Steam Machines and Steam Controller shipping to beta participants December 13th | Hacker News Reader

Steam Machines and Steam Controller shipping to beta participants December 13th

steamcommunity.com

95 pointsCrazedGeek57 comments

>We’ve had to make the difficult decision to limit our beta to the U.S. only, because of regulatory hurdles. This was not our original plan, and it means we can’t collect beta feedback from Steam customers world-wide, which is pretty unfortunate. All things considered, we’re sure it was the right decision, because the alternative was to delay the whole beta beyond the point when we’d be able to incorporate any feedback into the 2014 products.

For those outside the US who were hoping to participate.

Even if you're in the US there's only a very very small chance you'll be able to participate in the beta.

300 units, they said. Or literally 1 in a million for the US population :) Which makes sense, considering how much one of these units probably costs.

Well the average is 800$-sh if we assume even distribution of the classes.

So it is 240000$ - I think this is less than Valve's "Pay people to find new reasons not to do HL3" daily budget.

According to the eligibility page, I'm seeing:

"You are now one out of 421,843 eligible beta candidates."

It's been at that number since the registration ended, and I'm almost certain it's worldwide, since they didn't say U.S.-only until today.

I like my odds.

That was the number it showed when it ended and that is the number of worldwide participants. Your odds will now be significantly higher at the expense of my odds plummeting to zero :(

If it makes you feel better, I didn't get one, either.

The random number generator must have been racist against me specifically. Yeah, that's the ticket.

This seems like something they should have been able to foresee. I mean granted it is literally being told you didn't win a lotto for something other than extremely small odds not going in your favour, but they had everyone fill out forms giving over data.

Sounds about right. A delay would have hurt everybody. Of course I would have loved to get my hands on this in Europe but I'd rather they can release on time. That just means the US folks have to do all the work testing this and giving frequent feedback.

It's not like there is functionality that would require worldwide test data is there? Steam is the same all over the world. It's not like the XBone where 90% of it's features only work in the US.

I'm anxious to see how much Valve has learned since the launch of Steam a decade ago. Steam is great now, but it sucked pretty hard when it first came out. I remember the genuine shock I felt at having the friends functionality actually work the way I expected it to after about 4 years of pure shit. I have no doubt this will eventually be awesome, only question is how long it will take.

Ahh yes I remember friends list being offline for a solid year if not more. It sure has come a long way.

I remember almost not installing HL2 when it first came out, because I didn't want to install Steam. Back then, the concept of being locked in to a certain launcher felt so wrong. I didn't do anything with Steam (besides having to use it to install the Orange Box) again until that first massive sale they had back in 2008 or 2009 (Can't remember what year that was).

You can use Steam today to see how much they have learned about how to run Steam. I don't see what this has to do with that.

Steam today shows their software nous, hardware is a very different thing.

steam for still sucks in the way requires full control of my box to games. i only run it in a vm or spare machine, limit the games i can use. if that box is any good, it will be very good for people with a minimum security concern for their main computers

> steam for still sucks in the way requires full control of my box to games.

What? I run Steam as a non-Administrator user on Windows, and a non-root user on Linux. Would you elaborate on your objections?

The DirectX installer (and other dependencies) that most games want to run the first time they load requires admin access. A few games want to run the installer every time the game runs if the user isn't running as an admin. You can skip dependency installation sometimes, but other times games will crash until you take care of it.

I was really hoping to hear from the OP, but I'll address this tangent.

These installers would be an issue whether or not one was running Steam. If you were installing from a disc (heh) rather than using Steam, all that installation stuff would have happened while you were running the installer on the disc (with Admin privs).

So yes, installers that potentially install stuff under %PROGRAMFILES% and %SYSTEMROOT% will require admin privs. Unless you change permissions on the associated directories and registry trees, there's no way around that.

AFAICT, Steam itself (and, indeed, many games sold through Steam) does not require admin privs. Maybe I'm wrong about this. That's why I asked the OP for clarification.

Well, they've overcome the number one challenge, that's getting something launched. Valve does one thing great, and that's rolling out consistent updates and features, so I have faith they'll improve the platform if it gains traction.

In 2014 though, I think the biggest impact on their success will come down to marketing. Will people know what this device is? Will hardware manufacturers be able to get these units selling? Will people be raving about playing Steam games from their couch and telling all their friends to pick one up?

They're going to be a small player in the console world. I can't picture them stealing Xbox and Playstation users. Those Battlefield, Call of Duty, NBA, Madden, etc., fans make up a huge part of the console market, and I think they're incredibly loyal to their console, while being very unfamiliar with Steam. The Steam Machines might have a market with PC users looking to casually game in the living room, and they might have a market if they can offer cheaper hardware and lower game prices than consoles. Offering better visuals would be another huge plus. If someone visits their friend's house, and says, 'wow, why does XYZ game look so much better on your television?', that's going to drive some sales, since you always have a group looking for the best of the best.

It's a long road, and I think everyone is watching to see how this plays out. Valve is probably the only company that can make this happen at the moment, so it's exciting to see them taking a chance, and I applaud them for that.

They are not inin the game to compete with consoles. Their strategy is for the long run, to be free of winfows influence should windows one day become a closed system. obviously they dont have any of the marketing muscle needed to go mainstream and thats not their market anyway.

They better be in the game to compete with consoles, because they're going to be, whether they want to or not. If their long term plan is to simply run Big Picture Mode, and that's it, they'll fail, and never be more than a niche device used by a few PC gamers. The average person would much rather buy an Xbox or Playstation since they can use them for watching movies, Netflix, etc. Why buy a Steam Machine, if I need to buy something else to do the rest of the job in the living room? And if I'm spending a few hundred on an Xbox or Playstation, why bother with a Steam Machine for a couple of extra game titles?

They have a chance to make Steam Machines a huge success. If they pigeon hole themselves as strictly being a gaming device, and not a media device that competes with consoles, they should change the name of their UI, because they're missing the big picture.

However, I think you're wrong. Yes, they want a backup plan if they get kicked to the curb by Microsoft. However, if that was the only case, they'd be focusing on making Steam OS an actual operating system. Why spend the resources on a gaming controller, and Steam Machines? Why not go hard on a Linux OS that's a viable alternative for gamers? Their core audience is PC gamers, so why not make an OS for PC gamers, and skip the living room hardware all together?

I don't think I'm wrong :) If they were really in the game of competing with consoles, they would have had a very different strategy and not adopted a "anyone can make our box" kind of mode of operation. They are following the PC model, which means that they are not targeting mainstream consumers. Consumers who want to buy Steam Machines will have to be knowledgeable on what to buy, they will have to look at specs and understand that that box is more powerful that the other one, and get what that means for their games. It's very different from providing a single box and telling consumers " as long as you buy a PS4, you can play all PS4 Sony games". So, please tell me how they will solve that problem for the mainstream, because their strategy does not address it at all.

To me, they are still targeting PC gamers (and Gabe clearly said so, too) and they want to expand that segment from the top, not from the bottom. Maybe they will attract some PC gamers who did the switch to consoles a couple of years back, by providing something better than consoles while still "living-room" centric.

I think their strategy is however the following: - a Steam gamer currently is usually a guy in front of the PC in his room. - By putting a Steam machine in the living room, it offers the opportunity to open up Steam to family members. - Oh, by the way the family games management feature was introduced about the same time as the SteamOS release this year... what a coincidence huh ? - So they plan to expand through immediate social environment first, not necessarily targeting hardcore console users. That way they will create awareness progressively. That's why I am saying they are in for the long run.

Just like Steam was not made to replace from Day1 the box retail business, SteamOS's strategy will not replace consoles from Day1 either. Maybe it will be a big player in 10 years from now, but I don't expect any significant market share any time soon.

