Last week, Valve made a series of announcements that could dramatically impact how people play games on the PC. But slick micro-sites aren't created in a vacuum. Valve have been hinting at SteamOS , Steam Machines and the Steam Controller for years, through interviews and information that goes all the way back to 2010. I've combed through these interviews, in order to find out what the future might hold for Valve's move into the living room.
Just how open will their open OS be? How will Steam Machines evolve to match more powerful tech? Could the Steam Controller be any stranger? And what do these announcements mean for Windows? Read on to find out.
Back in January , Gabe Newell admitted that he thought of Apple, not Microsoft or Sony, as Valve's biggest competitor in the living room space. Talking to students of the University of Texas' LBJ School of Public Affairs, he envisioned an unflattering future of a "dumbed down" living room, in which Apple was at the centre.
"I think that there's a scenario where we see sort of a dumbed down living room platform emerging — I think Apple rolls the console guys really easily. The question is can we make enough progress in the PC space to establish ourselves there, and also figure out better ways of addressing mobile before Apple takes over the living room?"
This wasn't the first time Newell was publicly critical of the mobile phone behemoth. In October 2011 , he appeared on a panel at WTIA TechNW, reported on by The Seattle Times , saying, "it's sort of ominous that the world seems to be moving away from open platforms."
"I suspect Apple will launch a living room product that redefines people's expectations really strongly and the notion of a separate console platform will disappear," he said, going on to call their closed nature the "wrong philosophical approach".
"I'm worried that the things that traditionally have been the source of a lot of innovation are going – there's going to be an attempt to close those off so somebody will say 'I'm tired of competing with Google, I'm tired of competing with Facebook, I'll apply a console model and exclude the competitors I don't like from my world.'"
Openness also came up during the Casual Connect talk . “In order for innovation to happen, a bunch of things that aren't happening on closed platforms need to occur. Valve wouldn't exist today without the PC, or Epic, or Zynga, or Google. They all wouldn't have existed without the openness of the platform. There's a strong tempation to close the platform, because they look at what they can accomplish when they limit the competitors' access to the platform, and they say 'That's really exciting.'
“We are looking at the platform and saying, 'We've been a free rider, and we've been able to benefit from everything that went into PCs and the Internet, and we have to continue to figure out how there will be open platforms.'”
To point out the obvious: Steam isn't an open platform. While there have been moves to make it more accessible to developers - specifically Greenlight, a service Newell isn't too wild about in its current form - Valve remain the gatekeepers for everything that is accepted onto Steam.
But given how many of Newell's comments praise openness, it seems absurd to think that SteamOS could be anything but. Certainly, it's open intentions are touched on by the announcement page. "With SteamOS, 'openness' means that the hardware industry can iterate in the living room at a much faster pace than they've been able to. Content creators can connect directly to their customers. Users can alter or replace any part of the software or hardware they want. Gamers are empowered to join in the creation of the games they love."
Again, though, it doesn't touch on how game creators not currently on Steam will be part of SteamOS. Will there be a separate area for non-Steam Apps, or will compatible games be able to integrate more seamlessly? Newell's previous comments do suggest a broad degree of flexibility. When The Verge asked about SteamOS running Netflix, he said, "absolutely. You can fire up a web browser, you can do whatever you want." Linux users will know that, thanks to its baffling reliance on Microsoft Silverlight, running Netflix is more complicated than opening a web browser, suggesting that the operating system will be similarly supportive of the workarounds needed.
The likelihood is that if a game can run on Linux, it can run on SteamOS, but it would be nice to think that Valve are planning to lower the barriers further: removing some of the requirements to entry for games, even while maintaining their own front-page list of approved titles. That's certainly something that's been hinted at in the past, when Newell made mention of user-created storefronts.
"An editorial filter is fine, but there should be a bunch of editorial filters," he told The Verge . "The backend services should be network APIs that anybody can use. On the consumer side, anybody should be able to put up a store that hooks into those services. Our view is that, in the same way users are critical in a multiplayer experience, like the fellow next to you is critical to your enjoyment, we should figure out how we can help users find people that are going to make their game experiences better."
That could look like a super-charged version of the current recommendations system, in which users become curators of Steam's huge and growing catalogue. The more dramatic and 'open' way to do it, would be to let any Greenlight-submitted game become a candidate for user stores, with the most popular being boosted into Steam's 'official' selection.
