Imperfected
Member
How often does Valve crunch, again?
-
Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.
Ryse crunching team served 11,500 dinners by ship date #rysefacts
- Thread starter Wario64
- Start date
This is not true. It's avoidable through proper planning, skilled execution and customer collaboration.
If you allocate enough resources, you shouldn't need to crunch. If you cannot afford to allocate enough resources... you can't really afford to crunch, either.
Visualante2
Member
Because Turn10 outsources lots of work to India where they can have probably hundreds of people working for less to get it done.
Plenty of studios/developers offer a more balanced take on work. Valve can't amass the talent they have and keep it by being in constant crunch time. If you read the various articles about Half Life 2's creation, Newell frets most about the health and attitude of his team more than the game itself. The simple fact is that not many places can afford to pay that much, and even fewer have the motivation to pay well or be balanced when there's a ton of unemployed devs looking for any type of work. Some games just don't require a whole lot of pure talent, and involve hiring a ton of people to just labor away at the game. See Ubisoft and their 900 AC4 devs.
Shit like this is why your work conditions will never improve.
Indeed.
Thing is that 2-3 week crunch can be unavoidable but half a year crunch or more for project that is 2 year project is not situation that couldn't be avoided and it is easily management or should we say expected course of action planned from start by people who started project. Naturally there can be situations extraordinary that could happen but i doubt in AAA industry something like that can happen unplanned.
dragonelite
Member
If you have the solution im pretty sure you can make millions if brought it to the embedded / high performance software development world.
Funchameleon
Banned
The entire joke is that it isn't funny or appropriate for lighthearted messaging, much like the original dinners tweet
It is called planning in advance that something can happen and giving product month or two of breathing room for either fix or improve game. We are talking here about AAA industry where games are created for milions or xx milions and they earn xxx or xxxx ons.
Imperfected
Member
The worst part is that it isn't even efficient. The American standard 40-hour work week already pushes the limit of how much productive work you can squeeze out of the average employee. There's a pretty sharp fall-off in productivity for most people beyond that point, and from an employer perspective you really want to sit in that "sweet spot" where you're getting the maximum efficiency for the wage you're paying. To go over it means you're effectively getting less of an employee - because let's face it, we all start to lose a few steps when pushed beyond a certain point - but still paying full price for them.
There are jobs where you can get away with having a dramatic decline in employee productivity without having it cause massive problems - basically any time you just need a warm body present and going through the motions - but software development really isn't one of them. The quality of work you're going to get from a programmer who's been at it for fourteen hours straight really isn't worth it.
And don't even give me the whole, "There's no other way, they only had so much time to get it done!" argument. They could have used a larger team and worked them normal hours for the duration of development. That costs money, but so does crunching - lost man-hours due to declining quality of work, creation or exacerbation of health problems due to poor diet, stress, and a lack of exercise, employee retention problems, and all those "free" dinners add up - and beyond a certain point it actually becomes more cost-efficient to just treat your employees decently and spend a little bit more on actually having the labor to meet your needs reasonably, rather than spending it shoveling Red Bull into them and then paying it out in health insurance claims, anyway.
There are jobs where you can get away with having a dramatic decline in employee productivity without having it cause massive problems - basically any time you just need a warm body present and going through the motions - but software development really isn't one of them. The quality of work you're going to get from a programmer who's been at it for fourteen hours straight really isn't worth it.
And don't even give me the whole, "There's no other way, they only had so much time to get it done!" argument. They could have used a larger team and worked them normal hours for the duration of development. That costs money, but so does crunching - lost man-hours due to declining quality of work, creation or exacerbation of health problems due to poor diet, stress, and a lack of exercise, employee retention problems, and all those "free" dinners add up - and beyond a certain point it actually becomes more cost-efficient to just treat your employees decently and spend a little bit more on actually having the labor to meet your needs reasonably, rather than spending it shoveling Red Bull into them and then paying it out in health insurance claims, anyway.
The way the game industry handles crunch time, I wonder what months of development aren't crunch time.
I like "disgust". Much less whiny and over-dramatic, but still has a flavor missing from a bland word like "disagreement".
It's not hard to fix, there is just no desire to amongst those who cause it. They would have to admit they were doing it wrong all this time, and would have to learn to allocate resources before they are needed rather than just dumping people in the shit.
Interficium
Banned
Not to burst anyone's bubble but Valve isn't some nirvana where everyone goes home at 5 every day.
Just because a studio isn't "in crunch" doesn't mean they don't still fetishize 60 hour workweeks.
Just because a studio isn't "in crunch" doesn't mean they don't still fetishize 60 hour workweeks.
BurnOutBrighter
Banned
Agreed. It's actually in developers' best interests to do so. Think how much they could save on overtime salaries.
Quadraphonic
Member
Much like Apple fanboys and the iPhone, they'd rather not pretend it comes from a shit factory and takes precious minerals out of the earth to get to them.
So instead of trying to improve the situation, they blame those who give it to you.
So instead of trying to improve the situation, they blame those who give it to you.
Alternatives
1) Delay game - people will be upset
2) Have buggy game - people will be upset
3) Have graphical issues - people will be upset
4) Cut content - people will be upset
Seems to me like strict deadlines are probably the biggest issue.
EDIT: Oh nevermind. I see we've got the experts here who have all the answers.
1) Delay game - people will be upset
2) Have buggy game - people will be upset
3) Have graphical issues - people will be upset
4) Cut content - people will be upset
Seems to me like strict deadlines are probably the biggest issue.
EDIT: Oh nevermind. I see we've got the experts here who have all the answers.
dragonelite
Member
If you plan everything out of the gate allocate enough resources how would solve an backfire on E3 like 4~6 month before shipping? Just ship the game and write off the losses?
difference between standard crunch and non standard one are expectation of your employer.
5) Plan for delays or production issues. If nothing comes up people will just simply deliver better product.
RagingAvatar
Member
Sorry but do you realise how you sound?
You're speaking about this issue like it doesn't affect *people*.
I've worked on AAA games and ended up crunching for huge amounts of time, losing much of my personal time for issues that could have been avoided by planning, stricter project control and forcing other departments to stick to their schedules.
It seems to me that art and programming have to crunch the worst at the end of the project - to not even mention test (poor guys) - this is a deplorable practice, that frankly, isn't worth it for 99% of games.
You have to remember that for every Red Dead Redemption that had excessive crunch, there is another 100 games that got crunched on that you never heard of.
And in regards to your point in the quote above, you need to realise that most of crunch is not fixing the design issues for a bad game - it's usually just making up for the fact that others didn't stick to the schedule.
Daffy Duck
Member
I just hope they all had a balanced diet.
Or, budget time for delays in the initial workflow plan. Every project will have problems, plan to allocate resources to those before commencing, so you don't end up looking like a fool who can't think of any way to plan for unexpected contingencies. Or yes, delay the game if needed. But if you don't announce an insanely optimistic launch date, that won't be necessary.
Design the timeline with the actual resources needed in mind, if you then finish early you are golden. Basically, have realistic expectations and deadlines for the resources you have. Don't hope that everything will be perfect, because it won't be. Acknowledge that with extra resources instead of just throwing overtime at the problem.
soxinthebox
Member
People realise that salaries are the norm in the IT industry right? Salary pay doesn't increase depending on hours worked but rather you can think of your hourly pay going down for any extra hours you do.
The gaming industry needs some kind of union to increase base working conditions as this kind of crunch (any really) is not healthy.
EDIT: not to mention your personal life being affected (as most people seem to understand this).
The gaming industry needs some kind of union to increase base working conditions as this kind of crunch (any really) is not healthy.
EDIT: not to mention your personal life being affected (as most people seem to understand this).
Aussie Legend
Banned
Wow, now that is a post. Agree with all said here (with my very light dev experience)
Imperfected
Member
Not to say it's a "Nirvana", but Valve actively discourages employees from working excessive overtime. There's quite a difference between flat-out telling employees they aren't going home before dark for the next month - or strong-arming them with thinly-veiled threats - and letting them actually set their own hours, especially if your corporate performance reviews take a dim view of employees who need to pull overtime to achieve project goals.
http://www.valvesoftware.com/company/Valve_Handbook_LowRes.pdf
Does the system always work? No, I'm sure there are times when peer pressure or other influences convince people at Valve to work more hours than they'd like. I think just recognizing "crunch" as a clear failure of planning is a damn fine start for fixing those sort of problems, though.
You need to understand that 11,500 dinners served isn't a total you end up with because you needed to change some things after a bad reception to an E3 demo, or a "normal" drive to hit a rigid release target (X-Box One launch window). That's a number you arrive at because someone in management drew out the schedule and literally said, "Okay, we can force everyone to crunch through all of September, so that's 25% more man-hours we can squeeze out for these tasks." You don't accidentally end up needing hundreds of thousands of extra man-hours on a project unless you're truly, phenomenally incompetent; that is a completely premeditated act.
Always easier said than done, I'm sure. And do you know that they don't already? That seems like basic project management 101.
What resources will be needed is probably hard to tell ahead of time. Especially with games that take more than a year to make.
I really just think you guys are making it sound like such a simple idea, while in reality its likely a far more complex and involving issue.
Night_Trekker
Member
Great post.
And I suspect I am not alone in feeling like you and the others making similar arguments are making it sound impossible just because it it is not the norm.
As far as I know every other industry is able to allocate resources for projects without requiring something like crunch time. Yes, some businesses within certain industries are very bad at it, but I don't know of any other industry in which this sort of thing would be accepted. At best it is exceptionalism to excuse in video game/software development, and I think it might often be a just a lack of imagination that creates the idea that there is no alternative.
There are alternatives, they are just harder for managers and project leaders to implement, so they won't.
Crunch time is the WORST thing ever happen to the video game industry.
http://legacy.igda.org/why-crunch-modes-doesnt-work-six-lessons
http://legacy.igda.org/why-crunch-modes-doesnt-work-six-lessons
NinjaCarrot
Member
All of this is so true, thank you for speaking up.
In my experience the delays and production issues are mostly avoidable, often nothing drastic happens in development yet there is still a huge crunch, Sure if some catastrophe hits then that is unavoidable. Let me put it simply for people.
Things that save on crunch time :-
*Pre production tools are built on the back of the past projects experience and needs
*Design locks their concepts and content as early as possible, meaning the basic structure of the game is for the most part locked in as early as possible, core mechanics are decided upon early.
*This gives art the time to create the designs, art lock should happen as early as possible. This means no new assets are checked in, existing ones are just polished. *Audio and UI have an easier time getting things in with all features, content and art finalized
*The game can be streamlined, frame rates etc early and there isn't a panic at the end to fit your game on a disk or make it run at more than 15 fps .
Things that make for un necessary crunch down the line:-
*Design deciding to add new features late in the game having knock on effects through all departments
*Poor cross communication between departments, one department not working on what they should be because they are waiting for content from another
*Publisher shifting release/core dates
*Management doing nothing until it is too late to effectively get the correct flow of production
*Tools that require you to do a lot of work or take a lot of time to make minor changes, or that require a specific person to make said change. For example getting a coder to write specific code to make something happen and hack together what you want
I could go on but I feel like there is not much that has not been said, those that work inside know what should be done, they can either put up with it or try to figure a way round it so they can make good games under less pressure and cynicism
TAJ
Darkness cannot drive out darkness; only light can do that. Hate cannot drive out hate; only love can do that.
Never. It's probably why they don't make any money.
Very well said, and something you can apply to many industries that suffer from crunch. and a big SMH to people saying it's unavoidable ITT, or that it's near impossible. Truly groan worthy posts. It's very avoidable.
Shhhhhh, it's ok.... They don't know what they are talking about. It's ok.....shhhhhh
There's a difference between crunching because you take pride in your work and crunching because if you don't, you'll be let go. Newell takes the health and morale of Valve employees pretty seriously: http://www.gamespot.com/articles/the-final-hours-of-half-life-2/1100-6112889/
I understand what you are saying, and I will be the first to agree that making a AAA game is not an easy task, of course resources are hard to plan, but when the same avoidable mistakes are made year in year out (like the ones I posted a few posts above) you would think that lessons would be learnt and not to take the easy route out of pre emptively scheduling and expecting an extra year of overtime across the project to get it done . I don't have all the answers I just know I echo a lot of sentiments from developers I have worked with over the past 10 years
Hah, we made a bad decision and now we have no recourse other that to cut corners. Nice corner you backed the poor abused Ryse devs into. Sounds like they fucked up in the planning stage and now have to adhere to a strict deadline of their own making. The entire Xbone launch initiative has been shoddily and haphazardly strung together, why would this be any different?
I never said it was impossible. I just don't think some of you guys coming in with 'if they'd just do 'x', everything would be fine' are really doing anything more than people sit and watch F1 races and say that 'Oh, if they'd just fire so and so, they'd be winning races' or some shit, ya know? Every industry is different and I think you actually have to be involved in it to understand where the areas are that need to be looked at to improve things.
That one post from an actual developer details a few different things that could be done to improve the situation, and even then he still says some crunch time is probably to be expected.
TAJ
Darkness cannot drive out darkness; only light can do that. Hate cannot drive out hate; only love can do that.
That was nine years ago.
In addition to what people already said, business quarters play a big role, which causes a huge problem with development schedule.
Let's say that a company wants to get a game in for their Q4 report.
They'll schedule the deadline on March 20th, and then they schedule the rest of the development backwards.
This is the reason why the industry is forced to go overtime. There's no allowance for delays a public stock company.
You can bet that Watchdog's delay is going to cost Ubisoft multi-millions in retailer penalties and lower stocks from poor performance reports.
The issue, I'd argue, is the scale of money involved. How many devs are there out there who have the financial independence to be able to release on their terms? For the average dev, it's a balancing act, trying to finish a given game before the money runs out. Since at the size of project we're talking about everything has to be funded by the publisher, the pub then gets a lot of control over when to release money to development; hence having to pursue milestones and - on occasion - having to make massive changes at the publishers' whims.
It's not necessarily better at the small-budget side of things, either. It's easier to be financially independent at that level because the numbers are low, but on the other hand - when you run out of money, that is it, there's no-one to fall back on. I'm toying with the idea of dabbling in independent development myself, and I don't see any way I could do it without fitting it in in my free time around a full-time job - which would mean slow development, of course, but secure.
A good point. Shareholders in general are the bloodline of business nowadays and it creates lots of inevitable short-term thinking, which is necessary but also not ideal from a consumer standpoint.
I'm lucky enough to work in a game studio but I'm not a developer myself so I can't comment on the ins and outs of game production but I believe the biggest issue for crunch is deadlines especially on larger titles. This is due to the deadline being set by the company due to financial reasons not production reasons. Most developers can't afford to delay a game for 2-3 months to give it that extra polish due to the costing of doing that so crunching is a cheaper way of getting the same content produce, maybe not the most efficient though.
Add to that the way a game is built often as different parts of the project comes together different issues arise usually most of this is to do with code affecting other aspects of the game. For example changing of some physics values can totally flip out the AI which will need some new code to get back on to the right track which then in turn might effect the mission script code etc. It's often something small that starts to snowball.
I think a lot of developers try to react to media and fan response as well these days as games are shown off so early in production (in the most case) with only a snippet shown yet impressions made of the whole game. Developers may react to some extent to the response if negative/positive and change their focus to other aspects of the game which in turn will cause some extra design/dev time in an already packed production schedule.
So generally from what I've seen the issue is a financial one not a production short coming. Add to that marketing, publisher and investor expectations and something has to give.
Many dev teams/publishers will have the luxury of income from other sources (Valve etc) but will still need to crunch for the E3 deadlines and such to get the product look as presentable as possible.
Add to that the way a game is built often as different parts of the project comes together different issues arise usually most of this is to do with code affecting other aspects of the game. For example changing of some physics values can totally flip out the AI which will need some new code to get back on to the right track which then in turn might effect the mission script code etc. It's often something small that starts to snowball.
I think a lot of developers try to react to media and fan response as well these days as games are shown off so early in production (in the most case) with only a snippet shown yet impressions made of the whole game. Developers may react to some extent to the response if negative/positive and change their focus to other aspects of the game which in turn will cause some extra design/dev time in an already packed production schedule.
So generally from what I've seen the issue is a financial one not a production short coming. Add to that marketing, publisher and investor expectations and something has to give.
Many dev teams/publishers will have the luxury of income from other sources (Valve etc) but will still need to crunch for the E3 deadlines and such to get the product look as presentable as possible.
With now over 5 years in the industry, in my experience, it's setting release dates before you actually know when it will be done and not willing to budge on quality/quantity.
