The Austin Game Developer's Conference, now under new management, kicks off today with a keynote from Blizzard President Michael Morhaime. He started off the event with a discussion of the potential of gaming in the 20th Century, and the lessons his company has learned from the long trial that has been World of Warcraft. Read on for notes from his presentation.Morhaime takes the stage to applause, and begins by expressing fascination with the possibilities of the modern era. Right now, he says, the ability for people to connect is unprecedented in the history of humanity. Looking back on the 20th century, this puts us now in a very unique situation. Advances in transportation, communication, and information storage have made the world a far smaller place. The rate of change is so fast that we have a hard time coming to grips with it. Future shock is everpresent, but he (at least) views this as a good thing.
For a moment, he takes a steps back to look at the world as of one hundred years ago. In 1907 there were only 8000 cars in the US, with no airforce, and only 8% of homes in the states even had a phone. In comparison, recent years have seen the density of transistors (as described by Moore's Law) increase exponentially. It makes for a really nice graph.
This is the context in which we're looking at online gaming this year, at the conference. The online gaming market is still very much in its infancy, many things are still possible, still unrecognized in the world. The goal of designers, studios, and publishers, should be to live up to that potential.
His focus during the talk itself will be World of Warcraft, the history of Blizzard, and lessons that the online games industry can take from their successes (and failures).
A trio of UCLA grads started the company in 1991, with little more than $20,000 and a pair of PCs. Both were 386s, without CD drives and 'mighty' hard drives that could be counted in the dozens of megabytes. Their first jobs were conversions; PC to Mac or Amiga translations. They learned much from this process, getting a real feel for how games were made. They made a racing game for the SNES' launch, which they completed in about four months.
With the fruits of those early payoffs they pulled some more people in and began to make new console titles. Eventually, with the decline of 16-bit consoles, they turned to the PC for their first RTS: Warcraft. The rise of the Internet made them think hard about players collaborating, competing online. Their sale to Davidson, an educational software company, led through a series of alleys and buyouts to their current relationship with Vivendi. Their initial relationship with the Davidsons has been maintained through all those buyouts, allowing them to keep tight control of their IPs.
WoW has just been their most recent success, though Morhaime admits it has changed the company in ways they never could have imagined. Through buyouts and company modifications, their fundamental concepts have stayed the same. Gameplay comes first. He pulls the 'it all starts with a donut' slide from Rob Pardo's talk last year. The center of the donut is the hardcore gamers, and the casuals are the sweetness all around. To this end, they make an effort to ensure that system requirements are low. "Easy to learn, difficult to master" is also a concern; he mentions Guitar Hero as another title that nails this feeling.
Outside of design, they see the Blizzard name as their most important property. "Blizzard", for gamers, should equal high quality, fun, and polish. A gamer walking into a store should be able to see the Blizzard name on a title and know that they can trust the game will be fun. They only want to make "brand deposits", not withdrawals; they only seek to add value to their name through choices in pricing, polish, and experience.
The biggest pressure is resisting the urge to release early. There is pressure from all sides; budgetary concerns, programmers tired of working on the game, analyst expectations, consumer expectations. They see this as a huge danger. You only have one chance to make a first impression. Missing that opportunity is not something you can get back. They try to think long term. Don't mortgage the future to meet the quarter goals, as it were.
Diablo was their biggest trial. Back in the day Diablo was in danger of 'missing Christmas'. When the time came to make the final call, they held it back to December 31st. Even though they released the game late, that's not how it is remembered. It sold great, and is now an amazing brand. They now point to that as an unqualified success. Most recently they had to do the same thing with Burning Crusade. After the problems with the initial WoW launch, they wanted to really nail that rollout. They were rewarded with the fastest selling PC title of all time: 2.4 million units in one day.
Another pressure they have to deal with is the concept of 'doing everything at once'. The company instead tries to focus in on the truly important things. Build on successes, gain expertise, and then once that's under your belt go for the more ambitious stuff. WoW was a huge undertaking, but it wasn't their first online game. They built on experiences with Battle.net to create that title.
