before the "NPC schedules" are introduced into it, again: a cool novelty if it's your first Bethesda game, but also a massive point of breaking/failure after you've played more than 1.

So even beyond the outsourcing and corner-cutting, the framework itself is rotted because the precedent was set: so long as Todd Howard does some bullshit self-deprecating "OUR GAMES SURE ARE BUGGY, HUH? LOLOLOLO" jokes at a press event, reviewers are fine with it and consumers still buy it!

It's baffled me since Skyrim that people still give these games a full pass (apart from Fallout 76 because SURPRISE: they couldn't rely on the community to fix their bullshit in an MMO) considering how basic they are in so many other ways.

Edit: This is to not even broach the topic of the graphical arms-race that is one of the clearest examples of marketing shooting devs in the foot year-on-year and the "bucket of crabs" metaphor that if any one company just declared "graphics are fucking fine, our people are being worked to literal death to make this shit!" their competitors would just seize upon that and declare "WE GIVE A SHIT ABOUT GRAPHICS BECAUSE WE ARE OF THE PEOPLE!"

Meanwhile the Switch continues to be a sales juggernaut with "perfectly fine" graphics, often excellent as they tend to prioritize style over the carrot-on-a-stick of "realism."

@bigsocrates: A couple things, developers for the Playstation were using, compared to now, horrendously slow workstations that could take hours to render out something before you could see it, realise it wasn't optimised enough to run on the Playstation and try again, these days you can render things out within the editor, export it to the engine and try it out in real time, there's also dynamic optimisation systems in-engine like LOD, texture and resolution scaling and UE5's Nanite, the modelling and animation tools are also faster and less likely to crash thanks to decades of progress, so while art is more detailed and complicated like animating the wrinkles on someone's face, we have the tools and hardware to do it and thousands of hours of tutorials telling you how, i'm sure it's gotten even easier than when this was made 10 years ago, but i guess we can just excuse this as "Bethesda jank".

Picking on Starfield again it's hard to compare it to the previous games in terms of scale, yes they've got space travel so there's probably a lot more space to run around on the various planets, but how much of that is randomly generated area? 2kliksphilip has a great demonstration of this, i haven't played the game so i can't speak to the variety of environments and sheer quantity of art compared to a Skyrim, or what kind of budget is required, but luckily the industry generates $110 billion more revenue since then.

Comparisons of Starfield's 7 year development to other... "similar" games.
The Witcher 3 - 3.5 years
Death Stranding - 3 years
oo here's an interesting one Cyberpunk - 9 years, how did that happen?

@cikame: Obviously the tools HAVE improved. Just not enough to keep up with the demands. Think about how many objects there were in a PS1 level vs how many in a PS5 level. It's at least 100 times as many, and in terms of polygons and texture density thousands or even millions of times as many. It just takes more time to make that many objects at higher fidelity, even if your tools are 100 times faster.

Atari 2600 games could be made in a matter of weeks and there they had virtually no tools. They were programming to the metal. More complexity takes more time.

Starfield and Cyberpunk took longer than most games because they obviously had various development issues that required retooling and rework. Both were publicly delayed multiple times. But even from developers who are hitting their targets games take a lot longer to come out now than they did in the 1990 or early 2000s. Back then you could pump out a yearly sequel to games like Spyro or Tony Hawk and have each one be a classic. Nobody even tries that now, and the games that still do come out annually either have multiple teams on them (COD) or very few changes (sports titles.)

It's just a lot more work to make much more complex games.

Now if we want to start looking at non tech reasons we can look to games like Hollow Knight. Yes it's a small team but it's taking them 6 years to do a sequel to a 2D metroidvania and there the tools have made things radically easier. I think the reason there is that they know a lot rides on this game and they want to make sure it comes out right, and I think that also explains longer cycles for non-tech reasons. These days most studios can survive at most one or two flops before folding. Back in the day you could churn out a ton of projects and keep trundling along if they weren't total disasters. There are fewer "big" games than there used to be (though more games overall) and each one is a huge bet that you really want to get right.

