Gamasutra article on how crunch makes Epic's games better
Adam Martin
“I am a believer that if you’re going to make a great game, and there
is that caveat, I believe that crunch is necessary,”
...
“Working later than 2 am is a net loss. The productivity of the person
who’s doing that to themselves ultimately ends us costing them at the
end of that week,” he says. Epic has put a “go home law” in the
company handbook as a result.
NB: this is a journalist's one minute interpretation of a 1 hour talk,
so take it with a pinch of salt. If anyone's got a writeup or a
liveblog of the talk, I'd be interested to see it?
Adam
Michael Lubker
I work when I need to. Most of my time is spent reading/researching or
playing games, IF something necessary comes up, I do it. Given, most
of my research is for my game company. ;)
~M
~ "Democracy is two wolves and a lamb voting on what to have for
lunch. Liberty is a well-armed lamb contesting the vote!" - Benjamin
Franklin
http://snowballz.joey101.net
Steven Davis
Does Epic know that Project managment tasks are scheduled by hours of effort, not just fixed time intervals?
So, my guess is that for each hour you work over 8 per day, you are getting, what 20 percent less productivity and this starts dropping really fast... (I know that there are good arguments against even 8 hour days).
One of the rare, sensible articles I saw out of some (casual) studio in the UK was that they addressed an awful lot of productivity by taking most people's desktop Internet access away. I've certainly seen this as well.
It is always good to see that the games industry is at least 30 (40?) years behind the rest of the IT industry in its development maturity.
Predictability is what you want to strive for most of all.
Steven B. Davis
CEO
IT GlobalSecure Inc.
c...@secureplay.com
+1.650.472.3771 main
+1.650.278.7416 direct
+1.202.478.1743 fax
secureplay skype
product: http://www.secureplay.com/
blog: http://www.playnoevil.com/
corporate: http://www.itglobalsecure.com/
Technology & Business Development Mail:
IT GlobalSecure Inc.
808 Comet Drive Suite 205
Foster City, CA 94404
Corporate & Official Mail:
IT GlobalSecure Inc.
114 Hobbs Street
Greensboro, MD 21639
To no longer receive email from Steven Davis & IT GlobalSecure Inc., please respond to this email: c...@secureplay.com
Michael Lubker
new-economic/business-model distributed teams.
I'd like to see some links on the productivity drop by hour.
~M
Richard Brown
playing WoW and not working whenever I've seen them :D
Doesn't work for other departments, to much lost productivity getting IT
to unblock a site because you need it to find background info or
references, both for code and art.
I could perhaps see it to block certain sites like email sites and forums
during work hours, maybe even flash to stop people playing games when they
shouldn't, but that'll depend on what kind of games your making, less
beneficial for casual makers etc.
There is always a balance though.
Steven Davis
>> new-economic/business-model distributed teams.
In "new-economic/business-model distributed teams" people are likely engaged in piece work (excuse me, milestone based payments), so their hours are not an issue. This is clearly for on-site employees.
>> I'd like to see some links on the productivity drop by hour.
Me too. It is certainly intuitively true (just as there is some minimum amount of time to get to the point of getting anything done).
Steve
Steven Davis
>> playing WoW and not working whenever I've seen them :D
>> Doesn't work for other departments, to much lost productivity getting IT
>> to unblock a site because you need it to find background info or
>> references, both for code and art.
For a studio / office, I tend to recommend a separate set of machines that have Internet access (full, unfettered). This makes it physically clear when someone is "on the Internet" - whether for work or play.
(your mileage may vary)
>> I could perhaps see it to block certain sites like email sites and forums
>> during work hours, maybe even flash to stop people playing games when they
>> shouldn't, but that'll depend on what kind of games your making, less
>> beneficial for casual makers etc.
>> There is always a balance though.
Bruce E Hennigar II
http://www.google.com/ventures/
Is this a chance for indie development studios to get off the ground and
actually be funded?
Bruce
Michael Lubker
~M
On Tue, Mar 31, 2009 at 1:24 PM, Bruce E Hennigar II
<bruce.h...@gmail.com> wrote:
>
> Just wanted to share this with everyone:
>
> http://www.google.com/ventures/
>
> Is this a chance for indie development studios to get off the ground and
> actually be funded?
>
> Bruce
>
> -----Original Message-----
> From: game-studi...@googlegroups.com
> [mailto:game-studi...@googlegroups.com] On Behalf Of Steven Davis
> Sent: Monday, March 30, 2009 11:59 AM
> To: game-studi...@googlegroups.com
> Subject: re[4]: Gamasutra article on how crunch makes Epic's games better
>
>
Eskil Steenberg
Many think that crunch is just an evil that needs to be removed by
better scheduling.
I don't think it is that simple. Here is my theory:
Anything that is possible to do, will be done on budget and schedule
+/- 10% disregarding of its schedule and budget.
7 years after 9/11 ground zero is still just a hole. The empire state
building was built under schedule and budget, in less then a year. We
have better technology, better materials, so don't tell me it is more
complicated now then in 1934. (BTW, Empire wasn't badly built, it
withstood a plane crashing in to it...). The U2 plane was built in
less then a year too, today a military plane takes 20 years to develop
and they still cant make anything better then a Hercules.
The only way to get the impossible, is to ask for it.
If you say "making this game is going to cost 30million and take 3
years" Guess what, people will crunch the last weeks. Extend that by a
year and you will just delay the crunch a year.
Games cost a lot of money and takes a lot of time to make because we
want them to.
The fact is that people crunch because they want to get more stuff in
to the game.
At GDC this year I went to a talk about Red Faction. They talked how
incredibly hard it was to build destructible environments. But if you
work hard enough on it you can do it in this generation. All i could
think was: Super Marion Bros. It was possible to do in the 80s, it was
even easy to do in the 80s, it will probably be even harder to do in
the next generation. We create these arbitrary rules for what a modern
game has to be. Things that make game development hard and expensive.
Rules like "Everything has to be bump mapped". When you corner your
developer with less money and time, you force them to innovate.
Also if i look back at my time in crunch its the time i remember the
best. No team building works as well as crunch (Often though the team
bands together against management...)
Do i think that crunch makes games better? No, but i think releasing
games do, and any reasonable deadline will produce some amount of
crunch.
Cheers
E
Bruce E Hennigar II
becomes "crunch" time because of people slacking off, "unforeseen
circumstances" and a myriad of other things that can happen, people get fed
up. You should always allot yourself a certain amount of "catch up" time at
the end or even in the middle and end of a project to get back on schedule.
If your project is going to take 3 years to do, that should include a good
amount of "oh crap" time. I don't think you should ever schedule something
in the hopes that everything goes properly, without issues. Doing so is
asking for trouble.
-----Original Message-----
From: game-studi...@googlegroups.com
[mailto:game-studi...@googlegroups.com] On Behalf Of Eskil Steenberg
Sent: Tuesday, March 31, 2009 5:26 PM
To: game-studi...@googlegroups.com
Subject: Re: re[4]: Gamasutra article on how crunch makes Epic's games
better
Adam Martin
>
> Hi!
>
> Many think that crunch is just an evil that needs to be removed by
> better scheduling.
>
> I don't think it is that simple. Here is my theory:
>
> games do, and any reasonable deadline will produce some amount of
> crunch.
Large game projects today are frequently run by under-qualified staff.
I have been more and more amazed year on year how few really good
people are in studio-level exec positions, and how many has-beens and
failures are instead.
And then there's Scrum, which neatly sidesteps your core argument:
there will be no crunch if there is no arbitrary deadline. What
benefit is the deadline, anyway?
I have a friend who used to manage strike teams for McKinsey, and
admitted a fondness to telling teams about their deadlines a mere
couple of days in advance. He called it "managing by artificial
emergency" IIRC and felt his talented arrogant teams were too lazy for
their own good and needed this kick up the ass to make their best
work. Huh. That's in an industry (consulting) where the deadlines are
a lot more real than they are in ours.
But, again, with Scrum, since you can ship any week you choose, with
zero advance warning, the external stakeholders who require a
particular, arbitrary, ship date can have that whilst not screwing up
the dev team.
Scrum done well is a perfect fit for most game developers, IMHO. And,
IMHO again, that's why it's been sweeping through our industry so
rapidly and somewhat rabidly :).
Adam
Adam Martin
>
> I don't think crunch time is bad per say, it's just that when ALL your time
I think crunch is worthless and nearly always caused by idiotic
managers who were never taught how to prjoect manage properly (it's
amazing how many so-called trained PRINCE2 practitioners declaim that
it's overly rigid; how the heck did these people qualify? Or did they
lie on their resumes? Hmm).
Crunch is where you steal free unpaid overtime from your staff. Yes,
it is stealing, because it's a breach of the employment contract where
you contract a certain amount of work for a certain amount of money.
Plain and simple theft.
Paid crunch is a very different matter. Funny how all the
institutional (*) active users / proponents of crunch never seem to
offer to pay overtime for it, isn't it? When they start paying for it,
I will start being open to the belief that crunch has some value in
making a game. Until then, Occam's Razor suggests the real
purpose/benefit is to reduce the expenditure on salaries.
(*) by which I mean the companies that deliberately rely on it, not
the people who are moderately defending it, like Bruce and Eskil
Adam Martin
>
> At GDC this year I went to a talk about Red Faction. They talked how
> incredibly hard it was to build destructible environments. But if you
> work hard enough on it you can do it in this generation. All i could
> think was: Super Marion Bros. It was possible to do in the 80s, it was
> even easy to do in the 80s, it will probably be even harder to do in
> the next generation. We create these arbitrary rules for what a modern
> game has to be. Things that make game development hard and expensive.
> Rules like "Everything has to be bump mapped". When you corner your
> developer with less money and time, you force them to innovate.
You see, Eskil, I think you secretly are wholly anti-crunch :), but
that other - positive - things have got aggregated into the term
"crunch" by people trying to create defences for their own terrible
practices, adn it's just that you see some good from some of those. I
believe you can get all those benefits better, at less cost, without
using crunch, but instead using other bits and pieces of methology to
achieve them.
You just gave a great example that suggests that crunch is the wrong
approach; more intelligent design and more intelligent project
management would be far more effective.
Eskil Steenberg
other, People who don't make anything, has no say in anything.
Do you know what you call a game developer who doesn't produce anything?
A producer.
E
Adam Martin
>
> I think I need to get back to my original rule, Hire the best, an no
> other, People who don't make anything, has no say in anything.
Nod. Again, a core tenet of Scrum :P. IME, that's one of the few parts
of Scrum that people actually remember to do.
> Do you know what you call a game developer who doesn't produce anything?
>
> A producer.
LOL.
Bruce E Hennigar II
just know better than to think that everything on a project map will go
smoothly and the way you plan for it. Regardless of how good your people
are, something WILL go awry. Somewhere. It may be a couple hours, may be a
day lost, but something will go wrong and it will push your deadline. I
suppose that's the good thing about having no deadline though, eh?
When I set out to set my own studio, I ran through all the different ways I
could pitch the way my studio was run, but in the end I kept coming up short
because the idea of a studio who asks for the money and expects the
investors to trust them without deadlines is too new-aged. Or so I thought.
Is it feasible to say "here's out idea, here's what we plan on doing and we
need X amount of money to do it," and completely leave out a timeline? If
they ask is "we aren't working on a deadline. When it's finished, it's
finished," a reason response? Blizzard seems to think so, but Blizzard's
products are funded by their success already. I don't think they have an
outside publisher breathing down their necks for their main products. I know
WoW had an outside publisher and look what happened at launch? They wanted
to hit the deadline so much (as is the fault of nearly ALL MMOs released
nowadays) that they sacrificed bits here and there, even scrapping entire
ideas just to get it out. It sucked at launch. There was no stability.
As to the other comments, I've seen what happens when you ask people to work
for no recompense. Unpaid overtime is a sham and should be illegal if it's
not already ;) I ask my people to work as much as they want and hope that
I've hired people who have the drive to make the vision a reality. But I
would PAY them for their productivity.
I dunno, there's so much to say but I'd rather somebody just take the
initiative and SHOW the big companies how it can be done. Show them that
straining deadlines stifle creativity and result in the lowest common
denominator, quality-wise.
-----Original Message-----
From: game-studi...@googlegroups.com
[mailto:game-studi...@googlegroups.com] On Behalf Of Adam Martin
Sent: Tuesday, March 31, 2009 5:40 PM
To: game-studi...@googlegroups.com
better
Adam Martin
> could pitch the way my studio was run, but in the end I kept coming up short
> because the idea of a studio who asks for the money and expects the
> investors to trust them without deadlines is too new-aged. Or so I thought.
> Is it feasible to say "here's out idea, here's what we plan on doing and we
> need X amount of money to do it," and completely leave out a timeline? If
> they ask is "we aren't working on a deadline. When it's finished, it's
> finished," a reason response? Blizzard seems to think so, but Blizzard's
The first 3rd party dev studio I worked with when I joined a publisher
was using Scrum, and saying exactly that.
The problem was that they were producing crap every single milestone,
and their sprints were achieving nothing.
At the time, the publisher didn't understand Scrum either, so wasn't
sure what to do. With hindsight, having had a lot more experience of
Scrum, I believe it would have been obvious for all concerned: judge
the team on the results, not the excuses.
(in fact at the time we did judge them, and hard, on the lack of
progress, and refused to accept this hand-waving about "we're using
this new thing called scrum" as an excuse for their poor builds. But
that was based on common sense and a feeling of "WTF is going on here?
how come their builds are so terrible? and not even slightly fun?")
So. convince an outside stakeholder/client/investor that this is
exactly what will happen - that they will have very early visiblity at
all times, and that they have only to object at any time - and I think
you'll have them on your side. You'd need suitable restitution
clauses, of course...
> WoW had an outside publisher and look what happened at launch? They wanted
> to hit the deadline so much (as is the fault of nearly ALL MMOs released
> nowadays) that they sacrificed bits here and there, even scrapping entire
> ideas just to get it out. It sucked at launch. There was no stability.
As has been said quite a few times, stability of an MMO has never
truly been a major commercial issue, except in the minds and dreams of
developers, and in extreme cases of instability at precisely the wrong
moment (AO refusing to accept credit-card orders being one of
those...but the point was it was more important "how easy is it to
spend money?" had gone wrong rather than "a server has gone down").
Adam
Eskil Steenberg
> initiative and SHOW the big companies how it can
> be done.
That's the spirit!
My feeling is that the way to do it is not to remove crunch, but to
remove "release". I release to my alpha group every two weeks or so. I
never feel that a release is the one i have to get everything in to,
so no crunch. Then again i already work 7 days a week...
Cheers
E
Adam Martin
> I never feel that a release is the one i have to get everything in to,
> so no crunch.
Exactly :).
Richard Brown
Eskil Steenberg
> HR and the boss...don't forget them :D
HR: How the hell do you judge a programmer without being one? I'm a
programmer and I still think its hard.
Boss: How do you get respect and the knowledge you need about what you
do to be a boss without diving in and working on the project?
Sorry, EVERYONE who makes a decision needs to work on the projects. I
can imagine hiring someone unqualified to answer the phones, fill the
coke machine, arrange plane tickets and look good in the lobby, but
only barely.
Cheers
E