Everything valve doing for linux is making such a huge impact.
The HL3 memes don't even seem fair to use anymore. I don't even want to un-seriously make joke fun of them at this point. They are just genuinely doing so much for the community.
Valve is one of the few companies regularly seen on HN where the headline is something like "[company] is secretly doing something really great" as opposed to "[company] is secretly doing something evil"
It's an interesting case study. They're essentially another 'App Store middleman' raking in a huge 30% cut for selling games digitally. But they do enough really good stuff to keep both gamers and developers generally very happy.
I have plenty of complaints about them.
The highly addictive gambling mechanics in their games,
the extortionate cut afforded them by their dominant market position
or the very rough UX in many parts of the Steam client (takes forever to startup, shows pop up ads on startup, is quite the resource hog, the store that is a pretty poorly optimized website and a lot of cruft in the less well trodden areas).
But they do make some very nice open source contributions.
"We will make linux a viable gaming before we increment that number to 3!"
But I totally agree, I still install windows for gaming on my machine, but it looks like that for my purpose of gaming I can stay with Linux (I play mainly older games or indie games).
I strongly feel it’s because Valve is not a publicly traded company where they’ll eventually give up their values to meet Wall Street analyst quarterly targets.
It genuinely makes me see the value in private companies. Public companies must grow. They're accountable to so many different interests. Private companies can be happy sitting at whatever profit level they want. They can take time to tinker on something that they care about. If it doesn't pay off, that's fine.
I think I would say it this way: private companies can be good or bad, but public companies must ultimately become bad.
That's more a property of the community than of the company. If the community were differently inclined then the comments would be about how Valve is making money by addicting children to gambling and so on and so forth.
That is why I bought a steam deck: to financially support Valve's Linux efforts. I barely play games anymore but thanks to the Wine devs, CodeWeavers, and Valve, I no longer have to listen to the knuckle-draggers claiming that "Linux sucks because it can't play games". In fact, now it is the opposite: Linux is outperforming Windows[0].
I have a near infinite amount of respect for Wine. It seems like for at least the last twenty years, Wine just keeps getting better and better with every release.
I don’t know for sure, but I suspect a lot of the work is spent sussing out weird edge cases with different binaries. This is tedious, thankless work, but it is necessary to have true Windows compatibility.
Wine and Proton have gotten so good that I don’t bother even checking compatibility before I buy games. The game will likely run just as well or better than on Windows and it is so consistently good that it’s not worth the small effort to check ProtonDB.
I do wish that they would get Office 2024 working on Wine. This isn’t a dig at the Wine devs at all, I am sure that it’s a very hard problem, but if I can get that then I will have even more ammunition to get my parents to drop windows entirely.
I love Proton and I like Steam and Valve definitely has done a lot of good for the FOSS world, but let’s not make the same mistakes we made with Google by worshipping a company.
All it takes is new management to change the policies to make the company horrible and evil, and in the case of Google people made the realization far too late, and now Google owns too much of the internet to avoid.
Valve seems more like Apple than Google: a well-liked company that has an obvious and not inherently exploitive business model. Google as an ad company was always destined to go bad in a way that most non-ad companies are not.
No company is your friend, and they are all fundamentally structures around making a profit. But providing goods and services in exchange for money is not inherently exploitive or evil.
Valve is not building all this Linux Compatibility out of the goodness of their hearts. They are doing it to avoid being shutdown by Microsoft, who effectively had a monopoly on the OS people used to play games.
It's a bit of miracle that Valve beat MS to the punch and built momentum behind Steam as the marketplace for games. They know this.
If gamers move to Linux and all the compatibility issues are solved, Valve is not going to pick a different passion project. Conversely, as long as Microsoft has a monopoly on OSes for gaming, Valve will support linux gaming.
A "company" is just an organizational model employed by people to pursue the intentions that those people have. It goes without saying that large endeavors involving many people will have a mixture of good and bad intentions.
Opposing all organized endeavors simply because they have the potential to pursue bad intentions essentially resolves to being against anything anyone is ever doing, which is more than a little bit pointless.
These rumors come from Tyler McVicker who regularly gets things wrong and makes stuff up in order to get clicks on youtube. I have no idea why anyone still takes him seriously in 2025.
Are the rumors still hinting at a VR-only experience as they did a couple of years ago when Half-Life: Alyx released, or is that no longer the speculation? Because that would be unfortunate for me, I'd have to play with a bucket in hand.
Your games are still not owned by you, they are locked inside your Steam account (liable to be suspended at any time) and app (as I've learned when I couldn't play when their pretend-but-not-really-offline mode broke; I now block it at firewall level most of the time). That part will never become "community" oriented.
>The HL3 memes don't even seem fair to use anymore.
It's absolutely fair to mock them for not releasing these games and keeping radio silence all these years. They managed to dethrone Duke Nukem Forever.
There were multiple times in which the internet was hyped for Episode 3 and where it would make sense to release even a basic game like they've did with Episodes 1&2 just to wrap things up. I'm sure plenty of people that make up various explanations to why that happened but the end result is that Valve has chosen to disappoint the fans who have been waiting for the conclusion to the story. It's not like doing that would prevent them from releasing an another new entry in the series that uses revolutionary new technology or whatever.
Valve is sort of like modern Bell Labs for software. It has almost-monopoly on PC game sales, which results in massive profits. Then it uses part of these profits for public good on projects that are at best tangentially related to their actual business.
When I read what you wrote, I immediately asked myself "Doesn't Gabe have children who could have been raised with the same values? Maybe that..." and then I caught myself thinking exactly the same way as many others before me, and the reason why we have so many shitty politicians in positions of power today.
I hope Gabe has setup Valve in such a way that they can pass on his mentality as a whole inside the business practices themselves. I think, after all these years, he must have surely thought about what leaving would look like for Valve. Considering this is a guy who seemingly thinks in decades, I feel maybe even optimistically calm about it.
And Valve has been deeply rewarded as a result. The stance that you must abuse customers to maximize economic success will be looked back upon as the stupidity it is.
> Everything valve doing for linux is making such a huge impact.
Some of it is counter-productive though. Proton made WINE commercially viable, and in doing so, disincentivized native Linux builds of games to the point that some studios that had been releasing games natively for Linux have stopped doing so, since the Windows version now plays well enough under Linux.
So it became more straightforward to release games on Linux? Sounds like a positive. Or, is the gripe about distinction of released for vs playable on?
half life releases were tied to new platforms, such as HL2 and its physics engine, or HL Alex and VR kits.
it's like Nintendo having a Mario game for their new hardware, e.g. Mario 64, etc.
there weren't that many teases, nor is it great marketing; CS:GO competitive e-sports is better marketed and probably made Valve more money than any HL wink-wink-nudge-nudge ever would.
> and modern multiplayer games with anti-cheat simply do not work through a translation layer, something Valve hopes will change in the future.
Although this is true for most games it is worth noting that it isn't universally true. Usermode anti-cheat does sometimes work verbatim in Wine, and some anti-cheat software has Proton support, though not all developers elect to enable it.
It works in the sense it allows you to run the game; but it does not prevent cheating. Obviously, Window's kernel anti-cheet is also only partially effective anyway, but the point of open-source is to give you control which includes cheating if you want to.
Linux's profiling is just too good; full well documented sources for all libraries and kernel, even the graphics are running through easier to understand translation layers rather than signed blobs.
These things do not prevent cheating at all. They are merely a remote control system that they can send instructions to look for known cheats. Cheating still exists and will always exist in online games.
You can be clever and build a random memory allocator. You can get clever and watch for frozen struct members after a known set operation, what you can’t do is prevent all cheating. There’s device layer, driver layer, MITM, emulation, and even now AI mouse control.
The only thing you can do is watch for it and send the ban hammer. Valve has a wonderful write up about client-side prediction recording so as to verify killcam shots were indeed, kill shots, and not aim bots (but this method is great for seeing those in action as well!)
Anti-cheat is a misnomer; it's much more about detecting cheats more than it is preventing them. For people who are familiar with how modern anti-cheat systems work, actually cheating is really the easy part; trying to remain undetected is the challenge.
Because of that, usermode anti-cheat is definitely far from useless in Wine; it can still function insofar as it tries to monitor the process space of the game itself. It can't really do a ton to ensure the integrity of Wine directly, but usermode anti-cheat running on Windows can't do much to ensure the integrity of Windows directly either, without going the route of requiring attestation. In fact, for the latest anti-cheat software I've ever attempted to mess with, which to be fair was circa 2016, it is still possible to work around anti-cheat mechanisms by detouring the Windows API calls themselves, to the extent that you can. (If you be somewhat clever it can be pretty useful, and has the bonus of being much harder to detect obviously.)
The limitation is obviously that inside Wine you can't see most Linux resources directly using the same APIs, so you can't go and try to find cheat software directly. But let's be honest, that approach isn't really terribly relevant anymore since it is a horribly fragile and limited way to detect cheats.
For more invasive anti-cheat software, well. We'll see. But just because Windows is closed source hasn't stopped people from patching Windows itself or writing their own kernel drivers. If that really was a significant barrier, Secure Boot and TPM-based attestation wouldn't be on the radar for anti-cheat vendors. Valve however doesn't seem keen to support this approach at all on its hardware, and if that forces anti-cheat vendors to go another way it is probably all the better. I think the secure boot approach has a limited shelf life anyways.
Arc Raiders is a great example of a modern and popular multiplayer game that works with proton. I haven't heard about it having a problem with cheating.
Marvel Rivals, Age of Empires 2 DE, Path of Exile 1/2, Last Epoch, Fall Guys are other such examples. In fact, Marvel Rivals even explicitly mentioned Bazzite in one of their changelogs! I can't recall an instance when a major game name-dropped a (relatively) minor Linux distro like that.
I think a big portion of that is the rather poorly made anti-tamper solution they are using called 'Theia' most cheat developers are too unintelligent to correctly reverse engineer this kind of binary obfuscation
Valve is the only company I'd let inject anti-cheat software directly into my veins if it meant I could play CS and be sure others were not cheating haha.
As a former cheat developer, I think it is impossible since it is digging into some specific stuff of Windows. For example, some anti-cheat uses PsSetCreateThreadNotifyRoutine and PsSetCreateThreadNotifyRoutine to strip process handle permission, and those thing can't be well emulated, there is simply nothing in the Linux kernel nor in the Wine server to facilitate those yet. What about having a database of games and anticheat that does that, and what if the anticheat also have a whitelist for some apps to "inject" itself into the game process? Those are also needed to be handled and dealt with.
Plus, there are some really simple side channel exploits that your whitelisted app have vulns that you can grab a full-access handle to your anticheat protected game, rendering those kernel level protection useless, despite it also means external cheat and not full blown internal cheat, since interal cheat carrys way more risk, but also way more rewardings, such as fine-level game modification, or even that some 0days are found on the game network stack so maybe there is a buffer overflow or double-free, making sending malicious payload to other players and doing RCEs possible. (It is still possible to do internal cheat injection from external cheat, using techniques such as manual mapping/reflective DLL injecction, that effectively replicates PE loading mechanism, and then you hijack some execution routine at some point to call your injected-allocated code, either through creating a new thread, hijacking existing thread context, APC callback hijack or even exception vector register hijacking, and in general, hijack any kinds of control flow, but anticheat software actively look for those "illegal" stuff in memory and triggers red flag and bans you immediately)
From what I've seen over the years, the biggest problem for anticheat in Linux is that there is too much liberty and freedom, but the anticheat/antivirus is an antithesis to liberty and freedom. This is because anticheat wants to use strong protection mechanism borrowed from antivirus technique to provide a fair gaming experience, at the cost of lowering framerates and increasing processing power, and sometimes BSOD.
And I know it is very cliche at this point, but I always love to quote Benjamin Franklin: "Those who would give up essential liberty to purchase a little temporary safety, deserve neither liberty nor safety". I therefore only keep Windows to play games lately, and switched to a new laptop, installed CachyOS on it, and transfered all my development stuff over to the laptop. You can basically say I have my main PC at home as a more "free" xbox.
Speaking of xbox, they have even more strict control over the games, that one of the anticheat technique, HVCI (hypervisor-protected code integrity) or VBS, is straight out of the tech from xbox, that it uses Hyper-V to isolate game process and main OS, making xbox impossible to jailbreak. In Windows it prevents some degree of DMA attack by leveragng IOMMU and encrypting the memory content beforehand to makd sure it is not visible to external devices over the PCIe bus.
That said, in other words, it is ultimately all about the tradeoff between freedom and control.
companies will go where the money is. If Valve enables, say EA, to have their yearly franchise and in-game-stores on mobile devices, they will find a way.
I honestly don't know why so many people say that anti-cheat with Proton or SteamMachines won't work. SteamOS is an immutable Linux - especially with their own SteamMachine they can enable SecureBoot and attestation that you are using the SteamOS verbatim efi boot file, kernel, and corret system fs image - all signed by Valve. Just as Battlefield 6 does on windows (relying on SecureBoot). That would still allow you to install other OSes on your SteamDeck/SteamMachine, but it would fail the anticheat attestation. I personally see the push in hardware from Valve particular so that they can support anti-cheat on linux.
I think if Linux gaming becomes popular someone may come up with a solution where you run a native linux kernel-mode anticheat. That somehow connects to the wine-hosted game.
I'm not sure how I feel about that, but it's what I think will happen.
I think Valve is not some kind of God who free the human from the hand of Microsoft, they are a private company, they are just protecting their business and protecting their business "accidentally" also protecting the customer's benefits. The movement towards Linux benefits Valve the most since they have invested on Linux Gaming for 10 years now and that movement "conveniently" benefits the gamers too. That's a win-win situation, users can escape themself from a bloated Windows and Valve has the pioneering advantage.
> they are just protecting their business and protecting their business "accidentally" also protecting the customer's benefits.
part is wrong. From my observation, they are protecting their business through protecting their customers' benefits.
