How Project Atlas could kill the console and change gaming forever

When Ankur Pansari explains his startup video game company’s mission, many people shake their heads, roll their eyes, or dismiss the ideas outright

Screengrab via Artillery Games/YouTube

When Ankur Pansari explains his startup video game company’s mission, many people shake their heads, roll their eyes, or dismiss the ideas outright. It’s a nice dream and Silicon Valley can be awfully nice to dreamers but, when Pansari’s company is discussed around the Internet, the word “impossible” often appears.

Pansari is the co-founder and CEO of Artillery Games. On the eve of the release of the latest generation of game consoles—the Xbox One and PlayStation 4—the 11-man team at Artillery is hoping to replace the console entirely. They believe that high quality, blockbuster video games can and should be available for play for free in a Web browser on your computer, TV, phone or tablet.

Artillery’s proof-of-concept is Project Atlas, a hardcore, competitive, free-to-play real-time strategy game in the vein of StarCraft 2 and League of Legends. They talk hopefully about one day having 10 million users playing fully 3D games using HTML5 on Google Chrome or Mozilla Firefox instead of running out to buy $400 consoles or downloading games to their PC.

HTML5 is the newest, most powerful version of the Web’s core programming language. It’s behind some of the most advanced websites you’ve ever seen—works of art like Arcade Fire’s and Universeries—and is quickly challenging Adobe’s Flash as the platform of choice for Web game designers. Most powerful of all, it works instantly on billions of devices.

But despite that utility, it’s easy to understand why the public is skeptical. There have been HTML5 startups before that have come and gone. Even Pansari’s old boss, Facebook CEO Mark Zuckerberg, said his company’s biggest mistake was “betting on HTML5.”

But from the day Artillery was founded in late 2011, the company was already on working on acquiring a secret weapon: Sean Plott, an immensely popular StarCraft 2 commentator known online as Day[9]. He would provide hardcore gaming expertise, credibility in the gaming community, and outright smarts.

Plott has been widely coveted by game designers, so it’s no surprise that Pansari was after him. Plott says he gets approached with a lot of big ideas but, all too often, they’re only ideas.

When Pansari sat Plott down and explained what he wanted to make, Plott sighed and put his hands on Pansari’s shoulder.


Pansari started up his laptop, tethered it to his smartphone for Internet access, and quickly loaded up a demo of a graphically intensive, high quality game running smoothly on his Web browser.

“Want to make a game?”

The answer? “Yes.”

It wasn’t just that easy. Plott owns his own personal burgeoning media empire,, with an overall audience in the millions. There’s even a forum dedicated to him on Reddit, boasting 7,300 subscribers, which was created because there were so many posts about him in the main StarCraft forum that they overwhelmed everything else.

“I was mentally on board with the concept,” Plott said about Artillery Games’s proposal.

But is a fully in-motion company. I couldn’t suddenly drop it, drop my shows and do this game. I had lots of concerns about that.”But I voiced them and Artillery was totally on board. We produced a work-flow to minimize my stress, do work on plus work on Project Atlas. It’s a really wonderful position.”

The marriage between Plott and Artillery was inspired by some other very successful games. In 2007, Riot, the developers behind League of Legends, hired Steve Mescon, known as Pendragon, a major community figure from the DotA 1 days. Looking further back, Pansari said that Valve’s decision to hire the original development team behind Team Fortress in 1999 inspired him to seek out Plott.

Plott’s fame comes from his talent as analytic commentator, not a game developer. But he’s actually got some serious game-making chops. He received his graduate degree from the University of Southern California’s Interactive Media and Games Division working on—you guessed it—games. He’s been involved in educational research games for high school students and has created several small indie titles for festivals.

“It’s so fun to put your work in front of playtesters to look for bugs,” he said, “and then to see someone have actual, real fun. It’s an amazing mix of problem solving and creativity.”

Pansari and Plott got to know each other over StarCraft. Plott created the After Hours Gaming League which was an eSports twist on old corporate softball teams. Big Silicon Valley firms like Facebook, Google, and Microsoft (the first season’s champions) put up teams of employees to play StarCraft and League of Legends against each other. Pansari joined the Facebook team and then helped to advise the league.

The two talked on and off for some time before Pansari made the memorable pitch to Plott and got him on board. Artillery had already received $2.5 million in venture capital funding and was building out a talented team, luring employees from heavyweight tech firms like Google, Facebook, Adobe, Intel, and Zynga.

The team. Photo via Artillery Games

“One of the biggest things that attracted me — other than that these guys are all brilliant — is that they’re gamers, too,” Plott said. “That was amazing for me. I knew Ankur because he played StarCraft for Facebook.

“The fact that I go into work and it’s my people, it’s my tribe is so great.”

Plott was sold. Now, he and Artillery have to convince the rest of the world.

The journey toward Project Atlas began seven years ago. Pansari and co-founder Ian Langworth played StarCraft all the time. One Friday night, they were stuck without StarCraft because they’d left the game’s CDs at their office. They spent their days in the Facebook offices building new technology and applications for browsers, so why not have hardcore games work in a browser as well?

Of course, most games don’t have old fashioned CDs these days. You can go onto Steam or to buy, download, and redownload any game you need. But that takes time, memory and often money. If Pansari and company have their way, the monstrously successful Steam model will soon be old fashioned.

On Thursday, the campaign to convince the rest of the world of their dream began in earnest when Artillery released a tech demo of Project Atlas.

This is the first time anyone in the public has seen the work. Artillery wants to give the world a peek at what advancing technologies like WebGL, WebSocket and modern JavaScript can produce in the browser. So far, the Internet is impressed, even if still pretty skeptical. The video’s already racked up nearly 40,000 views and a mountain of excited comments (“Holy fucking shit this looks amazing!”).

