Steam Machines and The Risk-Free Bet of Valve

If you have been following Boiling Steam for a while you know that I have had reservations about the Steam Machines. To sum it up, I was not sure that Valve had a proper value proposition in place when they first announced them. No exclusives for that platform, no real differences versus running a Windows box in Big Picture mode, and a number of issues relating to marketing itself (not clear who is being targeted, not clear how consumers will know what machine to buy to play which games and so on…). In the beginning I attributed that approach to a lack of preparation, and I was kind of relieved when they decided to postpone the release of the said machines. And I was expecting more than an half-baked plan for the GDC 2015.

Now, the GDC 2015 has come and gone, and the good news is that the Steam Machines are still part of the big Steam Universe plan. Valve even has a fixed release date, along with dedicated pages on Steam to announced the different options and vendors supporting the initiative.

But I am more concerned now than before – unless we are missing some big, upcoming announcements, it seems like they are not more ready than before to support this initiative properly. And this reminds me more and more of the 3DO debacle as the release date grows near. Note that I’m not one of those nay-sayers who want the Steam Machines to fail – on the contrary I think the concept has true value in itself, but unless something is done I fear it will fail miserably on the market and hurt both Valve and their partners and in the worst case cause a (temporary?) setback for Linux Gaming.

The primordial issue in this whole launch is that they have not figured who is going to buy this. And this is a debate that’s been going on on social media and articles for a long, long time following the initial announcement. Are they targeting existing PC gamers, but non Steam users ? Existing Steam users? Console users ? Casual gamers ? Non gamers ?


Oh, wait, I know who they target: EVERYONE.

Every Steam Machine runs SteamOS and ships with a Steam Controller, but that’s where the similarities end. Each manufacturer has chosen to optimize for your needs in their own way, so there’s something for everyone.

That’s even worse than the 3DO because the 3DO was an unified standard with clear minimum specs. That’s worse than the Shield which positions itself a set top box which can also do light PC gaming on the cheap, for people with a budget who still want to game. Trying to please everyone is probably the worst marketing strategy ever, because you don’t have unlimited resources to cater to all market needs. That’s not even a positioning and you can’t build a Brand on this.

Looking for a lightweight gaming machine to cuddle up next to your DVR? Maybe you’re interested in running the latest release at 4K? Are you looking for an all-in-one solution to power your VR headset?

And their advice is to look at the attached page with the list of upcoming hardware… let’s go and check then.


Oh, ok. So let’s pretend for a second I am completely clueless about PC hardware, and I have to choose the “right one” for me. Uh…

Alienware Steam Machine (which is just the rebranded Alpha, basically):

The Alienware Steam Machine combines the awesome experience of console gaming with a massive library of over 980 games, all in full 1080p HD on your TV.

Really ? So why do you have like 4 different models ? What are the differences in performance ? Which one is the right one for me ? Can they all play all games in Full HD like you say ? Because man, you are talking about “awesome experience of console gaming” where everything is MADE to run on a fixed hardware base. That sounds like you may need to pay more for some reason, but nothing is specified anywhere on this page. Well done Alienware!

How about the Gigabyte Brix ?

The BRIX Pro takes the signature compact design philosophy of the GIGABYTE BRIX series, adding a high-performance Intel® Core™ i7 processor featuring Intel® Iris™ Pro graphics 5200. As well as a featuring a true high-performance processor that is ideally suited to CPU-intensive applications like image design and video editing, the superior graphics performance of the BRIX Pro means it’s also great for 3D gaming.

Oh, so the Brix pro is also great for gaming then ? I guess so, since it has “true high-performance” stuff “great for 3D gaming”. I hope the guys who purchase this one won’t be too surprised to get a slideshow kind of framerate when they launch Metro on this machine. I would feel a lot more comfortable if they gave some examples of games that should run well on this and some that would not.


Let me add some topping on the cake, as the BRIX is announced with 2 slots for RAM and optional storage. You guys want to sell a console-like experience but you expect people to grab by themselves RAM sticks and an additional hard drive! Ha!


The product description for the SBX from Ibuypower is even more of a joke. Check it out for yourself:

SBX was bred to enhance your living room. This sleek, award-winning design was made to blend the love of games with the enjoyment of a media center. You can now play all your favorite games on your gorgeous big screen. It’s small, it’s sleek. It fits anywhere and matches everything. Whether you prefer an upfront, in-your-face vertical orientation, or a more relaxed and easy going horizontal approach, your SBX looks great with unlimited color and animation combinations. Prefer something more subtle? The lights can be dimmed or turned off completely.

Are you guys aware that you are selling gaming hardware and not neon lights for interior decoration ? There is no mention ANYWHERE what this product is for in terms of gaming. I wonder why… Oh wait:

GPU: ATI Radeon™ R7 250X 1GB GDDR5

Oh. Right. You better not talk about gaming anything on SteamOS with a AMD GPU onboard – that can only end with sore disappointment.

The Maingear Drift seems completely oblivious to the issue of AMD on Linux/SteamOS :

The MAINGEAR DRIFT is the ultimate customizable Gaming PC that is designed from the ground up for living room gaming. Experience immersion that no console can provide with Ultra HD beyond 1080p gaming, powered by the latest cutting edge hardware from Intel, Nvidia and AMD.

Guys, it’s always better to try your own products out before selling them. Just a tip, but do what you want.


The Origin Omega Steam machine is not even trying to play the game of SteamOS at all:

Dual Boot: Still have PC Games only that are only playable on Windows? With an ORIGIN PC Steam Machine, you can customize it to have it dual-boot into Steam OS and your choice of Windows inside one system!

Oh, so you mean once your users boot on Windows they will realize that they have no reason to come back to SteamOS ? Great idea guys to promote the new OS ! I can’t even understand how Valve can allow such positioning on pages which are supposedly dedicated to machines running SteamOS.

I actually like Syber’s idea of having an upgradeable console… but…

Hands down, PC games are higher quality and higher performance than traditional consoles because they are designed to maximize the constantly evolving gaming hardware options. The Syber Steam Machine is the only fully upgradable Steam Machine at its price range. Utilizing standard PC components, you can upgrade with off the shelf hardware and components quickly and hassle-free.

yet why even bother selling their first machine at 499 dollars with…

GPU: AMD Radeon R9 270 2GB

Again, this is just a rehash of what they had prepared back in 2014, with no assessment of what works well on SteamOS so far.

I’m not just picking on a few, go and check for yourselves in this jungle of a page and you’ll see that there is no way for a random user to actually grasp what machine is needed for what they want to play. So, Valve, seriously, how do I pick the “right one” ? And on top of that, I would have expected AT LEAST a price/specs comparator when landing on such a page instead of a simple never-ending list of all options out there. Have they actually tested that page with real life users to get some feedback before publishing this ?

And that’s the next big issue. The price point. I’m not among the ones to think that we should limit the price to match the current consoles’ or something. It’s fine to go for higher specs hardware and ask for more cash, to be a premium option in the field of cheap consoles (kind of like the Neo Geo of the 16 bits era – it was a niche market but it had a market on its own!). But there’s some serious delirium at work from the different vendors. First, if they are trying to target casual folks who have no idea what hardware to buy, then they are doing an extremely poor work at educating them. So let me assume that they target people who apparently know PC hardware, and if that’s the case they should clearly realize how they are being ripped off. Apart from a few exceptions, almost everything machine proposed in that list is overpriced. Overpriced, as in you can make something of the same specs by yourself for a cheaper (or even MUCH cheaper for certain cases) price. That’s just plain wrong. It is clear that vendors need to make money out of the hardware since they are not making money on the platform (i.e. Steam, unlike Microsoft and Sony which are manufacturers and platform owners). Yet, as vendors they should have access the cheaper parts than the general public and scale should bring more efficiency as well on how to bring everything together for a better deal. At most, I would expect that they could sell their hardware at ABOUT the same price as what it could cost if you were to pick up the parts and do it yourself. No, they are clearly overcharging you – so buying the hardware from them is just a matter of convenience (and form factor), but there’s nothing economical in doing that. Some of these platforms are not future proof as well (i.e. non upgradeable like the Alienware one), so when the specs are not good enough anymore you need to get rid of it and buy a new one. Welcome to the future of gaming, where PC gamers will not be interested to purchase something more expensive than what they can get elsewhere, and console gamers will not find much appeal in replacing their machine every 2 years to play the latest Steam games – since they are used to a console being able to play games well for 5-6 years or so.

What’s missing in this whole page is the one that started it all, and hence my reference to “the risk free” aspect of this initiative. Valve is not there with a flagship machine, while they are clearly going to be there with the Steam link for streaming and the Steam Controller. While they seem to be committed to the project, they don’t seem to be committed enough to go out there and show “this is what a Steam machine should look like” both in specs and design.


You cannot bring PC gaming forward unless you take an active part of it, instead of relying on third parties who have no vested interest in growing your own platform. Valve is not showing courage here, it’s showing hesitation, as if they already know it won’t grow big anyway and they should not invest any efforts into making one by themselves. And let’s not ask them to take the front seat and set up factories and all from scratch, but they could at the very least take the Google approach, and work with a specific third party to create and design a “Nexus-like” Steam Machine stamped by Valve which should represent the state of the art of the expected platform. That’s a proven model after all in the Android marketplace, which is conceptually quite similar to what Valve is trying to create with SteamOS.

So it is now clear in my eyes that Valve is taking a back seat with the whole Steam Machines thing. They did not try to establish minimum specs to ensure most games would run well with any hardware out there sold on their own pages. We still have not heard of any way to know whether games will run well on X or Y machine. We still have to hear how SteamOS will let users know that their machine is not powerful enough to play certain games, and how users are expected to react in such cases. They have not put pressure on their partners to have reasonable price points vs specs (which should have been the greatest learning from the 3DO failure). They have chosen to let manufacturers publish whatever they wish on their Steam Machines pages without oversight or lead as to how they should present their hardware, and let users find “the right one” by themselves.