On the next page: Steam Machines and Gabe's vision of a connected future.
Yesterday Valve revealed the specs for their prototype Steam boxes. For an idea of how long SteamOS and its companion box have been in the pipeline, take a look at our series of interviews with Gabe and other members of Valve from September 2010. "So my job is always changing, right? That's the nature of the industry; it changes for a lot of people here. So right now I'm thinking a lot about longer term stuff," Newell told Tom , listing, among other things, "thin client architecture" as an avenue of exploration.
You can already see the results of that in SteamOS's ability to communicate with a Windows PC to stream incompatible games, but Newell seems to be planning to extend such capabilities in the future. In his interview with The Verge , he talked about the then "Steam Box" being able to send data through to multiple monitors.
"The Steam Box will also be a server. Any PC can serve multiple monitors, so over time, the next-generation (post-Kepler) you can have one GPU that's serving up eight simultaneous game calls. So you could have one PC and eight televisions and eight controllers and everybody getting great performance out of it. We're used to having one monitor, or two monitors — now we're saying let's expand that a little bit."
Clearly that's not a feature that you'll find out of the box, but then, that's the benefit of Steam Machines' malleability. Essentially, any device running SteamOS becomes a de facto Steam Machine, and - as their announcement post explains - "The specific machine we're testing is designed for users who want the most control possible over their hardware."
Pay particular attention to the set up mentioned in Newell's previous scenario. "One PC and eight televisions and eight controllers." That's specifically in relation to a Steam Machine as a server, not as a regular PC with an absurd number of HDMI slots. But what Newell doesn't mention is the eight PCs you'd presumably need receiving a signal from the main box. A few possibilities exist, of which:
Theoretically it should be possible. All it would need is a version of the OS specced purely to receive a stream from the central PC/Steam Machine. Lord knows, TV manufacturers are in desperate need of a valid reason to flog us more flatscreens (unless you want an expensive new 3D telly ... anyone? No?). Then there are small devices like Google's Chromecast and the Raspberry Pi, variations on which could network home entertainment servers with multiple screens/controllers. "Thin client architecture" doesn't have a defined shape yet.
That's not even the full extent of the futurism we can wring out of Newell's scenario. An eight player set up doesn't make sense as a luxury split-screen replacement, because PC games are traditionally not designed to support simultaneous players offline. To be of use outside very specific examples, the Steam Machine would need to act as a bespoke LAN, simultaneously running the profiles of multiple Steam users, and their copies of the game.
I'll admit that seems a little too far in the realms of wishful thinking. Even so, Steam's Big Picture mode can already support multiple accounts, which opens the possibility for them to one day take simultaneous logins. The flexibility of the system wouldn't even need to be that dramatic. If running multiple games is too far-fetched, it's possible other accounts could make use of some form of variation to Steam's family sharing - kids logging in to their profile from their own TV, accessing a house-wide Steam account with an added parental filter. If Valve do want Steam Machines as the server powering household entertainment, they're going to want to make the experience be as smooth and natural as possible.
On the next page: Steam's new controller might look a bit odd, but things could get a lot weirder.
That controller, eh? It's a big enough departure from the now standard 360 pad that I struggle to imagine how well it will work. Even so, it's a lot tamer than I was initially expecting, thanks to comments made about Valve's experiments with the relationship between game and player.
"Oh, I'm also thinking about biometrics," Newell told us back in 2010.
"So when you look at our games, more and more we have this representation of player state, where we think we know how you feel, essentially," he continued. "And with biometrics, rather than guessing, we can actually just use a variety of things like gaze tracking, skin galvanic response, pulse rate, and so on.
"Through combining those pieces of information, we can get a much more accurate indication of player state. So that's something we're super interested in. We've done some experiments in that space, and feel like there's some easy wins for customers and for developers."
Of course, being Valve, their plan for biometrics didn't end at the internal testing phase. Newell went on to tell Tom, "we think there are several people out there with interesting approaches on the hardware side [of biometrics]. Enough that we have confidence that the hardware side will be a sort of resolved problem in the not too distant future."
A wild flight of fancy, which soon afterwards they stopped thinking about, right? Nope. January, 2013 : "Maybe the motion stuff is just failure of imagination on our part, but we're a lot more excited about biometrics as an input method,” Newell said . “Your hands, your wrist muscles, and your fingers are actually your highest bandwidth, so to try and talk to a game with your arms is essentially saying, 'Oh, we're gonna stop using ethernet and go back to 300 baud dial-up.'"
It may contain a touchscreen and have fancy haptic feedback, but the Steam Controller announcement made no mention of biometrics. Given that, you could assume it was too difficult, or too expensive to fit inside a commercial controller. But if that is the case right now, it won't be forever.
As the announcement page claims: "We plan to make tools available that will enable users to participate in all aspects of the experience, from industrial design to electrical engineering. We can't wait to see what you come up with." In short, it's designed to be hacked, modified and iterated upon. As PCs with replaceable components, Steam Machines won't have a defined shelf life, which means we'll almost certainly see multiple revisions and new versions that build upon its controller's initial concept. Don't be surprised if one of them knows what makes you sweat.
Beyond that? One research project that's been teased on Valve employee Michael Abrash's Ramblings in Valve Time blog is 'wearable computing'. "By 'wearable computing' I mean mobile computing where both computer-generated graphics and the real world are seamlessly overlaid in your view," Abrash wrote in his introductory post . "There is no separate display that you hold in your hands (think Terminator vision)." Alternatively, if you didn't grow up in the 90s, think Google Glass.
Currently only at the R&D stage, Valve must still have some idea of how such a device could integrate into their existing products. SteamOS compatibility would seem like the perfect fit. It could easily show information peripheral to the game - everything from alerts, chat notifications and friends lists, to server browsers, Workshop lists and Steam guides. Expand out further, and it could offer video chat, or a Portal 2 style feed of your co-op partner's display, fitting nicely into Newell's previous server scenario.
On the next page: how SteamOS plays into Gabe's war against Windows.
We'll end with one of Newell's most memorable quotes from the last few years. No, I'm not talking about the time that he said he would "like to be super transparent about the future of Ricochet 2."
"I think Windows 8 is a catastrophe for everyone in the PC space," Newell told attendees of the Casual Connect conference in July of 2012. He was talking about why Valve were eager to get Steam's then 2,500 games running on Linux.
"I think we'll lose some of the top-tier PC/OEMs [original equipment manufacturers], who will exit the market," he continued. "I think margins will be destroyed for a bunch of people. If that's true, then it will be good to have alternatives to hedge against that eventuality." You can see more transcribed excerpts from Newell's talk courtesy of All Things D .
While Microsoft's maligned OS hasn't quite brought about the End Times, we can already see interest in SteamOS among one of the big two GPU manufacturers. In the last week, not only have Nvidia reached out to open source driver creators , but they've also revealed their part in the creation of SteamOS. Meanwhile, AMD's technological advancements have seemingly been driven more by their position inside both new consoles, than by any loyalty to Windows.
"The thing about Windows 8 wasn't just [Microsoft's] distribution. As somebody who participates in the overall PC ecosystem, it's totally great when faster wireless networks and standards come out, or when graphics get faster. Windows 8 was like this giant sadness. It just hurts everybody in the PC business.
"Rather than everybody being all excited to go buy a new PC, buying new software to run on it, we've had a 20+ percent decline in PC sales — it's like 'holy cow that's not what the new generation of the operating system is supposed to do.' There's supposed to be a 40 percent uptake, not a 20 percent decline, so that's what really scares me."
In this context, the focus of SteamOS - and its accompanying Steam Machines - makes sense. Realistically, SteamOS won't act as a replacement for Windows. Your desktop PC can already run Steam, and, if you're on Windows 7 or 8, the entirety of its catalogue. There's little reason for the majority of existing users to switch. By focusing on the living room, Valve are hoping to expand hardware growth by offering some of the advantages of PC gaming in a new setting. In doing so, there's every chance they'll further increase the already strong sales of PC games through their platform.
For now, expect SteamOS to create added value to the PC by bringing its flexibility into living rooms, and making it appealing to users used to closed systems. That's the goal, it seems, in the short term at least. That said, this is just the beginning of an operating system that will likely be tweaked and changed as often as Steam itself. SteamOS may not be a Windows killer on release. In the years to come? I wouldn't bet against it.
What would you like to see from SteamOS, Valve's new controller and Steam machines?