I suspect this leads to a trend where you have disproportionately more mediocre product/vision people at the large gaming companies, because if you have really good product sense in a space with as much passion as video games, and with a relatively low barrier to getting your foot in the door (if you have a small core team of very talented, dedicated people), what on Earth are you doing being a cog in someone else's machine?
If you want to work on a AAA level product (or even a AA product nowadays), that's increasingly becoming out of reach with a small team. Some people want to work on games that millions of people will play and enjoy. I know a guy who made and released a few small games on his own (I helped with one of them) and they sold like 1000 copies at most, but he maintains a day job in the industry for the past 12 years and is currently a senior engineer at Epic Games working on AAA titles.
Meanwhile I'm working on a small game in my spare time this past couple of years and I'm worried I might not break 1000 copies with its release either (not the prettiest game and a bit of a niche genre, but I'm on my own in my spare time). But at least I have a fairly well paying day webdev job so I'm not screwed if it doesn't do well.
https://kotaku.com/inside-rockstar-games-culture-of-crunch-1...
https://kotaku.com/red-dead-redemption-2-developers-speak-ou...
But it still seems to be a mixed bag of experiences (with the expected "certain locations have shit management" and "QA always gets shit on"). And at least they shipped a finished product, that was arguably improved by crunch.
In contrast, there are a ton of games that don't measurably improve/ship and people are asked to crunch to cover management screw-ups.
I'm baselining at things like Star Citizen and Beyond Good and Evil 2, of the "How are these people even allowed to call themselves professional managers/directors?" level.
Games have massive cost efficiencies from scaling. Studios are incentivised to always make the biggest games they can with their budget. Games team sizes vary strongly with the project period. They can’t handle more than a small team early and can absorb almost any number of people at any skill level before release.
End result is that game development is always boom and bust. Small core teams build a game up slowly, then more and more are added until just before release when a mad scramble for anyone and everyone they can be bought borrowed or stolen to get the game released on time. Then there’s no work except for the core team on the next game.
The obvious solution is to have many games running in one studio to even out the team sizes and absorb the crunch, but that implies either studio sizes beyond anything we see currently (and no increase in per project scope) or much smaller scope for games.
I’ve seen models like they work, and be much less painful to work with for everyone involved, but it did require different market conditions to justify the alternative model, I’m not sure the entire industry could switch.
That’s not to say fixing the problem is impossible, just that the bad incentives will always be there to push back against improvements..
Which inevitably leads to lying, poor communication, and surprise boom/bust layoffs.
In contrast, most other software development industries have figured out how to square the budget-time-people circle, most of the time.
It's a shame Rockstar doesn't do more interviews, as they appear to have a working incremental project development model, that reliably (if slowly) delivers results. My suspicion is they just hire/train professional PMs.