Gamasutra: The Art & Business of Making Gamesspacer
View All     RSS
September 17, 2014
arrowPress Releases
September 17, 2014
PR Newswire
View All





If you enjoy reading this site, you might also want to check out these UBM Tech sites:


Analysis: Are Long Development Times Worth The Money?
Analysis: Are Long Development Times Worth The Money?
September 10, 2010 | By Chris Morris

September 10, 2010 | By Chris Morris
Comments
    33 comments
More: Console/PC



[In this Gamasutra editorial, editor-at-large Chris Morris looks at Rockstar's Red Dead Redemption and its much-delayed development cycle, examining whether taking your time in completing development on your game makes creative and business sense.]

Before Red Dead Redemption became the year’s best-selling console title to date (and an earnings savior), Take-Two Interactive Software took a lot of heat over the game’s five-year development span.

To be fair, it seemed legitimate at the time. Take-Two (and especially its Rockstar division) has never been the speediest of companies when it comes to getting games on store shelves.

By the time Redemption finally shipped, expectations were low among investors (and, to some extent, players). Ultimately, it proved to be the catalyst that Take-Two says will propel the company to report its first non-GTA year profit in a decade.

So it’s time to turn the table on those critics. Are extended development cycles, in fact, justified?

It’s not a black and white matter, of course. Was Daikatana’s extended build time worth it? Obviously not. Will Duke Nukem Forever be worth a 13-year wait? It’s hard to imagine, no matter how good the game turns out to be, but we’ll find out next year.

The list of titles that have stumbled after long development cycles is extensive. Games like Tabula Rasa and even Alan Wake spring to mind. But for every horror story, there’s a Half-Life 2 or Grand Theft Auto IV that resets the bar for the industry.

Investors, as you might guess, don’t care too much about that sort of thing. They care about sales. And while a slow simmering game that turns out to be a smash is great, in the back of their minds, shareholders still do the rough math to determine how many other titles could have been made in that time – and whether they, collectively, would have had a higher profit margin.

It’s a problem that has haunted – and will continue to haunt – Take-Two. While investors are a bit more patient with games that have Grand Theft Auto in their name, they get antsy when something like Mafia II takes six years to complete. And they’re starting to feel the same way about Max Payne 3.

The issue boils down to art versus business. Great works take time, but for every year a title is under development, it becomes harder for a studio to recoup its investment. And these days, as work is about to begin on the first round of titles for next generation console systems, publishers are likely to see their costs spike once more. And that could make it harder for them to justify giving a marginal or questionable title more time to bake.

Activision, while it has its fair share of detractors, has come up with a viable solution to the extended development cycle issue. By rotating a franchise through multiple studios, developers have time to craft their game, while the publisher keeps its investors happy with regular releases of major game lines. The downside, of course, is the quality of the game is less consistent.

It’s a model more companies are starting to pay attention to, though. In July, Capcom announced it plans to increase the number of titles it ships each year – by cutting development time on its flagship titles, such as Resident Evil, from four years to two years.

The move’s meant to stabilize earnings after Capcom saw profits drop 73 percent in fiscal 2009 and will likely result in a fair bit of outsourcing to external developers.

Capcom could be the first of several publishers to go this route. As game development costs continue to rise – and the industry faces increased competition from external forces, such as Apple’s iOS devices and the expanding footprint of social gaming – it’s going to be harder for executives to justify cycles of four, five and six years unless every single game that takes that long is a blockbuster.

For developers, that’s undoubtedly frustrating news. The race to set new standards of excellence is a marathon, not a sprint – but bean counters are only concerned with the bottom line, which is inherently focused on the short-term.

Stock prices throughout the video game sector are flat, though. And some iPhone games, like Angry Birds, are starting to become as well known as select console franchises.

That’s going to keep those bean-counters in power positions -– and could continue to reshape the timeframes developers are given to make games for years to come.


Related Jobs

Cloud Imperium Games
Cloud Imperium Games — Austin, Texas, United States
[09.17.14]

Lead Network Engineer
Cloud Imperium Games
Cloud Imperium Games — Santa Monica, California, United States
[09.17.14]

Animation Programmer
Cloud Imperium Games
Cloud Imperium Games — SANTA MONICA, California, United States
[09.17.14]

Art Director
Cloud Imperium Games
Cloud Imperium Games — SANTA MONICA, California, United States
[09.17.14]

Audio Director










Comments


Robert Marney
profile image
To be fair, Capcom's heyday came when it was pumping out sequels and updates like clockwork. When your fan base will buy a new Street Fighter every year, why not make them one?

