Gamasutra: The Art & Business of Making Gamesspacer
arrowPress Releases
April 19, 2014
PR Newswire
View All





If you enjoy reading this site, you might also want to check out these UBM TechWeb sites:


Linux-based SteamOS is Valve's big play for the living room
Linux-based SteamOS is Valve's big play for the living room
September 23, 2013 | By Kris Graft

September 23, 2013 | By Kris Graft
Comments
    83 comments
More: Console/PC



As part of Valve's positioning for a more living room friendly PC game experience, the company dropped a bombshell: It's introducing the Linux-based SteamOS.

"We’ve come to the conclusion that the environment best suited to delivering value to customers is an operating system built around Steam itself," Valve said in its much-anticipated announcement.

"SteamOS combines the rock-solid architecture of Linux with a gaming experience built for the big screen."

Valve's timing for the announcement comes just a couple months before Sony and Microsoft launch their next-generation video game consoles.

Valve founder Gabe Newell has been vocal about his concerns regarding the possibilities of a more closed PC ecosystem, particularly with Microsoft's Windows 8, which he previously bluntly called a "catastrophe." Linux is an open source OS that Newell has said just needs more game support in order to make it a more viable platform, and Valve is spearheading that support.

Valve said SteamOS will be available "soon" as a standalone OS that's tailored for living room use.

SteamOS is just the first Valve living room announcement that's coming this week. Two more announcements are scheduled.

One of those announcements could well be Valve's Steam Box, a console-like platform that is TV-friendly, but Linux-based, and also plugged into Steam's 50 million-user-strong distribution platform.

But if Valve does announce its own box, don't expect it to be the only SteamOS device. Following Newell's vision of platform openness, Valve said it expects hardware makers to "iterate in the living room at a much faster pace than they’ve been able to." Users will also be able to alter the hardware in any way they like.

Valve said its OS will also facilitate closer relationships between game developers and players, and encourage more of a community among players.

SteamOS has been under tight wraps, but Valve says it already has game developers creating games for it. "In SteamOS, we have achieved significant performance increases in graphics processing, and we're now targeting audio performance and reductions in input latency at the operating system level. Game developers are already taking advantage of these gains as they target SteamOS for their new releases."

Valve said "hundreds" of games are running natively on SteamOS currently, and announcements about triple-A titles for 2014 are on the way.

The living room-friendly features that SteamOS will implement include the recently-announced family sharing option, in-home streaming of games from an existing PC or Mac to your SteamOS device, music and TV services (Valve said expect more about that soon) and more family-friendly settings.


Related Jobs

Activision Publishing
Activision Publishing — Vancouver, British Columbia, Canada
[04.19.14]

Principal Graphics Programmer
Activision Publishing
Activision Publishing — Santa Monica, California, United States
[04.19.14]

Executive Producer-Skylanders
Insomniac Games
Insomniac Games — Burbank, California, United States
[04.18.14]

Associate Engine Programmer
Insomniac Games
Insomniac Games — Burbank , California, United States
[04.18.14]

Senior Engine Programmer










Comments


Phil Maxey
profile image
Next two announcements, Steambox and HL3 in that order? Anyway price is the key to Steambox, if they make it $199 and less it will sell like hotcakes, if they don't, it won't. SteamOS great idea though, is it just me or can you feel the computer games world tectonic plates moving?

Hugo Cardoso
profile image
I may be cynical and jaded but I really doubt it's Half Life 3. While Half Life is their main franchise if they announce a game I would guess it would be L4D3.
It's been leaked that they're working on it and in terms of promoting the SteamBox I think a multiplayer couch co op title is a better bet than a great single player game.

Phil Maxey
profile image
But can you imagine if they did make HL3 exclusive? that would really put the cat amongst the pigeons.

Karsten Schlachter
profile image
according to my interpretation of those cryptic signs:
() [() ] () + ()
i'd also gues it's going to be someting like software, box, software
..and even thoug i'm afraid hugo is right i still hope sooo much it's going to be hl3

Vinicius Couto
profile image
I liked your interpretation.
And, going a bit further, I'd say the "() + ()" thing is about sharing and cooperating with software.
I guess that's the announcement I'm most anxious about.

Bob Johnson
profile image
Well if Steambox is only a streaming receiver then it should be cheap.

$199 a bit much. Needs to be $100.

sean lindskog
profile image
It is not just a streaming receiver. It's a linux-based "console" that can play linux-ported games. The streaming is an extra feature so that your steambox isn't limited to linux-ported games only.

warren blyth
profile image
Well, these announcements are on the Steam service website, not the Valve Software (games) website. I doubt they'd give any internal game the unfair spotlight by pimping it as the third pillar of Steam's evolution.

I'd bet the third announcement will be something about a mobile OS, or input UI, which works with the SteamOS (thus the plus sign between two circles). hopefully it'll be some combination!

Mobile is the grossest of walled gardens festering out of control, so I really hope they're have some ideas to rock that boat.

but I also hope they have ideas for enabling new inputs for more interesting gameplay (from tablet surfaces to biometric monitoring to vr goggles). I used to struggle to get game devs to support a funky 6DOF input device, and they never had time/budget to consider it - so it'd be nice if steam offer input standards that hardware innovators could tie into.