The success of StarCraft and WoW has turned Blizzard into a global company. This necessitates a number of considerations. The games can't just be geared to the US market. Initially they were very much a states-oriented organization, and this led to some frustrations with launching titles in localized languages. Grey market imports of the US version of the game led to lack of interest in brand new launches in Europe. In Asia, meanwhile, the importance of game rooms makes the market completely different than in the west. Accounting for all three of these markets is required for a simultaneous worldwide launch.
A slide showing the global launch growth of their titles shows a huge spike in Asian interest around the time of StarCraft's launch. To this day there isn't a Korean version of SC; they're all playing in English in those well-publicized online matches. They managed worldwide launches with Diablo 2 and WarCraft III, but held back for WoW. They'd never rolled out a Massive game before, of course, so they really wanted to make sure they had successes under their belt before taking the show abroad.
What they don't do is localize games to 'regional tastes'. They see gamers as having different styles of play, but they exist in all regions. They don't make different versions with each being tailored to a specific area of the world. Instead they make sure that all players find something to meet their gameplay style needs. Blizzard is very much their target market; if they like it enough, odds are (they think) that others will feel the same. Just the same, they do need to be culturally sensitive. The Pandarens from WarCraft III were an oopsie along those lines. Pandas dressed in Japanese clothing and having Samurai sensibilities did not go over very well. They responded quickly, changing the image of the Pandaren to a more culturally appropriate style.
By the time WoW was in the works, they had already started to think globally. They identified North America, Europe, and South Korea as the three areas they wanted to target directly for WoW's launch. To do that, they established a full control center, with customer support, server infrastructure, marketing, sales, and administrative control in each area. Other areas saw them teaming with companies that have strong understanding of local customs. The9 in China, for example, handled localization of the game for that market, as well as server maintenance for the game in that country.
Their big challenge for WoW was determining the demand for such a title. He quotes Tom Watson, CEO of IBM in 1942: "There is a worldwide market for perhaps five computers." How much infrastructure do you need for a Massively multiplayer game? How many units at launch? The local Fry's launch tipped them off that things might be different with WoW. Way too many people showed up for their local meet and greet and launch party. "Wow. We might need more hardware."
They had sales information for WarCraft III, as well as the stats for EverQuest's lifetime. They thought of WarCraft as a ceiling, and were dead wrong. At several times in WoW's early lifetime they had to stop shipping boxes to retail because they couldn't have supported the new players.
H.R. is really important, they learned. The company needed to scale up across the board starting in 2004; they went from under 1000 employees to over 3000 in the three years since WoW's launch. They needed more IT folks, additional sales, marketing ... they didn't even have the capacity to hire folks as fast as they needed them.
The big lesson: running a MMOG is not just game development. It's also a service. They thought they had enough experience running Battle.net to handle WoW, but moving to a subscription service was a very different experience. These backend elements are just as important as game design, if they impact the player's experience.
Communication was one of their other big learning curves. They need to make sure to communicate with the players, or 'people would make stuff up.' The Community team needs a process for keeping players informed, and keeping international staffers on the same page. This is especially important when there is a 'fire'. This lead to silence on their part when there was a problem, because they didn't want to say something 'wrong'. As a result, they developed lists and staffers that would ensure less 'wigging out'.
Morhaime briefly delved into the dark underworld of gold selling, credit card fraud, and trojan-laced websites that steal user information for WoW accounts. They see this as a core part of their mission: this adversely affects players across the board, and has to be stopped as best they can.
Testing is all-important to the company. "Never trust version 1.0." Everyone at Blizzard tests, as a result. Public Betas ensure that they eliminate 'cheese'; boring ways to play the game that are most efficient. WoW launched without a test site, and in retrospect they regret that. Patches are now thoroughly reviewed by a wide player audience before they go live. When Burning Crusade rolled out, they took the lessons from WoW's launch to heart. They upgraded their entire infrastructure for the game, and ensured they had extra capacity for their first day of new gameplay. The servers withstood the extra load well, and CS folks said they viewed it as a 'smooth patch release.' This time they did a worldwide release, with midnight openings of the game all around the world. 'Every hour they opened a new market.' They viewed it as something of a New Year's celebration for the folks at Blizzard.
To close, Morhaime displays a video showcases the colorful characters that showed up for the midnight launch of the game in Europe, everywhere from London to Stockholm, from Paris to Dublin. "It's an exciting time to be in the game industry. Good luck."