Alex K
profile image
When it is costing games millions to produce, and that high cost is being passed to the consumers, perhaps cutting back on development time would help shave off some of that cost. There has to be a way to make high quality games without emptying out pockets.

Ujn Hunter
profile image
But... the whole point of this (and every other) industry is to empty pockets.

Tom Baird
profile image
But cutting back on Time cuts back on quality and features. Either that or you have to have more staff (and therefore more money).



Companies are generally good at trying to cut costs. If you want to spend less on making games, you would have to accept that you are going to get less as a result.

David Delanty
profile image
The price of games is not relative to the budget of those games. In the 90's, back when a friend and I started a lemonade stand for four weekends in a row to raise enough money to buy a shared copy of LucasArts' 'Tie Fighter,' we had to rake up fifty-five bucks. The price since then has only gone up five dollars (roughly 8%) but the budget to produce games has increased by way more than 8%. Very likely, way more than 100%. Some titles nowadays, 1000%.



But they're all still sixty bucks at retail.



When Tie Fighter came out, a cup of coffee was 45 cents. A gallon of gas was one dollar. But game prices have remained very constant for almost two decades.

[User Banned]
profile image
This user violated Gamasutra’s Comment Guidelines and has been banned.

Mathieu MarquisBolduc
profile image
Quality games that makes a big profit pass by methodical game design and proper management, two things that are often lacking.

Andrew Glisinski
profile image
Totally agreed.

Alan Rimkeit
profile image
Truer words were never spoken. Valve, CCP, Bioware, and Blizzard have all proven that in spades.

Jose Enrique D'Arnaude
profile image
Long development cycles increase risk a lot, but if you have success on your goal, it will turn out to be the most profitable business out there in the gaming industry. Blizzard, Rockstar, Bungie, etc. all these studios share the same concept and make the most profitable games in industry. Yet that doesn't mean it has no risk, from my point of view the risk of that attitude for non known titles is too high, but rewards pays back if successful.

Michiel Hendriks
profile image
"As game development costs continue to rise"



Why is that?



Yes, the console became more complex a bunch of years ago. But that's the past. Development tools are constantly becoming better, and thereby reducing development time. These things should cancel eachother out, right?



I simply cannot believe that game production costs are still rising.

Andrew Glisinski
profile image
Pixar takes 4-5 years of time to produce 1-2 hour films. Other film studios take just as long or longer. (Some) Bands take 5-7 years to work on an album of 1-2 hours.



Games range from 5 to 40+ hours of (story-based) content. And you want to cut the cost of production and time frame?



Honestly, I think 5-6 years is the very least a studio should spend on crafting a game.

Christopher Braithwaite
profile image
So? TV production studios produce 20 - 24 hours of content in less than a year. What does this tell us? It tells us that using an hour as a metric for entertainment value is useless.

E Zachary Knight
profile image
Actually, most films are filmed in less than a year. If the film is special effects heavy, it will take a while longer to add those in, but not often does a film take more than a year or two.



Pixar animated films take so long because they have to create and animate models that are far more polygon heavy than those found in games. They add a lot of facial movement, hair, clothing, liquid effects etc that really add in to the development time.



But I do agree, hours produced is not an indicator of how long it should take to produce the entertainment.



With the proliferation of middle ware in the games industry, it really shouldn't take as long as it does to make some games. I understand that some developers like making their own engines and I am fine with that, but what really bugs me is when they insist on doing so for every game they make.

Ryan Duffin
profile image
Yes, there is is a higher quality bar for feature animation then games but Pixar animated films mostly take so long primarily because they are religious about pre-production, something the games industry could do a lot better at. 5-6 years is the total time but much of that with smaller, cheaper teams ironing out story and then R&D before actual production begins. They figure out what they're making while the stakes are low.



Games on the other hand, often rush into full production with a full team only to change course multiple times, sometimes quite dramatically, throwing away a lot of time, effort, assets and money in the process.

John McMahon
profile image
But they get to reuse assets like sets, actors, and audio. They then spend the majority of the money on SFX and guests. That is similar to reusing an engine between studios. the problem becomes the feeling of "ownership" by a studio like Infinity Ward and less teamwork between the studios on the engine's development. Which increases risk

Francis Page
profile image
OK, 5 years to make 1-2 hour of entertainment. So, a 40h game should take at least 100 years to complete. Simple math.