Right now I see the only reason to get an Xbone or WiiU (instead of sticking with PC) is because you want the unique feel of kinect or the wierd experiments with fast gamepad input. So if they can kneecap that selling point, it'd make steambox a real competitor?

Bob Johnson
profile image
@Sean

I don't think its a console per se. It is a "Shield OS." The games it can run directly will depend on the hardware it is on. The stuff which the hardware can't run can be streamed from your pc.

So you will have ~$100 low end mobile hardware that accepts AAA games streamed from your pc and you will have $1k+ SteamOS pcs that can play everything directly.

Eric Pobirs
profile image
Considering that the hardware side of the platform isn't as rigidly defined as a traditional consoles, there may not be any one SteamBox and it may not be produced by Valve. Instead, there could be a certification program that PC vendors submit hardware designs to receive the right to carry the SteamBox badge on the case. This would allow for a very wide range of price and performance levels. At the lowest level might be a box somewhat close to the new consoles in that it will have an AMD APU at its core. You can build a decent A8 machine for a pretty low price and with a gaming optimized OS it could preform pretty well for the money.

Such a product is inevitably going to be fragmented compared to traditional consoles but such is the cost of choice.

John Paduch
profile image
Everything about this makes my inner-PC fanboy wanna SQUEE! Everything mentioned on the page sounds legit, and could be a potential game-changer not just for PC but for the console environment as well. Damn, I'm excited for this. \m/ >.<

Bob Johnson
profile image
If this means I don't need to buy Windows when I build a gaming pc then I am all for it. Not too interested in playing pcgames on the couch.




Dave Kay
profile image
I'm on the fence with this one, too.

On the one hand, I'd love to be able to sit on my couch and play any of the couch-friendly games effortlessly... but on the other hand, most of the games I really want to play, for me, just aren't the same without a keyboard and mouse.

What I really need is something that is a true replacement for a mouse that can be used comfortably from a couch. So far, nothing comes close. Sorry, analog sticks for a shooter? Hell no. Maybe it's time for the trackball to make a comeback.

Jay Anne
profile image
@Dave
Valve has been working on a mouse replacement for quite some time now. They looked into the Novint Falcon, Razer Hydra, etc. I suspect they were disappointed in those and decided to make one internally, hence Gabe's comments about their disappointment in input device innovation. I really hope one of their announcements will verify this, because a pointer-oriented couch input device would have benefits for both TV watching as well.

Luis Guimaraes
profile image
Thumb trackballs are pretty fine (90-95% as good as mouse) for FPSes and clicking on stuff, but ideally for gaming it should be the size of a pool ball to give more room for the thumb to work, so the player doesn't have to trade-off steadiness for speed and vice versa.

Then add the analog stick for movement and make everything else shoulder buttons (scrap the grip system and make it so it can be holden just by using both palms) to use 3 or 4 fingers on each side, with 6-8 buttons each side. If physically possible, add a scroll wheel too on the right shoulder.

Leave the face for small function buttons (from ESC to F12) and it'll be perfect.

Curtis Turner - IceIYIaN
profile image
90-95%? I'd like somebody to develop an app to put that theory to the test. If we're lucky, Valve will do it and reveal the global statistics of the input wars. Then finally we can all go, not only will YOU suck, but you will suck THIS MUCH, n00b!

I'd like at least four fingers on the back of the gamepad aswell. I've used ASDF touch typing movement since 2000... Thank you Nox, you made me a beast at Counter-Strike, World of WarCraft, and countless other video games.

A = Back
S = Forward
D = Strafe Left
F = Strafe Right

Dane MacMahon
profile image
Not to let my PC gamer flag fly too high, but an analog stick is a truly horrible way to control a camera or first-person view. Yet the vast majority do so without complaint, because its comfier.

As much as I love my mouseaim all Valve have to do is use a tolerable method to replace it and if people want the box and platform in general they will adapt to it.

Luis Guimaraes
profile image
@Curtis Turner - IceIYIaN

Well that's my guess of course, but I have only around 110hs (sparsely spread across 2 years) or so using it for FPS, about 30 on multiplayer and the rest on single player campaigns. But I work with it all day everyday for 2 years.

I might be completely wrong and it could be 100% as precise/versatile as a mouse if you play with it for a decade as it's the case for mice. I don't know if you can't run accurate tests without selecting those player and also having them being on equivalent skill levels and being on their best day.

I have become a thumb trackball evangelist recently, making some videos to raise awareness and prove to people how good it is:

http://www.youtube.com/watch?v=RKmIXJs80Tk
http://www.youtube.com/watch?v=ydj141KdsjM

Dane MacMahon
profile image
@ Luis

While I am sure I would get used to it, like console gamers got used to analog sticks, I can see just from watching half the first video there is a lot of jittery short movements and an inability to turn 180 degrees in one smooth movement. These things would annoy me greatly.

Adam Bishop
profile image
I know I'm not the average case, but I already sit on my couch playing PC games effortlessly; my PC uses my TV as a second monitor, I have a 360 gamepad, and I just flip Steam over to the TV if I want to play a couch & gamepad game.

