Latest 30 Comments

The rocks shall be stoned!

View this comment - View article - View full comments - Report this comment

@Eike, because than the other gamers will come and visit your house when you cheat.
I prefer Kernel anti-cheat personally.
Also permabans can be more perma.

View this comment - View article - View full comments - Report this comment

How sim-heavy are these games ? I'm highly interested in a good Rally game, as long as it's fun with a controller and doesn't require the steering wheel etc.

Open to other Rally game recommendations as well

View this comment - View article - View full comments - Report this comment

For Linux / SteamOS fans, going by ProtonDB the early reports there are nice and positive too with multiple reports on it working great out of the box with Proton.

For how long?

View this comment - View article - View full comments - Report this comment

@Purple Library Guy
On an AMD machine this actually makes sense.
See the primary problem with your solution is that it's still hard to confirm that you run what you think you're running, but AMD's competitor to SGX actually works with mini kernels in vms.
The real question is. Wouldn't it be easier and just as good/bad to dual boot? Maye its cheaper, Windows is expensive.

Edit:
Correction it would be easier on an AMD machine this feature I just referenced does require exactly 0 reboots.

View this comment - View article - View full comments - Report this comment

Say, by modern standards of file sizes, the kernel isn't actually very big, is it? So OK, this is kind of ludicrous, but . . . imagine that the Linux version of Easy Anti-Cheat or whatever DOWNLOADED A CUSTOM KERNEL every time you started logging into the game, and you played the game (and only the game) on that, probably in a sandbox of some sort, and it would get deleted after your session was over. And the custom kernel would be constantly changing so they could tell whether you were using the latest one (and then, yeah, EAC's servers got hacked and malware got put in, but that was only that one time ).

I mean, clearly it would turn the "rootkit" problem up to the max, but from the game developers' perspective it would be the most trustable anti-cheat in town. And you'd have to wait for the download every damn time you wanted to play the game, but . . . if the kernel is pretty small, it wouldn't be that bad.

View this comment - View article - View full comments - Report this comment

Im willing to be infinitely more patient with indie game jam title that hits it big and suddenly has more exposure than they were expecting than I am with a AAA title that has spend hundreds of millions of dollars generating buggy garbage.

View this comment - View article - View full comments - Report this comment

@Uso, @Shotm7 and many others who suggest custom kernels as the solution.
Custom kernels are the problem, because people can legally and practically make their own kernel, make it return whatever they want and install it on their own device.
Anti-cheat makers, which are people who are defending against the owner of the device their software is running on don't trust the kernel to help defend against its master and on Windows Microsoft at least attempts to claim that title on Linux a principled stance has been taken to not claim that title.
Yes even signed kernels are treated as suspicious UEFI level cheating already exists.

Feature wise Linux wins all the rounds in anti-cheat.
Acceptable KASLR, tainted kernels, great hardware security modules integration, etc.
This just doesn't matter, because one could've and probably someone has made a kernel that showed all the signs of having these features, but didn't actually help.


If you ask me the technical solution isn't in the Kernel at all.
It's in the things the anti-cheat developer does trust: the development tools and the trusted execution environments.
To express this belief I hereby release [EASLR(Executable Adress Space Layout Randomization) under the gplv2.1](https://codeberg.org/Informeli/EASLR).


Edit:
Yes, I value this freedom flowing from this principled stance too and the lack of monopolistic power originating from it.
As such EASLR doesn't stop anyone from modifying anything on their computer.
It hampers undetectable mass distribution of this capability on the specific programs we're trying to protect.

View this comment - View article - View full comments - Report this comment

It might be worth grabbing this just to run the in-game benchmark?

View this comment - View article - View full comments - Report this comment

This mod is really the best for Heroes 2. It brings to the game every little quality of life improvment brought by Heroes 3 and the community patches. The perfect way to play Heroes 2 for me, and it's one of the great episode of the serie (with great soundtrack).

View this comment - View article - View full comments - Report this comment

Yall really think 1080p is the common res for living room tvs? Here in the US, I don't think I know anyone with a 1080p tv. Can you even still buy them? I mean like if I go to costco, every single set is 4k. And you can get the low end ones for around $200. I would wager highly that the common tv res in the states is 4k.

I don't know, but... Don't mistake what's being sold for what people have at home. Most, by very very far the most, people didn't buy their TV set this year. Or the last.

My own TV is 768p. 1360×768. So make of that whatcha will, gang

View this comment - View article - View full comments - Report this comment

I'm making a note here.

View this comment - View article - View full comments - Report this comment

What is wrong with me?
Why am I not excited anymore?
When did I lost my passion for gaming?