Now, imagine how long it should take to make the next tetris-like never ending puzzle...

Jane Castle
profile image
Ryan you must work at the same company I do in another department.... That's exactly what we do on EVERY game....

Andrew Glisinski
profile image
This is what I'm getting at, the fact that a video game has multiple layers, multiple stories that impact the design at every stage. Having to go back and redo multiple levels of detail is much more time & cost heavy.



If I could paraphrase, I would say that the development cycle for a film primarily revolves around one idea, then building every "scene" to work with that. Whereas if you take a game like Red Dead Redemption, you not only build out that one linear story-line to perfect it, but then all the branching storylines and dev. cycles for everything else in the game. This impacts level design, event triggering, battle tactics, director issues. Having enough time to work these out roughly, like Naughty Dog with Uncharted 2, did as well have a system in place, is what's required. Most certainly, this takes up more time than just singular scenes in a movie.



Why don't investors understand that in order to produce a great title, that is virtually free from encounter-able flaws, you need revision. Ideal revision, is more key than the initial ideas put on the table. Revision in a video game development cycle is much more heavy, and I would say stressful than if in the first year or two of a storyboard describing a linear "experienced" event.



I hope that everyone who's replied and mocked my comment, realizes that I am taking a creative and design-based stance on this issue. That I make an allusion to the fact I respect that video game creation is a complicated multi-faceted process that needs to be monitored at every stage. The companies that do care to monitor each stage carefully are most likely the ones to get burned by an investor demanding a project. Therefore, they require more time.

Ryan Duffin
profile image
Jane: No, just the same industry :)



Andrew: Preproduction IS about revision and iteration. It's about revision on a low-impact, low cost scale. It's planning your trip on paper before you get in the car. Sure, things might happen along the way but if you don't even know where you're trying to get, you're going to waste a lot of gas.



I'm not saying that if games just "plug in" an animated feature's production process, that everything would be sunshine and rainbows but there's a lot to learn from the way they do things. On the design/gameplay side, greyboxing is the equivalent of storyboarding but it's shocking how few studios do it. There is no reason you need a fully realized racetrack or city level to get your driving physics right, for example.



Whether you agree with me or not, remember the gist of the article: is it worth it? Yeah, it'd be nice to spend 3 times the time it takes on any artistic endeavor (that is actually a lie; I love deadlines. They make people act decisively and that's a big part of why these long projects often only come together in the end) but we are in the end, a business and we need to make a profit if we want to make our next game.



It's good for our industry creatively too. No publisher with any sense would agree to fund game that they *know* from day one that they won't see a return on for 6+ years, other then Call of Halo Theft Auto Craft.



Besides, if we *planned* on taking 5-6 years to finish a game, we'd wind up taking 10-12 anyway.

Keith Moore
profile image
I'm wondering if some games succeed better when sequels are released year to year based on their content. Street Fighter is basically two characters beating the crap out of each other, so creating new content in yearly cycles doesn't seem like such a stretch. Most sports games simply update the graphics, the rosters, a few improvements from the previous version? But when you have expanded adventure games like RDR and WOW you can't crank them out in a year when there are so many elements to account for: World, writing, action, etc. In those cases I would think it's just like making a major animated film...takes time.

E Zachary Knight
profile image
This topic really reminds me of the following:



http://en.wikipedia.org/wiki/Project_triangle



In it you have 3 points of a triangle representing the 3 goals of software development (really any kind of development project)



1: Time, the length of time it takes to finish the project. In development the goal is generally to finish quickly.

2: Cost. The money it takes to finish a project. Generally you want it to cost the least to finish.

3: Quality. The amount of polish on a project. You generally want it to be the best product available.



In this situation, choosing any 2 of the 3 will result in a reduction of the 3rd. So if you want something fast and cheap, the quality will suffer. If you want something of quality and cheap, it will take longer. If you want something fast and of quality, it will cost more.



You can try to shoot for the middle and come out with something that is reasonable in all three points. It is just a matter of finding that balance in your own game projects.

Francis Page
profile image
I'm not sure to get it. In software develpment, time and cost are basically the same. If it takes longer, you'll have to pay the developers for a longer time, increasing the cost.

Glenn Storm
profile image
Ephriam may be (indirectly) referring to the old development adage:



Good, Fast, Cheap: pick two.

Michiel Hendriks
profile image
That triangle doesn't really apply to creative projects. Because quality cannot really be measured in that case. How do you measure quality gameplay? You can measure the software quality, sound quality texture quality. But gameplay...