People forget that videogames are big business, and we're talking BIG business. We're used to thinking that you hire as many people as you need and you do your best to accomplish what you wanted, but that's not how big business works. Big business starts at the other end with the question, how much money could we possibly make, and then ask how can we accomplish that.

A good analogy is high-end sports. When they build a stadium they know they will always have an overflow of demand, and for every extra seat they squeeze in the profits will increase over time. So they don't think about how big they want the stadium to be, they think how big can we possibly make it. It doesn't matter how much it costs because the return on the investment is always far bigger.

That's exactly what's going on with AAA games these days. The big games are multiplying their investment several times over, so the only question is "how much more can we invest in these games ?". The demand for these ultra high production-quality titles is just so high that investors are just throwing money at them, telling the developers to "make it bigger", and hoping for an even bigger pay-off.

@ares42: I agree with your overall point that large-scale AAA ends up being the business people looking at their profit and loss sheets and saying “how can we make the maximum amount of profit on our projects so all stakeholders are exceedingly pleased,” but I don’t think I agree with the sports stadium example.

Not a sports expert, but I imagine you very easily could build a stadium that is far too big for a team in a smaller market where real estate is cheap in the city. Granted, yes, some teams are so popular that they could probably never feasibly build a big enough stadium to meet demand for tickets. In that case, sure every seat you squeeze in is extra revenue.

As for the analogy, a stadium is a one-time construction project, that once it’s done, the only other product for decades to come is “do a season of sports each year that draws a crowd”, and the requirements of that product don’t change dramatically year to year.

In video game publishing/developing, I don’t see the analogy to the stadium: the publisher could think “this studio always puts out hits, so surely the more money we invest in them, the higher the ROI is”, but any studio can put out a flop if the creative vision isn’t there, executive meddling goes too far, etc. See late stage BioWare, and how badly Anthem did despite an experienced developer and significant investment from the publisher. The entertainment business, whether it’s TV or Hollywood or gaming is full of hits and misses, so I don’t imagine the money people treat it the same way you would a sports franchise/stadium. The audience demand is much more fickle outside of sports.

@bisonhero: Stadium building is very economically complex because most teams don't actually pay the whole (or sometimes any) cost of the stadium. There also ARE ongoing costs including maintenance and frequently refurbishment, it's a whole thing.

I also think the idea that there is limitless demand for big budget games is wrong for a number of reasons. For one thing more and more development is going after HUGE profits from live service games. GTA IV was a hit that sold a lot of copies. GTA V is bigger than sections of the industry were as a whole in the 90s. Fortnite bigger than that.

The second issue is...big publishers are putting out fewer and fewer big games each year. Activision puts out Call of Duty, Blizzard stuff, and a couple other random releases every year. EA is basically EA sports + Battlefield.

Ubisoft is Assassin's Creed/Far Cry/ The Crew (for some reason) and some random other stuff.

This year has had a lot of big games that have been well received so it feels pretty packed but if you compare it to the number of bigger releases we had in the PS3 era there aren't that many.

2000 developers across 14 studios worked on Last of Us Part 2 it was the most polished game I've ever played incredible visuals with zero pop in. Bungie alone has ~900 full time employees.

Bethesda itself was never that big ~200 made Skyrim and ~300 for Starfield considering Nintendo had ~400 on Zelda over a longer period of time.

Bethesda for some reason have always been small for AAA.

Cyberpunk was not in development for 9 years. Even if you start counting when the project began in 2012, that's still 8 years until its 2020 release. But that's not accurate either, since it was just a skeleton crew of pre-production until development finished on Blood and Wine in early 2016. Full development of that game took place across a little over 4 years.

A small anecdote here. I work in SW development (not games). In a previous life we had a remote development office in Bangalore, India. Where, frankly many major SW corporations have an office there. This remote office we not given ownership over projects, but was mainly used for maintenance/legacy code, QA, and SOME small SW features.

Was it company culture? Was it talent? Was it the shear effect of asking for solutions from people whom were quite detached from the problem?