OK. There’s still time until November.

In theory.

I certainly hope I’m plain wrong about everything and the above Steam pages are just the result of a lack of preparation (again) and that most aspects will be fixed by November.

At this stage, no matter how much I’d like the Steam Machines to succeed, I’m cautiously pessimistic on their prospect. At least at launch. It seems as half-baked as last year.

But hey, it’s Valve. Back when they launched Steam everyone was laughing and saying it was a shitty service. Yet it ended up working and growing to lead the world of online game retail. Maybe something similar will happen in this case: a crappy beginning, then better iterations and more focus in the coming couple of years, until it finally takes off ?

BoilingSteam lets you access our content for free, but writing articles is a constant investment. We don't use ads or sponsporship, help us make our activities sustainable by donating via Patreon or LiberaPay if you prefer it anonymous. You can follow what we do via our newsletter, our RSS feed, our Mastodon profile or our Twitter feed. We also have Peertube, Youtube and LBRY channels. If you'd like to chat, you can also find us on (what is Matrix?)

0 0 votes
Article Rating
Notify of

Oldest Most Voted
Inline Feedbacks
View all comments

Futuremark began quasi-benchmarking Steam Machines days after the lost generation (i.e. the 2014 models affected by Steam Controller revisions – were announced ( I first learned about it while visiting Futuremark’s website ( It was brought to my attention that Futuremark had updated the benchmarks by the article comment by James Gallagher of Futuremark at (the article itself, “Valve’s VR thrills, but Steam Machines are looking like a flop”, is worth a read too). The “Valve is Working on ‘a Couple of Versions of Tools’ to Simplify Steam Machine Choice for Consumers” article was also mentioned in the… Read more »


Futuremark’s figures are based on the specifications provided at, but I noticed, for example, that’s current Steam Machine web page ( lists: CPU: Intel Core i5-4460 (listed as i5-4440 on Steam) GPU: NVIDIA GeForce GTX 760 OC (listed as GTX 960 OC on Steam) Memory: 8GB DDR3 1600MHz Storage: SSHD – 1TB To (8GB Nand) The following article probably gives a fairly good idea of what to expect from the Alienware Steam Machine: I’m not enamored with the GTX 750 Ti (and the Alienware Alpha actually has a GTX 860M) when compared to the performance the PS4,… Read more »


Futuremark has updated based on the latest information. It’s obviously not precise because they don’t have branded Steam Machines to test, and the 3DMark scores are based on 3DMark Fire Strike scores from Windows systems with similar hardware.

Perhaps Valve will deliver on the ideas quoted in

The Steam Machine initiative will fail or it will fly, but I won’t be holding my breath in anticipation.


I think they should have two levels, Steam Gaming Machine, say minimum core i5, 8GB RAM, Nvidia GTX750+, anything lower than this should be labelled Steam Streaming Machine. The manufacturers will sort themselves out with their pricing. Of course you would need to re-evaluate the minimum specs each year. Say a minimum of 30FPS on the most demanding game without having to run it below full HD?


I am also deeply concerned by the advertising. Listing just the names and pictures and prices leaves even experienced users going elsewhere to look up info. A rather big fail, which I hope won’t compound itself moving forward.

Peteris Krisjanis

I think we should throw this “win or bust” attitude away. It’s Valve. They have never followed traditional ‘winner takes it all’ pacing which everyone seems to be trying to dictate (a bit desperately I might add, burning much more money than required in the end). So yes, I think they know about issues, they know about price range, they know they have to figure out classification, etc. Some of this stuff will clearly be done till November (recommendation for example – Valve would not want to screw that up). However rest of it…I wouldn’t really bother that much. Also… Read more »


I agree with those who think Steam needs a clear performance comparison system. You release in 2015, create a benchmark called Steam2015, and force every Steam machine to get ranked. Games can simply say “Minimum: Steam2015 70 / Best: Steam2015 180.” Or something to that effect. Every year, two, or three, rev the benchmark, and allow users to run it on their own systems. It would probably be best to give an integer number to the most important subsystems: GPU, CPU. Possibly disk and memory. IMO, an “overall” score should be avoided because mediocre machines could hide in plain sight… Read more »


Full, automatic backwards compatibility has been ended though. Sure, some games have been wrapped in an emulator or ported because there was enough demand. You can’t put your PS2/PS1 game disk into the PS4. On the PC, games from even 30 years ago are still hardware compatible – it’s just some software layers that need to be in place (like Dosbox, Wine, etc.). When you re-buy games on the PC with GOG or Steam, you are paying for convenience. When you re-buy games on consoles, it’s because you have no other choice.


“Overpriced, as in you can make something of the same specs by yourself for a cheaper (or even MUCH cheaper for certain cases) price. ”

This is true of ANY name branded computer system. Dell, Asus, whatever. It’s not an insight, whatsoever. Write an article on how Steam will succeed.