Luis Guimaraes
profile image
@Dane

There three causes for that jittery:

a) Muscle insecurity: It was a lot worse the first time I played with it two years ago, my hand looked really scared of that alien thing. It gets better with practice. Today I don't shake as much but it takes a bit of warming up. This player have used it for his entire life just like we've been using mice: http://www.youtube.com/watch?v=sorCllWOQDY

b) Sweaty hands: it's very, very hot where I live, some days worse than others. Whenever it looks like I'm struggling to keep my aim steady it's because it really is, my finger is sliding all over the trackball and it sometimes doesn't move or I over compensate. Not sure there's a solution for that apart from air-conditioner and a towel...

c) These thumbs trackballs are too small and not made for gaming. As I said, this makes it so you either have to put sensitivity too high and beyond your shakiness (a) threshold, or too low that you have to be rolling it to turn around, sometimes more than once specially if your hand is sweaty which makes the ball sticky (argh) and it doesn't roll freely as it should.

EDIT:

Something like this would solve the area and sweat problem, but can't be rolled freely as a trackball for 180 turns and such, so I believe I'd still prefer a trackball of a poll ball size: http://www.youtube.com/watch?v=kiUUlaysqF8

Another aspect of it is that it's barely noticeable while playing. Only after watching the videos I was quite impressed of how shaky my aim was.

[User Banned]
profile image
This user violated Gamasutra’s Comment Guidelines and has been banned.

Dane MacMahon
profile image
@ Ronny

It comes down to making it as convenient, easy and consumer friendly as possible. You might think booting Windows on a TV and using a mouse on a lap-pad is awesome, but a ton of people find that cumbersome.

If the SteamOS box boots right into a Steam UI that is 100% controller-focused then it is just as easy as a console. This is attractive to consumers. Consoles sell more than gaming PCs (in the US) because they are more convenient.

The issues come after that: how many games will natively support SteamOS? How will you play mouse and keyboard games? Etc. etc.

Luis Guimaraes
profile image
How could I be forgetting: split screen!

Harry Fields
profile image
I have a "steambox" I'd love to put SteamOS on, but then, won't most of my games stop working? I dunno'. Proof is in the pudding, and this is a massive undertaking that will cost $$$. This will be Gabe Newell's Austerlitz or Waterloo. Before, Valve was a nuisance to many publishers. This changes everything up so much. Considering they release a game a year or so, they're betting the farm on other publishers and developers supporting this. At an Indie level... sure, I can see how compelling this is. For AAA stuff, I'm not so sure. MSFT is going to do what they can to sink this and quick. Apple and Amazon are probably looking at this a little apprehensively or at least cautiously and likely do not wish them well. Will EA change their stance and see this as another platform to release on or will they see it as continued competition for Origin? What about Acti-Blizz? If MSFT gives them favorable terms on licensing for Xbox in order stay console/PC exclusive, then SteamOS becomes very, very niche.
But again, maybe this is the birth of something bigger... A lot of folks think Steam can do no wrong. I'm not surprised by this announcement, but I sure as hell don't know what to think of it yet.

Bob Johnson
profile image
MS is in a weird position because they have let pcgaming go for so long. They have really wanted the consumer to buy an xbox for games and buy Windows for other stuff.

Improving Windows for gaming, especially living room gaming, conflicts with their Xbox business.

Kujel s
profile image
@Harry this is a very reasonable response, myself I'm very skeptical this will pan out but we shall see.

Alex Covic
profile image
I myself use Linux for 20 years (since kernel 0.99) and have seen ups and downs towards game development for Linux. Loki and others tried it in the past. Carmack once in favor, did a 180 not that long ago.

Valve hired a bunch of well-known Linux developers, who - in the shortest time - not only "finished" SDL 2.0, being in development for many years (look it up, to understand what it is), but also really went for the "SteamBox" in the only way that makes sense?

Have a(n open source) platform, but restrict it to specs every dev can count on and work with and for? I know, there are Richard Stallman types going to object to this. But Linux without the "GNU", without the strict liberal ideas of "everyone write your own drivers" and "choose the package managers and liberaries you want", is kryptonite for developers, who have to plan and rely on the idea, that their game in 2-3 years will still be playable on more than just their own system?

The uprising in smart phones, progress in web development and the success of Apple made OpenGL interesting again, for many developers. It is partly responsible for the "new" push to cross-platform development. Only this background, made a "Linux" version of recent games an afterthought worth having?

I don't know how far Valve can push this. It takes more than just one wealthy company to turn this boat around. It takes many, MANY million Desktop Linux users more, to make the Linux platform attractive for developers, to make more than just a tip-jar amount of (additional) money, form a Linux port.

If Steam pushes their name as a brand, consumers don't really need to "care" on what OS their game is running, as long as they can get the games they want?

Most people don't even know, their cars are running with Linux, their Smart TV's have already embedded Linux systems on them, traffic systems, powerlines, refrigerators, even the NYSE is "powered" by Linux. Consumers don't care. They don't need to know.

And THAT is, where Valve can succeed with their SteamBox. "It's just a device that plays your games" is a smart idea. They compete with other boxes (and consoles). Partners can be found. The hardware manufactorers want to sell more. Here is another way how to do it.

I wish Valve/Steam all the luck. Because, I don't want to use Windows 8, 8.1, 9, 10 ... to just play games, while I do everything else on Linux ... for the last 20 years.