All I know, is I've worked at companies with 5,000 sw developers, and at ones with 40.

There is so much waste once a team gets too big. There is no such thing as "Hey do this, have it done in 3 months", when the thing they need to integrate with is a moving target. You can't just throw work/problems over a wall. Smaller teams will always outproduce massive fleets of ... not top talent. I know that last statement seems unlikable. But I've seen what 4-5 rockstar developers can do vs. 30 decent developers.

i never thought too hard about the outsourcing cause i'm suire its different for each studio .

the first time i recall Hearing about it was deus ex :HR , imo as someone who never played previous games i thought the game was great but the boss fights were iffy ... turns out the boss fights were that way cause they had outsourced them so instead of being like the rest of the game where you had a certain amount of freedom to go at a fight/lack of with however you approach things the bosses were pretty much hope you speced out well enough for this arena battle of sorts cause your fighting this boss right here.

the next time i recall really hearing about it was xenoblade chronicals 2 .. all those weird oversexualized characters? alot of them were suposabily outsourced like that owl humanoid character with oversized jugs.

this is just 2 examples i recall off the top of my head but we've heard of plenty of them over the years .. between playtesting art assets and other areas. honestly idk how i feel about it in terms of if it helps or hurts a game. plenty of little indie devs outscource parts of games that they don't know how to do and it works out great . other times you get something big and it can go either way too.

As for starfield ... i think bethesda got too big in the head as a company . i've always enjoyed BGS games sure over the last decade or so they've started to get a lot of flak for their bugs and slightly outdated tech n what not but at the core i always thought they're games were good enough to deal weith it . ...... till now . i've played 14 hours of starfield and personally , i might sorta hate it. the world feels rather soulless, i know they intended to a point design-wise of going for the more blan space aesthetic but thats not the issue . its not even the ai generated barren copy pasted random worlds that are my issue cause frankly i haven't gone to them yet tho i don't want it . no ... my issue as their handcrafted city/towns just feel heartless. it used to be they would put enough care into their locations that they felt lived in , every place had a story of some sort , every npc had a name, or a job at least. starfield tho ? you get to new atlantis and more often than not you look at an npc and they are just 'citizen' no random overly played comment like a city/town gaurd like skyrim for example. then on top of that even the ones who are characters/named they just feel blan. sure the dated animations and graphics don't help matters but i don't think thats the real issue here when older titles had the same issues but still had heart to their characters even if they were sorta throwaway ones.

Idk , with each game BGS puts out i feel like the games loose a little something each time. morrowind was great , tho i watched a video about a dude complaining about all that was missing from daggerfall before it. personally i never played any entries previously so i can't compare. but what i can compare is morrowind to oblivion .. we went from indiviual armor pieces to less parts and no robes over armor . skyrim? removed the ability to have clothes And armor on period. we won't speak of weapon types. i believe each entry had less factions each time also. those are just 2 examples of things being 'lost', that said i still thought they were good games anyway. fallout isn't my prefered setting so i fell off of both 3 and 4 but 4 definately didn't feel as enjoyable to be in. and 76 was a different beast by design .

To me i think they've just gotten too big to actually put the care they used to into the games they make. idk if outsourcing has alot to do with it or if its a design and management issue or a money issue, but the care just doesn't feel like its there the way it used to be. starfield might've added some layers but it feels very much not worth it and thats a shame, but more of a shame is idk if i care about elder scrolls 6 if this is the care they're putting into a new IP they've been wanting to make for gods know how long .


Anyway this got longer than i actually planned on typing, i just woke up and all i've done is type lol.

TLDR: idk if outsourcing is the major issue but i'm sure it doesn't help things .

@lord_anime: It's great to hear from someone who's actually experienced it, all i can do is try to take everything into account and guess and it always felt like software projects, especially the hardest parts like coding and compiling, would just fall apart if too many people touched it.

