Facebook’s new game streaming service is best-defined by what it isn’t.
Unlike Microsoft’s xCloud, Google’s Stadia, and Amazon’s Luna, Facebook’s service isn’t trying to replicate the experience of game consoles or gaming PCs. The company isn’t working with huge studios like Activision or Electronic Arts to make their console games playable on phones or lightweight laptops, and it hasn’t even come up with a flashy brand for its effort.
Instead, Facebook is merely using streaming to deliver free-to-play mobile games faster and in more places. The initial launch will offer games such as Asphalt 9 and Mobile Legends directly through Facebook’s Android app and the Chrome and Safari desktop browsers, so players don’t have to waste time downloading a game before they can play.
That might seem less ambitious than what other tech giants are doing, but in a way it makes more sense. Facebook isn’t pushing its casual gaming audience to suddenly pick up Call of Duty or trying to convince PC and console gamers to set aside their dedicated hardware. All it’s doing is reducing the friction of trying new games without leaving Facebook, where 380 million people already play games every month.
“It doesn’t immediately strike you how different this is, but it’s radically different when a place you go 20 times a day—Facebook—can offer this kind of jump in and [play games] engagement,” says Jason Rubin, Facebook’s vice president of play. “It’s a really new world.”
Rubin might be right, but Facebook’s shot at cloud gaming also has its own challenges, including Apple’s chilly attitude toward game streaming on its mobile devices. And then there are Facebook’s own privacy issues.
Beyond basic browser games
Although Facebook already offers games on its website and Android app, those are either based on Adobe Flash or HTML 5. Adobe is discontinuing once-mighty Flash at the end of this year, and HTML 5 is too limited to support modern mobile games. Facebook is turning to streaming as a solution, running the games on its own servers and streaming the output to the player’s screens.
This approach has some clear advantages over downloading apps. On Android, the racing game Asphalt 9: Legends takes up 2 GB of storage and took me four minutes to download on a speedy 200 Mbps Wi-Fi connection. In Facebook, the same game loaded in about five seconds.
Streaming also has some inherent drawbacks. Facebook’s games can’t run offline. They won’t work well with a slow connection, and playing them will eat into data limits on wireless plans. They also exhibit more latency—the lag between pressing a button and seeing a response—than games that run on the device itself.
But instead of pretending those limitations don’t exist, Facebook is trying to work within them. Casual racing games like Asphalt 9 or turn-based role-playing games such as Mobile Legends don’t require super-fast reflexes, so latency isn’t a problem. They’re also free to play, so players won’t have sacrificed much if the experience falls short. And if players decide to download a game that they first tried though Facebook, they can link their account to transfer progress from the streamed version.
Still, Rubin says Facebook’s system could work with any kind of game, and the company plans to add more genres—and possibly free-to-play PC games—in the future.
“Over time, latency will come down, your proximity to a data center will go up, and you simply are on this march towards a point at which each type of game becomes possible,” he says.
That could even extend to games that couldn’t exist today on PCs or game consoles. Imagine, for instance, a multiplayer game that involves hundreds of simultaneous competitors, or a role-playing game with photorealistic graphics. By using more powerful cloud servers, developers could make games that otherwise wouldn’t be possible.
While cloud gaming proponents have hyped up those kinds of ideas before, Rubin believes Facebook has a better chance to make them happen because it already has millions of players on its platform. Once Facebook has proven that there’s a market for game streaming, developers might use it to make better games.
“Some developers will say, ‘Now that there’s this successful platform out there, I’m going to go into a hole for a year and a half and come out of it with a totally new type of game, into a platform with a user base,'” Rubin says. “My bet … is that those games will revolutionize gaming.”
Apple is a road block, privacy is a concern
Promising as all this might sound, Facebook has a few obstacles in its way.
The biggest is Apple, which wants to maintain tight control over how people play games on its platform. The company said last month that if services such as Google Stadia and Microsoft xCloud want to offer their services through the App Store, they’ll have to distribute each game individually. (Amazon has come up with a web version of its Luna app as a workaround.)
Rubin says Apple’s stance is a non-starter for Facebook, so it won’t offer game streaming on iOS at all. Apple has already rejected earlier versions of the Facebook Gaming app for hosting its own games, and Rubin believes Facebook’s app wouldn’t be allowed to launch games through a web browser as a workaround. Facebook could offer some version of the experience through Safari, but that would ruin the business proposition—the whole point is for people to discover games through the Facebook app, which they’re already using. Rubin isn’t even sure that would work from a technical standpoint.
“We’re exploring all the various ways that we might build this, but the rules are uncertain, and they seem to be always changing,” he says. “It’s a very frustrating process.”
The other issue is Facebook itself, and its never-ending tendency to create fresh privacy concerns. I ran into a couple of them while testing out the new games myself.
First, the games won’t work unless you enable Facebook’s “Future Off-Facebook Activity” feature, which allows apps and websites to send data they collect about you back to the social network. When this setting is off, you usually can still approve data sharing on an app-by-app basis. But in this case Facebook is requiring blanket approval. If you had disabled Future Off-Facebook Activity before, you can’t play these games without re-opening yourself up to all kinds of additional tracking. (Facebook says it’s exploring an option to let users approve specific games, but doesn’t have a timeline for doing so.)
Also concerning is that by default, Facebook attempts to share your gaming activity not just with friends, but with anyone who plays the same game. You can modify this setting on a game-by-game basis, but there’s no way to set up a default sharing option and have it apply to everything. Without extra caution, your gaming activity will be on display to the world.
Keep in mind that the main reason Facebook is even getting into game streaming is for advertising purposes. It wants mobile game developers to pay a premium for playable ads, and for users to spend more time communicating among themselves on Facebook, exposing them to more ads in the first place.
“As people adopt more games, and find things they like and play them, the value of ads goes up, because you’re more likely to be selling things people are buying,” Rubin says.
Although Facebook isn’t saying why its privacy options are so inflexible in this case, getting people to turn over more data and share more often is what makes the proposition work. But it’s also what makes Facebook’s cloud gaming push less exciting than it ought to be.