Steam Machines Delayed Until 2015

HardOCP News

[H] News
Joined
Dec 31, 1969
Messages
0
Those of you anxiously awaiting Valve's Steam Machine will have to wait until sometime in 2015 to get your hands on one. No big deal really, most of us will be building our own anyway. ;)

We’re now using wireless prototype controllers to conduct live playtests, with everyone from industry professionals to die-hard gamers to casual gamers. It's generating a ton of useful feedback, and it means we'll be able to make the controller a lot better. Of course, it's also keeping us pretty busy making all those improvements. Realistically, we're now looking at a release window of 2015, not 2014. Obviously we're just as eager as you are to get a Steam Machine in your hands. But our number one priority is making sure that when you do, you'll be getting the best gaming experience possible.
 
The controller is the lynchpin on this obviously. Valve has no experience with hardware on this scale and quite frankly seems to be in a "learning as you go" pattern. So it is good to see it pushed back, because with the execution, or lack there of, it was almost assuredly going to fail.

I think Valve is screwing up not adopting tech that is already fully supported in the controller market instead of trying to reinvent the wheel. I am not saying moving tech forward in this category is a bad thing, but it seems to me this is new tech just for the sake of new tech and more of a lateral move, and it certainly gives the user of it a barrier to entry as gamers are hit with a new learning curve.
 
I completely agree. The thing is, it's not hard to build an HTPC and load it with Steam/Big picture mode and grab an Xbox 360 controller and get the same experience, without the unknowns and hassle of Linux and an unproved controller from a software company.

I mean, I understand that PC controllers are generally mediocre (even the 360 controller is thoroughly meh), but that doesn't require a whole new OS, dedicated hardware models etc. And considering they've already nerfed the vision of the controller from earlier iterations ( no more touchscreen, lower quality triggers and housing, no more modularity etc), it really doesn't bode well. At this point I'm expecting:
- A Linux OS with a selection of games more limited than the Mac version of Steam
- A controller that screams "cheap MadCatz crap" more than it does "premium gaming experience."
- A set of OEMs that will release a single generation of Steam Machines and then jump ship as the things gather dust in warehouses before being re-purposed as low end Windows Machines.

Meanwhile, Microsoft is hinting that GFWL, while on its way out as a brand and platform, will be replaced with something with tighter integration into XBL. Considering MS's momentum with recent products, it might not suck. And that's not to mention the impending release of the XBO controller for PC.
Additionally, with the consoles now on x86, there are fewer barriers to porting, and greater performance parity, which means that not only does Steam OS and it's shitty controller have to compete with Windows, but Mac, and the consoles. And none of those are on Linux, with its headaches and limitations.

I see why they started this whole thing (fear of MS making a walled garden app system for games like iOS/Mac Store), but I'm not sure where they thing its going.
 
What really blew my mind is a popup I got on steam the other day about a steam stream feature or something like that. It allows you to play games from your main pc on low power systems similar to what the nvidia shield does.

I haven't used this yet but this could allow them to offer a very low priced steam machine for people who already have a nice gaming rig. (Probably most of the people interested in Steam box in the first place.)

Pushing the the launch back will also allow them to use more powerful, and probably cheaper hardware. It seems to me one of the biggest complaints about steam boxs was the price. I think by the end of 2015 we will probably see next gen console performance in a cheap apu setup. (it's not like it doesn't already exist.)
 
What really blew my mind is a popup I got on steam the other day about a steam stream feature or something like that. It allows you to play games from your main pc on low power systems similar to what the nvidia shield does.

I haven't used this yet but this could allow them to offer a very low priced steam machine for people who already have a nice gaming rig. (Probably most of the people interested in Steam box in the first place.)

Pushing the the launch back will also allow them to use more powerful, and probably cheaper hardware. It seems to me one of the biggest complaints about steam boxs was the price. I think by the end of 2015 we will probably see next gen console performance in a cheap apu setup. (it's not like it doesn't already exist.)