On this topic of outsourcing art i think about Noclip's documentary on C&C Remastered where they talked to the Malaysian studio responsible for the art and those guys seemed really nice, great people to work with, but you know that's a 2D sprite game with very low requirements and probably a super low budget, the control required to handle as many studios as the AAA games must be such an important factor to the overall quality.

Let's pick on Volition this time as an example of it maybe not working (i use maybe and possibly a lot because all of this is just me theorising), the long serving lead game designer and project director at that studio left after Saints Row 4, which many will agree was the last fully realised game they put out, you could point at that being one of thousands of possible reasons why the quality at Volition declined, but i feel like it becomes an even bigger factor when you're also corralling the work of external studios. Maybe Keyword play a part too and make the whole process easy, but i'm sure most of you have experienced times at work when it feels like everyone is making it worse and you have to roll up your sleeves and do it yourself, and for something as technically challenging as video game development that must be impossible to deal with, unless the director of the project is maintaining control, maybe :P.

from all i've absorbed, one of my personal takeaways re: contemporary AAA releases is-

your production and project management needs to be as good (if not better) than your design doc. for better or worse the table stakes for publisher investment (asset resolution, features like multiplayer, etc) dictate a certain scale of operation- so you probably have a lot of studios that just haven't developed those internal competencies.

to be fair that's kinda the world changing around them, even if they're a great studio.

but it is what it is- they either scale up to match expectations or scale down to scope (or in what feels like many cases, do neither and end up with a product that, despite blood, sweat and tears doesn't ultimately hit its marks).

pure speculation, but it's probably why you see people like Hermen Hulst and Angie Smets from Guerilla getting major jobs at Sony leadership. you have 2 people that delivered massive scale games more or less on time with solid performance in software and sales. when you hear people say it's a miracle games get done at all- i imagine to do that consistently is a rare talent indeed.

Project management must be a dark art in this industry. How many big games hit their intended release window? Something always goes wrong and I don't know if you can plan with that in mind.

This has been going on for a while. Look at the number of Ubisoft studios involved in Ubisoft games, even as far back as the early 2010s. (I don't know if Vinny's baby video when Rex was born is still accessible, but my memory says there are some comments in there from Vinny on the length of credits for Rayman origins while giving the thousand yard stare into the camera.)

Project management is a dark art once things get beyond a certain size no matter what the industry -- we just don't see it/don't pay attention to it. And software development in general is especially fraught with peril because at massive scale -- when you're dealing with hundreds of thousands of unique objects and actions that can all interact with each other in weird ways, you can't test it all, and you can't fall back on institutional knowledge / previous experience for every interaction. Something unexpected is going to get through and there's nothing you can do about it.

When you reach a point when a single individual can't keep all the details in their head -- in other words, once your application, game, house, research lab, power plant, highway network reaches a certain size -- you need help, and your chance of something going wrong goes up exponentially to the number of people involved.

Masahiro Sakurai has on Youtube a series about his life/experiences creating games in short bits, including a bunch of stuff on bugs and why bugs get in there. Worth checking out for an insider perspective. (I don't work in games, but I do work in software development, and I've been all up and down the chain.)

And one final note that some others have alluded to -- it's true 5 top-notch developers can do more work than 30 mediocre developers, but only up to a certain point. You can't play American football with 5 star players - you need to suit up 53. Similarly, if your project runs at any sort of scale, you're going to need to throw more people at it if you want to meet schedules. If nothing else, there are only 24 hours in a day, 7 days in a week... and if you're pushing your top developers with EA developer schedules to try and accommodate a lack of coders, they're not going to be top developers for very long.

Project management on the HR side isn't about getting the best performance from your top people -- it's about getting adequate performance from everybody else. Finding something useful they can do and letting them do it, and putting in place the framework and structures to assign, guide and correct their work while working to a large design that's constantly shifting.

To summarize: I don't think size of a development team is a real measure of success or failure. Small games can be good or bad, big games can be good or bad. If anything, if larger games are worse, it's not because there's a large number of people behind it, it's because if you can afford to hire a large number of people, you probably have a certain type of corporate culture that makes it hard to produce good games. EA is probably the most obvious poster child for this.