“We want to show people that it’s not impossible to create AAA games in HTML5,” Pansari said. Instead, “people have been using [HTML5] the wrong way.”

The Artillery Platform, built from scratch, is impressive even though the art has yet to be finalized. The graphics are clearly solid, the multiplayer capabilities look smooth, and the load time looks near instantaneous with a single click — and all that without any plugins or extra fuss.

But don’t just get caught up in how pretty it looks. The developer tools, which don’t make an appearance until the video is halfway done, look particularly powerful. These allow new units, abilities, and ideas to be built and implemented with incredible speed. That’s exactly how Artillery wants to run Atlas—new units, abilities, and ideas will be added all the time, a first for a major competitive real time strategy game. For all of us wondering about the actual game that Artillery will produce, the developer tools provide a telling peek into the future.

“We’re not limited by the technology anymore,” Pansari said.

Plott elaborated:

I haven’t felt any restrictions with Project Atlas. I’ve worked with other game engines like Unity. It’s typical to have conversations like, ‘We can’t have that many sprites, we can’t render that model.’ For me, this is total carte blanche with Project Atlas. Any experiment I’ve thought of running, we’re able to build and implement.”

The experiments in gameplay and technology have been coming fast and furious—a rate of 10 per week, because the developer tools allow them to try anything.

“The software development kit shows up in browser. So, we have a million potential approaches,” Plott said, “and we can iterate really quickly.”

The experiments are complemented by biweekly playtests. Private beta testing is scheduled to begin in winter 2013 with a select group of players. But even when the game is out of beta, the plan is to keep it changing all the time.

In fact, that’s how Artillery plans to make money.

The development tools will allow the regular creation and release of new units, abilities and other additions that can be purchased in microtransactions, thus helping to make the game profitable and long lasting.

“Microtransactions” has become four-letter word in a large part of the gaming world. These in-game purchases are a great way for developers to milk more and more revenue out of their titles, but occasionally at significant cost to the gameplay experience. Only a few titles have done microtransactions well—League of Legends, Dota 2, Counter-Strike: Global Offensive and Team Fortress 2. Beyond that, the list of good competitive games with microtransactions is awfully shallow.

Iif Artillery is depending entirely on them, will that dilute the overall experience by making gamers pay for every small feature?

Not according to Plott. “The game should be incredibly fun to play even if you don’t put in a dime,” he said. “We don’t want to begin thinking of microtransactions first. We want to make a fun game first and then think of a way microtransactions can live in that.”

Plott says he greatly admires the way Riot implemented microtransactions in the design of League of Legends. In fact, he insists, starting with a small number of units lowers the barrier of entry. That makes the game easier to understand and gets players hooked more quickly. Players can then pay or grind for more.

The lower barrier of entry underscores the greater goal of Project Atlas. Artillery says they want to make the world a better place through games. They want to make something that is, above all, fun.

If that sounds obvious—we are talking about games after all—maybe you haven’t played an ultra-competitive game like StarCraft 2. The game dominates the real-time strategy genre, but there are also a considerable number of players who simply can’t adapt to its highly competitive nature. Many players who attempt to make it up the games competitive rungs report shaking hands, sweaty palms and a sense of feeling overwhelmed. There have been thousands of words writtenincluding a scholarly article dedicated to “getting over ladder anxiety.”

Project Atlas will employ a number of different strategies to fight anxiety and promote fun. First, they’re emphasizing team play because “regardless of the game, you can feel good, like you contributed and get a good social experience.”

On the one-on-one side, Atlas is stressing comeback opportunities as a means to promote fun.

“In fighting games, you can have a fourth of your health versus an opponent with full health,” Plott said. “But if you hit the full combo, you can come back and kill him. One-quarter health doesn’t mean you play worse, it just means you’re closer to a loss. You can definitely still win.”

I asked Plott and Pansari what other innovations they want to bring to the RTS genre. Pansari said they’ll be unveiled on another day.

Plott chimed in:

“We’re not going to go from three races to four races and call it innovation,” he said, taking a little jab at Blizzard’s Warcraft series. “Project Atlas is unlike any major [real-time strategy game] you’ve ever played.”

“When most people approach real time strategy games, they wonder about sweet races, awesome units, great abilities. We spent months perfecting pathfinding, optimizing the way units move. How do you code the correct amount of stupidity into a unit so they have the right weight and feel? The same way a basketball with air feels just right to dribble.”

By bringing Plott on board and calling their game a “hardcore, competitive RTS”, Artillery has invited the scrutiny of an army of gamers, all wondering if Project Atlas is aiming to become the next great eSport—video games played competitively for cash and fame.

Project Atlas would face some stiff competition. Right now, the biggest eSport in the world is League of Legends, with 12 million people playing it every day. The annual League of Legends World Championship Series finals attracted 8.2 million unique viewers last year and there’s no reason to think this year’s event won’t top all previous efforts.

StarCraft 2, Dota 2, Street Fighter 4 and a few other competitive games round out the current eSports landscape. It’s a highly competitive market.

StarCraft 2 in a browser, Ankur Pansari’s dream.

“Becoming an eSport is something we can’t determine,” Pansari said, explaining that the community of gamers is the ultimate arbiter of what does or does not become an eSport. “If we’re lucky, it becomes an eSport and something we follow for decades. We hope it gets iconic status but we can’t make that happen.”

If the community wants it, Artillery will build it. LAN, or local networking capabilities, won’t be built into the game on launch but Pansari says they can be built if needed for competition. Competitive ladders, tournaments, clan capabilities and other features catering to eSports can all be made if there is a demand from the players.

If the platform works as well as Artillery hopes, it could change the way blockbuster games are made and played. A viable browser platform could change eSports in a big way.

Most fans are a little less ambitious. They just hope it’s a good game.

About the author