Tim Turcich
profile image
If the main point is games, this box might not be very constructive for years. A new Linux based platform is a major hurdle to overcome. SteamOS Big Picture is to your living room like Android is to touch screen phones. Lots of Linux platforms have failed in the past even if steam network is more proof positive into the future. Like most new platforms the software line up in the beginning will be horrible, valve games aside. It will have to be cheap. Just like android phones were at first. On a side note, it's really sketchy(to me) shoehorning games that were designed for PC into the living room equation. This sounds far better as an cheap alternative to a windows based gaming PC at this stage in the game, but your gonna need to install windows to get the full game line up back. I think the most people will inevitably buy into whatever has their favorite game line ups. Long road ahead for this! I wanna see games! Platform Smatform!

sean lindskog
profile image
I think there are a few types of people who should be really excited about this.

1. PC game developers and players
Valve/Steam have done tons to promote PC gaming, and this should further the cause. The only downside (or upside, depending on how you look at things) is that devs will need to port their PC games to linux. If SteamOS succeeds, very likely most major game/graphics engines will evolve to support linux ports as a base feature.

2. Linux game devs/users
It's unclear how much SteamOS will differ from other linux distros. But it is still linux based. This will inevitably lead to more game linux ports, which likely should work (maybe with some modifications, or maybe none will be required) on both SteamOS as well as standard linux distros. Wide adaptation of SteamOS should lead to improvements in linux in general (e.g. better linux drivers from computer hardware companies) and perhaps wider adaptation of other linux distros as well.

3. Open source / open platform advocates
If you buy in to Newell's concerns about closed platforms, this is good stuff. It's not clear to me yet whether SteamOS will itself be open source. Although I would imagine the linux OS licenses would require large parts of it to be. Regardless, if SteamOS results in Linux becoming a major port target for games, this drastically changes the landscape towards a more open source and open platform game dev environment.

You gotta remember, if you make games for Windows, you gotta play by Microsoft's rules, which they could change at any time (e.g. requiring certification, and/or only selling windows programs via a closed windows app store). That is already the case for consoles. Not just anyone can arbitrarily create and sell games for XBox/PS/Nintendo. Alternatively, if you make games for SteamOS, and Valve (uncharacteristically) became some draconian evil empire, the gaming world could easily migrate to another linux-based platform.

Re: Big Picture Mode
I can understand some of the mixed enthusiasm towards big picture mode. There will be some pressure to make PC / SteamOS games more compatable with a controller, rather than keyboard & mouse. However, I'm pretty sure there will still be lots of folks who play SteamOS with a keyboard and mouse as well - particularly for genres of games (e.g. RTSs, MMORPGs) which naturally lend themselves to a keyboard & mouse. This is assuming SteamOS will support using a keyboard and mouse - I don't know if that has been confirmed. But I would be very surprised if it does not.

Dane MacMahon
profile image
I'm two of those three, PC gamer and open platform advocate, but in not very excited. Main reasons: 1) I play a lot of classic and older games, and I don't want to stream them. This new OS wont play all my favorites natively for a long time, if ever. 2) I don't really have a drive to move my gaming to the living room. If I did I would probably buy a PS3 and play GTA5 six months earlier.

How unique I am among those audiences I don't know. I think it's important to remember though that Windows is still an open platform and will be for a long time, and PC gaming is different from console gaming in ways PC gamers like.

sean lindskog
profile image
Fair enough.

For me personally, here's why I'm most excited. I make (and play) indie PC games. Anything that might increase the audience for these types of games is good, regardless of whether it is on a linux, windows, or mac PC. If this wins over some of the console crowd, fantastic.

The open platform stuff is also pretty significant to me. True, Win8 is an open platform. It seems to me that something has spooked folks like Newell into believing this may not be true forever. At the very least, the ball is in Microsoft's court, and they can do anything they like with it.

Even if Win were to remain an open platform, but Microsoft's app store became the dominant market for PC games (like Steam is now), I would consider that worrisome. Digital distribution, which has been great for both indies and PC games, may have become a very different thing had it evolved under a company with different core values than Value.

I like windows. I use windows. SteamOS, if successful, could exert the kind of pressure needed to keep windows an open platform, while simultaneously providing an important open source alternative.

Dane MacMahon
profile image
Lots to reply to here.

I like indie games. Not the ones the press usually rave about, but I like them. And I do hope the PC keeps its indie success rate going in the face of broader indie pushes from Microsoft and Sony. If this SteamOS helps with that, that's a good thing. For me personally though I don't really have any drive to play those indie games on my TV. A lot of people do though, fair enough.

Newell is definitely worried, but I also think he's a businessman. Even if Win9, Win10 and Win11 are all open platforms I still think Newell would rather people played on Steam without Microsoft being involved or being competition. Which, you know... capitalism. I get it. It's fine.

I have to kind of chuckle at the core values thing. I suppose for developers Valve is a lot better to deal with than others, but as a consumer I just see a long and nasty EULA that says they can take my games away, activation and account-based DRM that takes my games away if I ever lose my account or Steam shuts down, a forced client I don't really need or use, forced updates to games even if I turn off automatic patching and a bunch of other annoyances, etc. etc.