I would REALLY like to try this, but can't come up with a reason to drop the money on a lower end machine, in case it doesn't work. I'm still not 100% clear on what the system requirements are for the lower end machine. I have the AC network to do it, there are just too many unknowns for me.
 
Pushing the the launch back will also allow them to use more powerful, and probably cheaper hardware. It seems to me one of the biggest complaints about steam boxs was the price. I think by the end of 2015 we will probably see next gen console performance in a cheap apu setup. (it's not like it doesn't already exist.)

Yeah, I'm currently in the process of replacing the 1037u board in my HTPC with a j1900 board. The 1037u will then become a 'Steam In Home Streaming receiver-end' machine. To see similar 'receiver-end' machines on the market utilizing say ARM and Steam-OS at a cheap price would be cool.
 
It was always my assumption the purpose of them developing the controller were because of the vast amount of games that don't have native controller support. Their controller would work around that; making all those non controller supported games playable with their controller.

If they didn't have that, then another limitation of the Big Picture/Steam Machine experience would be limiting you to games that only have native controller support (obviously you could always use m&kb, but Valve is going for a console type experience in all this - making m&kb support always viable, but not the intended experience).

Like I said, that way always my understanding of why they were developing the controller/why their controller is the way it is.

What you are talking about there is a software layer. What I am talking about is executing a totally new hardware.
 
Im not surprised that they are pushing back the release on this. It's probably for the better plus there is a lot of optimizing that needs to be done along with testing. If steam is really serious about pumping out a quality product to consumers to drive the market (like they want) they will need this to work like a dream, maybe even better.
 
Right, but for the games with no controller support - or even a dev company around anymore to retroactively add controller support - you would have steam just create their own dx-input type driver to add controller support to non controller games? Then you run into issues like the XBOX 360 controller not having enough buttons to accommodate the game. It's much more feasible to develop your own controller and driver/software than just a driver/software for an existing controller in that situation.

Just to add, I really don't expect the Steam Machine initiative to take off. I expect it to flop. But for what they're trying to do - it makes more sense to me for them to try and make a controller that works with all games, then to try and make all games work with the 360 controller.
 
AKA noobtards that don't know how to setup a HTPC and stick steam and a xbox controller on it.

That's the kind of community comment that has kept me away from Linux all these years. Don't bring it over here! :eek:
 
Right, but for the games with no controller support - or even a dev company around anymore to retroactively add controller support - you would have steam just create their own dx-input type driver to add controller support to non controller games? Then you run into issues like the XBOX 360 controller not having enough buttons to accommodate the game. It's much more feasible to develop your own controller and driver/software than just a driver/software for an existing controller in that situation.

Yes. But what you are talking about is NOT the issue for the push back. It is the controller hardware. The departure from traditional controller hardware is the issue.
 
Steam should have left this to the pros. They are practically printing money and have a great business model already. They make shitloads of money on their marketplace alone where there is a huge global trade market on in-game virtual items they sell lol (make a percentage off every sale especially in TF2, DOTA2, and CS:GO).

IMO they are overextending themselves and its too big of a risk for them to take.
 
Steam should have left this to the pros. They are practically printing money and have a great business model already.

I would agree on the whole. I do not think the company is being overextended though, at least not monetarily.
 
I might pick up a low end, power efficient steam machine down the road, but the real piece of hardware I'm interested in is the Steam controller. I had a chance to try it out at PAX East this year and it was GREAT!
 
^ I mean if that's where we disagree, then that's where we disagree :)
 
Lol I hate this no edit in the news section.

Or Kyle, is you're objection the form of controller they are developing, and would much rather have them create a new controller but more based on/resembling the 360 and existing type controllers?

Or is it like I'm assuming you just think they shouldn't be making a controller, period?
 
There are two reasons for the delay. Firstly, the OS is a mess. Specifically the drivers. Even Valve had to donate code to help speed things up. The binary drivers for AMD are just not ready at all. Intel's open source doesn't even support OpenGL 4.0 yet, and is considerably slower then Windows.

