Valve's SteamOS Not a Replacement for Windows 8
Although this was touched lightly on Monday, Engadget's hands-on report regarding Valve's Steam Machine prototype revealed an interesting tidbit about SteamOS itself; it's not a replacement for Windows 8. This is likely bad news for PC gamers looking for an alternative operating system that not only plays Linux-based Steam games, but allows them to manage files, work on documents and use the Steam Machine as a typical desktop.
As previously reported, SteamOS is similar to Steam's Big Picture Mode except that this interface is the basis for the entire hardware system. Engadget reports that the same Steam splash page washes across the screen when it launches, and the same tile-based layout of games and the Steam store are visible at launch. The platform is also built on pure Linux, not Canonical's Ubuntu, making it a custom platform instead of a spinoff.
The report goes on to state that SteamOS is not a replacement for Windows 8, that it offers little functionality outside what's described above. "Beyond basics like browsing the web, there's little in the way of standard OS functions," Engadget reports. "While Valve reps showed off slides of the box's vanity shots using a Windows PC, I asked how I'd view such shots from within SteamOS -- the answer is that there's no real way to do so, as there's no file browsing system or image viewing application."
The report points out that customers of Valve's Steam Machine initiative aren't really shopping for a desktop PC, but essentially a game console that focuses on PC games rather than the typical Xbox/PlayStation envelopes. These machines will ship with a game controller and the SteamOS platform, thus allowing Valve to say that the device is capable of playing the entire Linux-based Steam library. However, the report puts an emphasis on what a Steam Machine really is: PCs posing as game consoles.
What's surprising is that, based on the report, there won't even be base level support for media playback, or streaming options like Netflix, Hulu Plus and so on that are offered on the current console crops. That will likely change, as Valve already indicated that movies and TV shows were coming to Steam; Linux-based software is also likely on the horizon. Unfortunately, the game streaming aspect wasn't available at the time of the report.
"We're working with many of the media services you know and love," reads the SteamOS page. "Soon we will begin bringing them online, allowing you to access your favorite music and video with Steam and SteamOS. 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."
We're probably just scratching the surface of what's going to be possible with SteamOS. We're also betting even more juicy details will be provided during CES 2014, and we'll be right there front-and-center!
Check out all our SteamOS coverage below:

That's the big problem with SteamOS, really. If it doesn't run DirectX (or incorporate something like Wine), it can't play most of the games on the market anyway.
Windows really doesn't have a lot of overhead. There's a lot of stuff loaded in memory which isn't strictly necessary for a gaming console, but this manifests itself mostly in the form of a fixed memory footprint rather than as a relative impact on some other performance metric.
Windows 8 has a memory footprint of around 300MiB, about a quarter less than Windows 7 which weighs in at around 400MiB.
Even if all the stuff unrelated to gaming were to be stripped away, the total memory footprint would be reduced to around 200MiB; this is negligible on systems that are sporting 8GiB-16GiB on average if not more. Stuff that is sitting in memory as part of a service or some other operating system component is doing just that, sitting there until it's needed.
That said, there's very little difference between an OpenGL game running on Windows and an OpenGL game running on Linux. Benchmarks have shown that there's almost no difference between them and that's to be expected. A properly coded application won't care what platform it's running on as long as the libraries that it needs are present and the APIs/ABIs remain the same. As long as the libraries and OS are reasonably efficient (and they are) the microprocessor will spend the bulk of its time inside of the OS neutral application code, as it should.
So how will SteamOS perform compared to Windows? Exactly the same
That's the big problem with SteamOS, really. If it doesn't run DirectX (or incorporate something like Wine), it can't play most of the games on the market anyway.
That's the big problem with SteamOS, really. If it doesn't run DirectX (or incorporate something like Wine), it can't play most of the games on the market anyway.
It wont have DirectX as DirectX is MS and they wont allow it on any other OS, if they did Linux would have had it years ago.
But that overhead you speak of, its pointless as the hardware these days is well over what games utilize. A clean Windows 8/8.1 uses maybe 800MB of RAM at startup and most systems have a minimum of 4GB, most gaming systems have 8GB and some 16GB.
Most games still use a 32bit exe which can only access up to 2GB unless they use PAE which gives up to 3GB. Very few use more than that.
That's all thanks to the current consoles as they are still very old technology. Even with the new consoles I don't think a system like mine, 2500K 16GB of RAM HD7970, will have much to worry about since it is still far ahead of the new consoles in computer power.
If you optimize it enough, DX and OGL games will run very similarly. Of course it also depends as OGL is sometimes behind in what features are available for it so you have to try to put them on equal footing which is hard since DX may support one feature and have it run that OGL does not and vice versa.
That's the big problem with SteamOS, really. If it doesn't run DirectX (or incorporate something like Wine), it can't play most of the games on the market anyway.
Windows really doesn't have a lot of overhead. There's a lot of stuff loaded in memory which isn't strictly necessary for a gaming console, but this manifests itself mostly in the form of a fixed memory footprint rather than as a relative impact on some other performance metric.
Windows 8 has a memory footprint of around 300MiB, about a quarter less than Windows 7 which weighs in at around 400MiB.
Even if all the stuff unrelated to gaming were to be stripped away, the total memory footprint would be reduced to around 200MiB; this is negligible on systems that are sporting 8GiB-16GiB on average if not more. Stuff that is sitting in memory as part of a service or some other operating system component is doing just that, sitting there until it's needed.
That said, there's very little difference between an OpenGL game running on Windows and an OpenGL game running on Linux. Benchmarks have shown that there's almost no difference between them and that's to be expected. A properly coded application won't care what platform it's running on as long as the libraries that it needs are present and the APIs/ABIs remain the same. As long as the libraries and OS are reasonably efficient (and they are) the microprocessor will spend the bulk of its time inside of the OS neutral application code, as it should.
So how will SteamOS perform compared to Windows? Exactly the same
Anyone interested in a replacement will likely look to Ubuntu 12.04LTS or Mint and just run Steam for Linux. It should play the entire Linux / Steam OS library and provide a replacement for Windows 8 (such as it is).
IB
It's Linux. Of course you can do that. You can do whatever you want with it.
Surely you didn't actually expect that Valve was planning on replacing an incumbent OS of 2 decades with the first release of what is really nothing more than another Linux distribution.
We've always been able to do that, if we wanted to give up half our standing game libraries and jump through hoops. The same goes for user-modification of SteamOS—sure, it may prove possible, but without official support its just going to be the same Linux experience we already have.
The hope was that it might be different this time if Valve could get the other industry heavy-hitters in the same room and convince them that its no longer in their collective best interest to stake their businesses on how much effort Microsoft puts into their OS and graphics API.
With the first release? Of course not. But the Windows business model of today is structured around why we need it, not why we want it. I think Microsoft is complacent and vulnerable, and if someone with some industry clout threw money at it and tried for a few iterations, its entirely possible that we could've ended up with a practical (if imperfect) alternative to Windows.
Now, even after all their statements about how terrible Windows is, I guess Valve's only interested in a basic functionality game library UI for tech-apathetic console gamers. Which wouldn't have been such big news if they'd just told us what it would and wouldn't do to begin with.
For the life of me I can't figure out why Google won't release an X86 desktop version of Android. Sure it's imperfect, but man, at least that would give Microsoft a run for their money.
We've always been able to do that, if we wanted to give up half our standing game libraries and jump through hoops. The same goes for user-modification of SteamOS—sure, it may prove possible, but without official support its just going to be the same Linux experience we already have.
The hope was that it might be different this time if Valve could get the other industry heavy-hitters in the same room and convince them that its no longer in their collective best interest to stake their businesses on how much effort Microsoft puts into their OS and graphics API.
With the first release? Of course not. But the Windows business model of today is structured around why we need it, not why we want it. I think Microsoft is complacent and vulnerable, and if someone with some industry clout threw money at it and tried for a few iterations, its entirely possible that we could've ended up with a practical (if imperfect) alternative to Windows.
Now, even after all their statements about how terrible Windows is, I guess Valve's only interested in a basic functionality game library UI for tech-apathetic console gamers. Which wouldn't have been such big news if they'd just told us what it would and wouldn't do to begin with.
The Windows model is based around add-value and has been for a while. Windows 7 is a great OS and Microsoft developed themselves into a corner. It's hard to monetize a product that's defined by periodic upgrades when there's no incentive to upgrade.
Companies have been throwing money at every attempt to dethrone Microsoft in the consumer OS and productivity software market segments for decades with no luck. The Office suite, Visual Studio, and Windows are well entrenched incumbents and will remain that way for the foreseeable future. Competitors have managed to keep Microsoft at bay in the workstation and server segment (although the incredibly lucrative Windows Server has made huge gains in the past couple of years) and largely shut Microsoft out of the mobile segment (again, Windows Phone is making slow but steady gains). Microsoft knows how to defend their market position and serves their developer base very well so don't get fooled into thinking that a handful of small publishers with no incentive to bite the hand that feeds will start throwing money at a problem that doesn't exist.
I do hope that you're being sarcastic