Steam works well and is convenient, so people love it, but it's not digital delivery in a nutshell. I always get shouted down when I complain about it, and I own 400 games on it so obviously it's not the devil or anything, but at the end of the day stronger competition would be a GREAT thing. Not sure I want Microsoft to provide that, I would rather Amazon or GOG, but I'll take what I can get. A repeat scenario of Amazon pushing Apple to go DRM free and higher bitrate on their music would be ideal.

Anyway, long almost unrelated topics going on here. My fault.

At the core I am probably not the person to ask because I don't get the living room thing. One big reason I don't have a PS3 already to play Red Dead, GTA5 and Dragon's Dogma is because I don't want to hook it up on my TV and play it in the living room, and my monitor is 16:10. I'm not the guy to ask about what sacrifices people will make to enjoy their couch.

I just think, on the surface, this doesn't offer a compelling argument for PC gaming on the TV until it has a large library of native Linux games. Whether Newell can pull that off or not I have no idea.

sean lindskog
profile image
Yep, I get your critiques. I am also a fan of DRM-free games, and released my own DRM-free.

So when I talk about Steam's core values, I am not saying they are 100% perfect. But some things they do seem to value:
1. PC gaming
2. indies, to a certain extent (much more than traditional publishers did).
3. fair profit splits to developers. In modern digital distribution, devs get most of the money. This was absolutely not the case when publishers controlled the gateway into the retail market.
4. marketplace competition. Steam has done nothing to crush the competition, and in many ways has embraced it, for example by providing a Desura link on greenlight pages. You rarely (never?) hear of Steam purposefully taking moves to crush smaller rivals, in the way many large corporations often do. True, SteamOS/SteamBox does signal a march to battle against Microsoft in some ways, but that was largely in response to a perceived threat from Microsoft to their own existance.

That's all pretty respectable stuff, imho.

Dane MacMahon
profile image
You're not wrong, not at all. I just want then to have competition, mainly. If that means a better version on Games for Windows Live then so be it. I would rather it be someone else.

Also I have a lot of memories in my head of pointing out flaws Steam has and people responding "why don't you like digital delivery?" I endeavor to point out they are not one in the same. If I could buy every game GOG style I would.

David Rico
profile image
I definitely think this is great news for both developers and gamers alike. For independent developers, Steam Greenlight is a far more viable alternative than publishing on Xbox One or PS4. In addition, there are so many more tools to publish to Linux like Unity, which allow for painless porting of games to SteamOS. I don't even know how convoluted the process of publishing to Xbone or PS4 is.

Yes, controller support may start being pushed more. That is both a plus and a minus. Certain games just work for controllers better than others and I'd hate to see something like a realistic flight sim require dumbing down just to support a gamepad (although I doubt it, far more likely that it just won't be offered, or the user will get a big warning telling them to grab a keyboard).

The other nice thing that I think people don't think about, is that likely this will be THE platform for the Occulus Rift in the console space. They already said they have no plans to support consoles, but PC is already supported. Given Valve's stance on the Rift so far, I'm going to make an educated guess that they will support the Rift from day 1 of SteamOS release.

As the dev of a game that's been recently Greenlit (Rebirth, survival horror sandbox game) I'm very happy with this announcement as it means I'll be able to play my game on my TV and scare the bejeezus out of my wife.

Win/Win!

Jay Anne
profile image
Valve is ruthless about making sure to only release products they've meticulously tested that tell them that success is ensured. As their products get more and more ambitious and leave the realm of just "games that are fun", it becomes harder to run tests and gather data that tells you that success is ensured. I would love to know what kind of testing and data verified that stepping into the ring with Sony and Microsoft had a good chance of succeeding. If the 2nd announcement will be console hardware, they will also have greatly stretched the boundaries of their core competency by an order of magnitude. This is very interesting news to anyone interested in how businesses are run, even if you are not a gamer.

Harry Fields
profile image
Indeed. Manufacturing and logistics can be a real nightmare.... an expensive one, too... especially if you go with the make your money on the backend approach most consoles do.

To that end, I see them simply licensing this to vendors. Official "Steamboxes", accordingly, will be at least as expensive as a console. I don't know... As mentioned in another post, I'm surprised, baffled and don't know what to make of it all.

So any games can stream or just games that aren't Linux compatible? They're stepping into the ring with Onlive/Gaikai? The streaming business model hasn't yet proven very lucrative. It's expensive to run GPU intensive applications in the cloud.

Jay Anne
profile image
@Harry
I assume the streaming is similar to the Nvidia Shield. It is not a business or product in and of itself, but merely a software compatibility trick that widens the effective catalog and gives the existing install base a door into the walled garden without buying prohibitively expensive hardware.

Harry Fields
profile image
Just reread the article. I missed the end of it the first time. So it's streaming from another PC. We'll see how it works. That would be ideal for me, since my steambox is a little underpowered compared to my workstation in the office upstairs.

Merc Hoffner
profile image
Hah! Called it. http://www.gamasutra.com/view/news/200368/Valve_still_sees_Linux_
as_the_key_to_PC_game_success.php

And still it begs the question: DirectX ? Solution?

E Zachary Knight
profile image
Last I checked, Linux has no native Direct X support. So the solution, if you want to target Steam OS, is to ditch Direct X for OpenGL and SDL.