I agree, it's a long term play, and their chance of winning over hardcore console users is zero at this time. However, they will be directly competing with consoles for space under the television.

As for solving your problem with hardware confusion, they could have manufacturers ship Steam Machines with some kind of rating system. Remember how Windows had those performance ratings for memory, processor, etc? They could do something very similar. You look at a Steam Machine, it says processor 5.7, memory 4.5, graphics, 5.1. Then you look in the Steam game library and see a certain game recommends graphics 4.5 or greater. Or this indie title plays great on anything 3.4 or higher. You use those numbers to guide your purchase.

Changing hardware will come to their advantage though. Xbox and Playstation will ramp up over the next few years, and we'll see some improvements and major titles heading their way. However, what's happening 5-6 years from now? Xbox and Playstation are dated, Steam OS has the bugs worked out, and Steam Machines are available with the latest advances in hardware. Where are you going to buy Call of Duty now, for your 6 year old console, or pickup a Steam Machine that can run it twice as fast, at Ultra HD resolution? That's going to be a big plus for Steam, and if people are buying in 5 or 6 years from now, they're less likely to jump on the next console launch, since they already have powerful hardware in their living room.

Another huge plus for Steam in the living room is Oculus Rift. If it's released below $300, and exclusive to the PC and Steam Machines, they have a unique experience the consoles can't offer.

Nothing is short term, I think Steam Machines will make a crawl the first few years, but I think they're going to be pushing to replace consoles, and I think they might succeed. Cheap games, big sales, Oculus Rift, low cost or high performance hardware to choose from, and eventually an open store available for anyone to publish towards.

Edit: Oh, one other huge advantage. They're backed by every hardware manufacturer they get involved. Dell is going to be dropping flyers in every mailbox around the world, with a couple of pages showing off their Steam Machine line-up. Dell wants money, and if Steam Machines succeed, they get rich as well. PC sales are down, they'd love another piece of hardware to sell. Get every other hardware company doing the same, and that's a lot of free marketing.

I also think, like you, that OcculusRift will be the big game changer that may spell the doom for the consoles (because are far from being powerful enough for these kind of things).

As for the ratings system, sure, it's possible but it does not mean it's not confusing for consumers, and it adds another barrier to adoption - that's basically NOT mainstream. And you mentioned Windows performance ratings as a reference, but that's a failed example since nobody ever used these numbers anywhere.

If they go for categories/benchmarking, then they will need to have a kind of authority in place to attribute such numbers, and Valve's message so far is very against centralization. So I don't expect them to lead such efforts. If they ever did, I believe they would make it simple: SteamBox Level A to play these games only and do streaming, SteamBox Level B to play all GPU intensive games... etc...

EDIT; I don't think Dell will get rich, and neither most of the hardware manufacturers involved for that matter. SteamOS is no more no less than an Android OS-like for the PC, and that will push prices down for manufacturers which will fight hard to make small margins on large volumes. It will only be profitable for very few players.

The current consoles are plenty powerful for this type of thing, don't forget they are basically just gaming PCs in compact cases this generation and share almost the exact same hardware (Xbox One and PS4).

Really ? Even the Occulus Rift co-founder mentioned they are far away from where they need to be.

WHen you want to do virtual reality, you need constant 60 fps (or more, ideally) in order to give a good impression of movement in all directions. XBone and PS4 are already struggling to display a SINGLE 1080p screen at 60 fps already, let's not dream too much. They are massively underpowered, and the AMD APU is notorious for being mid-range in terms of performance at best in 2013. Give it one more year and even the cheapest gaming PCs will be more powerful than them.

One of my coworkers has a Occulus dev kit.. I'm pretty sure it's just a single screen in there so 1080p Occulus would only require a single 1080p stream as far as I know, which current consoles are more then capable of.

It's not just about rendering a single screen, it's about rending two different scenes even if it's one screen.

Bring back WON!