What has worked for me when I get in this very exactly the same situation (where no games "click" with me and I just wind up putting them down after a tiny bit over and over again) ... I try a different medium. I will pick back up any one of the bazillion scifi books I have on my to-read list, go through 2 or 3 of those and BAM I've got the bigtime itch for a good story heavy turn based rpg. Or start watching a series with my wife (I almost never watch tv normally). Stuff like that. Small breaks always work wonders for me.

View this comment - View article - View full comments - Report this comment

IMO cheating in casual gaming is more of a social problem and requires social solutions. An obvious solution is to stop making free-to-play games. Another one is to link accounts with real identities using eIDs or such − I'd personally be OK with that if done properly, your opinion may vary.

How would any of this solve the cheating problem, without technical anti-cheat?

View this comment - View article - View full comments - Report this comment

One question, could we not just sign the linux kernel and other parts, so that the anti cheat program would know that nothing has been altered? I am thinking about something like steam hosting a list of trusted hashes and that any company that is trusted by steam can push new hashes for theier compiled linux kernels there.

What you're describing is measured boot + remote attestation (typically implemented using Secure Boot and a TPM). It is indeed the technical solution to ensure that a remote peer only runs approved software. I'd agree to use such a solution for official competitions with a cash prize but not for casual gaming where I want to keep full control on my system.

IMO cheating in casual gaming is more of a social problem and requires social solutions. An obvious solution is to stop making free-to-play games. Another one is to link accounts with real identities using eIDs or such − I'd personally be OK with that if done properly, your opinion may vary.

View this comment - View article - View full comments - Report this comment

I see so many people saying it's not an issue because they don't play these types of games, or anticheat is bad/unescure etc. Or even that kernel level anticheat is not effective.

And I do agree that (kernel level) anticheat is... not great, and obviously has the potential to be a really dangerous backdoor. But the truth is that many, many players DO enjoy these types of games - I personally love battlefield and still play BF4 frequently; I used to enjoy COD and played WWII up until it was no longer safe to play and I spent years waiting for MW2019 to work on linux before it was obvious that would 'never' be possible.

I love Counter Strike, and I'm very glad it has had a native client for as long as I can remember - but the fact is that it does have a large cheater problem. CS2 in particular had the top ranks of premier filled with cheaters (whose name advertised their cheat configs) and I have personally seen several cheaters throughout my gameplay. Even 'ragehackers' were fairly common, though it has gotten a little better in the last few months. Cheaters stay cheating unbanned for months sometimes. I use leetify to track when cheaters are banned.

Valorant? barely has any cheaters. Players are notified that the odd cheater they do find is banned the same game they play against them. Their anticheat clearly does work, and unfortunately that is a fact. FaceIt for CS2 is pretty similarly cheater-free from my understanding, and that is Windows only due to anticheat reasons.

For highly competitive games, it can be a necessary evil - MW2019/Warzone players begged for a kernel level anticheat in the hopes it would even alleviate their issues.

What I'm trying to get at too is that this is not just a problem for the people that play those games - Linux just won't be able to grow as much as it should if there is not some way or another to get round these issues. It is a problem that will honestly effect every linux user, not just competitive FPS/MOBA players. Serverside anticheat is not as easy as it sounds, and it's not just because of server costs that companies don't rely on it. I hope it will one day solve the issue though.

View this comment - View article - View full comments - Report this comment

I don't understand why an anti-cheat driver inside the kernel would not work.
I mean Red Hat develop tainted kernel a long time ago: https://docs.kernel.org/admin-guide/tainted-kernels.html
and a big part of anti-cheat is just improvement over that.
Of course, you would need signed kernel, which would be a deal-breaker for a lot of user.

That would also make anti-cheat unusable with proprietary drivers (so no Nvidia GPU), and would mean that a big part of the anti-cheat would be Free software, and I'm sure peoples working on anti-cheat have no idea how to contribute code.

But if some company starts doing that, I guess this is doable.
Also, kernel-cheat might be a problem today, but kernel anti-cheat can't do anything against hardware cheat, so at some point most anti-cheat will move from kernel-side to AI side. (no sure if it is good news though)

View this comment - View article - View full comments - Report this comment

Personally, I think running an operating system that can't be easily exploited the way anti-cheat developers want is a good thing -- improved security is one of the reasons I use Linux in the first place -- but then I have almost no interest in multiplayer games, so this issue doesn't really impact me.

----------

@ElectricPrism

If your game doesn't FULLY work on the new 2026 STEAM MACHINE, you are ineligible to be in the FEATURED GAMES BILLBOARD on the HOMEPAGE.

Great solution... assuming your goal is get Valve sued for monopolistic practices.

View this comment - View article - View full comments - Report this comment

That FacePunch Rust Reddit post is pure b*llsh*t. Cherry-picked marketshare usage figures - They took percent of Linux users *after* killing off the Rust Linux client. There were more than 0.01% Linux before that. The 'Developer' person spewing the Reddit figures is probably in the PR Dept, regurgitating Garry Newmann's age-old drivel.

From my own viewing of Rust gameplay videos, it is clear to me that the game suffers loads of cheaters, entirely from the Windows side of the game, because they can't play (on those servers) without Anti-Cheat in place. My impression is that even if the (estimated) 1% marketshare of Linux players was playing Rust, that number of players in the system would still be dwarfed by the total Windows cheaters.

Garry - It's only an attack vector if it seriously dents the marketshare of Windows cheaters (which it clearly doesn't, especially if we go with your 0.01% Linux marketshare).

View this comment - View article - View full comments - Report this comment

Fixed short freezes happening every 5 min in Deadlock.

That sounds really anti-thematic.

View this comment - View article - View full comments - Report this comment

I still vastly prefer FoundryVTT, but that's me :P

View this comment - View article - View full comments - Report this comment

Kernel updates come at least on a weekly basis ...

yes, this is why I was thinking that trusted companies could push the hashes to a server. This is something that could be easily automatized.

View this comment - View article - View full comments - Report this comment

@logge:
At first I didn't even understand how this could work, but they have a good documentation and are very helpful on the LVRA discord. Once you see it's like a client-server-like thing, it becomes easy. Yet, setting up Envision can be a bit tricky. You just have to try and learn while testing around. It's a bit of a learning curve, eventually once it's setup is done, it just works.
Unfortunately not all games work with Monado (for OpenXR) and/or OpenComposite/XRizer (for OpenVR), but here you can check the protondb-like vronlinux database. It has entries per game for Steam VR, Monado, WiVRn and ALVR.
I do see though that Envision is not made for the average user, but which Linux user is this "average user" anyways, hm?

edit: Ah yeah, for me it doesn't matter if I start Envision or Steam first, at least I didn't run into any issues so far, and I have started it in arbitrary order many times. But before starting Monado via Envision, VR controllers and basestations have to be turned on. They are working on detecting them on the fly, but for now you have to start up things before the software itself.

View this comment - View article - View full comments - Report this comment

Kernel updates come at least on a weekly basis ...

View this comment - View article - View full comments - Report this comment

I played it on the Switch.
The game itself is pretty good. The requirement to be logged in with a Ubisoft account is horrible.

I cannot recommend the game therefore.

(On Switch, at least, you could work around the login-requirement by disconnecting from the internet. I don't know if that's an option on PC.)

View this comment - View article - View full comments - Report this comment

One question, could we not just sign the linux kernel and other parts, so that the anti cheat program would know that nothing has been altered? I am thinking about something like steam hosting a list of trusted hashes and that any company that is trusted by steam can push new hashes for theier compiled linux kernels there.

View this comment - View article - View full comments - Report this comment

WRC generations started working, and perfectly out of the box now at that! Hell yeah!! Too bad this happens one day after Assetto corsa rally came out but it's still a great game.

View this comment - View article - View full comments - Report this comment

I may seem like an amateur, and I am, but why not create a specific linux kernel for this kind of games ?
An other option is that when you enter the steam gamemode, and a game is launched, the steam environment with the game is sandboxed.
Third solution: reimplement user kick voting

These only try to fix the symptoms, not the problem. The problem isn't that they cannot keep all the cheaters out. Cheaters will always exist. The problem is that we cannot control the servers and kick them ourselves. And the solution to that is dedicated servers.

View this comment - View article - View full comments - Report this comment

Kernel Anti-Cheat is generally a sign of a game you'd be better off not buying/playing in the first place.
Linux users can't play Battlefield 6? Good for them!

The only difficulty is just that sense of exclusion, I think. "Everyone" plays yearly release shooters and the like.
It's like being excluded from TwitX. Yeah, it has a massive presence and you'll see people talking about a service you can't use. But you'll realistically avoid a lot of drama and misery that comes with using it.

Besides, remember why GOL doesn't promote games with AI content: There's an inexhaustible supply of excellent games without AI. The same goes for games without Kernel Anti-Cheat.

View this comment - View article - View full comments - Report this comment

I realize most Linux folks don't like the Ubi launcher

Not only Linux folks. Ubisoft Connect was the bane of my existence and one of the more annoying things I had to deal with when I was on Windows. It needed updates every goddamn time I wanted to play R6 Siege. And it was obnoxiously amensic, asking me for my credentials in what feels like every single time I have to launch it to play the game, the 'Remember me' button didn't work. Not to mention the extra 1-2 min I'd have to sit for it to initialize or steal my data or something before Siege properly launches. The overlay is also garbage as it's very slow and unresponsive at times.

Probably the shittiest launcher out there as I had better luck with the likes of EA launcher which was notorious itself but admittedly less so recently, as they must've gotten their shit together.

I haven't used the Ubisoft launcher in months, so all of what I said could very well be outdated by now.

View this comment - View article - View full comments - Report this comment