Wyatt Epp
profile image
Maybe, maybe not... ;)
http://lists.freedesktop.org/archives/mesa-dev/2013-July/041900.h
tml

Harry Fields
profile image
Solution is: Valve has to get enough installs to make it worth developer's while to include an OpenGL renderer for ease of port to Linux. Unless there is a market there, there's really no compelling reason to support OpenGL on Windows, especially if you already have significant code assets written for DirectX.

Merc Hoffner
profile image
To a developer if you build on Direct X then I guess you reach PC and Xboxone (and 360), and that's the current driving force right? But if PS4's OpenGL variant and whatever Valve are building and OpenGL ES are similar enough, could that become the preferential route? In a new generation, how significant is your legacy investment?

Harry Fields
profile image
Well, your people are your biggest investment so it's all on where their skillsets are. Now sure, programmers are smart cookies and can switch apis, but if they've been doing directX their entire career and have to learn OpenGL, well.. that eats at profitability. That said, there are a lot of people who work with OpenGL/WebGL and so in the end, it's where your employees are at skill-wise.

Damian Connolly
profile image
The next big piece of news is that HL3 will be linux-only :P

Mike Griffin
profile image
Valve, the software company, would lose tens of millions of dollars if they shunned other platforms with their imminent major threequels, HL3 and L4D3.

It just wouldn't make good business sense to limit your games' potential reach, and Valve has deep investments/relationships on the PC and traditional console side, as a game developer.

On the other hand, this doesn't preclude them from releasing their mega-sequels to SteamOS first in a limited window of exclusivity, to sweeten the platform's appeal and ignite early sales of their hardware partners' SteamOS Certified-Linux boxes.

[User Banned]
profile image
This user violated Gamasutra’s Comment Guidelines and has been banned.

Dane MacMahon
profile image
Double phone post of doom.

Dane MacMahon
profile image
The streaming aspect makes up or an almost complete lack of Linix support from publishers in the short term, sure. However I don't see it as a long term solution to a lack of games that run natively. Hopefully they are pushing Linux support hard to publishers and indie devs, though they might not have the numbers to make their case until this launches.

Even then... Color me skeptical. If you're going to stream games you lose a decent amount of the graphical fidelity and framerate that makes PC gaming a superior experience. And of course you're not using a mouse on the couch. Why then would this pull console gamers away from consoles?

Seems to be a limited product for PC gamers who want to stream to the TV then for occasional play, just like nVidia's Shield, then. That's fine, and I might even want it, but it's pretty underwhelming compare to this idea of Valve revolutionizing PC gaming.

Maybe I'm just not seeing it?

Jay Anne
profile image
Sign up Minecraft, Skyrim, Call of Duty running natively and you've got a decent foothold to build off of. Very curious if they ever figured out how to play Dota 2 on a TV.

Dane MacMahon
profile image
Something your comment made me think of... this is kind of doing two things at once, isn't it? The SteamOS and the living room initiative are almost completely separate things on some level.

They could have launched the SteamOS for PCs in general, then made the living room push after a library was built up.

Chad Wagner
profile image
Ding Dong the Walled Garden is dead! Long live the Steam Walled Garden!

Dane MacMahon
profile image
*tips hat*

Jeff Leigh
profile image
My question is - is this actually an open platform for game developers the way Windows currently is? I.E. - developers make and deploy non-steam games that runs on Steam OS?

Can other companies run their own app stores on Steam OS - Google, Amazon, etc.? This was one of their big reasons for considering Windows 8 "catastrophic", was that Steam couldn't compete with Microsoft's app store.

Have they made any sort of promise to keep the platform open to developers that are not interested in distributing through steam (and paying Valve a royalty for slapping their name on Linux?) Do all games have to bee "greenlighted" to run? etc.

If Steam OS is just Valve's flavor of yet another "walled garden", that's not the same as 'open' the way Windows is currently open.

sean lindskog
profile image
That's a great question.

Given the nature of linux, it may be difficult legally to make SteamOS a walled garden.

Even if it were, I suppose we're better off (as consumers and game devs) to have multiple walled gardens to choose from. The competition keeps the prices and royalty rates fair, and also gives us a choice of walls to choose from - for example walls that are more friendly to indie developers.

It is an investment in the future of PC gaming, regardless of the nature of the walls.

Jay Anne
profile image
@Jeff
I would not be surprised if it is a walled garden like Steam is. As much as businesses like to complain about walled gardens from the outside, it makes perfect business sense when you are the gardener.

Vinicius Couto
profile image
Karsten Schlachter up above interpreted those announcement symbols as:
( ) = software
[( ) ] = hardware.

Going with that idea, maybe the ( ) + ( ) is the part about sharing and offering software.
You know... how Gabe's been saying he doesn't want Valve to be the Guardian of Which Games Are Sold. He's been wishing for a more open environment...
Maybe that's what the third announcement is about. That's the one I'm most anxious for.

Jeff Leigh
profile image
"Walled" is just a synonym for "closed" that sounds a little better. If the platform is a "walled garden" then it is not open. An "open" platform would mean "no walls".

I think we definitely need more facts on this one. If Steam OS is Valve replicating point-for-point Microsoft's (failed) attempt to create a more closed platform but Valve calls it 'more open'... is that actually true or false?

Vin St John
profile image
"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. SteamOS will continue to evolve, but will remain an environment designed to foster these kinds of innovation."