Plus, they're building a moat collectively and from an open source stack. So, given the stack gets enough momentum, having Valve or not as a company won't matter anymore.
It's trying to get the elephant out of the bag, and once it's out, then there's really no way to put it back, because it's being out is better for everybody. Game companies and gamers alike.
I think this calls out a subtle, but significant difference between private and public companies.
Public companies as an asset class have to compete with an open market of other investments, so the incentives drive a min-maxing approach to revenue and value. The shareholder mandate dictates the company pursue maximal return in order to stay competitive amongst a sea of other potential investments.
A private company doesn't have this same concern. They still need to pursue profit, but not necessarily MAXIMUM profit. This means that in a sea of hypothetical directions, they are free to choose one that is slightly less profitable but has an abundance of positive externalities, vs. one that is maximally profitable but carries many negative externalities.
> From my observation, they are protecting their business through protecting their customers' benefits.
Yeah that's what I mean too, that's why I put the "accidentally" in a double-quote.
This sounds like what Red Hat is doing, they created an open-source software, prove the importance of it in the community then sells the support package to enterprise who interested in using it.
Hope that they will not close the door when Microsoft, AWS or Oracle making their own GabeCube and call it SatyaCube, BozosCube or LarryCube
You are absolutely correct. Valve's linux push was driven by developments in the windows platform, specifically around the release of windows 8. Microsoft was pushing a windows store similar to Apple's app store, and Valve was unequivocally stating that they were worried Microsoft would basically lock down the platform and only allow software sales through their own store, destroying their steam business. Gabe said it plainly himself (https://www.bbc.com/news/technology-18996377):
> Mr Newell, who worked for Microsoft for 13 years on Windows, said his company had embraced the open-source software Linux as a "hedging strategy" designed to offset some of the damage Windows 8 was likely to do.
> "There's a strong temptation to close the platform," he said, "because they look at what they can accomplish when they limit the competitors' access to the platform, and they say, 'That's really exciting.'"
> This is seen by commentators, external to be a reference to the inclusion of a Windows Store in the Microsoft operating system.
Having an open platform is good for consumers, but Valve is primarily looking out for themselves here. Gabe realized that windows could take Apple's IOS route (i.e. https://blog.codinghorror.com/serving-at-the-pleasure-of-the...) and lock down their OS, and everything he's done since has been an effort to protect his company against that existential threat.
GabeN was the lead developer on Windows 1, Windows 2, and Windows 3. When Windows 95 launched, he was a bit upset that no one was making games for Windows. He did a rough port of Doom to prove the viability. Around the same time Alex St. John, Craig Eisler, and Eric Engstrom were building DirectX, GabeN saw the potential, left to create Valve, and proceeded to try and making Windows gaming a great thing.
I can only imagine that he was heartbroken to see Windows go the way it did with Windows 8, 8.1, 10, and now 11.
Well actually they tried with Windows Phone, Windows RT and Windows 10 S but failed miserably. Even Apple didn't even try to lock their macOS from installing 3rd party app.
Valve's Steam platform is well liked and actually wanted by gamers. Gamers move their games to the steam launcher and often wait for games to come to steam (e.g. with Anno games from Ubisoft).
This is in contrast with EA's Origin, Microsoft's Xbox PC and Ubisoft's Connect, which everyone hates.
I always think Valve as the "ideal" capitalist company, because what they do fits the idea of "invisible hand" perfectly, that each individual acting in their own self-interest end up benefiting everyone.
And you'd be right, that Valve is nothing special, if that idea is correct, because in that case most companies will be like Valve. But just look around, do you see many companies like Valve? No, that's because capitalism is bullshit and that makes Valve stand out.
I have been a big fan of Valve since the Orange Box days and I always dreamed of working there, but let’s not kid ourselves. This is all enabled by the massive monopolistic cash-cow that is Steam that requires a tiny team to maintain. Similarly their top games have minuscule teams and still rake-in millions in microtransactions, fueled by a shadow economy of gambling and speculative trading aimed at kids.
Yes to a large extent they got those monopolies by building truly outstanding products in good faith and by being pioneers in quite a few areas. And certainly they are an exemplary case of investing that wealth into legitimately innovative and widely appreciated long-term endeavors.
My point is that Valve is not all that special for being nice, many organizations do crave to be like that but they don’t have the luxury to have hit that jackpot. For people with mountains of money, they are among the best, but it’s not exactly a high standard, and they are remarkably inefficient in leveraging that advantage.
They’ve long lost the organizational know-how to make good games, and they have delivered remarkably few public facing successes in the last decade: mainly Valve Index and Steam Deck, both still relatively niche and wide apart, both primarily attempts at expanding Steam’s dominance to fairly uncharted markets, with mixed success. The first iteration of Steam Machines was dead on arrival, as was their long-anticipated game Artifact. CS 2 was not a significant enough upgrade to Go to really count. Half-Life Alyx was popularish I suppose. Anything else of note?
Valve cuts 30% of your revenue no matter how much you earn. They also cut 15% of the transaction by being the middleman on the market.
They also ignored the gambling/trading plague for too long, until a lot of countries threatened them to stop indirectly promoting gambling (which definitely hit them financially).
They are sitting on a money printing machine and their job is making it print no less to buy GabeN another yatch. They are like the cigarette company who donates shit load of money to the charity and cancer prevention lab while making more cigarrate then ever because people love smoking it.
I don't think they wanted or planned to be monopolized, but they are definitely taking the advantage of being it.
> acting in their own self-interest end up benefiting everyone
I'm so sick of people acting like Valve is some saint that does no wrong. Their market dominance means game developers wanting to reach the PC gamer market must comply with Valve's terms. Why do you think every Japanese visual novel released on the platform is a cut down, all-ages version that requires an off-site patch to restore the full game (and often even then it's censored in weird ways)? They got sick of being delisted while Valve turns a blind eye to all the trash porn games.
Ask yourself, does a marketplace that exerts creative control over specific studios' works while threatening financial repercussions if they don't comply benefit everyone? That sounds more like the mob to me.
It’s funny, because Zach Barth (of SpaceChem and many other wonderful games fame) worked at Valve and then described them as the ideal anarcho-communist type of organization.
They are a significant actor in the market, and as a non-pc-gamer I am glad that their business goals align with Linux users. I don't believe that they do it out of kindness, but that's actually a good thing for a long term investment.
The trick is to get your egoistic interests to align with what is good for the world.
Environmentalists also only do what they perceive as doing good (and may as well be objectively good) because not doing so won't jive with their self-image. But that in itself doesn't devalue the thing they did.
If Valve shows anything it is that not being a pushover and trying to align your business interest with your values (and not vice-versa!) can also pay of economically.
Yep. I know Apple has little motivation to support such a project but it would be great to see them work with Valve on this. Having the majority of Steam games "just work" on modern Macs, like they do on the Steam Deck, would be fantastic.
Apple leadership cares more about "games on the Mac App Store built for Metal on a Mac" than it cares about "games on the Mac". This won't change until leadership changes.
I think it's more than "little motivation" if we're being honest. Right now Valve is quietly targeting MS' attempt to create a walled garden for gaming on Windows and (probably) cut them out. Their very clever approach has been a full end-run around the OS by using Proton, which I'm sure genuinely thrilled Apple... as long as Valve is only doing that to MS.
Why would Apple ever invite Valve to potentially do the same to them?
Main issue is the lack of Vulkan support on macOS. Currently, solutions like MoltenVK have to be developed to add Vulkan support, which isn't as clean as just supporting it.
For some reason the prospect using Wine, Rosetta 2, and DXVK with MoltenVK on top just to run some games doesn't inspire a lot of confidence that this whole thing will be performant and/or stable.
This isn't an "issue" so much as a feature. Apple had some vulkan support until move to the full A1 architecture had them only make Metal a first class citizen to the GPU. Concurrently happening was a pretty nasty breakup of Apple with the Khronos group.
This wasn't an inconvenience, it was a deliberated decision.
No, the main issue is a fundamentally different rendering pipeline (tile based deferred rendering) that makes "Vulkan support" a conceptually difficult square peg in a round hole problem, since everything is made for immediate rendering, like all the other mainstream GPUs use.
Are you looking for Crossover? It's a bit annoying to not run Steam natively (no cmd+H to hide, etc) but it's got a lot of support. Performance is decent on my M2 mini, and even cross-platform stuff like Baldurs Gate 3 is comparable performance to native.
Especially anything that Mac Steam natively calls out lack of 32bit support has good support.
Sadly, that's not true—for instance, I was trying to run the Shadowrun Returns series the other day, and while it launches, it will hang indefinitely when you try to actually start a game. (M4 Max)
I previously played through Returns, Dragonfall, and part of Hong Kong on Mac before the 32bit-apocalypse.
Some crossover games perform better than the native ports. I play Path of Exile on Mac using the Windows client with a translation layer, and it plays better than the native release.
Valve employed Alyssa Rosenzweig while she developed the graphics stack for asahi linux. That's a very simple statement that masks the size of the achievement and its impact on the world. No, we haven't entered a golden era of gaming on macs, but the world has been shown the way. And no, the software challenges are not insurmountable.
bummed she left for intel she was next level. asahi lina commented after she left that there's currently no one doing graphics work anymore on asahi. projects kinda got some uncertainty to it now
The last time I can remember a collaboration between Valve and Apple was for the SteamVR support on macOS back in 2016. Sadly it fell apart a year(-ish) after that.
But… one can dream!
Unfortunately, this will not happen. Even if they have it all working:
Above all, Apple wants to show that their hardware is awesome, especially because it really is. Running x86 games or compatibility layers even with great emulation will make that $3000 Mac look half decent at best, against a $1500 gaming laptop. Simply not the story Apple want to tell.
It would be neat if Valve would fund having Steam Client run on Apple Silicon without Rosetta 2 so arm games like Baldur's Gate 3 can be fully supported.
GPTK is mostly a bunch of developer tools for converting to Metal, and the closest it gets to anything like Proton is an "evaluation environment" that is nothing close to Proton's performance. Proton is mostly Wine, and Wine on macOS uses MoltenVK, so it's probably easier to just port Proton.
I think it's a pretty reasonable wish for more macOS + Apple Silicon support of games, including more native FEX & Proton ARM support within the steam client. (We're lucky Steam works, it's a better games client than the Mac App Store dreams to be, but that's also not saying much either.)
Apple Silicon has no UEFI support except as provided by Asahi, so that would be needed at a minimum to boot Windows 11 natively. Then there's the whole issue of having native Windows drivers for the Apple Silicon-specific hardware.
You’d run FEX with WINE/Proton, no windows needed. If you did use a VM, I’d think it would be a Linux VM. But, Linux VM on macOS could already use Apple’s Rosetta2 for x86_64-to-arm64 translation.
Speaking of which, maybe you could just run the games with Apple’s WINE “game porting toolkit” direct with Rosetta2. Worth a Google.
A funny detail is that Microsoft's mostly fruitless ARM efforts unintentionally ended up being a boon for Valves ARM effort. From MSVC 2019 they started augmenting x86 binaries with undocumented metadata specifically to assist the Windows x86-on-ARM emulator, but then the FEX team figured out how that works and implemented the same optimizations in their emulator, greatly increasing the performance of most recent Windows games on ARM Linux.
I don't understand what you mean by this. My wife's HP laptop has an ARM processor with Windows and it.. just works? Like everything works. The computer is super fast, quiet, great battery usage, ultrathin and even scary affordable for what you get. All software works. I've not found x64 programs to run noticeably worse, at least not the ones we use, and plenty programs have an ARM build.
As a Windows user you don't even need to know that it's an ARM computer. Just use it like you'd use any other Windows computer.
Don't just say stuff, man. This is not Twitter. Try to at least figure out whether you're vaguely directionally correct before writing snarky comments.
I think they meant that by most metrics, Windows on ARM has not yet had a market impact. Not that the product doesn't work.
That being said, the mindshare well of "Windows on ARM" was poisoned by Windows RT, then later the objectively terrible performance of Windows 10 on ARM at launch.
I think the problem is that, until recently, there was little impetus to actually run Windows on devices where ARM actually has a meaningful advantage over x86. The Windows ARM laptops out there today don't impress, not just because of the software, but because the hardware itself isn't "better enough" than Intel or AMD to justify the transition for most people the way Apple Silicon was, especially for games. That is to say nothing of desktops, where battery life isn't even a concern.
Valve is using ARM to run Windows games on "ultra portable" devices, starting with the Steam Frame. At least right now, there isn't a competitive x86 chip that fits this use case. It also feels like more of an experiment, as Valve themselves are setting the expectation that this is a "streaming first" headset for running games on your desktop, and they've even said not to expect a great experience playing Half-Life: Alyx locally (a nearly 7 year old title).
It will be interesting to see if Intel/AMD catch up to ARM on efficiency in time to keep handhelds like the Steam Deck and ROG Ally from jumping ship. Right now it seems Valve is hedging their bets.
> At least right now, there isn't a competitive x86 chip
I don't think there will ever be a competitive x86 chip. ARM is eating the world piece by piece. The only reason the Steam Deck is running x86 is because it's not performant enough with two translations (Windows to Linux, x86 to ARM). Valve is very wisely starting the switch with a VR headset, a far less popular device than its already niche Steam Deck. The next Steam Deck might already switch to ARM looking at what they announced last week.
x86 is on the way out. Not in two years, perhaps not in ten years. But there will come a time where the economics no longer make sense and no one can afford to develop competitive chips for the server+gamers market alone. Then x86 is truly dead.
The key difference is that Valve isn't trying to make Windows work, just desktop gaming, which I'd imagine is a large part of why Microsoft's efforts failed. As much as Linux desktops haven't particularly had much polish over the years, there's still an advantage to taking something bare bones and trying to flesh it out in a way that's works well compared to taking something that's already pretty bloated and then trying to retrofit it into something new.
It turns out the best API for gaming on Linux and gaming on ARM was Win32 and x86_64. Who knew?
Well, compiling ARM game binaries is actually super duper easy and just totally fine. The issue Windows actually has with ARM is GPU drivers for the ARM SoCs. Qualcomm graphics drivers are just super slow and unreliable and bad. ARM CPU w AMD GPU is easy mode.
It's mostly because Microsoft have lost focus & interest in the desktop OS market and have shifted priorities to cloud service (Azure). Right now Microsoft is a sleeping giant that doesn't see the writing on the wall regarding Valve's efforts.
Oh I think they see it, the problem is that they can’t execute anymore. They are pivoting in the same direction but nobody wants to use the XBox Store on Windows and nobody trusts Game Pass anymore.
Any leads on when the next generation of Steam Deck will be released? Hoping it could be sometime in 2025, but suspect it will be more like 2026.
Over the holidays I was playing GTA: San Andreas on a Nintendo Switch. It's fun but so underpowered for a game released in 2004 (Yes, 21 years ago! Damn..). I'm really craving something more.
As a sidenote, it's really cool Valve allows installing SteamOS on any hardware. There are some alternative comparable form-factor devices:
* Lenovo Legion Go S
* Asus ROG Ally
But I have yet to see any of these in real life, so not sure how good or bad they really are.
It won't be for a while, since Valve is releasing the Steam Machine next year and has commented that they are waiting until they can build a Steam Deck successor that is significantly better than the original.[1] My guess is 2027.
Just get a Steam Deck. it's an incredible value for what you get and for what it can do. I'm no expert, but I do pay as close attention as I can to what's going on with gaming hardware because of my limited budget, and I'm guessing Steam Deck 2 is more like Q1 2028, not any time sooner. I'm ok with that. I play all the games I want on my Steam Deck OLED, and I see plenty of life left in it, even this "late" in the game.
> Just get a Steam Deck. it's an incredible value for what you get and for what it can do
I'm still kind of flabbergasted that we're in a world where the cheapest Steam Deck model literally costs less than the Switch 2. Sure, neither of them are exactly powerhouses as far as console hardware goes, but at least on one of them you literally can just use the system however you want as a desktop OS as a bonus...
I would assume Steam Deck 2 isn't dropping before at least H2 2026, if not later, if they didn't bring it out with the announcement of the other devices.
Valve's only official statement as far as I know is that it will come when they see a significant enough hardware upgrade to warrant a new system. If they don't move to ARM, AMD's Medusa APUs are their next architecture with major upgrades, so I would guess that Valve would order another custom AMD chip but based on Medusa, which won't release until at least 2027. I would guess at least H2 2027 but probably early 2028 for an AMD-based Steam Deck 2.
Valve has said they want at least double the capabilities, while still fitting in a similar power envelope. Unsaid is that it also needs to fit in the same price budget, but I tend to believe that's their intent. It's gonna be a while. Valve got a stellar deal on some somewhat unusual Zen2 APUs, orginally built for Magic Leap; finding a similar good deal is going to take time. I sort of hope Valve isnt going to put out a $1600 Halo system (but probably would buy such a next-gen Gorgon Halo system). Maybe Gorgon Point is good enough for them, in which case yeah 2026H2 is reasonable.
Are you aware that the year is 2025, and that it is 92.2% over? There is next to no chance of a Deck2 this year. I would really really not hold my breath for 2026 either.
I recently bought a Legion Go S because the primary way I game nowadays is streaming from my desktop to a handheld, and the higher quality display (1900x1200 resolution with 120 Hz over the 1280x800 and 90 Hz on the Steam Deck OLED) seemed worthwhile given that my desktop can easily provide enough throughput to play with relatively high graphics settings. It came with SteamOS preinstalled, and from a software perspective, everything does seem pretty close to identical. The only things I've slightly missed from the Steam Deck are slight hardware nits with the Legion Go S; the placement of the equivalent of the two SteamOS-specific buttons (not sure exactly how to refer to them, but they're labeled "Steam" and "..." on the Deck) just above the Start and Select buttons while looking and feeling the same mixes me up sometimes in a way that never happened on the Deck, and I miss having four unmapped buttons on the back of the device that I can set up however I like in games rather than only two. I also tend to prefer having symmetrical thumbsticks higher up on the device rather than having one high and one low; I've noticed that my hands aren't quite as comfortable when using the D-pad for extended amounts of time, which is unfortunate given my preference for it when playing stuff like emulated GBA games (incidentally one of the few things I tend to do locally instead of streaming; the low power profile setting already is an easy battery life win when streaming, and in practice it make the battery life when GBA emulation also much more tolerable, along with keeping the fans much quieter without seeming to impact performance of the emulation, given that even with this setting the fast-forward function can go far faster than I'd ever need it to).
If you're considering getting an alternative handheld, a better OS would be either Bazzite or CachyOS Handheld edition. SteamOS is not bad, but it uses an older kernel+graphics stack which doesn't make it very ideal for running on recent hardware. Plus, dedicated gaming distros like Bazzite have additional hardware support (like thirdparty game controllers) which may not be supported in SteamOS.
Currently, AMD Strix Halo based handhelds are the most powerful portable gaming devices out there, with the top three being the GPD Win 5, the OneXPlayer OneXfly Apex, and the AYANEO Next 2. Of these three, the GPD Win 5 has already started shipping. Problem is they're stupid expensive.
Personally, I will wait until I can run FSR4 natively on these portables, because FSR makes a pretty significant QoL improvement on these handhelds.
FWIW it's fairly straightforward to set up FSR4 with Proton-GE nowadays, assuming you're comfortable with editing one config file or manually specifying an env var for the game[1]. I'm not sure if using an alternate version of Proton would be considered "native" though, or if you mean for the default version of proton (or for Linux builds of games specifically), but setting it up is a fairly straightforward process even for people who might prefer not to use the terminal if you use something like ProtonUp to manage the installation for you. I imagine that the process for using a custom Proton isn't much different on Bazzite and CachyOS, although I'm not sure whether it would be something commonly done on CachyOS given that they have their own Proton distribution.
I wouldn't be surprised if they don't. Valve don't want to sell hardware, they want to sell games. They only make hardware as flagships for new markets, then they want other hardware manufacturers to take over.
the legion go is more powerful and a has a nice screen, but is heavier, boxier, and has a worse batteyr life than the steam deck
Valves moving into hardware more than ever right now, not moving away from it. They've already sand multiple times a deck 2 is on the cards, but only when theres enough of a hardware bump to make it make sense as a product. Slapping a tiny bit newer cpu in there and calling it a Steam Deck 2 isn't what Valve are about.
They definitely are working on it. They announced the steam machine, steam controller, and the valve frame (standalone vr headset with seamless screen sharing from a PC), and in their reveal video the first thing they rather coyly say is “we’d love to share information about our next Steam deck, but that’s for another day!” and announce a bunch of other cool stuff.
Can someone tell me how much more power efficient is ARM actually? Like under load when gaming, not in a phone that sleeps most of the time. I've heard both claims, that it's still a huge difference and that for new AMD Zen it's basically the same.
I think that's still highly debatable. Intel and AMD claim the instruction set makes no difference... but of course they would. And if that's really the case where are the power efficient x86 chips?
Possibly the truth is that everyone is talking past each other. Certainly in the Moore's Law days "marginal impact" would have meant maybe less then 20%, because differences smaller than that pretty much didn't matter. And there's no way the ISA makes 20% difference.
But today I'd say "marginal impact" is less than 5% which is way more debatable.
It's workload-dependent. On-paper, ARM is more power-efficient at idle and simple ops, but slows down dramatically when trying to translate/compose SIMD instructions.
You seem to have conflated SIMD and emulation in the context of performance. ARM has it's own SIMD instructions and doesn't take a performance hit when executing those. Translating x86 SIMD to ARM has an overhead that causes a performance hit, which is due to emulation.
I like what Valve is doing for the Linux world, but I'd like to mention that box86 and box64 have been running x86 games onto arm (incl. android phones) for a long time too... And it does that on Risc-V and LoongArch too...
There is not so much support from companies to this project that I know of, but the people behind box64 manage to make it a solid and fast solution to running windows game on arm.
Hypothetically, if Valve made a strong push to make SteamOS compatible with all Windows programs, not just games, could they make a serious run at knocking down Windows?
And why would they care? Not even Microsoft really cares about Windows licensing for consumers and businesses are never going to use computers running fake Windows.
I don't need Android apps that often, but it would be neat for the options here to expand and improve. I want to say much as Proton has accelerated things, but man, I am pretty lost now tracking which projects Proton encompasses and the history of where Valve backed/helped these efforts.
I still really want to believe it's collaborative. That good work is going to flow upstream, to collaborated Valve + crowd spaces.
* The most skilled workers are the most undervalued
* Make products to serve the customer
* Management is a skill, not a career path
* The only people they consider themselves to be unable to compete with are their customers, so enabling the customer to produce better content in their ecosystem is the most efficient way of producing things.
Have we seen a commercially available high performance 64-bit RISCV chip at production scale yet?
There’s a lot of work and experience built up for ARM through Proton and other tech (that can be reverse engineered to see how it works) like Rosetta. A lot of that would have to be redone for RISCV. Seems like a lot of risk in the short term for what’s not an obvious product benefit.
I would expect the high-end RISCV market to mature before a company like Valve dives in.
Sure, it's not open source or anything. But ARM doesn't seem to be a typical greedy incumbent that everyone hates. They don't make all that much profit or revenue given how much technology they enable - there isn't much to disrupt there.
RISC-V is severely lacking in high-performance implementations for the time being.
From the last interview question in the article (pertaining to Arm):
> We don’t really try to steer the market one direction or another; we just want to make sure that good options are always supported.
Sounds like their priority is to support Steam on the hardware consumers are currently using. Given that, it makes sense they'd go Arm in the Steam Frame, because Fex alone is already a massive undertaking, and Snapdragon is a leading mobile chipset for performance and power efficiency.
Valve is just trying to outflank Microsoft here. And they're doing a magnificent job of it.
Microsoft has on at least half a dozen occasions tried to draw a box around Valve to control their attempts to grow beyond the platform. And moreover to keep gaming gravitas on Windows. Windows Store, ActiveX, Xbox, major acquisitions ... they've failed to stop Valve's moves almost every time.
Linux, Steam Box, Steam Machine - there's now incredible momentum with a huge community with more stickiness than almost any other platform. Microsoft is losing the war.
The ARM vs RISC battle will happen, but we're not there yet. There also isn't enough proliferation for it to be strategic to Valve.
RISC-V was developed at UC Berkeley. It's roughly as Western as West realistically gets, short of being made in Hawaii.
> That's a geopolitical question
Sure, but that's not actually about where RISC-V is from. It's that it's a purposely open platform -- so much so that its governing body literally moved to Switzerland.
The reason it's a geopolitical question is more to do with what we did to their supply chains with sanctions on companies like Huawei and ZTE, and what COVID did to everyone's supply chains independently of that. Both of those things made it really evident that some domestic supply chains are critical. (On both sides -- see: the CHIPS Act)
Where RISC-V comes back in is that open source doesn't really have a functioning concept of export restrictions. Which makes it an attractive contingency plan to develop further in the event of sanctions happening again, since these measures can and have extended to chip licenses.
(Edit: I'm not saying any of this is mutually exclusive with valid concerns about Huawei, raised by various other sources. I'm less familiar with ZTE's history, but my point in either case is more of a practical one.)
I suspect that many projects—such as BOOM—have stalled as a consequence of this situation. If it continues, the long-term impact will be highly detrimental for everyone involved, including stakeholders in Western countries.
RISCV is at least a decade, if not two from being useful enough for mainstream adoption. Neither the hardware or software is anywhere close to being ready.
> PC gaming is going fast to requiring controller support.
Hmmm I don't know. I did play Elden Ring & Clair Obscur with a controller but I also played Baldur's Gate 3 on mouse and keyboard. I also play VR games with controllers or hand tracking. Basically I play with whatever the game recommends.
To me it's like saying PC game is dead because mobile phone is so popular. Sure a LOT of players, and thus money, goes into mobile gaming but that doesn't prevent proper AAA and indies games on PC to have interesting mechanics.
PS: if you are into these kind of things checkout exotic controllers in events like Amaze. I remember a collaborative where you had to blow in pipes to push a rocket in the right direction and plenty of weirder stuff. Really cool but totally niche.
The one very visible trend in the last 30 years of game development was about reducing input complexity. It has nothing to do with complexity of games themselves. Now instead of fighting clunky controls like in good old times you fight game challenges, where the input tries hard to be as transparent as possible
With the 2 options you have left because those are all the buttons :)
And autoaim because those sticks aren't precise enough.
But it's not first person shooters I worry about, because those have devolved into competitive multiplayer IAP fests that create toxic communities.
I worry about strategy games and anything with a whiff of complexity. Reduce options because going through menus with a controller is slow and clunky. Reduce options because when playing at TV distance you can't read a serious list of properties like wargames have.
Controllers provide analogue controls (eg. thumbsticks and triggers) that most keyboards don't have.
If, as you suggest, the control schemes of video games are becoming less complex (Forward, down, forward, high punch) then surely the result would be more games that are playable with only a keyboard, not fewer?
I downloaded an unoptimized game and tested it (Stray, a few years old at this point) and got some graphical glitches, but between 10-20 FPS on a three-year-old phone. That it’s even possible is impressive, and I imagine it will only get better as more games get optimized.
I find it kinda ironic that they phase out 32bit at the same time. I’d guess it would be easier to emulate 32but x86, although the difference perhaps goes away with a JIT.
There's no point in emulating just 32-bit x86 when modern games are all built for x86-64 plus several extensions. If they could make devs target an arch of their choice, they'd ask for aarch64 and skip funding FEX.
I suppose this will is in order to be able to push lower powered hand helds to penetrate further into global markets. Not everyone can afford an X64 Nvidia usd5000 gaming workstation.
Given how expensive PC gaming is, between GPU and memory price explosions, I imagine everything Valve is doing, including recently announced hardware, is all in service of keeping Steam relevant.
> and that’s when the Fex compatibility layer was started, because we knew there was close to a decade of work needed before it would be robust enough people could rely on it for their libraries.
In a recent Linus Tech Tips video, Linus Torvalds (original Linus) was asked, "if you could go back in time and start the Linux project from scratch, what would you do differently?" He had two answers, one was "nothing," and the other was "if I knew how much work this was going to take, I never would have started this project."
It makes me wonder, is there some kind of blissful ignorance required to kick off a project that will take you years to see through? How many times have I self limited myself, stopped myself from starting something, because I put on my lead hat and did some estimations and thought eh, not worth it?
> When you get into lower power, anything lower than Steam Deck, I think you’ll find that there’s an Arm chip that maybe is competitive with x86 offerings in that segment.
At which point does this pay off the emulation overhead? Fex has a lot of work to do to bridge two ISAs while going through the black box of compiler output of assembly, right?
afaia emulators like Fex are within 30 to 70% of native performance. On the fringes worse or better. But overall emulation seems totally fine.
Plus emulator technology in general could be used for binary optimization rather than strict mappings, opening up space for more optimization.
Does anyone know what the limfac is? The machine code produced is of course different on different CPU arches, but isn't this handled at the compiler level? I.e. lower level than game devs worry about.
> isn't this handled at the compiler level? I.e. lower level than game devs worry about.
But game devs (at least of a certain type) are notorious for thinking about low-level hardware performance right from the start. As a class I'm pretty sure game devs use godbolt much, much more than your typical developer.
Sadly, nowadays very few, if any, game developers care about performance or optimizations. Look at recent headline about "Helldivers 2 devs slash install size from 154GB to 23GB" and it was done by simply deduplicating assets. Gone are the days of finding inredible ways to use less opcodes that game would feel smoother.
The best thing Valve could do is nuke Wayland/X11/Xwayland from orbit. Wayland is a mess that apps still don't support and doesn't work with NVIDIA GPUs. X11 is ancient and screen tears. Xwayland is the worst of both worlds.
Is the problem in this relationship Wayland of Nvidia? It is a shame that GPU's are pretty much the one big part of your computer that doesn't really conform to the general "ownership" model.
I don't use Linux on laptops or desktops, only servers. I've been through all that pain. The userland has a long way to go to be a good OS everyone can use.
RISC-V SoCs are still underpowered for the job, and the software ecosystem hasn't matured compared to ARM, specially for Windows/x86-64 emulation. The focus on RISC-V chips is simply too recent and has lacked enough demand for it. Your question is at least a decade too early.
Unless I misunderstand something (not quite awake fully yet...),
that's good right? Aka "play games on any platform" as goal. A
bit with the inofficial goal of scummvm, to rescue old commercial
games from vanishing for young, future generations.
What Valve hasn't announced yet, but which has leaked, is "Lepton", a fork of Waydroid they're working on to enable Android gaming for Steam—in particular, getting Oculus Quest games going on the Steam Frame.
The future actually looks pretty good for indie gaming development on the Android platform, Google's shenanigans notwithstanding.
I cannot speak for the Steam console and I don't care about playing PC games on my phone: it's not the form factor for me.
But I'm really grateful for Valve and Steam.
Increasingly, more and more Windows-only games "just work" with Linux (or work with minor tweaks taken from ProtonDB).
I bought a Lenovo Legion a couple of weeks ago and I'm having a terrific experience with Linux+Steam so far. I don't claim to play the latest AAA games, but I don't feel the need to live at the edge anyway.
One game that has resisted running so far is Space Marine 2. Eventually I'll get it going. Some people report success.
Thank god. Microsoft has shown they don't care about their users as anything other than eyeballs to shove bullshit to for _years_ and Gabe called them out on it back with Windows 8, and Valve has been working on this since.
Steam Deck is fantastic to use. Good riddance to Windows.
Back when iOS and iPad were eating Microsoft's lunch in mobile, Microsoft freaked out and released Windows 8 with that new tiles UI framework ("desktop and tablet are going to converge so we need to dumb down the interface") and the Windows Store that was supposed to be their response to the App Store. Microsoft wanted all future Windows software to be released through the App Store. Of course, this was an existential threat to Valve/Steam, so Valve vociferously pushed back.
The Windows Store and its apps were so bad that Microsoft eventually scaled back their ambitions, but Valve has not forgotten.
He made some critical comments to the media about the Windows app store and Microsoft's position of trying to turn Windows in to an iOS type situation with everything locked down. Remember that Microsoft had just released Windows RT and later Windows 10 S which could only run apps from the Windows app store.
This could have pushed Steam out of the market if it had succeeded. Valve then spent the next decade building up Linux gaming almost from scratch to reduce their dependance on Microsoft.
valve is just as ambitious as microsoft or google, they just appear to be the good guy for now because theyre trying to gain some ground on evil corp, and linux is one of the only openings they (or anyone) has got.
Did they at the same time announced that they had been funding open source ARM compatibility for over a decade? Maybe it was mentioned somewhere, but the article had new details for me at least, even if I consider myself somewhat up-to-date generally.
Agreed. I saw the Steam Frame announcement, and plan to get one as soon as it's available.
I saw the mention of Fex then too, but absolutely nowhere, before now have I seen any information that they'd been working on this for the best part of a decade.
I thought for a moment from the title that Valve has finally started funding game developers to make content from SteamOS, but no, this is just another case where Valve pays some contractors for open source projects and force developers to foot the bill for verifying compatibility.
> force developers to foot the bill for verifying compatibility
How are they forcing developers? If developers don't think it's worth it to make their game compatible with Steam Deck, can't they just avoid doing that?
They are forcing developers to be the one to pay for it if they do it because there is no other player in the space that would financially benefit from games having SteamOS support. Practically every other company with an game platform, Playstation, Xbox, Nintendo, iOS, Android, etc have programs to fund bringing content to their platform. Also developers can't avoid supporting SteamOS because there is no way for them to 100% opt out of being on that platform.
The last studio I worked at where the Steam Deck came up, the rendering lead muttered “ew, no! we don’t have time to figure that out!” and that was the end of the conversation.
A week after launch, the Proton devs pushed a hotfix and the binary’s been compatible with Linux ever since.
Why the vitriol? This is one of the rare cases where a company actually puts money in open source development. Of course they ultimately do it for business reasons but everyone benefits from it as a whole, so I fail to understand the issue here.
It's not just a larger install base. Those users may require extra support, those users may tank your reviews, those users may have a worse looking game or one that crashes a lot that can result in reputational damage.
Everything valve doing for linux is making such a huge impact.
The HL3 memes don't even seem fair to use anymore. I don't even want to un-seriously make joke fun of them at this point. They are just genuinely doing so much for the community.
Valve is one of the few companies regularly seen on HN where the headline is something like "[company] is secretly doing something really great" as opposed to "[company] is secretly doing something evil"
People complain about the gambling/loot box stuff, and yeah there's legit ethical concerns there.
But overall Valve just seems straightforwardly less shitty towards the consumer than other major companies in their space, by a long shot.
196 replies →
It's an interesting case study. They're essentially another 'App Store middleman' raking in a huge 30% cut for selling games digitally. But they do enough really good stuff to keep both gamers and developers generally very happy.
3 replies →
I have plenty of complaints about them. The highly addictive gambling mechanics in their games, the extortionate cut afforded them by their dominant market position or the very rough UX in many parts of the Steam client (takes forever to startup, shows pop up ads on startup, is quite the resource hog, the store that is a pretty poorly optimized website and a lot of cruft in the less well trodden areas). But they do make some very nice open source contributions.
22 replies →
This is because it’s still majority owned by the original founder(s).
3 replies →
"We will make linux a viable gaming before we increment that number to 3!"
But I totally agree, I still install windows for gaming on my machine, but it looks like that for my purpose of gaming I can stay with Linux (I play mainly older games or indie games).
Incentives happen to be aligned on this part. That’s all.
Its scary that nowdays a company is simply doing "good business" and it is so unusual that its worth praise.
I strongly feel it’s because Valve is not a publicly traded company where they’ll eventually give up their values to meet Wall Street analyst quarterly targets.
It genuinely makes me see the value in private companies. Public companies must grow. They're accountable to so many different interests. Private companies can be happy sitting at whatever profit level they want. They can take time to tinker on something that they care about. If it doesn't pay off, that's fine.
I think I would say it this way: private companies can be good or bad, but public companies must ultimately become bad.
That's more a property of the community than of the company. If the community were differently inclined then the comments would be about how Valve is making money by addicting children to gambling and so on and so forth.
That is why I bought a steam deck: to financially support Valve's Linux efforts. I barely play games anymore but thanks to the Wine devs, CodeWeavers, and Valve, I no longer have to listen to the knuckle-draggers claiming that "Linux sucks because it can't play games". In fact, now it is the opposite: Linux is outperforming Windows[0].
[0] https://www.youtube.com/watch?v=CJXp3UYj50Q
I have a near infinite amount of respect for Wine. It seems like for at least the last twenty years, Wine just keeps getting better and better with every release.
I don’t know for sure, but I suspect a lot of the work is spent sussing out weird edge cases with different binaries. This is tedious, thankless work, but it is necessary to have true Windows compatibility.
Wine and Proton have gotten so good that I don’t bother even checking compatibility before I buy games. The game will likely run just as well or better than on Windows and it is so consistently good that it’s not worth the small effort to check ProtonDB.
I do wish that they would get Office 2024 working on Wine. This isn’t a dig at the Wine devs at all, I am sure that it’s a very hard problem, but if I can get that then I will have even more ammunition to get my parents to drop windows entirely.
11 replies →
> knuckle-draggers claiming that "Linux sucks because it can't play games"
they still do it because you can't play all the multiplayer games with kernel level anticheats
1 reply →
I wonder when games will start supporting Linux natively, especially after the Steam Machine is released.
22 replies →
They sold the Deck hardware at a loss, so I hope you've bought several full-price games to play on it since.
1 reply →
I love Proton and I like Steam and Valve definitely has done a lot of good for the FOSS world, but let’s not make the same mistakes we made with Google by worshipping a company.
All it takes is new management to change the policies to make the company horrible and evil, and in the case of Google people made the realization far too late, and now Google owns too much of the internet to avoid.
Valve seems more like Apple than Google: a well-liked company that has an obvious and not inherently exploitive business model. Google as an ad company was always destined to go bad in a way that most non-ad companies are not.
No company is your friend, and they are all fundamentally structures around making a profit. But providing goods and services in exchange for money is not inherently exploitive or evil.
19 replies →
Valve is not building all this Linux Compatibility out of the goodness of their hearts. They are doing it to avoid being shutdown by Microsoft, who effectively had a monopoly on the OS people used to play games.
It's a bit of miracle that Valve beat MS to the punch and built momentum behind Steam as the marketplace for games. They know this.
If gamers move to Linux and all the compatibility issues are solved, Valve is not going to pick a different passion project. Conversely, as long as Microsoft has a monopoly on OSes for gaming, Valve will support linux gaming.
1 reply →
A "company" is just an organizational model employed by people to pursue the intentions that those people have. It goes without saying that large endeavors involving many people will have a mixture of good and bad intentions.
Opposing all organized endeavors simply because they have the potential to pursue bad intentions essentially resolves to being against anything anyone is ever doing, which is more than a little bit pointless.
3 replies →
I mean you're completely correct.
But if we treat all companies the same regardless of their behavior, they don't have any incentive to change their behavior.
So I'll keep rewarding the good behaviour and punishing the bad.
4 replies →
I always took the HL3 memes more as a good-faith joke. Like it's part of gaming culture more than a serious jab at them.
I personally can't wait for "SteamOS 2: Episode 2 part 1" :)
There's pretty strong rumours that they actually have been working on a new Half-Life. People are hoping it releases with their new hardware in 2026.
These rumors come from Tyler McVicker who regularly gets things wrong and makes stuff up in order to get clicks on youtube. I have no idea why anyone still takes him seriously in 2025.
Are the rumors still hinting at a VR-only experience as they did a couple of years ago when Half-Life: Alyx released, or is that no longer the speculation? Because that would be unfortunate for me, I'd have to play with a bucket in hand.
18 replies →
Yes, I too have dabbled in that strongest of drugs called hopium.
Imagine HL3, Portal 3 et L4D3 but all linux only. oh my
9 replies →
They are merely trying to commoditize their complement https://gwern.net/complement
Your games are still not owned by you, they are locked inside your Steam account (liable to be suspended at any time) and app (as I've learned when I couldn't play when their pretend-but-not-really-offline mode broke; I now block it at firewall level most of the time). That part will never become "community" oriented.
Steam Games can definitely be DRM free too. Its the developer/publishers choice.
3 replies →
>The HL3 memes don't even seem fair to use anymore.
It's absolutely fair to mock them for not releasing these games and keeping radio silence all these years. They managed to dethrone Duke Nukem Forever.
There were multiple times in which the internet was hyped for Episode 3 and where it would make sense to release even a basic game like they've did with Episodes 1&2 just to wrap things up. I'm sure plenty of people that make up various explanations to why that happened but the end result is that Valve has chosen to disappoint the fans who have been waiting for the conclusion to the story. It's not like doing that would prevent them from releasing an another new entry in the series that uses revolutionary new technology or whatever.
I prefer not to have HL3 rather than a half-baked one, or a very short episode.
They have only one chance of publishing HL3, and I hope put in it the same love and care they put in 1 and 2.
I'd be very disappointed if they just released it just for the sake of releasing it.
2 replies →
Valve is sort of like modern Bell Labs for software. It has almost-monopoly on PC game sales, which results in massive profits. Then it uses part of these profits for public good on projects that are at best tangentially related to their actual business.
Honestly dreading the day Gabe has to pass on the torch. Under him valve is such a consumer focused company
When I read what you wrote, I immediately asked myself "Doesn't Gabe have children who could have been raised with the same values? Maybe that..." and then I caught myself thinking exactly the same way as many others before me, and the reason why we have so many shitty politicians in positions of power today.
I hope Gabe has setup Valve in such a way that they can pass on his mentality as a whole inside the business practices themselves. I think, after all these years, he must have surely thought about what leaving would look like for Valve. Considering this is a guy who seemingly thinks in decades, I feel maybe even optimistically calm about it.
5 replies →
He lives on a yacht and fills his days diving and doing marine research. I'm pretty sure Valve is mostly running itself.
1 reply →
And Valve has been deeply rewarded as a result. The stance that you must abuse customers to maximize economic success will be looked back upon as the stupidity it is.
From what Ive read his son is pretty actively involved day to day already at valve.
> Everything valve doing for linux is making such a huge impact.
Some of it is counter-productive though. Proton made WINE commercially viable, and in doing so, disincentivized native Linux builds of games to the point that some studios that had been releasing games natively for Linux have stopped doing so, since the Windows version now plays well enough under Linux.
So it became more straightforward to release games on Linux? Sounds like a positive. Or, is the gripe about distinction of released for vs playable on?
1 reply →
Their absurdly high 30% cut combined with having the only otherwise decent store with real network effect driven market share is a very real criticism
maybe instead of HL3 they will deliver linux on the desktop for the masses.
because that's the foreseeable trajectory at this point
>The HL3 memes don't even seem fair to use anymore.
And people have forgotten that they existed. I mean it is 18 years since the orange box.
It's because Valve is privately owned. No shareholder value to maximize.
To be fair, without the HL3 memes, would Valve ever become as massive as they are now without them constantly teasing and playing into it?
(answer: probably, but I would like to believe that this is one of the greatest unintended marketing tactics of the 21st century).
Valve had near-total dominance over PC gaming distribution before HL3 even became a meme.
half life releases were tied to new platforms, such as HL2 and its physics engine, or HL Alex and VR kits.
it's like Nintendo having a Mario game for their new hardware, e.g. Mario 64, etc.
there weren't that many teases, nor is it great marketing; CS:GO competitive e-sports is better marketed and probably made Valve more money than any HL wink-wink-nudge-nudge ever would.
> and modern multiplayer games with anti-cheat simply do not work through a translation layer, something Valve hopes will change in the future.
Although this is true for most games it is worth noting that it isn't universally true. Usermode anti-cheat does sometimes work verbatim in Wine, and some anti-cheat software has Proton support, though not all developers elect to enable it.
It works in the sense it allows you to run the game; but it does not prevent cheating. Obviously, Window's kernel anti-cheet is also only partially effective anyway, but the point of open-source is to give you control which includes cheating if you want to. Linux's profiling is just too good; full well documented sources for all libraries and kernel, even the graphics are running through easier to understand translation layers rather than signed blobs.
These things do not prevent cheating at all. They are merely a remote control system that they can send instructions to look for known cheats. Cheating still exists and will always exist in online games.
You can be clever and build a random memory allocator. You can get clever and watch for frozen struct members after a known set operation, what you can’t do is prevent all cheating. There’s device layer, driver layer, MITM, emulation, and even now AI mouse control.
The only thing you can do is watch for it and send the ban hammer. Valve has a wonderful write up about client-side prediction recording so as to verify killcam shots were indeed, kill shots, and not aim bots (but this method is great for seeing those in action as well!)
29 replies →
Anti-cheat is a misnomer; it's much more about detecting cheats more than it is preventing them. For people who are familiar with how modern anti-cheat systems work, actually cheating is really the easy part; trying to remain undetected is the challenge.
Because of that, usermode anti-cheat is definitely far from useless in Wine; it can still function insofar as it tries to monitor the process space of the game itself. It can't really do a ton to ensure the integrity of Wine directly, but usermode anti-cheat running on Windows can't do much to ensure the integrity of Windows directly either, without going the route of requiring attestation. In fact, for the latest anti-cheat software I've ever attempted to mess with, which to be fair was circa 2016, it is still possible to work around anti-cheat mechanisms by detouring the Windows API calls themselves, to the extent that you can. (If you be somewhat clever it can be pretty useful, and has the bonus of being much harder to detect obviously.)
The limitation is obviously that inside Wine you can't see most Linux resources directly using the same APIs, so you can't go and try to find cheat software directly. But let's be honest, that approach isn't really terribly relevant anymore since it is a horribly fragile and limited way to detect cheats.
For more invasive anti-cheat software, well. We'll see. But just because Windows is closed source hasn't stopped people from patching Windows itself or writing their own kernel drivers. If that really was a significant barrier, Secure Boot and TPM-based attestation wouldn't be on the radar for anti-cheat vendors. Valve however doesn't seem keen to support this approach at all on its hardware, and if that forces anti-cheat vendors to go another way it is probably all the better. I think the secure boot approach has a limited shelf life anyways.
49 replies →
> though not all developers elect to enable it.
Looking at you Rust.
Edit:
And the rest of you. If even Microsoft's Masterchief Collection supports it, I Don't understand why everyone else does not.
https://areweanticheatyet.com/
First i thought you meant the video game Rust.
Then I saw the arewe…yet url and thought you meant Rust the programming language
Then I visited the arewe…yet link and realized it was the Rust game you meant after all
10 replies →
> I Don't understand why everyone else does not.
It's because the Linux versions of those anti-cheats are significantly weaker than their Windows counterparts.
8 replies →
Wow, what a cool site. Just learned that Hunt: Showdown is supported in Linux. And it wasn't the first time I checked. Will love to give it a try.
Arc Raiders is a great example of a modern and popular multiplayer game that works with proton. I haven't heard about it having a problem with cheating.
Marvel Rivals, Age of Empires 2 DE, Path of Exile 1/2, Last Epoch, Fall Guys are other such examples. In fact, Marvel Rivals even explicitly mentioned Bazzite in one of their changelogs! I can't recall an instance when a major game name-dropped a (relatively) minor Linux distro like that.
1 reply →
I think a big portion of that is the rather poorly made anti-tamper solution they are using called 'Theia' most cheat developers are too unintelligent to correctly reverse engineer this kind of binary obfuscation
4 replies →
Valve is the only company I'd let inject anti-cheat software directly into my veins if it meant I could play CS and be sure others were not cheating haha.
Online cheaters are first against the wall when I become dictator...
Maybe they'll secretely fund an open source project to emulate only the windows kernel calls that Anti Cheats use.
As a former cheat developer, I think it is impossible since it is digging into some specific stuff of Windows. For example, some anti-cheat uses PsSetCreateThreadNotifyRoutine and PsSetCreateThreadNotifyRoutine to strip process handle permission, and those thing can't be well emulated, there is simply nothing in the Linux kernel nor in the Wine server to facilitate those yet. What about having a database of games and anticheat that does that, and what if the anticheat also have a whitelist for some apps to "inject" itself into the game process? Those are also needed to be handled and dealt with.
Plus, there are some really simple side channel exploits that your whitelisted app have vulns that you can grab a full-access handle to your anticheat protected game, rendering those kernel level protection useless, despite it also means external cheat and not full blown internal cheat, since interal cheat carrys way more risk, but also way more rewardings, such as fine-level game modification, or even that some 0days are found on the game network stack so maybe there is a buffer overflow or double-free, making sending malicious payload to other players and doing RCEs possible. (It is still possible to do internal cheat injection from external cheat, using techniques such as manual mapping/reflective DLL injecction, that effectively replicates PE loading mechanism, and then you hijack some execution routine at some point to call your injected-allocated code, either through creating a new thread, hijacking existing thread context, APC callback hijack or even exception vector register hijacking, and in general, hijack any kinds of control flow, but anticheat software actively look for those "illegal" stuff in memory and triggers red flag and bans you immediately)
From what I've seen over the years, the biggest problem for anticheat in Linux is that there is too much liberty and freedom, but the anticheat/antivirus is an antithesis to liberty and freedom. This is because anticheat wants to use strong protection mechanism borrowed from antivirus technique to provide a fair gaming experience, at the cost of lowering framerates and increasing processing power, and sometimes BSOD.
And I know it is very cliche at this point, but I always love to quote Benjamin Franklin: "Those who would give up essential liberty to purchase a little temporary safety, deserve neither liberty nor safety". I therefore only keep Windows to play games lately, and switched to a new laptop, installed CachyOS on it, and transfered all my development stuff over to the laptop. You can basically say I have my main PC at home as a more "free" xbox.
Speaking of xbox, they have even more strict control over the games, that one of the anticheat technique, HVCI (hypervisor-protected code integrity) or VBS, is straight out of the tech from xbox, that it uses Hyper-V to isolate game process and main OS, making xbox impossible to jailbreak. In Windows it prevents some degree of DMA attack by leveragng IOMMU and encrypting the memory content beforehand to makd sure it is not visible to external devices over the PCIe bus.
That said, in other words, it is ultimately all about the tradeoff between freedom and control.
A similar concept, trusted computing: https://en.wikipedia.org/wiki/Trusted_Computing
companies will go where the money is. If Valve enables, say EA, to have their yearly franchise and in-game-stores on mobile devices, they will find a way.
I honestly don't know why so many people say that anti-cheat with Proton or SteamMachines won't work. SteamOS is an immutable Linux - especially with their own SteamMachine they can enable SecureBoot and attestation that you are using the SteamOS verbatim efi boot file, kernel, and corret system fs image - all signed by Valve. Just as Battlefield 6 does on windows (relying on SecureBoot). That would still allow you to install other OSes on your SteamDeck/SteamMachine, but it would fail the anticheat attestation. I personally see the push in hardware from Valve particular so that they can support anti-cheat on linux.
I think if Linux gaming becomes popular someone may come up with a solution where you run a native linux kernel-mode anticheat. That somehow connects to the wine-hosted game.
I'm not sure how I feel about that, but it's what I think will happen.
I think Valve is not some kind of God who free the human from the hand of Microsoft, they are a private company, they are just protecting their business and protecting their business "accidentally" also protecting the customer's benefits. The movement towards Linux benefits Valve the most since they have invested on Linux Gaming for 10 years now and that movement "conveniently" benefits the gamers too. That's a win-win situation, users can escape themself from a bloated Windows and Valve has the pioneering advantage.
I believe the same but,
> they are just protecting their business and protecting their business "accidentally" also protecting the customer's benefits.
part is wrong. From my observation, they are protecting their business through protecting their customers' benefits.
Plus, they're building a moat collectively and from an open source stack. So, given the stack gets enough momentum, having Valve or not as a company won't matter anymore.
It's trying to get the elephant out of the bag, and once it's out, then there's really no way to put it back, because it's being out is better for everybody. Game companies and gamers alike.
I think this calls out a subtle, but significant difference between private and public companies.
Public companies as an asset class have to compete with an open market of other investments, so the incentives drive a min-maxing approach to revenue and value. The shareholder mandate dictates the company pursue maximal return in order to stay competitive amongst a sea of other potential investments.
A private company doesn't have this same concern. They still need to pursue profit, but not necessarily MAXIMUM profit. This means that in a sea of hypothetical directions, they are free to choose one that is slightly less profitable but has an abundance of positive externalities, vs. one that is maximally profitable but carries many negative externalities.
> From my observation, they are protecting their business through protecting their customers' benefits.
Yeah that's what I mean too, that's why I put the "accidentally" in a double-quote.
This sounds like what Red Hat is doing, they created an open-source software, prove the importance of it in the community then sells the support package to enterprise who interested in using it.
Hope that they will not close the door when Microsoft, AWS or Oracle making their own GabeCube and call it SatyaCube, BozosCube or LarryCube
2 replies →
You are absolutely correct. Valve's linux push was driven by developments in the windows platform, specifically around the release of windows 8. Microsoft was pushing a windows store similar to Apple's app store, and Valve was unequivocally stating that they were worried Microsoft would basically lock down the platform and only allow software sales through their own store, destroying their steam business. Gabe said it plainly himself (https://www.bbc.com/news/technology-18996377):
> Mr Newell, who worked for Microsoft for 13 years on Windows, said his company had embraced the open-source software Linux as a "hedging strategy" designed to offset some of the damage Windows 8 was likely to do.
> "There's a strong temptation to close the platform," he said, "because they look at what they can accomplish when they limit the competitors' access to the platform, and they say, 'That's really exciting.'"
> This is seen by commentators, external to be a reference to the inclusion of a Windows Store in the Microsoft operating system.
Having an open platform is good for consumers, but Valve is primarily looking out for themselves here. Gabe realized that windows could take Apple's IOS route (i.e. https://blog.codinghorror.com/serving-at-the-pleasure-of-the...) and lock down their OS, and everything he's done since has been an effort to protect his company against that existential threat.
He didn't just "work on Windows."
GabeN was the lead developer on Windows 1, Windows 2, and Windows 3. When Windows 95 launched, he was a bit upset that no one was making games for Windows. He did a rough port of Doom to prove the viability. Around the same time Alex St. John, Craig Eisler, and Eric Engstrom were building DirectX, GabeN saw the potential, left to create Valve, and proceeded to try and making Windows gaming a great thing.
I can only imagine that he was heartbroken to see Windows go the way it did with Windows 8, 8.1, 10, and now 11.
Well actually they tried with Windows Phone, Windows RT and Windows 10 S but failed miserably. Even Apple didn't even try to lock their macOS from installing 3rd party app.
Valve's Steam platform is well liked and actually wanted by gamers. Gamers move their games to the steam launcher and often wait for games to come to steam (e.g. with Anno games from Ubisoft).
This is in contrast with EA's Origin, Microsoft's Xbox PC and Ubisoft's Connect, which everyone hates.
I always think Valve as the "ideal" capitalist company, because what they do fits the idea of "invisible hand" perfectly, that each individual acting in their own self-interest end up benefiting everyone.
And you'd be right, that Valve is nothing special, if that idea is correct, because in that case most companies will be like Valve. But just look around, do you see many companies like Valve? No, that's because capitalism is bullshit and that makes Valve stand out.
I have been a big fan of Valve since the Orange Box days and I always dreamed of working there, but let’s not kid ourselves. This is all enabled by the massive monopolistic cash-cow that is Steam that requires a tiny team to maintain. Similarly their top games have minuscule teams and still rake-in millions in microtransactions, fueled by a shadow economy of gambling and speculative trading aimed at kids.
Yes to a large extent they got those monopolies by building truly outstanding products in good faith and by being pioneers in quite a few areas. And certainly they are an exemplary case of investing that wealth into legitimately innovative and widely appreciated long-term endeavors.
My point is that Valve is not all that special for being nice, many organizations do crave to be like that but they don’t have the luxury to have hit that jackpot. For people with mountains of money, they are among the best, but it’s not exactly a high standard, and they are remarkably inefficient in leveraging that advantage.
They’ve long lost the organizational know-how to make good games, and they have delivered remarkably few public facing successes in the last decade: mainly Valve Index and Steam Deck, both still relatively niche and wide apart, both primarily attempts at expanding Steam’s dominance to fairly uncharted markets, with mixed success. The first iteration of Steam Machines was dead on arrival, as was their long-anticipated game Artifact. CS 2 was not a significant enough upgrade to Go to really count. Half-Life Alyx was popularish I suppose. Anything else of note?
Valve cuts 30% of your revenue no matter how much you earn. They also cut 15% of the transaction by being the middleman on the market.
They also ignored the gambling/trading plague for too long, until a lot of countries threatened them to stop indirectly promoting gambling (which definitely hit them financially).
They are sitting on a money printing machine and their job is making it print no less to buy GabeN another yatch. They are like the cigarette company who donates shit load of money to the charity and cancer prevention lab while making more cigarrate then ever because people love smoking it.
I don't think they wanted or planned to be monopolized, but they are definitely taking the advantage of being it.
> acting in their own self-interest end up benefiting everyone
I'm so sick of people acting like Valve is some saint that does no wrong. Their market dominance means game developers wanting to reach the PC gamer market must comply with Valve's terms. Why do you think every Japanese visual novel released on the platform is a cut down, all-ages version that requires an off-site patch to restore the full game (and often even then it's censored in weird ways)? They got sick of being delisted while Valve turns a blind eye to all the trash porn games.
Ask yourself, does a marketplace that exerts creative control over specific studios' works while threatening financial repercussions if they don't comply benefit everyone? That sounds more like the mob to me.
Stop deifying companies.
It’s funny, because Zach Barth (of SpaceChem and many other wonderful games fame) worked at Valve and then described them as the ideal anarcho-communist type of organization.
2 replies →
Many of steams consumer benefits were a direct result of Valve getting sued and losing court cases. For example refunds and forced arbitration clause.
They are a significant actor in the market, and as a non-pc-gamer I am glad that their business goals align with Linux users. I don't believe that they do it out of kindness, but that's actually a good thing for a long term investment.
> They are a significant actor in the market, and as a non-pc-gamer I am glad that their business goals align with Linux users.
Until MS (or worse, Oracle) shows up with their half-baked clone (like Xbox Machine or Larry Cube) and ruins everyone's party
The trick is to get your egoistic interests to align with what is good for the world.
Environmentalists also only do what they perceive as doing good (and may as well be objectively good) because not doing so won't jive with their self-image. But that in itself doesn't devalue the thing they did.
If Valve shows anything it is that not being a pushover and trying to align your business interest with your values (and not vice-versa!) can also pay of economically.
Would love to see it on MacOS X -- Steam works great on my Mac Mini for the games it supports, would be great to see everything run on it.
Yep. I know Apple has little motivation to support such a project but it would be great to see them work with Valve on this. Having the majority of Steam games "just work" on modern Macs, like they do on the Steam Deck, would be fantastic.
Apple leadership cares more about "games on the Mac App Store built for Metal on a Mac" than it cares about "games on the Mac". This won't change until leadership changes.
9 replies →
Apple already made it, it's just that it targets developers rather than end users: https://developer.apple.com/games/game-porting-toolkit/
I think it's more than "little motivation" if we're being honest. Right now Valve is quietly targeting MS' attempt to create a walled garden for gaming on Windows and (probably) cut them out. Their very clever approach has been a full end-run around the OS by using Proton, which I'm sure genuinely thrilled Apple... as long as Valve is only doing that to MS.
Why would Apple ever invite Valve to potentially do the same to them?
19 replies →
Main issue is the lack of Vulkan support on macOS. Currently, solutions like MoltenVK have to be developed to add Vulkan support, which isn't as clean as just supporting it.
For some reason the prospect using Wine, Rosetta 2, and DXVK with MoltenVK on top just to run some games doesn't inspire a lot of confidence that this whole thing will be performant and/or stable.
LunarG is working on first party Vulkan driver for MacOS https://www.lunarg.com/lunarg-achieves-vulkan-1-3-conformanc.... My understanding is performance is worse than MoltenVK at this point, but it's getting there.
This isn't an "issue" so much as a feature. Apple had some vulkan support until move to the full A1 architecture had them only make Metal a first class citizen to the GPU. Concurrently happening was a pretty nasty breakup of Apple with the Khronos group.
This wasn't an inconvenience, it was a deliberated decision.
1 reply →
It's Android where you need Vulkan. Overwhelming majority of PC games that can run on Vulkan can also support Metal due to the engine they use.
The real barrier is DX games.
No, the main issue is a fundamentally different rendering pipeline (tile based deferred rendering) that makes "Vulkan support" a conceptually difficult square peg in a round hole problem, since everything is made for immediate rendering, like all the other mainstream GPUs use.
4 replies →
But why? Valve is doing this because they don’t want to have the OS vendor exert total control over them and the gaming industry.
Apple is a terrible choice by that metric.
It'll never happen for a whole bunch of reasons, but a phone with Apple's hardware, Valve's OS, and Nintendo's game library would be amazing.
1 reply →
Are you looking for Crossover? It's a bit annoying to not run Steam natively (no cmd+H to hide, etc) but it's got a lot of support. Performance is decent on my M2 mini, and even cross-platform stuff like Baldurs Gate 3 is comparable performance to native.
Especially anything that Mac Steam natively calls out lack of 32bit support has good support.
Sadly, that's not true—for instance, I was trying to run the Shadowrun Returns series the other day, and while it launches, it will hang indefinitely when you try to actually start a game. (M4 Max)
I previously played through Returns, Dragonfall, and part of Hong Kong on Mac before the 32bit-apocalypse.
CodeWeavers, the developers of Crossover, also do most of the development on proton under contract for Valve.
This is speculation but I suspect there's something in that contract that prevents Valve from competing with Crossover on MacOS.
3 replies →
Some crossover games perform better than the native ports. I play Path of Exile on Mac using the Windows client with a translation layer, and it plays better than the native release.
Valve employed Alyssa Rosenzweig while she developed the graphics stack for asahi linux. That's a very simple statement that masks the size of the achievement and its impact on the world. No, we haven't entered a golden era of gaming on macs, but the world has been shown the way. And no, the software challenges are not insurmountable.
bummed she left for intel she was next level. asahi lina commented after she left that there's currently no one doing graphics work anymore on asahi. projects kinda got some uncertainty to it now
The last time I can remember a collaboration between Valve and Apple was for the SteamVR support on macOS back in 2016. Sadly it fell apart a year(-ish) after that. But… one can dream!
Unfortunately, this will not happen. Even if they have it all working:
Above all, Apple wants to show that their hardware is awesome, especially because it really is. Running x86 games or compatibility layers even with great emulation will make that $3000 Mac look half decent at best, against a $1500 gaming laptop. Simply not the story Apple want to tell.
If apple wanted to show that they have good hardware they wouldn't gimp the iPad pro with iOS. They really don't care.
2 replies →
I'm not sure what FEX could offer on macOS that Rosetta 2 doesn't already, with better performance thanks to Apple Silicon magic.
Running x86 code on ARM macOS is the most solved part of the stack, if anything needs work it's the API translation layers.
Aren’t most Mac issues now around Metal vs OpenGL and DirectX?
Rosetta 2 is going to be EOL'd within the next few years. A more permanent solution would certainly be welcome.
6 replies →
It would be neat if Valve would fund having Steam Client run on Apple Silicon without Rosetta 2 so arm games like Baldur's Gate 3 can be fully supported.
I wonder if Apple's GPT (Game Porting Toolkit) could added to the macOS Steam client as a compatibility tool, like Proton is in the Linux client.
GPTK is mostly a bunch of developer tools for converting to Metal, and the closest it gets to anything like Proton is an "evaluation environment" that is nothing close to Proton's performance. Proton is mostly Wine, and Wine on macOS uses MoltenVK, so it's probably easier to just port Proton.
2 replies →
Apple's GPTK only supports D3D12 -> Metal. In addition, it's ambiguous if 3rd parties can distribute the D3DMetal dylib, as there's no license.
Are you expecting to run Windows 11 ARM version on your Mac Mini directly, or within Parallels?
I think it's a pretty reasonable wish for more macOS + Apple Silicon support of games, including more native FEX & Proton ARM support within the steam client. (We're lucky Steam works, it's a better games client than the Mac App Store dreams to be, but that's also not saying much either.)
Apple Silicon has no UEFI support except as provided by Asahi, so that would be needed at a minimum to boot Windows 11 natively. Then there's the whole issue of having native Windows drivers for the Apple Silicon-specific hardware.
You’d run FEX with WINE/Proton, no windows needed. If you did use a VM, I’d think it would be a Linux VM. But, Linux VM on macOS could already use Apple’s Rosetta2 for x86_64-to-arm64 translation.
Speaking of which, maybe you could just run the games with Apple’s WINE “game porting toolkit” direct with Rosetta2. Worth a Google.
EDIT: indeed, you can already play x86 windows games on Mac using software written by Apple: https://gist.github.com/Frityet/448a945690bd7c8cff5fef49daae...
I think they're wishing for something like the Proton/Fex combination for running x86 Windows games on ARM Macs, like they already do for Linux.
kinda funny that Microsoft has tried and failed multiple times to make Windows on ARM work
and then valve is probably going to succeed, to Microsoft's detriment
A funny detail is that Microsoft's mostly fruitless ARM efforts unintentionally ended up being a boon for Valves ARM effort. From MSVC 2019 they started augmenting x86 binaries with undocumented metadata specifically to assist the Windows x86-on-ARM emulator, but then the FEX team figured out how that works and implemented the same optimizations in their emulator, greatly increasing the performance of most recent Windows games on ARM Linux.
I don't understand what you mean by this. My wife's HP laptop has an ARM processor with Windows and it.. just works? Like everything works. The computer is super fast, quiet, great battery usage, ultrathin and even scary affordable for what you get. All software works. I've not found x64 programs to run noticeably worse, at least not the ones we use, and plenty programs have an ARM build.
As a Windows user you don't even need to know that it's an ARM computer. Just use it like you'd use any other Windows computer.
Don't just say stuff, man. This is not Twitter. Try to at least figure out whether you're vaguely directionally correct before writing snarky comments.
I think they meant that by most metrics, Windows on ARM has not yet had a market impact. Not that the product doesn't work.
That being said, the mindshare well of "Windows on ARM" was poisoned by Windows RT, then later the objectively terrible performance of Windows 10 on ARM at launch.
I think the problem is that, until recently, there was little impetus to actually run Windows on devices where ARM actually has a meaningful advantage over x86. The Windows ARM laptops out there today don't impress, not just because of the software, but because the hardware itself isn't "better enough" than Intel or AMD to justify the transition for most people the way Apple Silicon was, especially for games. That is to say nothing of desktops, where battery life isn't even a concern.
Valve is using ARM to run Windows games on "ultra portable" devices, starting with the Steam Frame. At least right now, there isn't a competitive x86 chip that fits this use case. It also feels like more of an experiment, as Valve themselves are setting the expectation that this is a "streaming first" headset for running games on your desktop, and they've even said not to expect a great experience playing Half-Life: Alyx locally (a nearly 7 year old title).
It will be interesting to see if Intel/AMD catch up to ARM on efficiency in time to keep handhelds like the Steam Deck and ROG Ally from jumping ship. Right now it seems Valve is hedging their bets.
> At least right now, there isn't a competitive x86 chip
I don't think there will ever be a competitive x86 chip. ARM is eating the world piece by piece. The only reason the Steam Deck is running x86 is because it's not performant enough with two translations (Windows to Linux, x86 to ARM). Valve is very wisely starting the switch with a VR headset, a far less popular device than its already niche Steam Deck. The next Steam Deck might already switch to ARM looking at what they announced last week.
x86 is on the way out. Not in two years, perhaps not in ten years. But there will come a time where the economics no longer make sense and no one can afford to develop competitive chips for the server+gamers market alone. Then x86 is truly dead.
9 replies →
The key difference is that Valve isn't trying to make Windows work, just desktop gaming, which I'd imagine is a large part of why Microsoft's efforts failed. As much as Linux desktops haven't particularly had much polish over the years, there's still an advantage to taking something bare bones and trying to flesh it out in a way that's works well compared to taking something that's already pretty bloated and then trying to retrofit it into something new.
It turns out the best API for gaming on Linux and gaming on ARM was Win32 and x86_64. Who knew?
Well, compiling ARM game binaries is actually super duper easy and just totally fine. The issue Windows actually has with ARM is GPU drivers for the ARM SoCs. Qualcomm graphics drivers are just super slow and unreliable and bad. ARM CPU w AMD GPU is easy mode.
Shows how a stable API will beat the hell out of bleeding-edge improvements every time.
It's mostly because Microsoft have lost focus & interest in the desktop OS market and have shifted priorities to cloud service (Azure). Right now Microsoft is a sleeping giant that doesn't see the writing on the wall regarding Valve's efforts.
Oh I think they see it, the problem is that they can’t execute anymore. They are pivoting in the same direction but nobody wants to use the XBox Store on Windows and nobody trusts Game Pass anymore.
Why don't they just ask Copilot to do it?
It may have taken them a while, but it does now work fine.
Define 'fine'
5 replies →
They can't even make Explorer work. They're pathetic.
Any leads on when the next generation of Steam Deck will be released? Hoping it could be sometime in 2025, but suspect it will be more like 2026.
Over the holidays I was playing GTA: San Andreas on a Nintendo Switch. It's fun but so underpowered for a game released in 2004 (Yes, 21 years ago! Damn..). I'm really craving something more.
As a sidenote, it's really cool Valve allows installing SteamOS on any hardware. There are some alternative comparable form-factor devices:
* Lenovo Legion Go S
* Asus ROG Ally
But I have yet to see any of these in real life, so not sure how good or bad they really are.
Source: https://www.pcmag.com/picks/the-best-handheld-gaming-devices
It won't be for a while, since Valve is releasing the Steam Machine next year and has commented that they are waiting until they can build a Steam Deck successor that is significantly better than the original.[1] My guess is 2027.
1. https://www.theverge.com/2023/9/21/23884863/valve-steam-deck...
Just get a Steam Deck. it's an incredible value for what you get and for what it can do. I'm no expert, but I do pay as close attention as I can to what's going on with gaming hardware because of my limited budget, and I'm guessing Steam Deck 2 is more like Q1 2028, not any time sooner. I'm ok with that. I play all the games I want on my Steam Deck OLED, and I see plenty of life left in it, even this "late" in the game.
> Just get a Steam Deck. it's an incredible value for what you get and for what it can do
I'm still kind of flabbergasted that we're in a world where the cheapest Steam Deck model literally costs less than the Switch 2. Sure, neither of them are exactly powerhouses as far as console hardware goes, but at least on one of them you literally can just use the system however you want as a desktop OS as a bonus...
I would assume Steam Deck 2 isn't dropping before at least H2 2026, if not later, if they didn't bring it out with the announcement of the other devices.
Valve's only official statement as far as I know is that it will come when they see a significant enough hardware upgrade to warrant a new system. If they don't move to ARM, AMD's Medusa APUs are their next architecture with major upgrades, so I would guess that Valve would order another custom AMD chip but based on Medusa, which won't release until at least 2027. I would guess at least H2 2027 but probably early 2028 for an AMD-based Steam Deck 2.
5 replies →
Valve has said they want at least double the capabilities, while still fitting in a similar power envelope. Unsaid is that it also needs to fit in the same price budget, but I tend to believe that's their intent. It's gonna be a while. Valve got a stellar deal on some somewhat unusual Zen2 APUs, orginally built for Magic Leap; finding a similar good deal is going to take time. I sort of hope Valve isnt going to put out a $1600 Halo system (but probably would buy such a next-gen Gorgon Halo system). Maybe Gorgon Point is good enough for them, in which case yeah 2026H2 is reasonable.
Are you aware that the year is 2025, and that it is 92.2% over? There is next to no chance of a Deck2 this year. I would really really not hold my breath for 2026 either.
Sorry, good catch - yes, I meant 2026 or 2027!
I recently bought a Legion Go S because the primary way I game nowadays is streaming from my desktop to a handheld, and the higher quality display (1900x1200 resolution with 120 Hz over the 1280x800 and 90 Hz on the Steam Deck OLED) seemed worthwhile given that my desktop can easily provide enough throughput to play with relatively high graphics settings. It came with SteamOS preinstalled, and from a software perspective, everything does seem pretty close to identical. The only things I've slightly missed from the Steam Deck are slight hardware nits with the Legion Go S; the placement of the equivalent of the two SteamOS-specific buttons (not sure exactly how to refer to them, but they're labeled "Steam" and "..." on the Deck) just above the Start and Select buttons while looking and feeling the same mixes me up sometimes in a way that never happened on the Deck, and I miss having four unmapped buttons on the back of the device that I can set up however I like in games rather than only two. I also tend to prefer having symmetrical thumbsticks higher up on the device rather than having one high and one low; I've noticed that my hands aren't quite as comfortable when using the D-pad for extended amounts of time, which is unfortunate given my preference for it when playing stuff like emulated GBA games (incidentally one of the few things I tend to do locally instead of streaming; the low power profile setting already is an easy battery life win when streaming, and in practice it make the battery life when GBA emulation also much more tolerable, along with keeping the fans much quieter without seeming to impact performance of the emulation, given that even with this setting the fast-forward function can go far faster than I'd ever need it to).
If you're considering getting an alternative handheld, a better OS would be either Bazzite or CachyOS Handheld edition. SteamOS is not bad, but it uses an older kernel+graphics stack which doesn't make it very ideal for running on recent hardware. Plus, dedicated gaming distros like Bazzite have additional hardware support (like thirdparty game controllers) which may not be supported in SteamOS.
Currently, AMD Strix Halo based handhelds are the most powerful portable gaming devices out there, with the top three being the GPD Win 5, the OneXPlayer OneXfly Apex, and the AYANEO Next 2. Of these three, the GPD Win 5 has already started shipping. Problem is they're stupid expensive.
Personally, I will wait until I can run FSR4 natively on these portables, because FSR makes a pretty significant QoL improvement on these handhelds.
FWIW it's fairly straightforward to set up FSR4 with Proton-GE nowadays, assuming you're comfortable with editing one config file or manually specifying an env var for the game[1]. I'm not sure if using an alternate version of Proton would be considered "native" though, or if you mean for the default version of proton (or for Linux builds of games specifically), but setting it up is a fairly straightforward process even for people who might prefer not to use the terminal if you use something like ProtonUp to manage the installation for you. I imagine that the process for using a custom Proton isn't much different on Bazzite and CachyOS, although I'm not sure whether it would be something commonly done on CachyOS given that they have their own Proton distribution.
[1]: I don't think there's a way to link to it directly, but `PROTON_FSR4_UPGRADE=1` (or a specific different version if you'd like) is documented in the README in this table: https://github.com/GloriousEggroll/proton-ge-custom#modifica...
MLID says 2028+: https://www.youtube.com/shorts/Srvv_Zd_k4c
No earlier than 2027, it's valve you're talking about, they don't need to rush
I wouldn't be surprised if they don't. Valve don't want to sell hardware, they want to sell games. They only make hardware as flagships for new markets, then they want other hardware manufacturers to take over.
the legion go is more powerful and a has a nice screen, but is heavier, boxier, and has a worse batteyr life than the steam deck
Valves moving into hardware more than ever right now, not moving away from it. They've already sand multiple times a deck 2 is on the cards, but only when theres enough of a hardware bump to make it make sense as a product. Slapping a tiny bit newer cpu in there and calling it a Steam Deck 2 isn't what Valve are about.
They definitely are working on it. They announced the steam machine, steam controller, and the valve frame (standalone vr headset with seamless screen sharing from a PC), and in their reveal video the first thing they rather coyly say is “we’d love to share information about our next Steam deck, but that’s for another day!” and announce a bunch of other cool stuff.
Can someone tell me how much more power efficient is ARM actually? Like under load when gaming, not in a phone that sleeps most of the time. I've heard both claims, that it's still a huge difference and that for new AMD Zen it's basically the same.
The instruction set has marginal impact. But many power efficient chips happen to be using the ARM instruction set today.
I think that's still highly debatable. Intel and AMD claim the instruction set makes no difference... but of course they would. And if that's really the case where are the power efficient x86 chips?
Possibly the truth is that everyone is talking past each other. Certainly in the Moore's Law days "marginal impact" would have meant maybe less then 20%, because differences smaller than that pretty much didn't matter. And there's no way the ISA makes 20% difference.
But today I'd say "marginal impact" is less than 5% which is way more debatable.
14 replies →
It's workload-dependent. On-paper, ARM is more power-efficient at idle and simple ops, but slows down dramatically when trying to translate/compose SIMD instructions.
You seem to have conflated SIMD and emulation in the context of performance. ARM has it's own SIMD instructions and doesn't take a performance hit when executing those. Translating x86 SIMD to ARM has an overhead that causes a performance hit, which is due to emulation.
1 reply →
I like what Valve is doing for the Linux world, but I'd like to mention that box86 and box64 have been running x86 games onto arm (incl. android phones) for a long time too... And it does that on Risc-V and LoongArch too...
There is not so much support from companies to this project that I know of, but the people behind box64 manage to make it a solid and fast solution to running windows game on arm.
Hypothetically, if Valve made a strong push to make SteamOS compatible with all Windows programs, not just games, could they make a serious run at knocking down Windows?
And why would they care? Not even Microsoft really cares about Windows licensing for consumers and businesses are never going to use computers running fake Windows.
There is no real business case.
>businesses are never going to use computers running fake Windows.
why not? If it's cheaper and compatible, why not?
10 replies →
Here’s an idea, charge money for it?
I’m sure there are lots of businesses that dislike Microsoft and the freemium model they’re using.
2 replies →
If you include all the low level works needed to make them work well which is a lot, then maybe.
I was also delighted to see Steam has an effort underway for Android on Linux, allegedly a fork of Waydroid, that they are working on. Tentatively delighted because it's unclear if this really will be open source, but hopefully! https://steamdb.info/app/3029110/info/ https://www.gamingonlinux.com/2025/12/valves-version-of-andr...
I don't need Android apps that often, but it would be neat for the options here to expand and improve. I want to say much as Proton has accelerated things, but man, I am pretty lost now tracking which projects Proton encompasses and the history of where Valve backed/helped these efforts.
I still really want to believe it's collaborative. That good work is going to flow upstream, to collaborated Valve + crowd spaces.
Wow this is exciting news!
For an understanding of what Valve are doing, here is a 1 hour talk by Gabe Newell (CEO):
Gabe Newell: On Productivity, Economics, Political Institutions, and the Future of Corporations https://www.youtube.com/watch?v=Td_PGkfIdIQ
TL;DR:
* The most skilled workers are the most undervalued
* Make products to serve the customer
* Management is a skill, not a career path
* The only people they consider themselves to be unable to compete with are their customers, so enabling the customer to produce better content in their ecosystem is the most efficient way of producing things.
Given how arm license is know to be less than friendly.... Wouldn't it be preferable to explore a RISCV architecture.
As far as I know RISC provides similar power efficiency and sleep that is like ARM.
Have we seen a commercially available high performance 64-bit RISCV chip at production scale yet?
There’s a lot of work and experience built up for ARM through Proton and other tech (that can be reverse engineered to see how it works) like Rosetta. A lot of that would have to be redone for RISCV. Seems like a lot of risk in the short term for what’s not an obvious product benefit.
I would expect the high-end RISCV market to mature before a company like Valve dives in.
>at production scale
You can even omit that part and the result is the same: nothing
>arm license is know to be less than friendly
Sure, it's not open source or anything. But ARM doesn't seem to be a typical greedy incumbent that everyone hates. They don't make all that much profit or revenue given how much technology they enable - there isn't much to disrupt there.
RISC-V is severely lacking in high-performance implementations for the time being.
From the last interview question in the article (pertaining to Arm):
> We don’t really try to steer the market one direction or another; we just want to make sure that good options are always supported.
Sounds like their priority is to support Steam on the hardware consumers are currently using. Given that, it makes sense they'd go Arm in the Steam Frame, because Fex alone is already a massive undertaking, and Snapdragon is a leading mobile chipset for performance and power efficiency.
Agree but I would argue RISC is catching up fast.
1 reply →
No one has yet produced a RISC-V CPU or SoC with truly competitive CPU and GPU performance and compatibility to the current state of arm64 or amd64.
It’s a catch-22: why build a RISC-V CPU if there’s no software for it, and why write software if there’s no CPU to run it?
1 reply →
That's a geopolitical question.
ARM is Western
RISC is China / Eastern
Valve is just trying to outflank Microsoft here. And they're doing a magnificent job of it.
Microsoft has on at least half a dozen occasions tried to draw a box around Valve to control their attempts to grow beyond the platform. And moreover to keep gaming gravitas on Windows. Windows Store, ActiveX, Xbox, major acquisitions ... they've failed to stop Valve's moves almost every time.
Linux, Steam Box, Steam Machine - there's now incredible momentum with a huge community with more stickiness than almost any other platform. Microsoft is losing the war.
The ARM vs RISC battle will happen, but we're not there yet. There also isn't enough proliferation for it to be strategic to Valve.
> RISC is China/Eastern
RISC-V was developed at UC Berkeley. It's roughly as Western as West realistically gets, short of being made in Hawaii.
> That's a geopolitical question
Sure, but that's not actually about where RISC-V is from. It's that it's a purposely open platform -- so much so that its governing body literally moved to Switzerland.
The reason it's a geopolitical question is more to do with what we did to their supply chains with sanctions on companies like Huawei and ZTE, and what COVID did to everyone's supply chains independently of that. Both of those things made it really evident that some domestic supply chains are critical. (On both sides -- see: the CHIPS Act)
Where RISC-V comes back in is that open source doesn't really have a functioning concept of export restrictions. Which makes it an attractive contingency plan to develop further in the event of sanctions happening again, since these measures can and have extended to chip licenses.
(Edit: I'm not saying any of this is mutually exclusive with valid concerns about Huawei, raised by various other sources. I'm less familiar with ZTE's history, but my point in either case is more of a practical one.)
1 reply →
> RISC is China / Eastern
Imo this is a really strange characterization of RISC. I've never seen this before. I think you try to paint a misleading picture in bad faith, please consider this: - https://riscv.org/blog/how-nvidia-shipped-one-billion-risc-v... - https://tenstorrent.com/en/ip/risc-v-cpu - https://blog.westerndigital.com/risc-v-swerv-core-open-sourc... - https://www.sifive.com - ... - https://riscv.org/about/ -> "RISC-V International Association in Switzerland"
1 reply →
No: RISC is open ARM is closed.
I suspect that many projects—such as BOOM—have stalled as a consequence of this situation. If it continues, the long-term impact will be highly detrimental for everyone involved, including stakeholders in Western countries.
4 replies →
*RISC-V
ARM is a RISC: https://en.wikipedia.org/wiki/Reduced_instruction_set_comput...
RISCV is at least a decade, if not two from being useful enough for mainstream adoption. Neither the hardware or software is anywhere close to being ready.
The one thing that worries me is that PC gaming is going fast to requiring controller support.
Which is reducing the complexity of control schemes to either nothing or mortal kombat like combos.
Forward, down, forward, high punch to build a new city in a 4X in a few years?
> PC gaming is going fast to requiring controller support.
Hmmm I don't know. I did play Elden Ring & Clair Obscur with a controller but I also played Baldur's Gate 3 on mouse and keyboard. I also play VR games with controllers or hand tracking. Basically I play with whatever the game recommends.
To me it's like saying PC game is dead because mobile phone is so popular. Sure a LOT of players, and thus money, goes into mobile gaming but that doesn't prevent proper AAA and indies games on PC to have interesting mechanics.
PS: if you are into these kind of things checkout exotic controllers in events like Amaze. I remember a collaborative where you had to blow in pipes to push a rocket in the right direction and plenty of weirder stuff. Really cool but totally niche.
> Elden Ring & Clair Obscur with a controller
I play soulsbornes on consoles with a controller, of course. Because From doesn't know how to do any other control scheme, they have "console DNA".
But how do you play Civilization (<= 5, the good ones) with a controller?
The one very visible trend in the last 30 years of game development was about reducing input complexity. It has nothing to do with complexity of games themselves. Now instead of fighting clunky controls like in good old times you fight game challenges, where the input tries hard to be as transparent as possible
> you fight game challenges
Like "kill 100.000 mobs" ?
With the 2 options you have left because those are all the buttons :)
And autoaim because those sticks aren't precise enough.
But it's not first person shooters I worry about, because those have devolved into competitive multiplayer IAP fests that create toxic communities.
I worry about strategy games and anything with a whiff of complexity. Reduce options because going through menus with a controller is slow and clunky. Reduce options because when playing at TV distance you can't read a serious list of properties like wargames have.
3 replies →
Controllers provide analogue controls (eg. thumbsticks and triggers) that most keyboards don't have.
If, as you suggest, the control schemes of video games are becoming less complex (Forward, down, forward, high punch) then surely the result would be more games that are playable with only a keyboard, not fewer?
Heh the MK combo mention was a joke. Forgot I’m on HN.
3 replies →
Oh so they did not "adopt" Fex, they actually financed the leading developer from the start.
Commoditize your complements (layers above and below).
I downloaded an unoptimized game and tested it (Stray, a few years old at this point) and got some graphical glitches, but between 10-20 FPS on a three-year-old phone. That it’s even possible is impressive, and I imagine it will only get better as more games get optimized.
How did you do that?
I followed the same steps the Verge reporter did: Downloaded GameHub, connected my Steam account. The initial boot was lengthy, but it worked.
I find it kinda ironic that they phase out 32bit at the same time. I’d guess it would be easier to emulate 32but x86, although the difference perhaps goes away with a JIT.
There's no point in emulating just 32-bit x86 when modern games are all built for x86-64 plus several extensions. If they could make devs target an arch of their choice, they'd ask for aarch64 and skip funding FEX.
2026 will be the year of the linux desktop
I used to have `echo "$((( $(date +%Y) + 1 ))) will be the year of the linux desktop"` at the end of my .bashrc
Linux desktop is getting better every year, meanwhile Windows and arguably MacOS are getting worse every year.
1 reply →
If you ever get Linux to boot on your Snapdragon notebook...
I’m interested in seeing Proton perform on Arm for Windows x86 games. That sounds like a real challenge.
I suppose this will is in order to be able to push lower powered hand helds to penetrate further into global markets. Not everyone can afford an X64 Nvidia usd5000 gaming workstation.
Given how expensive PC gaming is, between GPU and memory price explosions, I imagine everything Valve is doing, including recently announced hardware, is all in service of keeping Steam relevant.
Video tech is driven by pr0n and OS tech is driven by games.
> and that’s when the Fex compatibility layer was started, because we knew there was close to a decade of work needed before it would be robust enough people could rely on it for their libraries.
In a recent Linus Tech Tips video, Linus Torvalds (original Linus) was asked, "if you could go back in time and start the Linux project from scratch, what would you do differently?" He had two answers, one was "nothing," and the other was "if I knew how much work this was going to take, I never would have started this project."
It makes me wonder, is there some kind of blissful ignorance required to kick off a project that will take you years to see through? How many times have I self limited myself, stopped myself from starting something, because I put on my lead hat and did some estimations and thought eh, not worth it?
https://www.youtube.com/watch?v=mfv0V1SxbNA
> When you get into lower power, anything lower than Steam Deck, I think you’ll find that there’s an Arm chip that maybe is competitive with x86 offerings in that segment.
At which point does this pay off the emulation overhead? Fex has a lot of work to do to bridge two ISAs while going through the black box of compiler output of assembly, right?
afaia emulators like Fex are within 30 to 70% of native performance. On the fringes worse or better. But overall emulation seems totally fine. Plus emulator technology in general could be used for binary optimization rather than strict mappings, opening up space for more optimization.
Valve has done wonders for Linux. I've often thought about this: https://kaveh.page/blog/linux-valve
Don’t you think that if usage goes significantly up, MS will just use its lawyers to shut down all of these cute initiatives ?
No, the supreme court has settled it [0], there's no basis for a lawsuit.
[0]: https://en.wikipedia.org/wiki/Google_LLC_v._Oracle_America,_....
Does anyone know what the limfac is? The machine code produced is of course different on different CPU arches, but isn't this handled at the compiler level? I.e. lower level than game devs worry about.
The exception I see is if SIMD intrinsics.
This system allows playing unmodified production x86 executables on arm64. It doesn’t have anything to do with the developers.
That's great, but begs the question: why not just compile the games for ARM?
7 replies →
> isn't this handled at the compiler level? I.e. lower level than game devs worry about.
But game devs (at least of a certain type) are notorious for thinking about low-level hardware performance right from the start. As a class I'm pretty sure game devs use godbolt much, much more than your typical developer.
Sadly, nowadays very few, if any, game developers care about performance or optimizations. Look at recent headline about "Helldivers 2 devs slash install size from 154GB to 23GB" and it was done by simply deduplicating assets. Gone are the days of finding inredible ways to use less opcodes that game would feel smoother.
3 replies →
Yooo this is awesome, maybe we will finally get some game support on Macs now
Doubtful, the article is about FEX-emu and Apple has only shown interest in native ports of games to their platform.
Plus, it looks like upstream FEX doesn't play very nice with Apple Silicon in the first place.
The best thing Valve could do is nuke Wayland/X11/Xwayland from orbit. Wayland is a mess that apps still don't support and doesn't work with NVIDIA GPUs. X11 is ancient and screen tears. Xwayland is the worst of both worlds.
> Wayland is a mess that apps still don't support and doesn't work with NVIDIA GPUs
KDE supports Wayland: https://blogs.kde.org/2025/11/26/going-all-in-on-a-wayland-f...
Nvidia has had Wayland support for a while. Here are their latest beta drivers. The first item of the release notes is about Wayland: https://www.nvidia.com/en-us/drivers/details/258750/
Is the problem in this relationship Wayland of Nvidia? It is a shame that GPU's are pretty much the one big part of your computer that doesn't really conform to the general "ownership" model.
What do you use or recommend?
I don't use Linux on laptops or desktops, only servers. I've been through all that pain. The userland has a long way to go to be a good OS everyone can use.
6 replies →
No, thanks. That's the worst Android did, creating their own incompatible thing. We don't need another NIH like that.
Android is a much more successful platform than the Linux desktop.
3 replies →
This may be a naive question: why ARM and not RISC-V?
RISC-V SoCs are still underpowered for the job, and the software ecosystem hasn't matured compared to ARM, specially for Windows/x86-64 emulation. The focus on RISC-V chips is simply too recent and has lacked enough demand for it. Your question is at least a decade too early.
How many people do you know with RISC-V computers?
Now we just need qualcomm to sort out their linux snapdragon support
Unless I misunderstand something (not quite awake fully yet...), that's good right? Aka "play games on any platform" as goal. A bit with the inofficial goal of scummvm, to rescue old commercial games from vanishing for young, future generations.
It should be RISC-V...
What would be the benefit?
See the FAQ on risc-v site.
What Valve hasn't announced yet, but which has leaked, is "Lepton", a fork of Waydroid they're working on to enable Android gaming for Steam—in particular, getting Oculus Quest games going on the Steam Frame.
The future actually looks pretty good for indie gaming development on the Android platform, Google's shenanigans notwithstanding.
I cannot speak for the Steam console and I don't care about playing PC games on my phone: it's not the form factor for me.
But I'm really grateful for Valve and Steam.
Increasingly, more and more Windows-only games "just work" with Linux (or work with minor tweaks taken from ProtonDB).
I bought a Lenovo Legion a couple of weeks ago and I'm having a terrific experience with Linux+Steam so far. I don't claim to play the latest AAA games, but I don't feel the need to live at the edge anyway.
One game that has resisted running so far is Space Marine 2. Eventually I'll get it going. Some people report success.
Space Marine 2 runs on my desktop, running Manjaro. Never tried on a handheld. Works fine!
Any tweaks for Proton that you think were helpful? I tried stuff from ProtonDB but it hangs during the initial cutscene and never recovers.
Thank god. Microsoft has shown they don't care about their users as anything other than eyeballs to shove bullshit to for _years_ and Gabe called them out on it back with Windows 8, and Valve has been working on this since.
Steam Deck is fantastic to use. Good riddance to Windows.
> Gabe called them out on it back with Windows 8
Context?
Back when iOS and iPad were eating Microsoft's lunch in mobile, Microsoft freaked out and released Windows 8 with that new tiles UI framework ("desktop and tablet are going to converge so we need to dumb down the interface") and the Windows Store that was supposed to be their response to the App Store. Microsoft wanted all future Windows software to be released through the App Store. Of course, this was an existential threat to Valve/Steam, so Valve vociferously pushed back.
The Windows Store and its apps were so bad that Microsoft eventually scaled back their ambitions, but Valve has not forgotten.
Gabe Newell: "I think Windows 8 is a catastrophe for everyone in the PC space."
2012: https://www.pcgamer.com/gabe-newell-i-think-windows-8-is-a-c...
He made some critical comments to the media about the Windows app store and Microsoft's position of trying to turn Windows in to an iOS type situation with everything locked down. Remember that Microsoft had just released Windows RT and later Windows 10 S which could only run apps from the Windows app store.
This could have pushed Steam out of the market if it had succeeded. Valve then spent the next decade building up Linux gaming almost from scratch to reduce their dependance on Microsoft.
valve is just as ambitious as microsoft or google, they just appear to be the good guy for now because theyre trying to gain some ground on evil corp, and linux is one of the only openings they (or anyone) has got.
The only proprietary, steam client itself, you can fork their stack build your own store if you like.
fix anticheat!
Steam API doesnt even have an arm version. You cant build a native arm app with it.
This may be a naive question but why ARM and not RISC-V?
This is news how exactly since the announcement of arm Steam Frame?
Did they at the same time announced that they had been funding open source ARM compatibility for over a decade? Maybe it was mentioned somewhere, but the article had new details for me at least, even if I consider myself somewhat up-to-date generally.
Agreed. I saw the Steam Frame announcement, and plan to get one as soon as it's available.
I saw the mention of Fex then too, but absolutely nowhere, before now have I seen any information that they'd been working on this for the best part of a decade.
I think it's more revealing of how they've been playing the long game
I thought for a moment from the title that Valve has finally started funding game developers to make content from SteamOS, but no, this is just another case where Valve pays some contractors for open source projects and force developers to foot the bill for verifying compatibility.
> force developers to foot the bill for verifying compatibility
How are they forcing developers? If developers don't think it's worth it to make their game compatible with Steam Deck, can't they just avoid doing that?
They are forcing developers to be the one to pay for it if they do it because there is no other player in the space that would financially benefit from games having SteamOS support. Practically every other company with an game platform, Playstation, Xbox, Nintendo, iOS, Android, etc have programs to fund bringing content to their platform. Also developers can't avoid supporting SteamOS because there is no way for them to 100% opt out of being on that platform.
15 replies →
The last studio I worked at where the Steam Deck came up, the rendering lead muttered “ew, no! we don’t have time to figure that out!” and that was the end of the conversation.
A week after launch, the Proton devs pushed a hotfix and the binary’s been compatible with Linux ever since.
Why the vitriol? This is one of the rare cases where a company actually puts money in open source development. Of course they ultimately do it for business reasons but everyone benefits from it as a whole, so I fail to understand the issue here.
Because the title mislead me. It turned out that 0 windows games are receiving funding to add ARM compatibility.
6 replies →
I'm sure those developers hate getting a larger install base for free.
It's not just a larger install base. Those users may require extra support, those users may tank your reviews, those users may have a worse looking game or one that crashes a lot that can result in reputational damage.
5 replies →