Comment by PaulKeeble
21 hours ago
I think its interesting that mainstream PC gaming press is now talking about Linux. We have the benchmark Youtube channels doing some benchmarks of it as well and plenty of reports of "it just works", which is pretty promising at least for the games that aren't intentionally excluded by DRM. For me its still controllers and equipment incompatibility due to my VR headset and sim wheel/pedals setup, I use Linux everywhere else in my router and home servers. I just hope that Nvidia notices that there does appear to be a swing happening and improves their driver situation.
The last remaining roadblock is kernel level anti-cheat frameworks.
Pretty horrible technology, and unfortunately a good majority of the gaming industry by revenue relies on it.
I'd say there are two remaining roadblocks. First and biggest is kernel level anti-cheat frameworks as you point out. But there's also no open source HDMI 2.1 implementation allowed by the HDMI cartel so people like me with an AMD card max out at 4K60 even for open source games like Visual Pinball (unless you count an adapter with hacked firmware between the card and the display). NVidia and Intel get away with it because they implement the functionality in their closed source blobs.
This is kind of a niche problem. It only affects people with AMD GPUs running games at over 4k60 with HDMI. Get an NVidia or stay at 60 FPS or stay at 1080p or use DisplayPort and you will be fine.
It is not really a roadblock, more like a bump, and it is not the only bump by far. Some games just don't run on Linux, or quite terribly and they don't have a big enough community for people to care. Sometimes one of your pieces of hardware, maybe an exotic controller, doesn't like linux. Sometimes it is not the fault of the game at all, but you want to do something else with that PC and it isn't supported on Linux, and you don't want to dual boot. Overall, you will have less problems with gaming on Windows, especially if you don't really enjoy a trip to stackoverflow and the command line, but except for anti-cheat maybe, there is no "big" reasons, just a lot of small ones.
And sure, it is improving.
This is the first I learned of this since personally I have no need of anything over 4k@60 (that already borders on absurd in my mind). I'm curious if this is something that's likely to get reverse engineered by the community at large?
Outrageous that a ubiquitous connection protocol is allowed to be encumbered in this way.
Is HDMI really a roadblock to gaming when DisplayPort exists?
13 replies →
Does AMD not support Display Port? I'm not an expert on this, but that sounds to me like the superior technology.
16 replies →
I don’t understand why they can’t support AMDPort 2.1 which coincidentally has the same connector and protocol.
Competent cheat makers don't have much difficulty in defeating in-kernel anticheats on Windows. With the amount of insight and control available on Linux anticheat makers stand little chance.
The best Valve could do is offer a special locked down kernel with perhaps some anticheat capabilities and lock down the hardware with attestation. If they offer the sources and do verified builds it might even be accepted by some.
Doubt it would be popular or even successful on non-Valve machines. But I'm not an online gamer and couldn't care less about anticheats.
Anticheat is one of those things where I probably sound really old, but man it’s just a game. If you hate cheating, don’t play on pub servers with randoms or find a group of people you can play with, like how real life works.
For competitive gaming, I think attested hardware & software actually is the right way to go. Don’t force kernel-level malware on everyone.
29 replies →
Not a gamer, but it seems like super competitive games should be played on locked down consoles not custom-built PCs where the players have full control?
Also, for more casual play, don't players have rankings so that you play with others about your level? Cheaters would alll end up just playing with other cheaters in that case, wouldn't they?
2 replies →
Yeah this is also the model Microsoft is moving to. A separate attested vm for games, immutable to the rest of windows.
> The best Valve could do is offer a special locked down kernel with perhaps some anticheat capabilities and lock down the hardware with attestation.
That would require essentially turning it into a console or Android.
1 reply →
This seems both semi probably but also like maybe a bit of a critical moral hazard for Valve. Right now folks love Valve. They do good things for Linux.
Making a Valve-only Linux solution would take a lot of the joy of this moment away for many. But it would also help Valve significantly. It's very uncomfortable to consider, imo.