Jacek Wesolowski
profile image
...gameplay starts rough and then is improved iteratively, just like pretty much any other creative endeavour (e.g. a stage play, a film, a novel, a painting, a musical performance). Think editing, rehearsing.



In games, you can iterate on your design more in order to make it better, but it will cost you. You're going to have to rely on your gut feeling rather than taking a measure, but that only means you need to iterate more frequently and more carefully.



Besides, software quality etc. is part of a game's quality, unless you enjoy it when the game you're playing crashes every ten minutes. Technical considerations have a strong impact on your schedule, just like any other constraint.

[User Banned]
profile image
This user violated Gamasutra’s Comment Guidelines and has been banned.

Nathan Mates
profile image
I've seen projects that took slightly over 5 years to finish. I sized up the project's director as someone who was (1) a perfectionist, who (2) kept changing his mind. Either one can set the project's schedule back. Both together was a recipe for a mediocre game when finally shipped. Despite the 5-year dev cycle, things like gun and driving mechanics became usable in the last year of the cycle. So, the designers were scrambling at the last minute to get things polished/fun. But, they didn't really have the time to do that. So, mediocre game, dumped out that cost north of $40 mil (maybe $50 mil) never even hit the top-20 NPD lists.



Any project that hits the 3 year mark should have some long soul-searching meetings, with no-holds-barred feedback solicited from all team members. And, management needs to make changes based on that feedback, including management stepping aside and replacing themselves if they're the problem. Repeat those meetings yearly until shipped. (And I don't see management doing that often, but if studio management (or the investors/parent company's management) is going to get a return on their money, they should be looking into the true state of things.

Justin Lloyd
profile image
I've been on a few of those projects. They are not fun. They can destroy careers.



During the 8-bit era of home computer games there was a useful measurement that boiled down to a just two questions that you could apply to whether the game was in trouble.

1. Has the game taken longer than six months to reach release candidate state?

2. Has a second programmer been called in to help out on the development to get the project finished?



If the answer is yes to either of those questions, your project is in trouble. If the answer is yes to both of those questions, you're in very serious trouble.



Yes, it was a simple yard stick. Yes, it is a bit tongue in cheek. But it was surprisingly accurate. I've "helped out" on so many troubled projects I've lost count of them all. I know a lot of other long timers in the industry have also done those gigs.



I use similar yardsticks (though a little more scientific these days) in my own company when performing crisis analysis and critical analysis when engaged by various large publishers. My company has handled several crisis recovery projects of AAA titles, and we get to see the worst of the worst practices because of that.

Bart Stewart
profile image
What I find problematic is when private game development companies go public or (more commonly) are acquired by publicly-held publishing companies.



Obviously the benefit is that you gain access to a lot more money. But the downside is that you give up a lot of control over how that money will be spent. Now you answer to self-appointed "analysts" whose preference for short-term profit makes it extremely difficult to invest in products that need time for content creation, iteration to maximize fun, and polish.



This can't do anything but push games away from being deep, complex, interesting worlds that can be explored over hundreds of hours and toward being short, simple, "accessible" toys that you quickly beat and just as quickly forget.



Maybe, from a pure beancounter viewpoint, that's progress. Doesn't feel like it to me, though.

Jacek Wesolowski
profile image
This topic is a bit of a blanket for a number of issues.



One issue is the goldrush mentality, or always hunting the biggest pile of cash in sight while leaving smaller piles neglected. It's a naive approach to economics, in that the economic calculus only says you need to make good enough money to cover your expenses. What it certainly does not say is that, if everybody tries to make the best money, then everyone is going to end up actually making it.



Yet that is exactly what everybody is trying, even though the market has been clearly oversaturated with "best money", a.k.a. AAA console shooters, for some time now. It seems like everybody was betting on the potential of the gaming market to grow indefinitely -- now we're stuck with a bubble which, in my opinion, has already begun to burst. Long development cycles are preventing it from being as apparent as in case of, say, real estate, but hardcore big budget games have been a quirky business at least since the beginning of the century. You can expect less investment in big budget games in the future just for that sole reason.



Mature markets have learned to diversify. The games industry is only learning it now , with MS and Sony trying to find new audiences for their otherwise unsustainable hardware, retailers slowly giving way to digital distribution, and publishers equipping themselves with "social" game development studios. Starting "indie" branches also seems like a natural step. The good news is that, in order to diversify properly, you need to keep old sources of income while trying to find new ones. The investment into large projects is not going to cease, unless the entire AAA development system collapses.



As an aside, mature markets have learned to recognise corner cutting as an early symptom of one's business model going awry. We've been cutting corners for years, but there's no longer much spare potential for prettying one's balance up with unpaid overtime, novelty factor, shady public relations etc.



But that's just one issue. Quite another is development methodology, which is, essentially, a giant mess. Today's teams of 100, with budgets in the order of 20 million USD, are being managed in roughly the same fashion as ten years ago, only then those were teams of 20 with budgets in the order of 3 million. Again, long development cycles have prevented the industry from adopting better methodologies faster, but our long development cycles are a symptom of mismanagement, rather than a cause for it. Importantly, "mismanagement" does not stand for just the operations management, i.e. it's not all the producer's fault. Design branch is at fault as well.



For instance, ten years ago you could afford to build an entire level and then scrap it on the grounds of it not being enough "fun". Ten years ago a fully developed level looked like today's rough prototype. Just look at screenshots from Half-Life, Deus Ex, or any other major hit from that time period. It's not even alpha by today's standards. Hence the need for designers who can do the entirety of their job while only having pre-alpha content at their disposal. But I've personally worked with more than one designer who were unable to do just that. They couldn't get past the lack of polish and right to the core of the system they were designing. Apparently, it's not that easy to adapt, especially when you have a track record of successful projects that you have developed using your old ways.



Yet another issue is the business model of a typical game development studio. You set up a permanent team. You pitch your project to publisher. You get greenlit. You're expected to deliver milestones, roughly equal in size, at roughly equal intervals. The only time when you're more or less free to do whatever you want is in between contracts, but you have a good reason not to, because the team is not making any money, so you want to pitch another project as soon as possible.



The problem is, the bulk of work being done is production, particularly asset creation and level construction, but what actually makes or breaks the game most of the time is feature development. Feature development is, in terms of man-months required, 50 per cent design, 30 per cent tech, 15 per cent concept art work (i.e. sound and narrative as well as graphics), and 5 per cent business considerations. Now, compare the sizes of your typical design and asset teams -- 80 per cent of your project's value is provided with 20 per cent of your efforts!



The practice, however, is that you need to do vertical slices. Vertical slices cause you to produce parts of game in advance, before feature-related risks are handled. Then you end up throwing entire levels out the window, because some features you've dreamed up actually don't work together. You could have arrived at the same conclusion with just placeholder art, at a fraction of cost, but then your asset team would have been sitting idle, so there would be no real savings.



There are at least two ways to leverage this. The more obvious one is to have separate teams for preproduction and production. Preproduction team is the creative core of your project - they make the features, they create "the fun". Production team provides "firepower" - when it's needed. The main benefit is that both kinds of teams work on different schedules. It's possible for a production team to cooperate with several pre-production teams, and thus avoid ever being idle. It also liberates pre-production teams to submit their work "when it's done". Low maintenance costs would make it easier for them to take risks. A full development cycle would not get shorter, but it would get way cheaper.



The less obvious way is to have a traditional studio with a research workshop attached to it. The workshop never participates in actual projects. Instead, it keeps spawning working prototypes for features - perhaps even unrelated to what the rest of the studio is doing at the moment. Over time, these working, tested features form a kind of library of ideas. Each idea is proven to have the "fun" potential and work with the available technology. When a new project is commenced, the library is browsed for proofs of concept that bear relevance to the new game. The dev team then uses these as foundation for their new design. The main benefit is that, once your workshop has warmed up, you always have a surplus of working features that you can use with any game. This is a major risk reduction for your main project. With good technological background, you could even switch between feature sets on the fly. It would be like working on a dozen game variants at the same time, with next to no added cost, and then gradually narrowing it down to a single perfect combination. Any individual project would get shorter *and* cheaper, thanks to time and labour saved on failed production iterations, though there would be an added cost of maintaining the workshop.



There are also certain "social" benefits, such as using the research workshop as a staging ground for new employees, and keeping the exchange of personnel between teams so that everyone can get wildly creative from time to time. The question, however, is whether the industry has matured enough to recognise the value of such things.

Merc Hoffner
profile image
"Great works take time"



Is this actually definitely true?

Josh Foreman
profile image
@ Jacek: Good thoughts. Your second model is sort of like what Pixar does. They have a second department that does proof-of-concept work, though that department still releases their work as shorts.


none
 
Comment: