Are you going to try to tell me that The Witcher 3 was developed by having everyone work a standard 8 hour day and that the team was given as much time as they felt was needed to produce it? Of course not... it was developed with just as much of a crunch, pressure and "inhumane working conditions" [1] as Cyberpunk 2077, but because it was a resounding success on all platforms, no one batted an eye.
The issues with Cyberpunk 2077 have next to nothing to do with crunch time, they simply tried to make a game for a platform that was never going to be able to support it, regardless of how much time they took. The game on PC is a masterpiece of technology, it looks and feels amazing and it's a resounding success. The failure on PS4 and XBox is because those platforms were still very large markets and they wanted to take advantage of them and well push come to shove, there's just no way of making the game play well on them.
That's all there is to this story, making a game for a platform that can't support it is going to be a failure. Now that Cyberpunk has failed, people will use that failure as a way to argue about things that likely have nothing to do with the cause of the failure, and you know maybe that's a good thing and better working conditions come of it... I don't know... what I do know is even if CD Projekt Red decides to change its working culture for the next game they make, there's dozens of other new game studios rising up and it may be the case that the only way for them to compete with the existing giants is to work day and night to the bone on making the next masterpiece.
[1] https://www.gamebyte.com/cd-projekt-red-admits-crunch-period...
Take a look at the development for the game Hades. They took their time and polished everything, even having a pre release game in steam and incorporating community feedback. Rushing is not the only option
What happens between consenting adults ain't anybody else's business.
Eg I love Amazon as a customer, but I wouldn't really accept a programming job with them.
Doing so makes you at least complicit in that, and at most an active perpetuator of it.
Doing what? And who is 'you' in that sentence?
"Consenting adults" is a fine heuristic for what happens in a private bedroom between individuals with equal power. It does not apply to for-profit exploitation done by organizations with special legal privileges. Privileges they have only because we see commerce as generally beneficial and to be encouraged.
What do you mean by exploitation? When was it the norm and where?
While I appreciate your point and even agree to a large degree with your point, the stakes are very different between these two products.
This is part of the issue I think. AAA games are so high flying in their technical and artistic scope, and they have all the business and power bullshit that comes from their now massive, 100mil+ budgets. Their failures are spectacular because when there is so much complexity inherent in the system and teams, lots can go unnoticed, and it falls apart in many different ways.
You can't even really compare the clean delivery of a AAA war shooter from 2005 with one from 2021 for example, the technologies and disciplines involved have grown so much.
BTW I believe the best place for a game developer is something similar to ID before Activision. It's small and flexible, but not too small to develop big titles. Nowadays you might need some 50-100 developers for a big title but if you go over the number and/or go public the productivity and more importantly the focus is going to be lost.
Ah, so they leveraged unpaid quality assurance and design labour in order to cut financial corners.
Are you going to tell me that The Witcher 3 was a success because it was released May 19th 2015 instead of October 24th 2015 or even May 19th 2016?
It's unrealistic to believe there's no reason for deadlines. For a FAANG company, sure, they're making so much money they can -- mostly -- afford to do things without deadlines but few video game studios are at a level of funding that they can spend another 3, 6, 12, months.
The internet claims Cyberpunk 2077 had 500 to 5000 people working on it. I'm sure that 5000 number is not full time but it's not hard to believe the 500 number. Just paying 500 people $15 an hour, 40 hours a week $1.2 million a month, not counting rent/power/taxes etc... And of course some large percent of those people are probably getting closer or $30 per hour $60k yearly salary) to $60 per hour ($120k a year salary)
Witcher3 claims 150-250 employees so your delay, shipping October instead of May is easily $4 million needed and I suspect that number is low. $12 million if it took an extra year as you suggested and that's a low-ball estimate.
That The Witcher 3 ended up being a massive success is great but it was never a guarantee.
Development cost at least doubles, the graphics would start to look dated relative to other titles. Competition with other open world action rpgs increases.
The people responsible for making it the game that it is would have moved on, and other ideas would have come into the dev cycle.
Cyberpunk could have been delayed again, but it still would have been bad. The Witcher 3 could have had less crunch and released a year later and been as much of a success, but the studio would lose way more money.
Labor is extremely expensive and a year more of labor isn't worth it, so they rush. It just worked that time.
He spent three months just on planning the design for a completely new MMO to replace the current one. Then went to upper management and told them to choose - they could have him just manage the current game, fix the bugs and issues, and make it a really polished, but still fundamentally flawed game, that would never have the kind of audience like WoW, one that would forever tarnish the brand; or, he could make a completely new MMO from scratch in one year, while also supporting the existing one. They chose the latter. Today FFXIV is the world's most popular MMO.
The process of getting there has been documented in interviews and it is what Naoki-san says in this one [1], that I want to bring attention to. Paraphrasing, "we could only get there by aggressively cutting down features and micromanaging daily tasks for each employee, to the point that the tasks fit into a standard 8 hour work day, which generally has about 6 hours of actual work done, when you factor out meetings and breaks".
This is a man tasked with fitting 5 years of work into 1, while also supporting the existing product, and he didn't immediately reach for overtime. He absolutely worked himself to death, but he didn't push the team beyond their limits. It helps that he was an MMO fan and had a full design document for the game he wanted to build, but it still shows that with good management you can deliver a product better than everyone else while not doing overtime.
To be fair, 6 hours of actual work is still a bit more than I hit most days. Having to work at full capacity every day would still be straining, but nothing on the order of doing 12 hour days 6 days per week for months.
Compare being chased by the police in GTA vs Cyberpunk for example.
The game is simply better made than Cyberpunk. It's not really a knock against CDPR, it's their first attempt up against rockstar, but there is simply not enough polish, not enough game design etc. For me to say anything else.
I wanted to love cyberpunk, but the moment I picked my faction and then was immediately funneled into the same story as everyone else it kind of set the tone for the rest of the game.
I sincerely hope they keep developing it but at the moment I can't bring myself to play at all vs. GTA which is very satisfying just to drive about a bit before I go to bed.
It's a simple game, and as limited as GTA is.
The first link is filled with early-mid 20th century resources on this subject. The rest are all from the 80s - 00s because people... stopped studying this. Most of the studies covered the same thing, and it very quickly became an uninteresting problem in the field of Clinical Psychology because the answer was almost unanimously, unequivocally, "yes, this makes people output lower quality work product". It's one of the most settled problems in Psychology that exists in the modern day.
The only people still researching it are non-profits, and their research serves simply as advocacy, because otherwise businesses won't listen that this extremely, absurdly well-known and well-supported fact is true.
https://igda.org/resources-archive/why-crunch-mode-doesnt-wo...
https://www.tandfonline.com/doi/abs/10.1080/0267837890825693...
https://synapse.koreamed.org/articles/1125602
https://www.jstor.org/stable/40967587
Nevertheless, here is something specific to Game Development from 2016: https://www.takethis.org/wp-content/uploads/2016/08/CrunchHu...
Select quotations from papers quoted in that article:
"In a national Australian survey, employee cognition increased for hours worked up to 25 per week. Beyond that, cognitive abilities decreased. By 60-hours of work per week, cognition was lower than people who weren’t working at all (Kajitani, McKenzie, & Sakata, 2016)"
"In a large study of 4 US-based companies, productivity losses related to fatigue were estimated to cost the employers $1,967/employee each year (Rosekind, Gregory, Mallis, Brandt, Seal, & Lerner, 2010)."
"Lost productivity due to presenteeism (being at work while ill) is almost 7.5 times greater than that lost to absenteeism (Employers Health Coalition, 2000, p. 3)."
"In a study of a large, multi-employer, multi-site employee population, health care expenditures for employees with high levels of stress were 46% higher than those for employees who did not have high levels of stress (Goetzel et al., 1998)."
"People identified as workaholics were found to have higher rates of ADHD, OCD, anxiety and depression compared with those who weren’t ranked as being addicted to work (Andreassen, Griffiths, Sinha, Hetland, Pallesen, 2016)."
Cos if you can be max productive at 3 challenging but sufficiently different cognitive/creative tasks for 15-25 hours per week each, that sounds like a super fun, productive, efficient, sustainable, healthy, and balanced life
If what you are saying is true, you'd expect game companies that follow the advice to have a much easier time recruiting people (for less money even, probably) and deliver better games quicker?
What's going wrong? Is everyone involved an idiot or evil?
There are still people that don’t approach things rationally though. And there are ignorant people who get funding that run their teams extremely badly. There are also “evil” people who think crunch is necessary (forged through adversity etc) to make a good product.
I’ve heard more stories of long working weeks from tech startups than I have games recently. Loads of people posting here seem to regularly clock in long working weeks when the topic comes up.
I'd think though that the factors affecting the success of a company is very complex and to boil it down to one point is insufficient when talking about competition within the multiple markets that comprise 'games'.
If you don't deliver a better game quicker the change of process (I.e. less hours) and by extension you, will probably be blamed even if there are other legitimate reasons.
I see that this would apply to _most_ existing game studios. But if what fao_ says is true, studios that are explicitly set up to take advantage of the research should easily out-compete those that don't?
The article relating loss of sleep to loss of productivity is fair and robust, it's worth taking seriously.
The article relating suicidal ideation with work hours is devastating, but it's hard to see how it relates to your point since the highest percentage of suicidal ideation is among those working less than 40 hours (17.2% suicidal ideation). The study itself is in Korean so I am not able to dig too deep into it but there is a table on p342 which I think is pretty revealing. Basically what it shows is that suicidal ideation is highly correlated to one's level of education; people who have little to no education showing the greatest level of suicidal ideation whereas those with a college education have very little suicidal ideation. My suspicion is that those with little education either work very long hours, or work very few hours which is why suicidal ideation is bimodal. Those with a college education are most likely to work within the 40-60 hours a week which had the lowest suicidal ideation.
The research linked to jstor.org is behind a pay wall, but I managed to find another source for it at [1] and it contradicts your point. It specifically states that no difference was found among those working an 84 hour work schedule compared to those working a 40 hour work schedule, and I quote right from the study which I link to below:
"In conclusion, working of one's own free will on an 84-hour workweek schedule with physically and mentally demanding job assignments does not necessarily induce greater physiological strain than working with similar tasks on an ordinary 40-hour workweek schedule."
It appears the problem with working long hours is not the duration of time, but rather the manner in which those hours are scheduled. The article points out that when people are able to schedule their own extended hours, they perform as well as those who work a given 40 hour schedule, at least as far as their physiology is concerned.
All told, the arguments you presented should not be quickly dismissed and need to be respected, but they are not as strong you made them seem and certainly do not establish my point as being patently absurd. On the contrary one of your sources says that crunch time over short periods of time from 2-3 months can produce an increase in productivity, and another one of your sources says that working long hours is fine so long as one is in control of their schedule. It is somewhat strange that you'd present said sources while also claiming that I'm being absurd.
Yes, follow up on the citations (It was written in the 80s) and you'll find it quickly contradicted.
> All told, the arguments you presented should not be quickly dismissed and need to be respected, but they are not as strong you made them seem
As I already said, we stopped bothering to do psychological research on this en masse, because it's not an "interesting" problem -- it's mostly been settled that it's unequivocally a bad idea. The idea that you think that you can look at 5 links and suddenly emerge with a complete understanding is very interesting though! Those were left as breadcrumbs for you to follow up on that I could gather in the shorted time possible, not as the complete picture.
I think posting a few links that you likely didn't bother to read through yourself (given that one is behind a paywall and another is Korean) and then proclaiming absolute confidence and authority on a subject shows not only poor critical thinking skills on your part but also a propensity to discuss challenging topics in bad faith.
No, and it showed. Witcher 3 was a success because of previous games. 3 was a dissapointment for a lot of people, they took the Ubisoft formula of repetitive content and diluted the experience until I got bored after about 100 hours.
W3 was a huge financial success but the writing was on the wall and only people who did not pay attention where surprised at Cyperpunk disaster. They tried to force a good game out with W3 and even more so with Cyberbunk. The results where clearly different than W2 which I personally think was their peak - a human curated story pace that left you dreaming for more.
Compare this with GTA games. Rockstar knows you can't rush things nor can you force industry revolutions in AI over night. Slow and steady wins the race. To which company do you think will the investor money go next?
While it's great to see games/companies become widely recognized, I feel that it always slowly gets poisoned by investors. Anyways, cheers all - stay safe.
I know there are a lot of people who enjoyed it. For me though it was a sign of things not going in the right direction.
I’ve not played but I hear AI is still dumb af, it’s COD with a cliche Hollywood cyberpunk style.
Pretty sad our art is still tethered to 80s-90s fanboy stylings. I feel like the obligation for things to fit such a literal expectation constrains creators. Then again the stories coming out of the the dudes running these places. Egh
It's more about getting the right personalities on your team than giving blanket license for artists to take as long as they please.
The recent IGN interview was great. It was an hour of hot chocolate for my niche demographic. It also confirmed that the magic of Halo was what we thought it was all along. You want to make a great game? Watch the Halo Vid Docs and take notes.
(I played about 3 levels of the first one, it was ok for that time, I know people really love it, but so do Doom, or Diablo, etc)
As for storytelling in the campaigns: it was a space opera complete with mysticism and an epic journey. It's really hard to stretch that into a never-ending story, but the books have done well. The games have not. The first problem is in making the self-insert chief a human. It was obvious from the beginning. When I played Halo 4 I had a mental list of all the ways the creative direction dropped the ball. There's something that 343's culture has never understood about the Halo games. They think it's a Michael Bay film because the audience is young men but it's more like Star Wars V.
This is a highlight reel of what I'm talking about.
Eh, if you want to make a comparison, I would put most corporate programming into the bin of 'craft', much more than 'art'.
It's of course all subjective, and there's no one true category. They are just different ways of thinking about the subject.
A former corporate programmer who was a highly paid corporate programmer at the time when he told me this is that all they do is very fast simple arithmetic. Add and subtract.
Great hackers on the PG spectrum may get struck by inspiration and may outperform the average corporate programmer by 10x or 100x or more and work crazy hours while the creative electricity buzzes.
It totally depends. Most corporate programming is CRUD (https://en.wikipedia.org/wiki/Create,_read,_update_and_delet...), so not much arithmetic involved. But a lot of 'book keeping', and organising your information. Not rocket science, but more listening to your stakeholders.
> Great hackers on the PG spectrum may get struck by inspiration and may outperform the average corporate programmer by 10x or 100x or more and work crazy hours while the creative electricity buzzes.
See also Carmack etc. But I was talking about the typical programming tasks.
(In some sense, this reminds me of getting startup equity, which is basically worthless until the company goes public.)
All the crunch and pain and buggy games are a result of humans trying to force the other humans who are programming the games and content, and indirectly, the computers themselves into doing a rush job.
You can't rush art. Programming is an art since whatever is being programmed has never been made before, else you could go Copy + Paste and be done.
Yes, there are millions of paintings of meadows, just as there are millions of copies of very similar but ever-so-slightly different boring line of business applications. But if you want THIS painting, this one that you have commissioned to show the meadow in YOUR back yard, you can't rush the artist, or else the painting quality will be reduced. Depending on how much rush you add, the quality can be subtly bad but close enough, or horrifically bad as if, say, this pretentious commenter were painting it.
Let the artists (programmers) take the time they need to produce the artifact. Yes, the artists can sometimes get stuck in cycles of tweaks (goldplating) and need to be told this artifact is done, ship it, but it always seems like we lean too far to the other side: feature factories, churn churn churn, crunch crunch crunch.
Let the artists paint / program in peace!