You don't have to play these specific games though. I mean, what's your privacy, what's not being bombarded by ads in your OS worth to you? Have you taken an honest thought about this?
If you want to play games with friends, you have to play whatever the group plays. This is especially problematic as the group tries out new games, increasing the chance you can’t join because you’re not on Windows.
11 replies →
Yes, but sometimes it is nice to socialize with other people and they might play these types of games. I don’t enjoy Call of Duty, but I’ll play it from time to time so I can chat with my brother (this is the only way to get him on the phone/microphone for some reason). I value the time I am spending with him more than a bit of privacy (in that context).
I am very pro-Linux and pro-privacy, and hope that the situation improves so I don’t have to continue to compromise.
besides ads and privacy concerns it's been such a delight not having to deal with unwanted updates, hunting phantom processes that take up cpu time, or the file explorer that takes forever to show ten files in the download folder. I cannot be paid to use windows at this point.
Isn't it a more fundamental problem? I can imagine a cheating setup where you have a separate PC with a HDMI capture stick ("analog hole") and access to the controllers.
The Linux kernel has eBPF now so if they wanted to start spying on everything you do they can just do it.
> The Linux kernel has eBPF now so if they wanted to start spying on everything you do they can just do it.
Sure, except that anyone can just compile a Linux kernel that doesn't allow that.
Anti-cheat systems on Windows work because Windows is hard(er) to tamper with.
38 replies →
That would require that they actually make the effort to develop Linux support. The current "it just works" reality is that the games developers don't need to support running on Linux.
I always wondered. Isn't exactly what eBPF would allow you to do?
Assuming that cheats work by reading (and modifying) the memory of the game process you can you can attach a kprobe to the sys_ptrace system call. Every time any process uses it, your eBPF program triggers. You can then capture the PID and UID of the requester and compare it against a whitelist (eg only the game engine can mess with the memory of that process). If the requester is unauthorized, the eBPF program can even override the return value to deny access before the kernel finishes the request.
Of course there are other attack vectors (like spoofing PID/process name), but eBPF covers them also.
All of this to say that Linux already has sane primitives to allow that, but that, as long as devs don't prioritize Linux, we won't see this happening.
> your eBPF program triggers
but how does the anti-cheat know that the kernel is not modified such that it disables certain eBPF programs (or misreports cheats/spoofs data etc)?
This is the problem with anti-cheat in general (and the same exists with DRM) - the machine is (supposedly) under the user's total control and therefore, unless your anti-cheat is running at the lowest level, outside of the control of the user's tampering, it is not trustworthy. This leads to TPM requirements and other anti-user measures that are dressed as pro-user in windows.
There's no such thing in linux, which makes it inoperable as one of these anti-cheat platforms imho.
4 replies →
And native GNU/Linux games instead of depending on Windows.
I am wondering can game be shipped with their own "kernel" and "hypervisor", basically an entire VM. Yes performance will take a hit, but in my experience with my own VM, it's like 15-20%.
Yes, maybe.
Modern games already employ a bunch of VM-like techniques for tamper protection.
This has effectively killed PC game piracy.
Do you pass through the GPU? Or how does it work?
Another unresolved roadblock is Nvidia cards seriously underperforming in DX12 games under Proton compared to Windows. Implementing DX12 semantics on top of Vulkan runs into some nasty performance cliffs on their hardware, so Khronos is working on amending the Vulkan spec to smooth that over.
That's being addressed:
The problem is on multiple levels, so everything has to work in conjunction to be fixed properly.
What percentage of games require DX12? From what I recall, a surprisingly large percentage of games support DX11, including Arc Raiders, BF6 and Helldivers 2, just to name a few popular titles.
At the same time, Vulkan support is also getting pretty widespread, I think notably idTech games prefer Vulkan as the API.
5 replies →
Clearly, when there will be enough Linux gamers another solution to the kernel-level anti-cheat issue will be found. After all, the most played competitive shooter is CS and Valve has does not use kernel-level AC.
> After all, the most played competitive shooter is CS and Valve has does not use kernel-level AC.
Valve doesn't employ kernel AC but in practice others have taken that into their own hands - the prevalence of cheating on the official CS servers has driven the adoption of third-party matchmaking providers like FACEIT, which layer their own kernel AC on top of the game. The bulk of casual play happens on the former, but serious competitive play mostly happens on the latter.
2 replies →
The competitive CS leagues do use AC though. The big issue for these games is the free-to-play model does not work without anti-cheat. Having a ~$20 fee to cheat for a while before getting banned significantly reduces the number of cheaters, and that's what CS does with their prime server model.
And for what it's worth, I'm pretty sure Valorant is the most played competitive shooter at the moment.
Isn't it pretty much an open secret that JVM-based cheats can trivially bypass VAC?
How does their revenue rely on it? People won't buy/recommend their games if they can't solve a fundamental problem, without full control over the machine their product is running on? Then they can change their business model and/or game mechanics. Simple as that. The only reason that blatant security violation was ever considered a viable option is because Microsoft gave them the ability to actually do it with the click of a button. Those companies can adapt, or die.
Games being playable also rely on it.
Well, if you go by revenue, mobile gaming dwarfs all else.
But is that really a roadblock?
First, let's ask ourselves how many PCs have users play games with anti-cheat frameworks. I'm absolutely no expert, but if it's more than, what? 10%? let's even say 20% - I'd be surprised.
> and unfortunately a good majority of the gaming industry by revenue relies on it.
Well, it used to be the case that game makers relied on copy protection in floppy discs, and movie distributors on DVD/BluRay copy protection. Conditions changed and they adapted.
I actually think it’s better to exclude the AAA games from Linux.
This is a big reason I’m excited for Steam Frame - high quality VR on the Linux desktop.
AND high quality Linux desktop on the VR :)
The Quest 3 works offline with ALVR streaming over a private (non-Internet connected) WiFi network. Together with my 3090 I get 8k @ 120fps with 20ms latency over a WiFi6e dongle. I had to manually install the dkms for the dongle on PopOs, but apart from that it just works. ALVR starts SteamVR and then I use Steam to start the game. Proton seems to use Vulcan for rendering.
Overall, I had a pretty bad experience with ALVR. I never managed to figure out the cause of stuttering on mine. I wished Meta would support Linux.
> I just hope that Nvidia notices that there does appear to be a swing happening and improves their driver situation.
I firmly believe that Nvidia doesn't want the general public to ever have better hardware than what is current as people could just run their own local models and take away from the ridiculous money they're making from data centers.
In step they're now renting their gaming GPUs to players with their GeForce now package.
The market share for Nvidia of gamers is a rounding error now against ai datacenter orders. I won't hold my breath about them revisiting their established drivers for Linux.
> I firmly believe that Nvidia doesn't want the general public to ever have better hardware than what is current as people could just run their own local models and take away from the ridiculous money they're making from data centers.
You're underestimating them. They don't even want rich professional users to own hardware that could compete with their datacenter cash cow.
Take RTX 6000 Pro, a $10k USD GPU. They say in their marketing materials that these have fifth-generation tensor cores. This is a lie, as you can't really use any 5th-gen specific features.
Take a look at their PTX docs[1]. The RTX 6000 Pro is sm_120 in that table, while their datacenter GPUs are sm_100/sm110. See the 'tcgen05' instructions in the table? It's called 'tcgen05' because it stands for "Tensor Core GEN 05". And they're all unsupported on sm_120.
[1] - https://docs.nvidia.com/cuda/parallel-thread-execution/#rele...
I’ll keep repeating it: the more people vote with their wallet, the more game companies will deploy Linux - including the anticheat.
EAC has the support for Linux, you just have to enable it as a developer.
I know this, I worked on games that used this. EAC was used on Stadia (which was a debian box) for the division, because the server had to detect that EAC was actually running on the client.
I feel like I bring this up all the time here but people don’t believe me for some reason.
> EAC has the support for Linux
This does not mean it supports the full feature set as from EAC on Windows. As an analogy, it's like saying Microsoft Excel supports iPad. It's true, but without VBA support, there's not going to be many serious attempts to port more complicated spreadsheets to iPad.
I'm surprised to hear you are having trouble with wheels / pedals, we should be there already!
https://github.com/JacKeTUs/linux-steering-wheels
Hopefully vr headset support will get better
Funnily enough the most annoying things on my system at the moment is RGB and keyboard/mouse customisation.
I haven’t found a tool that can access all the extra settings of my Logitech mouse, not my Logitech speakers.
OpenRGB is amazing but I’m stuck on a version that constantly crashes; this should be fixed in the recent versions but nixpkgs doesn’t seem to have it (last I checked).
On the other hand I did manage to get SteamVR somewhat working with ALVR on the Quest 3, but performance wasn’t great or consistent at all from what I remember (RTX 3070, Wayland KDE).
Have you tried running the windows RGB utility via Wine with HIDRAW enabled for the device?
Alternatively, given you’re running NixOS you can just override the `src` of the derivation with a newer version. This is part of the point of running NixOS: making small modifications to packages in the fly.
I was annoyed recently because I replaced my GPU and I had to boot into Windows for the first time in months and install drivers just to turn off the RGB on the card because OpenRGB wouldn't find it.
For VR support Monado works very well for me with both Pimax (base-station tracked) and WMR (inside-out tracked) headsets.
When that steam deck clone came out and games played better on SteamOS than on Windows on the exact same hardware, it woke a bunch of people up. Microsoft scrambled to bring the startup time and footprint down but shots had already been fired.
You don’t want a vendor you have to publically shame to get them to do the right thing. And that’s MS if any single sentence has ever described them without using curse words.
I've got the Legion Go S with Steam OS, and that shit is great. It's stable, my games run well, the OS is pretty much entirely in the background, but I can still access it fully if I need to. Love it.
Gaming now works better on Linux than it does on Windows. This must be upsetting for Microsoft, but it was their game to lose.
I dont get the feeling they care. Microsoft is so lost under Satya at this point. Totally blinded by Azure and AI and stock price growth. At some point they're going to realize all the ground they've lost and it's going to be a real problem. They're repeating a lot of the same mistakes that cost them the browser and mobile market.
Yeah. MS must have been so hurt about losing to the iPhone, they really jumped the gun on AI as if to avoid a similar mistake. It's Satya's major play and I think they are already paying for that decision. xbox is hollowed out so that AI can be funded, while the pc/console hybrid project is doomed to fail because "windows everywhere" doesn't work if windows is crap. indeed, they might be left with just the cloud business in the end.
2 replies →
"Totally blinded by Azure and AI and stock price growth."
Stock price growth is their core business because that is how large firms operate.
MS used to embrace games etc because the whole point was all PCs should run Windows. Now the plan is to get you onto a subscription to their cloud. The PC bit is largely immaterial in that model. Enterprises get the rather horrible Intune bollocks to play with but the goal is to lock everyone into subs.
1 reply →
They don’t care, they’re defunding Xbox and even the Windows team is hollowed out.
2 replies →
If people were buying new PCs every year like they used to I'd be worth it. Turns out there isn't as much value having a "captive market" on a PC unless it's locked down.
The irony is that gaming on linux got better but the instigator was not the OSS community. All of it was funded by closed source software competing with other close source software. The OSS community by itself did not have the conviction to climb over this bulwark.
It's not so much conviction, as it is coordination and resources.
Conventional companies just have a lot more money, and it's easier for them to internally 'coordinate' when they want something to get done.
That said, yes, there are certain things that the broader/volunteer FOSS community simply isn't any good at.
But when Steam started to develop Proton, WINE was 90% there! Valve only had to provide the remaining 90%.
The strength of Linux and Free software in general is not in that it's completely built by unpaid labor. It's built by a lot of paid, full-time labor. But the results are shared with everyone. The strength of Free software is that it fosters and enforces cooperation of all interested parties, and provides a guarantee that defection is an unprofitable move.