Most of that is just talking up their community features, but the relevant part is "users can alter or replace any part of the software or hardware they want." That sounds to me like a clear-cut promise for actual "openness." I would imagine that, like Android, the OS is tailored to make it easy and obvious to install software from Steam, and only knowledgable users will ever bother adding software from other sources, until Amazon makes a popular enough app store for SteamOS.

Jonathan Adams
profile image
This is going to be a really interesting generation. The gaming media must be dancing in the streets from all of these new things to talk about.

Mike Griffin
profile image
I've been piecing together many of the breadcrumbs left by Valve since approximately 2011, when their Linux initiative really started to take shape.

Among other bits, here's an interesting older blog post from 2012:
http://blogs.valvesoftware.com/linux/faster-zombies/

Interesting -- in terms of Valve's work with the major PC hardware players on Linux support, and work improving OpenGL vs. Direct3D.

Earlier this year, Valve hinted that it had secured the support of 15 to 20 computer system vendors with regards to its proposed living room Steam device. Today they declared "hundreds" of games are coming to SteamOS, including many AAA titles.

Obviously strong technical game support was a pillar of Valve's convincing of these third-parties to hop on board within a Linux environment, and Valve used its own titles as a benchmark for that performance and viability testing over the last few years.

They've probably contributed more to Linux' potential legitimacy as a living room games OS in the last 3 years, than efforts of the last two decades. Of course that's what you do when you're Valve and you've selected which horse to run in the race. You take a long shot and make it a winner.

Michael Joseph
profile image
There is a vacuum of love for the PC in this day and age of portables and consoles. I think SteamOS will do very well because there is little reason for developers and for PC gamers not to love it. Valve has garnered a lot of goodwill over the years and gamers trust them.

And you have to appreciate the mentality and boldness of Valve to not let their fate or their maximum potential to be be decided or limited by others such as Microsoft. But we've seen this from Valve before. They stopped allowing their potential to be limited by publishers years ago and they are once again acting like a company that believes in itself and that makes it the type of leader that others will want to follow.

Andrew Syfret
profile image
It's ambitious and very cool, but I'm a bit concerned it's a desperation play in reaction to their primary host OS choosing to start their own walled garden and marketplace.

The reason Valve can take so long to build games and spend millions on hardware teams is because they own PC digital distribution. Windows Store, while it's not a problem for them right now, indicates that MS wants this territory and will begin pushing into it. Valve has to account for this risk, but there are no platforms left which will guarantee openness...except one. Hence they go Linux.

However, just because it's a desperation play doesn't mean they won't succeed =D Valve is definitely one of the company's that could revolutionize the gaming space (they have form).

At least they are honest about Linux wrt performance problems. It's infuriating trying to discuss these things with Linux fanboys as they won't admin it or claim it's not an issue.

Troy Walker
profile image
I'm not holding my breath that this will shake things up... if anything, I see this as being a huge money pit for Valve.

Samuel Prince
profile image
Am i the only one that thinks that the third anouncement will be an Open-Source Source Engine 2? For me it makes sense because it's a good way to encourage development for the platform. But i guess we will see...

Kujel s
profile image
This sounds very likely and very Valve like.

Karsten Schlachter
profile image
this really is a great idea!
at least they could at least open source the current source engine which is still "sort of state of the art"..at least good enough to make descent games..but that would only help indie devs..

..to enhance aaa support of linux it sure would have to be source 2..go valve!

Daneel Filimonov
profile image
https://developer.valvesoftware.com/wiki/SDK2013_GettingStarted They already have. For a couple of months now. It's a huge step because they have released the source code and everything, and they are managing it through GitHub so anyone (capable) can make a change and add to it.

EDIT: To make clear, this isn't the Source 2 engine but it does have fragments that will lead into the Source 2 framework when they are satisfied with the 2013 engine (or perhaps they will update the 2013 engine to use S2 assets).

warren blyth
profile image
Agreed, Samuel. I expect a source engine 2 announcement.

Dunno if it'll be open source. I figure the hook will be that it will make it very easy for devs to port their game across the three current OS options, and tie into the already announced SteamOS and SteamMachines.

[User Banned]
profile image
This user violated Gamasutra’s Comment Guidelines and has been banned.

Michael van Drempt
profile image
Has anyone considered that one of the announcements could be an open Steam API? That would be a rather dramatic change.

Karsten Schlachter
profile image
earlier this year gabe stated in an interview that they're working on that (or at least something like userers getting the power to create their own stores) so that's totally possible
...and would make sense to do right now - especially because of the current "just another walled garden" critics, which was quite predictable..it definitely would be the right timing

Daneel Filimonov
profile image
They are in closed beta (for those who have access to Greenlight and such) with Steamworks which is an early step into the open Steam API. Once all the kinks have been worked out, I'm sure they will release a more user/consumer-friendly version for the public.

Duvelle Jones
profile image
Why do I get the feeling that Ubuntu is the base for StreamOS? Why does that disturb me should it be true?

E Zachary Knight
profile image
Ubuntu is just a custom flavor of Debian, which has one of the largest market shares of all Linux distros when accounting for all branches of it. With Valve announcing Linux support almost exclusively for Ubuntu last year, it would make sense for Steam OS to be either derived from Ubuntu or its Debian parent.

