I can see a benefit to this, but also fear some major downsides.
Here's some areas for concern:
1) My first concern is that it's designed "for the livingroom". Does this mean that if I were to install it on a normal gaming PC hooked up to a normal keyboard/mouse/monitor on a normal desk that I would get some half-baked super-zoomed-in low-dpi experience? I hope that Valve will have a "desktop" and "tv" mode that caters to each setup type, akin to regular Steam vs big picture.
2) While I don't see them doing so, will Valve only permit games purchased through Steam to run on the platform? Are they going to treat this OS like a consoles' interface that can only run "singed" code, or will they keep the open-ness of Linux and treat it like a normal OS that just happens to have the Steam client built-in as a core OS feature/service?
3) This OS could cause platform fragmentation for PC game development, leading to developers needing to split their time and resources making two PC versions of the same game (in addition to the two console versions they're likely already making) causing the overall product to suffer. Either that, or they will need to pick-and-choose which OS they will support and either go Windows or SteamOS, effectively cutting off a portion of their PC audience who runs the platform that's getting snubbed.
4) Streaming games is well and good as a temporary solution, but Valve will absolutely need to find a way to get Windows apps to run natively or efficiently through emulation if they hope for this to be succedssful. Nobody's going to want to dump their back-catalogue of Windows games... This isn't a console where you can just dump everything that came before and start anew every 5-7 years. If people can't run Windows/DirectX games, they won't want to budge. The WINE project would be a good place to start for Windows emulation, but through such emulation you won't get anywhere near the performance you would by just running the game natively on Windows.
5) Drivers, drivers, drivers. Unless Valve has some plan to work with major gaming hardware manufacturers, specifically Nvidia and AMD, to entice them to develop better Linux drivers (or unless Valve can make their own that work well for the various hardware out there) this won't really take off. NVidia's linux support is absolutely atrocious and AMD sucks at making drivers for ANY platform (though their Windows offerings aren't as terrible as of late.) Hardware support has always been a pet peeve of mine with Linux, and I REFUSE to compile my own damn third party driver when the manufacturer should have made a decent one out of the gate.
There are some upsides to this though:
1) A dedicated gaming OS would be more efficient than Windows at running games natively developed for it. Valve has hinted to as much on their teaser page that they've seen such results. They can still make it a fully-fledged OS with all the bells and whistles of a Windows-class OS and ger such performance if they priortize the OS' processes and services properly to give the running game and its dependencies priortiy when a game is being played.
2) The streaming can be useful... in some cases. Personally, I plan to upgrade my current rig and use my old parts to build a HTPC for the livingroom to act as a media player and couch-gaming solution with Big Picture Mode. The streaming would allow me to use my new rig for processing the games and the older crappier hardware for displaying/control of the game to ensure that I don't have to take a huge graphical/performance hit when I game from the livingroom. Not that my old hardware is -that- bad, but I could run games on high/max in both setups instead of just at my desk.
3) Having a free OS to throw on there instead of having to buy another copy of Windows or cheat the Microsoft automatic Activation/grace period to let me activate my current copy of Win7 twice helps a lot too.
Log in to comment