Analysis: Why SteamOS probably won’t cause a PC gaming revolution

When I had a chance to talk to a Valve representative at the company’s Steam Machine partner unveiling event at CES last week, my first question was the one that I’m sure is on the minds of a lot of PC gamers these days. That question: what does a gaming PC with SteamOS provide over a similar Windows machine that has access to a much larger library of natively compatible games?

Valve engineer and business developer Kassidy Gerber noted that SteamOS has been “really built from the ground up to be part of the living room.” On a Windows machine, Gerber said, Valve “can’t control the initial boot-up experience like we can on SteamOS.” In addition, Gerber pointed out that the back catalog of Windows games on Steam will be supported through in-home streaming and that the Steam Controller allows players to control games from their couch in a new way.

Looking at these arguments as a potential consumer, none of them are really that convincing. The Steam client on Windows already has a Big Picture mode that makes it work just as well as a “part of the living room” as SteamOS does. In fact, you’d be hard pressed to tell at a glance whether a computer is running SteamOS or a Windows-based Big Picture mode. Controlling the boot-up experience and launching to Steam right out of the box is nice, but it’s trivial to get a Windows box to boot directly to Big Picture mode as soon as the system starts if that’s what you really want.

Streaming games from an existing gaming rig is nice as well, but that living room “receiver” can just as easily be a Windows machine that can stream games and run more games natively. Plus, this streaming feature is only really useful if you already have a high-end Windows machine somewhere in the house—it’s not really an option for console players who want to check out the PC gaming world they’re missing.

As for the Steam Controller, I have mixed feelings about its potential to revolutionize the concept of playing PC games on the couch. Regardless, the controller will also work with Windows and Mac-based machines and will be sold separately. Thus, it doesn’t really factor into the discussion of why PC gamers should consider switching to Valve’s new OS.

In short, the major selling points for SteamOS can largely be countered with an argument along the lines of “but Steam for Windows does that too… and more.” This is a major problem for anyone trying to argue that SteamOS will break the stranglehold that Microsoft’s operating systems have had on PC gaming for nearly two decades now. Overcoming the lock-in effects and sheer industry inertia of that lengthy monopoly is going to be incredibly difficult, even for a company like Valve.

It’s the games, stupid

Anyone looking to buy a SteamOS box this year is going to find it nearly impossible to avoid one major downside when compared to a Windows machine: the library of games that are playable natively on SteamOS is much smaller than that for Windows and will likely remain so for the forseeable future. It’s not a small difference, either—Steam’s online store currently lists 486 games that run on Linux compared to 2,483 that work on Windows. The 75 percent of the Steam library that doesn’t run on Linux includes an overwhelming majority of the titles on Steam’s top sellers list as well as a number of big name titles that most any potential SteamOS customer is going to want to play.

Sure, it might not always be this way. Steam’s library of native Linux titles has been growing quickly—it only included 50 or so games in February—and Valve is putting a lot of weight behind Linux game development, including tools to make it easy to port Windows games to SteamOS. Still, getting the SteamOS library to something approaching parity with Windows (say 75 to 80 percent of the best-selling new releases also having concurrent Linux releases) is going to be a gargantuan effort that will likely take years to happen, if it comes to pass at all.

Bringing Windows-level game selection to SteamOS will also require major third-party publishers to play ball and devote resources to Linux ports, something they’re unlikely to do until and unless there’s a proven market for those ports. In turn, the market of gamers on SteamOS is likely to stay small until there is a critical mass of games that the platform supports. This chicken-and-egg problem of developer adoption vs. player adoption is incredibly familiar in the console gaming space, where companies have to convince their customers to essentially throw out their old OS for a new one every five or six years.

Traditionally, console makers make this argument with exclusive games, those heavily sought “system sellers” that in and of themselves are enough to get players to look at a new platform. Valve is in a great position to make and market these kinds of exclusive games as a reason players should jump to SteamOS, as I’ve argued before.

But the company seems loath to take advantage of this position. Valve’s Greg Comer told IGN back in November that “you won’t see an exclusive killer app for SteamOS from us.” More than that,  Valve business developer Anna Sweet told IGN that the company would be encouraging companies “to put their games in as many places as possible, including not on our platforms.” In other words, forget about exclusives as a way to set SteamOS apart.

Face it, this box would be better for gaming if it were running Windows and its huge library of games.

When I put the question of SteamOS-exclusive software to Gerber, she said, “We want to do what’s best for customers. Our customers may want to play our games on Windows or Mac or whatever, so we don’t want to lock that down for them.” There’s some nobility in that answer, and perhaps a reflection that Valve doesn’t want to cut out its existing Windows/Mac client business to prop up its nascent SteamOS business. That’s understandable, but it doesn’t change the fact that without exclusives, the SteamOS game library will always be a subset of the Windows game library, no matter how large it grows.

The other trick console makers use to convince players to switch to a new platform is improved performance. That’s a bit harder to do in the PC world, where Windows machines and SteamOS machines can be built with exactly the same raw hardware parts. It’s not impossible, though. Valve has already made some hay out of the fact that its Linux port of Left 4 Dead runs at a slightly higher maximum frame rate than the Windows version, thanks to reduced overhead from the OS. This is the kind of difference that has a chance of being a true selling point for a lot of gamers, especially the significant subset of PC players that demand the very best gaming performance out of their high-end rigs. For some reason, though, Valve hasn’t been very vocal or upfront about promoting this advantage as of yet.

The one remaining area where Steam Machines can potentially compete with Windows-based gaming rigs is hardware price. Going with SteamOS eliminates the need for a Windows license, which costs anywhere between a (speculated) $50 or so for an OEM pre-install, right up to $200 for a full-price, Pro-level retail copy. That added cost isn’t necessarily a deal breaker, but it’s an edge for machines with an OS that’s absolutely free, especially on the lower end of the hardware power spectrum.

This price advantage could make SteamOS the obvious choice for the niche market of gamers looking for an extremely cheap machine meant primarily to stream games from the office Windows box to the living room TV. For the much larger market of PC gamers in general, though, I don’t think the small price advantage is going to be enough to make up for the major disadvantage of a smaller game library in most consumers’ minds, not to mention the hassle of having to learn and get used to an entirely new OS.

The revolution does not begin now

It’s easy to see the logic of SteamOS from Valve’s point of view. CEO and co-founder Gabe Newell has made it clear he considers Windows 8 a “catastrophe” for gaming and has hinted that he worries the openness of the platform is under threat from the Windows 8 Store, which could easily develop into something of a challenge to the entire Steam business model. Given that worry, it makes sense for Valve to start planning for what options it might have outside of Microsoft’s playground.

Those high-level corporate concerns don’t really factor in to the individual-level decisions of consumers, though. For those consumers, the differences in library size and the “Steam for Windows does that too” arguments are going to be hard to overcome.

That’s not to suggest that SteamOS won’t find a small market of players excited about a better gaming option on a totally open operating system, or there may be those who are looking for slightly cheaper hardware meant primarily for streaming to the living room TV. But anyone expecting SteamOS to cause a revolution in the PC gaming space, or for the OS to become the new de facto gaming standard in place of Windows, has to answer a lot of questions about how exactly Valve is going to overcome decades of lock-in and network effects that have given Windows a stranglehold on PC gaming.