Wyatt Epp
profile image
Because Canonical is increasingly showing themselves to be a hostile agent within the community? Reinventing huge wheels tends to be a fairly good indicator of a distro that's not playing nice. And they seem to be taking the Fedora approach to unilaterally using their user base as guinea pigs, which I can't approve of in the slightest.

@Zachary:
If you look under the hood, it's not really an "Ubuntu" thing. Steam is more akin to a package manager in the user's $HOME. Like CPAN with enormous modules.

Duvelle Jones
profile image
Wyatt, I would not say that is a Fedora approach. I have worked with that distribution for the last 9 years as a user, and frankly... Fedora makes it's mandate quite clear.
When it's says "First", that does mean that you are likely to get something that is stable but not finalize, as it something around the beta stage of development (there are acceptations to this, I have seen it... but generally, this is what you are to expect). So yes, expect bugs... nothing that is critical most days, but bugs all the same.

IF Canonical seems to be emulating any company, there is this one in Cupertino, California that seems to have everyone attention as of late.


@ E Zachary Knight:
I know, the thing that has bugged me about that is what Wyatt has mentioned. As far back as 5 years ago, Ubuntu begun diverging from the community... braking things with major components, developing it's own isolated components to the usual Linux Stack, etc. Mostly at the direction of Canonical Ltd. desperately trying to make some manner of income from the distribution.

As someone that does watch the direction of the community, I think that the latest turn between Canonical's Mir and the Wayland project is going to have some major consequences. Since as far as I can tell, Mir presents no real technical reason for existing... and there is some rather big money behind Wayland at this point that help with the development.

Michael van Drempt
profile image
Ubuntu may be more top-down than other flavours, but a mainstream audience was never going to make the jarring transition from a closed system like Windows or Mac to a bleeding-edge community-driven Linux-hacker distro overnight.

Ubuntu may also bring with it a big audience of consumers that aren't at all interested in contributing to Linux from a programming perspective, and that may feel like they're taking your baby away from you, but that's what it feels like when a baby grows up. That's the price of success.

Duvelle Jones
profile image
@Michael Van Drempt
That is not the issue here, you see the price of "moving from the reservation" is upkeep... Ubuntu is now personally responsible for components that they have broken, components that they themselves have build for no-one else but themselves, etc.

Like in the case of Unity, that is more of less a GUI shell similar to that of Window "Explorer". The problem being that it is based on the GTK2 tool-set, so any changes made to GTK2 for Unity can't really be merged into the upstream GTK code because Unity exists only for Ubuntu. And the Gnome project simply moved ahead with GTK3/Gnome3, leaving any changes to GTK2 code as maintenance and bug-fixes... which will be abandoned at some point.
In fact that isn't the only problem... Unity has been using old components that the community has abandoned within that that 5 year time frame.... namely the compiz compositor, and has had to try to update that alone with varying results.

That is a rather large maintenance burden that could have been rectified by simply having a better dialog with the rest of the community. But Ubuntu... the users and the developers... have to live with the solo course that Canonical is setting for itself or abandon the distribution.
That is the price of what they have been doing, the lost of support. And the more that you see Ubuntu do it (and not have something that is compelling)... the more the greater community is simply going to ignore the distribution.

"Ubuntu may also bring with it a big audience of consumers that aren't at all interested in contributing to Linux from a programming perspective. "

You might not like the idea but Ubuntu is in the position where it needs to begin to convert some more users into someone that contributes to the distribution. and it's in desperate need for those with coding expertise. As it stands right now, Ubuntu has to hold up some pretty major parts of the linux stack completely on it's own, and Canonical can't do that alone. The company is simply too small to do so nor has the resources to try.
Hell, Red Hat... a company that has been in the business of Linux for years now... doesn't attempt to do that when they don't have to. They are a billion-dollar company.

If Valve want's to work with a distribution like that, they are welcome to... but it will leave them in a lurch when they work with the greater community, because they have to work with a base that is isolated for the rest of the world.
And when you have companies like NVIDIA, AMD, Intel, Samsung, Qualcomm, etc... all with some form of linux dev community support? I don't if Canonical can expect to be operational for much longer... by acting the way that the company has.

Wyatt Epp
profile image
@Duvelle: I respect their stated aim. It may simply be that I disagree (vehemently) with their idea of "stable but not finalized". cf. KDE4.0, Pulseaudio, systemd, policykit, hal, ext4, wayland... there's a clear long-term trend of shipping untested crap to end users. I firmly believe there's such a thing as "too bleeding edge" for certain things. Things like init systems and things declaring themselves the "will eat your children" release. (This isn't to say I like their packaging policy for RHEL/CantOS, either. I think even Debian stable tracks upstream better.)

Karsten Schlachter
profile image
I may be naive, but even though steamos may be sort of closed (it can imagine it having no ui besides steam itself at all) i don't think valves aim is to create their one closed plattform

I see it more as a catalyst to create a solid linux user base for game developers...i expect any performance improvement they make to make it in to general linux dists, too...and i don't see any special apis making an steamos game not running on ubuntu or any other major linux dist..
..and because of that i think IF valve is going to make some exclusive titles to push the steambox or -os the games would not be strict steamos, but rather linux only (at first)...'cause according to the numbers valve would have not much to lose but (espacially publicity whise) a lot to win by supporting the hole linux community


none
 
Comment: