Right, exactly.

That's been the problem. We haven't really dove in to the forum discussions because there haven't seemed to be a lot of folks there who really want to discuss facts. I mean, just watching it. Most people didn't watch the free video available of the panel [Ed note: embedded below] to see and make judgments for themselves, so that was kind of disappointing. So the panel was a leadership panel of five different CEOs. It was the last session. We were drinking wine and kind of taking different stances on various topics. It turns out at least three of the people, maybe four at the time, were on the IGDA board. What's sort of coincidental is that the IGDA board had our meeting at the forum so therefore those were the CEOs who happened to be there at the event. If that makes sense?


Honestly, I'm not sure which of the various things that got everybody so upset. I think the main one was that if someone walks into the door and says, "I refuse to ever work past 5pm, I'll never work more that 40 hours a week and you can't make me", they're probably not a fit for us. Just the same way they wouldn't be a fit, I assume for you, if they said, "Well, I'll do E3 but I'm out at 5 and I'm not writing any articles till the next morning." Or a lawyer or a doctor saying, "I don't deliver babies after 5 and I never go out after 40 hours a week." I mean, our average number of work hours is what, 49, 50 in the US? So to have someone walk in and say they refuse to ever crunch for an E3 demo, it's kind of silly. It just shows that they're probably not passionate about what they do. That's very different from saying that we force people to work hard all the time.

That said, do you guys have stats for the number, average number of hours worked, let's say on a project, on a single project, for an Epic employee?

I don't think we track average stats. We require 40 hours a week here. We've got three rules on any given day: You have to work eight hours in the office. You have to be here between 1:30pm and 5pm since that's when we have all our meetings. Some folks crawl in at 1:15 and some folks come in at 7 am and are out by 5. Kind of depends on personality. And then you have to be out of here by 2am. We don't let people work too late because it starts causing problems where they'll roll their hours over to the next day, that sort of thing. Honestly, the rule I have the most trouble here with these guys is kicking them out at 2. That's the one that pisses folks off. It's not the 8 hours a day, it's the 2am and I'm still working and I'm on a "I've got a bug by the tail and I want to finish it." And we'll have someone going around banging on doors, kicking everybody out because they need to go home. During crunch time we definitely pick it up. I don't think we have stats but I know we were on 12 hour days, five days a week for Gears 2 for, I don't know, maybe six weeks. Something like that.

During those crunch times, is that 2am stop time still enforced?

Yeah, absolutely. Because it's dangerous, right? You don't want people writing code who've worked more than ... 12 hours a day is the most you can ever really get out of a programmer anyway. And you can't sustain that, right? There's no way you could work someone 12 hour days, 250 work days a year because they just, they'll break down. It's crazy. So, you know, even logic dictates you don't work people that hard but, in general, we crunch because we've got a sexy E3 demo that we want to get out. Our global illumination team wanted to show off our cool feature for GDC and they wanted more polish so those guys, the couple of them on that team, crunched for two weeks before the show. Nobody ordered them to come in extra hours or anything but it's kind of a pride thing.

Games like Gears, you know, it's one of the best reviewed games of all time on the platform. You don't get a game out like that with a bunch of people who don't have any passion about the quality of the product and don't want to spend that one extra night. We spend nine months before we ship a game saying, "It's a marathon, not a sprint. Hold back. Please don't give me everything you've got right now cause your gonna burn out." I mean, that's the biggest job our producer does – sending people home when he sees they're too tired and beating down that mantra of "It's a marathon, not a sprint. I don't need you to sprint right now." And then two months before we ship it – a month and a half – it's a fucking sprint and everybody goes. And we let them go. You know, that's the time we pull off the chains and everybody runs as fast as they can. We push hard. We leave it all on the field. And then it's done. By that time, we know when it's shipping so we've got a fixed end date in mind and so everybody knows "I can work this hard for a month, month-and-a-half, because I know I'm gonna get a break after that. You know, I'm gonna take a week off and then it will be slow after that and we'll work on DLC or bug reports or whatever else."

"We basically have one voluntary departure a year, or something like that."

One good metric to examine the participatory element of your crunch cycles with your staff would be your turnover rate. How is the turnover rate at Epic?

In 2006, our voluntary turnover rate was 1.3%. In 2007, it 1.1%. In 2008 it was 1.03%. We've got 110 folks here, so you can figure out what that means. And we've staffed up from 80 to 110 the past couple years, so we basically have one voluntary departure a year, or something like that.

And that's below industry standards?

Shockingly below industry standards. I think 12% to 15% is sort of the standard in the tech industry.

So, why do you think that is? Coupled with accusations of overworking, why do you think developers at Epic are less likely to leave than at another development studio.

You know, I wish I knew for sure. We try to track – we use anonymized climate surveys here – to try to track what's making them happy, what they like and don't like. How are the benefits? Do you feel our vacation and time off is fair? And we track year after year and when we see that people feel like – I don't know – our life insurance policy isn't making them happy – only 85% of the people like it – we figure it's a mix of they don't understand what they're getting, and we can improve it. And we work on that.

"The average person here made more on bonuses than they did in salary for Gears 2."

So my climate survey for last year, I think 97% of the people here felt that they were getting the same benefits they'd seen everywhere else, in terms of time off, vacation, sick, and so folks really like the benefits. And I don't know! That's all I do man, is try to make people happy here. Everyone at Epic is a volunteer, right? These are the guys that work on the core technology that hundreds of teams worldwide are depending on. So any tech artist here can walk into any company in the world and be heralded as a savior, right? So to keep them here, I pay them much better than most. The bonuses we saw on Gears 2 probably were more than – well, I know were more, on average, than the annual salary – not just for the top people in the building, but if you take off the top ten best paid people in the building, and it's still the average person here made more on bonuses than they did in salary for Gears 2. And we've only seen royalties for 2008. We haven't even seen this year's royalties. Gears 2 is still selling strong.

So, in other words, there's a financial incentive if you're working on a big project. If people know or are confident that that project is going to be successful, then they know that they'll be compensated well for crunching.

Yeah, exactly. If our next game doesn't sell well, they won't get bonuses. They understand that. Everybody here is incentivized. All of our money stays in the building. There's no "send it off to the ownership" or to some fund or something like that. They're paid extremely well and we're very creative about spending money rather than just giving out money. You know, spending on benefits and finding new ways to make the benefits here better.

But, honestly, these guys get to work on games they really want to make. There's no dog product here at Epic, right? You work on the engine and your code is seen by thousands of programmers and it affects hundreds of games all over the world. That's awesome. Much less your own products. And so our artists are doing the kind of games they want to do, our programmers are writing the kind of code they want to do and then, they're all paid really, really well for it. There's a lot of creative freedom here. I think it's a mix of all those things. It's the games you want to be making, it's a studio that has an excellent reputation for quality. People love not having to squeeze a game out in time to please the publisher. We get to decide when our games ship. And then the financial rewards have been – knock on wood – fantastic and we don't really see that going away. People aren't leaving because I think they love it here.

There's still a larger, industry-wide issue that I think – maybe the sensitivity to which – is responsible for this "kerfuffle" of crunch time ...

Agreed.

... of overworking employees. The IGDA, of which you were a board member, has a Quality of Life White Paper, all about how working less hours is not only better for the employees, it's better for the industry and better for the products. So, first off I wanted to know if you've read the white paper, and if you agree with its findings.

It's very difficult. Let me tell you a story: I had somebody come up to me after the leadership forum and basically curse me out for saying that it's okay to crunch when you need to, to make a great game. He said, "It's not because your guys aren't treated well, because they are. It's not because you're not sharing the royalties with them and everything else and they don't see why they're doing it."

My guys ask to crunch. They say, "Hey, we're not crunching yet. What's going on? Why isn't everybody crunching? This is really serious!" That kind of stuff. He said the problem was his management team would point to what we do and say, "Oh, see? Crunching is okay because they're doing it." Then that company would not be treating their employees the same way and would not be making a product based on passion rather than schedule and that sort of thing. Using what we said as an example to mistreat people. That was kind of a scary thought.

So, I definitely am trying to be a little bit more cautious about saying how we do things here and trying to caveat it a lot. It works for us here because we have highly skilled, experienced folks. Average is six titles shipped here. So, these are senior people who know what they're doing. They know their own limitations. Then we make sure they're rewarded for the work that they do.

But yes, I'm familiar with that [white paper]. In fact it's one of the reasons that I joined the Board in the first place. Because when I ran for the Board it was right around the time of "EA spouse" hitting and there were certainly organizations that were not taking quality of life seriously. But I thought that the efforts of the IGDA SIG task force were really misguided. I mean they wanted to grade employment contracts, and put out, you know, "Epic, it's a B- for its employment agreement." Or quality of life: "Ubisoft Montreal gets an A- for quality of life because they have five weeks of vacation instead of three" or whatever. I felt that sort of objective measurement, and then publishing that nobody should want to work at, say, Big Huge because of their C- rating in quality of life, was a bunch of crap.