The other problem is the unwillingness of developers to make a proper port to Linux. The Witcher game for example uses eON to bring their games over. Which acts very similar to Wine. So performance is going to be very slow. I think Valve is giving everyone more time before going ahead with SteamBox.
 
I still see no purpose to these other than for extending the Steam name/brand.
 
Maybe we are misunderstanding each other. Yes I am aware of that. But that comes with the territory of trying to create a brand new controller that works with all games.

My first reply to you was because you were saying that they shouldn't be making a HW controller and only developing software to make it work with existing controllers.

I explained I just find it much more feasible to create a controller to work with all games, rather than trying to make all games work with a pre-existing controller.

So in what they are trying to do - an all new controller that works with all games, I'm not all that surprised there are setbacks - but it is still the right course of action for what thay are trying to do IMO.

Building a custom controller with already available technologies and building a new controller with new controller technologies are two different things. I did not say a "pre-existing controller," you did.
 
I figured the Steam machines were bound to fail the moment the previews came out and there were systems ranging from a few hundred, to several thousand, dollars.

Steam should have specific hardware requirements to be labeled a Steam Box, at least initially. On of the big advantages of consoles is each PS4 will perform the same as all the PS4s out there, but with the wide range of specs this will not be the case Steam machine, and to the average person buying a Steam machine will be no easier then buying a computer, which many have no clue what they are doing when computer shopping.

Consider this, someone sees one friend playing a game on a PS4 and likes what they see, they can go buy a PS4 and assuming their tv is of similar quality as their friends they will have basically the same experience. Now imagine someone sees their friend playing a Steam game on a top end Steam machine and then goes and buys a $400 steam box, they will likely have a much poorer then expected experience.

And of course this doesn't even address the lack of games available for Linux, I mean when Mac beats you in game titles you should know you have issues.
 
I figured the Steam machines were bound to fail the moment the previews came out and there were systems ranging from a few hundred, to several thousand, dollars.

Steam should have specific hardware requirements to be labeled a Steam Box, at least initially. On of the big advantages of consoles is each PS4 will perform the same as all the PS4s out there, but with the wide range of specs this will not be the case Steam machine, and to the average person buying a Steam machine will be no easier then buying a computer, which many have no clue what they are doing when computer shopping.

Consider this, someone sees one friend playing a game on a PS4 and likes what they see, they can go buy a PS4 and assuming their tv is of similar quality as their friends they will have basically the same experience. Now imagine someone sees their friend playing a Steam game on a top end Steam machine and then goes and buys a $400 steam box, they will likely have a much poorer then expected experience.

And of course this doesn't even address the lack of games available for Linux, I mean when Mac beats you in game titles you should know you have issues.
Some of the Steambox's are suppose to be glorified Roku-esque streamers, or Like Nvidia's Project Stealth or whatever that streaming controller is.
 
Right, but for the games with no controller support - or even a dev company around anymore to retroactively add controller support - you would have steam just create their own dx-input type driver to add controller support to non controller games? Then you run into issues like the XBOX 360 controller not having enough buttons to accommodate the game. It's much more feasible to develop your own controller and driver/software than just a driver/software for an existing controller in that situation.

Pffft use this controller. PLENTY of buttons.

http://community.us.playstation.com...FEF4E5BEE1/image-size/original?v=mpbl-1&px=-1
 
Just want the controller.

Me too, it's Windows compatible if I'm not mistaken.

Well they missed a window of opportunity. By 2015 Windows 9 and DX 12 will be shipping and the argument for this platform is going to get weaker, assuming that Windows 9 is something of a hit which given the cyclic up and down popularity of Windows should be the case. And that's also more time for the PS4 and Xbox One to mature.
 
Me too, it's Windows compatible if I'm not mistaken.

Well they missed a window of opportunity. By 2015 Windows 9 and DX 12 will be shipping and the argument for this platform is going to get weaker, assuming that Windows 9 is something of a hit which given the cyclic up and down popularity of Windows should be the case. And that's also more time for the PS4 and Xbox One to mature.

I would like to see Microsoft create an actual Steam competitor with the XBOX brand/XBOX store/app (in Windows 8 and +). If only they could get their crap together and realize they have all the pieces there already for the taking.
 
Some of the Steambox's are suppose to be glorified Roku-esque streamers, or Like Nvidia's Project Stealth or whatever that streaming controller is.

I understand that, although I'm not sure why I should stream from my main gaming rig to a Steambox to play a game on my TV, as opposed to say, just connecting my computer directly to the TV?
 
Steam should have specific hardware requirements to be labeled a Steam Box, at least initially. On of the big advantages of consoles is each PS4 will perform the same as all the PS4s out there, but with the wide range of specs this will not be the case Steam machine, and to the average person buying a Steam machine will be no easier then buying a computer, which many have no clue what they are doing when computer shopping.
Fixed hardware is not a big advantage. If anything it's a catch 22. Fixed hardware means easier development, but you also limit developers on what they can do. Xbox 360's 512MB of ram seemed pretty big in 2006, but today most graphic cards have more memory then that. Which is why modern games still have washed textures and frequent loading times. It may even explain why games don't have big open environments with lots of detail. The CPU of the PS3 and 360 were so limited that no modern game explores physics, which would require a heavy duty processor. Which may also explain why games don't have destructible environments still.
Consider this, someone sees one friend playing a game on a PS4 and likes what they see, they can go buy a PS4 and assuming their tv is of similar quality as their friends they will have basically the same experience. Now imagine someone sees their friend playing a Steam game on a top end Steam machine and then goes and buys a $400 steam box, they will likely have a much poorer then expected experience.
A $500 SteamBox would out class the PS4. iBuyPower gave details of a SteamBox they were making and it's going to have some impressive specs. A HD 7790 or R7 260X has already been proven to be a more capable setup then the PS4.

But the SteamBox has been delayed until 2015, which means those specs are likely to go up. At that point I imagine even Intel's Iris Pro graphics will roflstomp the PS4.
And of course this doesn't even address the lack of games available for Linux, I mean when Mac beats you in game titles you should know you have issues.
That's probably the reason why the SteamBox was delayed. But weren't we comparing the PS4 to the SteamBox? Wouldn't even the PS4 lack titles compared to Mac? What does that say about the PS4?
 
I understand that, although I'm not sure why I should stream from my main gaming rig to a Steambox to play a game on my TV, as opposed to say, just connecting my computer directly to the TV?

Because your gaming computer might not be in the same room as the TV you want to play one. Unhooking your computer and moving it from room to room whenever you wanna play in 'x' room would be a pain in the ass. The Steam 'In Home Streaming way' is better in this regard.
 
Because your gaming computer might not be in the same room as the TV you want to play one. Unhooking your computer and moving it from room to room whenever you wanna play in 'x' room would be a pain in the ass. The Steam 'In Home Streaming way' is better in this regard.
The problem is going to be lag though. Even very small input delays can make you completely uncoordinated. Otherwise that would be the ideal setup for just about everybody. Buy one powerful desktop, and it can run everything to little micro-PCs hooked up to your TVs and what not. For streaming video, absolutely... games... no.
 
...but the real piece of hardware I'm interested in is the Steam controller. I had a chance to try it out at PAX East this year and it was GREAT!
Strange, as every single publication that received a hands-on with it has panned the Steam controller as a hunk of junk. Ars, PC Gamer, Rock Paper Shotgun, etc.

I can't help but feel that brand loyalty is clouding your judgement.

It's kind of odd that PC gamers swear by their mouse and keyboard, and are almost militant in their defense of it as a superior input method - yet heap praise upon a touchpad simply because it's made by Valve.
 
The problem is going to be lag though. Even very small input delays can make you completely uncoordinated. Otherwise that would be the ideal setup for just about everybody. Buy one powerful desktop, and it can run everything to little micro-PCs hooked up to your TVs and what not. For streaming video, absolutely... games... no.

And for games where the lag effects your game play, like Counter-Strike (or any other online MP or FPS game if you want) - you play them at your gaming rig. For games where that won't effect gameplay - Devil May Cry 4 or Limbo or any local SP non-FPS or whatever have you - you now have the option of playing in a different room easily.
 
And for games where the lag effects your game play, like Counter-Strike (or any other online MP or FPS game if you want) - you play them at your gaming rig. For games where that won't effect gameplay - Devil May Cry 4 or Limbo or any local SP non-FPS or whatever have you - you now have the option of playing in a different room easily.
I suppose, but honestly if you're hell bent on streaming to a TV, you could probably go HDMI/USB over ethernet relatively inexpensively, and then not worry about lag. Especially if the Steambox was wireless, I can't imagine how bad the delay would otherwise be.
 
I suppose, but honestly if you're hell bent on streaming to a TV, you could probably go HDMI/USB over ethernet relatively inexpensively, and then not worry about lag. Especially if the Steambox was wireless, I can't imagine how bad the delay would otherwise be.

And have a wire running along the floor? Upstairs to downstairs in my case no less? No thanks.

Besides, I have MoCa networking throughout my house (which is the best option aside from having ethernet drops installed in every room). Powerline would be the next best option. Then wireless is a last resort. Valve even explains this to users.
 
And have a wire running along the floor? Upstairs to downstairs in my case no less? No thanks.

Besides, I have MoCa networking throughout my house (which is the best option aside from having ethernet drops installed in every room). Powerline would be the next best option. Then wireless is a last resort. Valve even explains this to users.

Ethernet wiring is pretty standard for newly built homes. Go to your local hardware store and you can buy all the equipment needed to setup a home network. Usually for dirt cheap. Don't need expensive wiring. Cat 5e will do fine for gigabit.

I'm sure wireless is an alternative, but wireless is always flinky for everyone. You could run wire all over the walls of your home, but it becomes unsightly. It's pretty safe to assume that Ethernet wiring is here to stay.

Not that I stream gaming to my TV. If you can run wires, then you can put a computer to your TV. There's really no reason to put a HTPC on your TV. BTW, best remote for a HTPC setup.

dsc01422.jpg
 
As an Amazon Associate, HardForum may earn from qualifying purchases.
So far the steam streaming is completely unplayable for me on my Asus RT-N66U Black Knight from my Windows Box to my Nix laptop. Doesn't matter if I'm on the 5ghz on both, 2.4ghz on both, or one on each, all with no other network traffic.
The router never worked right with my Ouya and Kainey streaming games or video either, don't know if its just me or the software just isn't ready for prime time yet.
I'd like to hear some setups that managed to get the streaming working if anyone has managed yet.

BTW, I can honestly say AMD Linux drivers are still terrible, being that most low end Steam boxes are AMD based pushing the box back a year giving AMD and Nvidia some time to work out their drivers is probably a good thing.
 
what this really means is that 'Half Life 3 Delayed Until 2015'...

:rolleyes:sigh That is the first thing I thought of when I read this news, unless they decide to release it before Steam Box...which I very much doubt.

I agree with DarkStar_WNY they should have come out with specific hardware and then grew off of that.
 
One thing I believe Valve STRONGLY need's to focus on here is performance. If they are looking at a 2015 release I would venture that Steam Machines at this point will be able to play games at above PS4 performance.

They need to push this aspect. PS4 running a game at 1080p and 30FPS where the Steam Machine is playing the same game at 1080p and 60FPS for roughly the same price.

That and get more games on Linux.

Who knows though...they have a way to go and it will be a rocky road.
 
I get this vibe that the controller is moving away from a novel concept to basically a near carbon copy of the xbox controller. And I don't think that's a good thing because the xbox controller is awful for PC games and awful for FPSs. They have to do something different here if they want PC gamers to play in the living room. And, from what I understand, that's what they want. They're not competing for console players.
 
Back
Top