This is one of the reasons you see Linux everywhere, and *BSD, rarely.
3 replies →
It isn’t about conviction. Gaming takes tremendous resources and they were not there. But if this starts shifting the tides there is a possible future where game developers start building for Linux as a primary target and to run games on Windows or Mac you would use emulation. In fact this seems like a better overall approach given that there are no hidden APIs with Linux.
Money and resources suddenly materialized once someone realized that there was profit in it is pretty much the expected way this goes. OpenTofu happened not because of some OSS force of will but because a group of companies needed it to exist for their business.
This flow is basically the bread and butter for the OSS community and the only way high effort projects get done.
This still has a "sometimes" on it, there are more then a few games that need magic proton flags to run well, nothing you can't go look up on protondb, but lots of games you would want to play with friends might have some nasty anti-cheat on it that just won't let you play it at all.
Exactly. Battlefield 6 for example does not work at all in Proton.
This is a far better user experience for Battlefield players than in Windows.
Have you ever actually attempted to play that half-assed buggy piece of shit?
As long as Valve depends on the Windows ecosystem for content, they are quite safe.
Game studios will keep buying Windows and Visual Studio licenses, target DirectX, and let Valve do whatever they need for game content.
Gaming works fine with exception of things like BF6 that require kernel level anti cheat.
The one thing I haven’t been able to get working reliably is steam remote play with the Linux machine as host. Most games work fine, others will only capture black screens.
if you are running KDE you can whitelist Steam for remote desktop work, this is because of wayland.
Proton has gotten so good now that I don't even bother checking compatibility before buying games.
Granted, I don't play online games, so that might change things, but for years I used to have to make a concession that "yeah Windows is better for games...", but in the last couple years that simply has not been true. Games seem to run better on Linux than Windows, and I don't have to deal with a bunch of Microsoft advertising bullshit.
Hell, even the Microsoft Xbox One controllers work perfectly fine with xpad and the SteamOS/tenfoot interface recognizes it as an Xbox pad immediately, and this is with the official Microsoft Xbox dongle.
At this point, the only valid excuses to stay on Windows, in my opinion, are online games and Microsoft Office. I don't use Office since I've been on Unixey things so long that I've more or less just gotten used to its options, but I've been wholly unable to convince my parents to change.
I love my parents, but sometimes I want to kick their ass, because they can be a bit stuck in their ways; I am the one who is expected to fix their computer every time Windows decides to brick their computer, and they act like it's weird for me to ask them to install Linux. If I'm the one who has to perform unpaid maintenance on this I don't think it's weird for me to try and get them to use an operating system that has diagnostic tools that actually work.
As far as I can tell, the diagnostic and repair tools in Windows have never worked for any human in history, and they certainly have never worked for me. I don't see why anyone puts up with it when macOS and Linux have had tools that actually work for a very long time.
> At this point, the only valid excuses to stay on Windows, in my opinion
I didn’t see a performance increase moving to Linux for the vast majority of titles tested. Certainly not enough to outweigh the fact that I want EVERY game to work out of the box, and to never have to think if it will or won’t. And not all of my games did, and a not insignificant number needed serious tweaking to get working right.
I troubleshoot Linux issues all day long, I’ve zero interest in ever having to do it in my recreation time.
That’s a good enough reason for me to keep my windows box around.
I use Linux and OSX for everything that isn’t games, but windows functions just fine for me as a dumb console and I don’t seem to suffer any of these extreme and constant issues HN users seem to have with it from either a performance or reliability standpoint.
My VR glasses work on Omarchy, to my surprise, I plugged them and they work. I have XReal, older model.
Aren't the XReals just displays in the glasses? If they work with other devices, it's no surprise linux can just use a display standard.
But they work out of the box, which is my point. You can use a device that can be inbetween which places screen into fixed space in front of you for example. While it is cool, it is kind of a hassle to have this device inbetween. I just plug them directly and they work.