Valves big misstep with the Steam Machines was that they expected developers to port their games over to Linux natively, on their own dime. Needless to say that didn't end up happening at any significant scale, so when they resurrected SteamOS they refocused on Windows binary compatibility through Proton instead.
Valve launched Steam Machines with their own OS and started shipping a version of Steam on Linux with predictable library versions. At the same time, they started working with the Wine project and shipping things which is now called Proton but is actually the cumulative results of their own patches.
This paved the way for the success of the Steam Deck when adequate material became available.
I don’t think it makes sense to call the Steam Machine a misstep because there was no Proton. There would be no Proton nor Steam Deck without the ground work started with the Steam Machines.
>I don’t think it makes sense to call the Steam Machine a misstep because there was no Proton. There would be no Proton nor Steam Deck without the ground work started with the Steam Machines.
I’ve written before about how I think the Steam Deck is one of the best v1 products in recent memory, in large part because Valve learned so much (and so well) from the failures of the Steam Machines.
I don’t know if I would call it a misstep, but it was absolutely a failure. And a brutal one. Valve should be lauded for taking the right lessons from that failure and investing in Proton and doing the compatibility work themselves rather than expecting devs to do it (Apple is the only company that consistently gets developers to rebuild for their platform, and even game developers won’t do that), but we shouldn’t let the fact that it wound up on the right path years later diminish the fact that the original strategy —- if not the devices or idea itself —- was hugely flawed.
I have a Steam Machine, one of the Alienware ones. You're completely correct. SteamOS as shipped on those machines sucked. The controllers sucked. The PC hardware wasn't bad and could play games alright but the experience sucked. I made it into a Linux desktop I used for years and it worked much better for that than as a dedicated games console. But the SteamDeck very plainly incorporated lessons learned from Steam Machines.
But all the update you like in your SteamDeck also came to the Steam Machines plus nice remote play on your local network. You couldn’t see because you stopped using it as an actual Steam Machine. I personally deeply disagree about the controller too.
Personally, for such an early and unlikely product, I don’t view it as a failure at all. They ironed everything they had to using the platform as a stepping stone.
First of all they can make it easier option for OEMs go for Windows/XBox handhelds and thus push SteamDeck out of market, by making it a niche device only a minority cares about, netbooks attack plan style, which they are starting now.
They can also introduce Windows/DirectX APIs that Proton will have a hard time copying, by requiring additional hardware or subsystem features not easily copyable.
They can let their legal team have some fun.
Finally, companies don't last forever, and I am betting as the 2nd most valuable company in the planet, Microsoft will outlive Steam's current management and eventually deal with the problem in another way.
> First of all they can make it easier option for OEMs go for Windows/XBox handhelds and thus push SteamDeck out of market, by making it a niche device only a minority cares about, netbooks attack plan style, which they are starting now.
Steamdeck is already a niche device, and they have direct competition since years now. Optimizing the software and interface won't probably make it much worse for Valve.
> They can also introduce Windows/DirectX APIs that Proton will have a hard time copying, by requiring additional hardware or subsystem features not easily copyable.
Seems unrealistic that games will adapt a new restricted API, which at the same will be a longterm hazzle for Wine/Proton, and will survive that court-battles. At best, it's just some annoying money-sink, with Microsoft playing on time to make it worse for everyone without gaining any real advantage.
> Finally, companies don't last forever, and I am betting as the 2nd most valuable company in the planet, Microsoft will outlive Steam's current management and eventually deal with the problem in another way.
That's actually realistic, but there is also the chance that valve will be outliving Windows and has to battle with whatever will follow.
I don't think reducing the price of Windows for OEMs will do much as the SteamDeck uses Arch Linux specifically for better control of the software/updates than relying on Microsoft.
They can try making Proton's job more difficult, but I'd expect that major changes to the APIs would prevent a lot of existing games from working on Windows.
Legally, I don't think they've got a leg to stand on.
Basically Valve are doing with Steam OS and Proton against Windows, the same thing that Google did with Chrome against Internet Explorer. Microsoft's usual tactic with Embrace Extend Extinguish isn't working with Edge, as Chrome just has great development velocity and market entrenchment and distribution. Crossing my fingers that Valve will pull off the same thing. In the gaming market Valve have great distribution and they are strong developers, so maybe.