The important bit is that we will have steamos to play with soon. They say it will be downloadable when the hardware ships.

Yeah, I think I am more excited about the new linux distro than the hardware. Only two more days to go.

This is what I'm excited about as well, but also can't wait until more games are announced for SteamOS early next year.

I'm very hopeful about the prospect for improvement of linux graphics drivers. It seems to have a been a chicken and egg sort of problem in past; I really hope contributions from valve will create some momentum, especially for the FOSS drivers. It seemed hopeless to me only a couple of years ago, but then it wasn't so long ago that you couldn't suspend your laptop without "blowing your balls off" as Con Kolivas remarked...

I'm really concerned, though, about commercial vendors aligning with the one distro/graphics stack (apparently Ubuntu in this case?), such that I have to choose between putting up with Unity's "Shopping lens" and a ghetto of bad or unimplemented 3d support for distros that don't adopt the same graphics stack as Ubuntu as the distros move away from X...

Drivers are usually kernel-tied more than anything else.

I wouldn't expect too much improvement from the FOSS drivers like Nouveau (I suspect the performance required of modern games is still out of their reach), but the binary X drivers from should see some big improvements - any distributions that currently can leverage those would benefit.

In particular, I'm hopeful that the radeon and intel FOSS drivers will improve (although I guess that nvidia said they will help with nouveau now: http://arstechnica.com/information-technology/2013/09/nvidia...). I gather Valve has given significant feedback to all three vendors (http://www.pcworld.com/article/2049369/amd-nvidia-ramp-up-li... and: http://games.slashdot.org/story/12/08/31/1551230/valve-finds...), and that steam games have, perhaps, stressed the sort of hitherto rarely used code paths that would only show themselves to be broken or unreasonably slow when some compositing DM feature wouldn't work, for example when KDE 4 was famously broken on fglrx for a couple of years, and no one cared enough to fix the driver bug because who plays games on linux, right? Of course that's the closed driver but the same kinds of problems have happened for all the drivers for want of heavy use, and of course they won't be heavily used if they are way broken.

Please correct me if I'm wrong, but I think Wayland/Mir support must be added to any driver that will support them, and I believe it's not yet clear that that will be done. Intel has already refused a pull request for XMir support (http://slashdot.org/story/13/09/08/0431242/intel-rejects-sup...), and I think the proprietary drivers have no support for either of the new WM's, nor will will they publicly admit to having any plans to implement it.

KMS was added to the kernel five years ago this month, and the proprietary drivers still don't support it, although the intel driver does. Imagine a similar outcome for the new compositors, where the vendors don't implement support for a long time, or, worse, where there are factions comprised of certain combinations of driver/distro/WM... that would be bad.

AFAIK Intel is pushing Wayland. As for Mir, I'm not aware of any group outside of Canonical with the slightest amount of interest.

Yes, so the question is: if Valve is creating an Ubuntu derivative will they stay with X11 when Ubuntu switches to Mir?

As it is now, Steam is easy to install on Ubuntu, and possible, but maybe a bit of a pain for other distros, depending on which one. It would be sad if Steam eventually only worked well on SteamOS, because of fragmentation in the graphics stack. If I wanted to choose my OS solely on the basis of what works well for gaming, I would be running Windows.

Then again maybe Intel and Valve will sort of bring some adult supervision to the distros by putting some pressure on everyone to pick one graphics stack and live with it. I'm not sure whether I think this is a good thing or not.

According to Engadget it isn't based on Ubuntu or anything?

http://www.engadget.com/2013/11/04/valve-steam-machine-hands...

But I assume it'll be Debian, but we'll see.

I'd love a Gaming OS which I can do a few things in a terminal, e.g. media server.

Yep, that's basically my experience with my home desktop: a media server, and I can play a few games on it with the Nvidia proprietary driver. I'm a "casual gamer" if at all, so booting windows is too much trouble. And when I want to do some real work at home, the OS doesn't get in my way. This is probably a hard sell for serious gamers though, what with the lack of AAA game titles.

I figure I should be working anyway...

Looks nice. Just curious: does anyone know where Valve is making the hardware?

I hope this is a good place to launch this idea:

Is it possible to make an android app which acts as a steam controller through usb/bluetooth? Here is an example (JSON-ish) config file: http://gearnuke.com/steam-controller-configuration-file-alre...

> Is it possible...

Probably? It'd be best to wait until we get the hardware and software in hand before specing out the task, though. ;)

I want to hear veteran RTS and FPS player weigh in on that controller. Bringing PC-style gaming into the living room would be a dream.

I'd really like to see someone fire up Quake 3 Arena and play a few rounds with a mouse, 360 controller, and Steam controller. Give them a few hours practice with each device, and let's watch some comparison videos, against the same skill bots, on the same map.

Not exactly an RTS veteran, however Tommy Refenes (programmer from Team Meat) weighed in on his thoughts about it.

http://tommyrefenes.tumblr.com/post/62476523677/my-time-with...

Yeah, I saw that one, and it's reasssuring that this unorthodox gamepad can still be used as a gamepad (and it really excites me about how powerful the haptics can be if they can use the "clickyness" of the buttonpad make it feel d-pad-ish) it's kind of the opposite of what I was hoping to read. I want to see how it performs in the novel world of handheld pointers, not the proven territory of handhold joysticks.

I gave one a go for about 20 minutes a couple of months ago.

Honestly, I think they have some way to go, but I'm optimistic they can make it work.

Using one of the sides to control a regular mouse in a game like Civ 5 I found that it was not even as good as a regular laptop touchpad. I suspect that a lot of the reason for that may be software though. There is a lot of subtle stuff going on with the motion of a laptop touchpad that has been built up over a decade of iteration and they just didn't have it tuned right.

Using it to control an FPS was interesting, but I was unable to tell if it would be better or worse than, say, a 360 controller in the long term. You certainly can't just pick it up and expect to do well with it quickly. I'm certain that more muscle memory would develop in that area over time though.

At the time I tried it, the haptics were only used to emit a clicking sound that was faster the faster you moved your finger along the surface. It vibrates the surface which makes it feel sort of like you are moving your finger across lots of small bumps rather than a flat surface.

That's disappointing. I don't expect it to be a smash hit in the pro-gamer circuit - mouse-equivalent performance is way out of reach, but it needs to beat the ever-loving crap out of a thumbstick.

I have a 100 game library of games on Steam that I would love to play on my TV but the SteamBox will only play a tiny percentage of them due to being linux based. Seems great for those new to steam though...

Apparently the SteamOS can stream games, OnLive style, from your Windows computer to your TV.

Then build a Windows box. What's the point of wanting SteamBox to be a Windows box?

The Steam Machine is additive, not a replacement for your current game machine, your existing windows machine will just stream games to your steam machine over the network.

The Nvidia Shield does this kind of game streaming at 1080p and it looks great.

Maybe by the time we get SteamOS they'll finally let customers change their account names. Without losing access to their games. Pretty please?

You can change the name that other users see on Steam. It's been that way for years. Unless you mean something else?

Is there a (legal/branding/whatever) reason why they didn't call it Steam Engine? Seems like such an obvious name.

Sounds a bit too much like "Source Engine".

Ew.. Friday 13th... wait til it passes.

I don't believe this crap, but coincidently my own PC burnt with fire (literally), right after saying that and turning it on (a few years ago). Yeah, bad coincidence. However, I got part of the money back, because I sued the A/C Adapter company, whose part caused the fire. But I tell you, they were such big assholes, who looked up for the cheapest comparable parts on ebay for my hardware as a "base" price. Interestingly I had a $1200 worth collectors edition graphics card and they paid just $100 bucks for it, because the next cheapest and older model of the card on ebay was that much.

Reply on news.ycombinator.com