"You could come up objective measurements here for Epic and say that it's the worst place to work."

I really didn't like that. You could come up objective measurements here for Epic and say that it's the worst place to work, except we win Best Places to Work in North Carolina awards and employees never leave because they love it. You could come up with another set of criteria that says that Epic is the absolute best place to work. So, just the notion of objective criteria doesn't make any sense to me. Yes, that's part of the reason I joined. There really wasn't any traction with that Quality of Life effort. It hasn't gone very far in the past couple of years, unfortunately. That's it.

Do I think that you can get the same amount of work done and the same quality of products made if you restrict everyone to nine-to-five all year round, all the time? No, I sure don't. I think that's reflected in any creative industry in the US or in the world. That's not how people make movies. That's not how they make records. And that's not how they make video games. It just doesn't make sense being upset at us for saying up front, "This is the way we do it." Telling our employees before we hire them that this is the way we do it and then they stay here and they love it. Everybody points to us and says, "You guys are evil because you're propagating some negative management technique." It's kind of insulting really.

So, I am glad you brought up "EA spouse." "EA spouse" was the first scenario that brought all of this to the forefront and to the attention not only of the industry media but also the consumers who play the games. As you at Epic are no doubt very aware, as now-gen games started taking off, the increased amount of work that was required from publishers and developers found them unable to cope with the demands of these systems. So, this has obviously opened up a huge opportunity for Epic in the middleware business.

Right.

But you guys are also similarly tasked with satisfying a lot of external companies. So, one of the big things I see with companies and crunch and whether or not it's even necessary -- the holiday release schedule.

Yes.

Increasingly in the last several years, it's gotten more and more outrageous with more and more triple-A games being stacked on top of each other in a two to three month period.

There were some fantastic games that came out in holiday and just they were the number six best seller because there were five awesome sellers too.

Well, there were worse scenarios where there were some fantastic games that didn't show up on the top ten or twenty because they didn't get enough press because they were put out to die.

And you can't buy the marketing, right? Because it's already been snapped up at ridiculous prices. And I agree that holiday-centrism is terrible for us.

So what are your thoughts on, or I guess, more elaborate thoughts on the holiday release schedule – besides that it's terrible – and whether or not that has an affect on Epic as the provider of middleware for a lot of the games that are coming out during the holiday season. And do you think a balanced release schedule would help ease the burden on developers, not only at Epic but in the entire industry and help to reduce crunch time?

"It's all balanced around holiday. That's where the dollars are."

So, yes, the holiday schedule causes a major impact on how people design games. They are thinking: "I need to finish this at holiday." And unfortunately software products are notorious – especially entertainment-based software products not just games but all of it – are notorious for not being easy to schedule to land on time. So there are two things you can do: you can miss holiday or you can scramble and work really, really hard to try to make holiday. Unfortunately, I don't think there is anything we can do about it. I don't think that if all the publishers banded together and decided that they were ship only 1/3 of their titles at holiday sales would spread across the year because, unfortunately, the way our consumer system works here – at least in the United States and a lot of Western Europe – it's all balanced around holiday. That's where the dollars are. So, sure, would I love to be able to ship a game in April and do just as well as I do in holiday? Yes. Do I think that that's the case? Really only for a couple isolated incidents has it worked well to ship outside of holiday.

I look at certain publishers like Capcom which seems to have – maybe three plus years running – this uncanny knack for releasing games in Q1 and doing really well. And Grand Theft Auto 4 was able to sell an immense amount of units even past Q1. Bioshock was released in August, and was a huge hit.

Yeah, but if you took the Dead Spaces and the COD 5s, take three games that sold over a million units out of holiday and put them in February and then see how well other February titles do, right. I mean part of the reason they do well there is that there aren't a lot of other titles launching.

Right.

"If I'd shipped Gears in April, would I have done better? We don't know so, unfortunately there is a lot of fear to move out of holiday."

For instance, if you just put the games that ship in holiday and spread them out over the year, then I think you would see sales diminishing for the titles that are currently shipping out. It's all a "who knows?" If I'd shipped Gears in April, would I have done better? Maybe? Maybe not. We don't know so, unfortunately there is a lot of fear to move out of holiday.

Well, the big question there, with something like Gears. Grand Theft Auto 4 can sell 10 million units in the off-season. And then you have a title like yours, Gears 2: huge launch, huge anticipation, huge built-in audience, so if you were to release later in the year outside of the holiday season, you'd be more in the Grand Theft Auto-type of scenario. Where you would have a audience, built-in press, you're going to have the magazine coverage, you're going to get the coverage regardless of when you release.

Right. So, yes I agree that when you've got a major franchise you're kind of lucky enough to be able to go into a different time of year. And you will clear out space, right? I mean if we announce that we are shipping Gears 2 in April, I think, I mean if we announced early enough publishers would know. Okay, if I'm going to go outside of holiday, I'm not going in April. And nobody shipped on top of GTA, at least nobody did it on purpose, we all knew when it was coming.

I think a couple did.

Yeah, and how is that going? Because I sure didn't want to ship on top of the Warcraft [Wrath of the Lich King] expansion. I ended up doing that because they came into holiday with us. I wish they hadn't, but at some point you pick a date and stick with it. No it's a great point. Could we move out of holiday and be successful? I'm not sure. I know for Gears 1 it was crucial for us to be in holiday because we were part of the platform strategy for selling Xboxes. And they wanted to move Xbox selling in Christmas, and that's you know, they sell a lot more platforms in Christmas than they do the rest of the year.

For Gears 2, sure we probably could have moved on from holiday, I'm not sure if it would have done better for us but to get back to your original question, there is an emphasis on holiday whether or not it makes sense and yes that absolutely impacts schedules for studios. As a middleware provider for us, yes, in June, July its crazy from the support side. There's tons of people shipping games that holiday who just realize it's not going to work the way they built it and so, yeah, it's certainly busy for us but right now it's the nature of the biz.

At Epic, I'm curious about how things work internally. Certain teams go into crunch. Let's say they get the E3 demo done. At other developers, we've heard cases of devs being rolled off of one project and right into another and then right into another, kind of constantly picking up the slack on crunch time or crunches on different projects. So they are kind of in a perpetual crunch mode.

That works about as well as you'd think it would.

I would imagine so.

If you scoop someone off the front lines of Afghanistan and fly them to Iraq and you put them back out and you keep doing it, performance is going to suffer. No. We don't ship so many products that that's a problem for us. The guys who suffer the most are the gameplay programmers assigned to a game like Gears because that game ships and then they've got about two weeks of wait-and-see time while it's in certification where it might pop out. So they can't go to Aruba, but they can go home and sleep. And then if it pops out, they've got to spend either two or three concentrated days to fix whatever bugs came up in certification and send it back in. And then they get to rest some more. Then we start having first play results four weeks later. Once first play hits, God knows what could happen. Oh no! The shotgun's broken! You can cheat with it or something. So everybody has to scramble and get a patch out. So they probably have the worst of it, but still I think you hear a couple "Wait a few weeks. Wait a few weeks." And during that time, they're doing R&D for the next game or cleaning up the shortcuts they had to take and doing documentation.

We don't ship games fast enough around here that we have to roll people from game to game to game. Bigger studios have that ability, but of course they have the responsibility to not do it. You can't put someone on back-to-back crunches. I've had one person who ever had to do that and that person is a workhorse, knew they were taking it on, volunteered to jump on the crunch for another product because he felt really strongly that he could make an impact on it. We gave him weeks of downtime after that. That's something we avoid strenuously here.

Regarding the IGDA quality of life uproar, what could, or what should the IGDA do regarding the quality of life of developers, if anything? You said you ran for the Board in part to address this problem and don't think that it has really been addressed. Do you think that this is much ado about nothing industry-wide or just much ado about nothing in terms of specifically Epic Games?

That's a great question: what could we be doing? I think the IGDA is not a union. It's not there to protect the workers from evil companies or something like that. I think some people wish it were and so those are the people that got upset when I said we crunch when it's time to get a product out the door, but we're careful about it and we make sure we reward the guys. All they heard was crunch and IGDA Board of Directors and they said "he's the one who's supposed to protect me from evil crunch." But that's not the IGDA Board of Directors job. The organization is to try to share the best ways to make games and to take stances on issues. If their stance is that crunch is a very bad thing, they should be actively taking a stance on it. But they're not. The IGDA doesn't feel that crunch is a bad thing.

It's exploiting people. It's mismanaging people. That is a real problem and taking a stance against that. So the Mythic pop-up, you know when they were not crediting workers who left even though they put the time in, that makes a lot of sense for the IGDA to be speaking up about to say people should be credited for their work.

I don't think that the IGDA should be developing credit standards that specify if someone's work is at least 8% of the product life cycle that I should give them a full credit because I'm not going to give them a full credit. I don't think it's appropriate. I'm never going to sign onto that and never will other publishers. I think as soon as you start getting into specifics of here's the way the IGDA mandates how credits, or lifestyle, or quality of life should be done, I think they're getting into a real problem because they are an international organization. Saying that the way this MMO in Korea gets made should be done in the exact same way and credited in the exact same way as one in Iceland, as one in North Carolina. It just doesn't make any sense.

I think it's more that they should be looking out for the interest of the developer and advocating strongly when they see someone doing something wrong. I don't see the IGDA at all saying we've done something wrong, except for one board member who is very spoken out in disagreement with the way we do things. But that group doesn't think we're doing anything wrong so I'm kind of surprised to see the flap a month after that panel went off.

You said the IGDA is not a union. In terms of a union for the video game industry and video game developers, what are your thoughts on that? Would you be in support of one? Do you think that there is value for one? Do you think there is a place for one in the market and in the video game industry at least in the U.S.?

Well it's really tough, because I don't think a union would significantly change the way that most companies do things today. I can imagine there were some really rough situations that came out over the past couple years of companies not treating employees well, and I guess unions could have stood up to that. I'm not a big fan of unionization overall, but I can see how it could have helped in some situations. That's dicey. I definitely don't think an international union makes sense, so it's not the IGDA that should be involved. And my big concern is – I'm on the ESA board, right? So that's a really fun meeting where I'm sitting there with the head of all the major publishers in the United States. And we're all worldwide publishers, and we were talking about, one of them had a major large studio in – I'm trying to tell this story without identifying who it is – but say, in a Western European country. And that Western European country changed their laws on vacation and overtime and crediting all pretty rapidly in a row, and so they took those 1,500-odd people and just moved them to another country. Which is devastating, right? I mean, that affected a lot of people, I'm sure, and many people didn't move with the company, looked for other jobs, had to move. I don't want to see all of the work moving out of the United States because this is the place where you've not allowed to crunch. Epic would move to Canada if we were told that crunch is illegal and you can never work someone more than 40 hours a week, because our products would not be as good. Once it finally all comes together and we want that last little push at the end, I don't want a union representative leaning over my shoulder, telling my guys who want to be there working that they're not allowed to. That's silly. So I think a union could be really damaging to the games industry here in the United States, just because I think you'd see more work going overseas, and that's already a trend right now.

And you don't want people avoiding working with American game companies. I mean, you already hear it, for example, France with 8 weeks of vacation. There are folks who refuse to work with French game companies, because they're going to disappear, you know, all of July and August, and everything shuts down and it becomes a real problem. Obviously there are amazing products coming out of France, but there are folks who just refuse to do it.

It's not surprising that [French-based] Ubisoft is investing so much in Canada.

Yeah, and a lot of their really great products are coming out of there right now.

Montreal is their real bread and butter right now.

That and the 30% tax credits in Montreal doesn't hurt either, right? It's the same thing, but its legal structure has a massive impact on where game development is getting done and how it's getting done.

It's like Delaware for banks, but for games.

Exactly, yeah. We've got a couple of corporations in Delaware, it's just one of those things that you do.

It's amazing how Montreal is apparently becoming the North American capital of video game development.

Yeah, really while no one was paying attention.

Products aren't done when you ship them, not only in terms of that old chestnut of, "You're shipping a broken game!" but with everything from DLC to ongoing content. So you're not only creating new content and releasing it, you're fixing old content, let's say if the game actually is broken in any way, and you are modifying or tweaking or evolving existing functionality like multiplayer. So as a game goes through crunch, like you said, then it ships and there's a certain amount of downtime. But in reality it's a living thing and it's out there now, and you need to have people on staff who are able to respond to complaints or criticisms, cheats and bugs. And then you have other people who are still working on DLC, so how do you guys address that internally, when you do have to get that team onto new games but they're still supporting and working on old games?

Well, the cool thing is we've doing that for 18 years now. I mean ZZT, the first game we shipped, was actually an engine which had the ability to make new content with it and Town of ZZT was the game that Tim made on top of it and so we've been maintaining a community of people making games way before Unreal came out in '98 and that's 7 years prior. We're sort of used to it. If you think about Unreal Tournament, we shipped what, four bonus packs over a year and a half, two years. I couldn't tell you how many patches and updates to that, so that was 99 to 2001 or 2002, something like that, where we shipped tons of new content. I just shipped a gig of content for free for Unreal Tournament 3 and that game shipped in 2007, right? So that how Epic works. We're used to it. We build into the schedule the additional content. We build a lot of it before we ship the game in the first place and then we keep maintaining it. You don't have to crunch to ship bonus pack content.

There's the thing. I mean I want work for everybody to do. You don't want someone to work really hard, get out a great game, take a break and then come back and there's nothing to do yet because the design team is scratching their heads and staring at a blank piece of paper trying to figure out what the next IP is going to be. It's really natural and easy to have artists come back and start working on downloadable content or additional levels or whatever it is, bonus packs. So, it's actually really good for us, especially as a small studio, right? When you're bigger company, Ubi Montreal is a great example. They've got a pre-production team working on lots of games and once all the artists finished with whatever title, they're immediately able to move over and be productive on another game that's in the middle, the main chunk of production. For us, as a smaller studio, we're not a two-team studio. We've always been one-and-a-quarter, one-and-a-half, so some of our guys work on DLC and new content. Other people are working on prep-production on a new title, that kind of thing. The only time it gets tense if when it's a bug, like you say. When we find out there's a real problem in multiplayer or something like that and then we've got to get it done. We've got to get it over to cert right away. So other than that it's generally just a good job.

Like you said, the forums are on fire and have been for some time and there a huge amount of internal discord going on in all sorts of IGDA venues about that specific quote, that specific situation and Epic in particular. Why do you think that is and why this scenario in particular?

Golly, I need to say I wish I knew, but I know there's people who are really sensitive to overwork and I get that. There've been people who've had a really bad experiences and we've heard the stories, the EA spouse horror stories about folks being worked way too hard and never getting any time for their families, that sort of thing. Our guys vote on how they want to crunch and last time they chose having weekends off, so you could spend a whole weekend with their family and recharge. Other times, they've chosen six days a week, but fewer hours, you know that sort of thing. And hearing some of the horror stories, that's just frightening. How do you work someone a 100 hours a week and get anything useful out them. It doesn't even make sense, right? So, I get it. I get why people are scared and upset and worried about abusive employment. I understand that.

Why this particular issue came up with it? I think probably has a little bit to do with me being so confident that we are doing the right thing. I think they want everyone who crunches always to be contrite about it and say, "We're so sorry we fucked up, we didn't mean to let anybody work more than 40 hours a week. This is a huge disaster. We'll never do it again." We don't feel that way. We're very confident we did it the right way. We've done it 18 times and we're going to do it a 19th time and everybody here knows we're going to do it a 19th time. We're not lying and saying "Oh, we blew it!" I think that really bothers folks who have seen people do it, overwork someone, and then when we say "Yes, we crunch and we burst sometimes," confidently, like we know what we're doing and we're sharing that message. Because I really think Epic knows what it's doing and we've proven that again and again. We've been very successful. I think that really upsets people. It's not like we always crunch and that's the message to get across is when we do, we do it very carefully, rationally and we think it was the right decision. We don't regret it at the end. I had one or two that were, wow that went too long, we had a rough time, we made some mistakes in planning but that's not to say that crunching is the wrong thing to do.

I actually did a talk the year before on how me manage the company. I gave an hour long talk [video linked below] about how we manage people and performance and it included basically all these same topics and it didn't bother anybody that year. I don't know what it was about the panel the following year.


Fascinating to hear that you vote on how the team is going to crunch, which seems like a sensible way of doing it.

We have all of our leads in a meeting. Generally some of them bring up "It's time we should really think about crunching." A lot of my guys are already working late we should do the whole team at once. Because the last thing you want is the guys who are working late to be looking at the folks who aren't saying "How come they're not?" and getting the whole "Doesn't anybody know we're in trouble here? If we're going to get this feature in we've got to push push push." So the leads will get together and we will unanimously agree to do crunch at that point and that's when we take feedback from the team and choose how to do it. And we do it as a team, always and we'll only shut down entire teams at a time. So the whole Gears 2 team will crunch and then we'll say, "Okay, art is done. They're content complete." They're not even allowed to touch anything at this point. So they're off crunch now. Except for bugs that might pop up and we'll have one or two people assigned to on call for bugs. Then we'll shut down the level team and then finally the code team as we go through. So, we do it as a group. I think that's pretty important.

Thanks for the time, Mike!