From latest Bazzite release

ujust check-anticheat

Epic has announced that it’s bringing Fortnite to Linux and rebuilding its anti-cheat as Easy Anti-Cheat Linux v2 with Fortnite as the first pilot before it opens up to other developers.

EAC has supported Linux for years, but many games never opted in because the Linux version couldn’t offer the same protection as the Windows one. To close that gap, v2 will require a set of security features Linux already has: a signed kernel and modules, TPM 2.0, Secure Boot and measured boot, IOMMU, kernel lockdown in integrity mode, kexec disabled, and BPF LSM.

Here is the kicker:

They were kind enough to make us one of the accepted well-known signing keys.

So its a whitelist system for kernel level anticheat.

you are viewing a single comment's thread
view the rest of the comments
[–] 92 points 1 day ago* (last edited 18 hours ago) (24 children)

I don't want my OS designed with anti-cheat in mind. Anti-cheat is an endless arms-race that sacrifices control of our computer to anti-cheat creators so they can fight on our land, instead of on their own servers. If Microslop adds 'live camera feeds to AI' or creates a 'DRM network router system' will should Linux try to emulate those? Is this "taking anti-cheat seriously"? Supporting more Digital Restrictions Management is a step in the wrong direction - we need to support video games that respect users' software freedom.

(Edit) It's not Linux doing this, it's Epic.

  • source
  • hideshow 24 child comments
  • [–] 15 points 1 day ago (8 children)

    I agree. Incompetent game developers compensate with stuff installed on my machine running in the kernel their inability to detect that there is a player in their game in their server capable of popping heads through walls.

    The problem is also on the customer side, this is the result of democratizing linux: when only the tech nerds used linux such stuff was taboo and unthinkable, but now an entire generation of kids is fine with paying money to install crap in their kernel to do this.

  • source
  • parent
  • hideshow 8 child comments
  • [–] -2 points 1 day ago (7 children)

    That's not really what client side Anti-Cheat is for. You can't tell if a player can see through walls just from server side information.

    It is a result of democratising Linux but I'd go further: when the only multiplayer games you could play on Linux were ut2k4 and Wolfenstein: enemy territory, this was less of an issue.

  • source
  • parent
  • hideshow 7 child comments
  • [–] 5 points 9 hours ago (4 children)

    You can avoid sending to the client information about players they are not supposed to see.

    Once you treat all your clients as adversaries that is the pattern of thought a good developer should have.

  • source
  • parent
  • hideshow 4 child comments
  • [–] 1 point 9 hours ago (3 children)

    I don't think it is feasible to be doing that on the server, unless you want games to cost even more. But also it's not that simple: suppose you don't want the client to be able to draw enemies through walls. But you do want the client to be able to play their footsteps through walls. So you need to send enough information for that - and now it's a matter of doing a small amount of computation to work out where an enemy is from the sounds being played, which can be much more effective than a human player's ears and brain.

    I guess actually a simpler example is rendering enemies through smoke or fog - the game relies on them being hard but not impossible to see, and so the data needs to be there for a hack to potentially exploit.

  • source
  • parent
  • hideshow 3 child comments
  • [–] 1 point 7 hours ago (2 children)

    Look into how CS2:GO does it, valve does not rely on kernel anticeat, there is a good video about it on youtube.

    The case with the smoke you are describing indeed sounds like a challenge I have no immediate solution for, but even in that case you can statistically analyze what is a normal kill rate for a given skill level and suspicious rates can get special scrutiny.

  • source
  • parent
  • hideshow 2 child comments
  • [–] 2 points 7 hours ago (1 child)

    I don't think "a given skill level" makes sense because the point is that a hack should make you look higher-skilled than you really are, at least in one specific skill (like spotting enemies through smoke). While there are often obvious cases (the one time I saw someone hacking in a video game, everyone else on my team was dead so I was spectating them, and they happened to be AFK... yet as an enemy came into view, their character started turning to keep them exactly centred!) good hackers can just be subtle and present themselves as being at the absolute peak of human ability yet still within feasible limits. That's still unfair!

    I should say now we're however many comments deep (yeah, probably should have made it clear earlier) that I don't actually care for kernel-level anti-cheat myself. I think it's invasive and I'd rather not have it. BUT I think loads of people here just don't get why it exists, and there IS a valid reason for it.

  • source
  • parent
  • hideshow 1 child comment
  • [–] 2 points 4 hours ago

    Well, if the discussion is interessant I don't care about comment depth ;)

    My point with statistics is that either something looks like a real player or it does not. Even if a cheater looks like a better player but still human, i'd argue that it is lower priority as it does not ruin the game for others.

    In your example the ratio between time passed not exactly pointed on head and pointed wherever else could be measurable and everyone outside real human ranges could be flagged.

    I agree with your point on kernel anticheat, my argument was never that it was useless, just that there are ingame ways on the server to catch most stuff, without running software as root on my machine.

  • source
  • parent
  • [–] 7 points 1 day ago* (8 children)

    Isn't offering the option sort of the entire point of linux? If you don't want the feature then don't enable it or use a non-EAC kernel.

  • source
  • parent
  • hideshow 8 child comments
  • [–] 7 points 19 hours ago (7 children)

    Did you hear about the game Doom Eternal and the backlash when they included Denuvo DRM with it? The DRM was removed shortly after, to the benefit of all users. If there were people who wanted "the option" of having Denuvo on their computer to play the game then they would be people who are very confused about their own best interests.

    Does an EAC kernel do more than restrict what the user can do under the guise of preventing cheating? That is exactly not a user feature, it is an anti-feature. I can't play a game I paid for called Escape from Tarkov on Linux because there are no (legal) servers without BattlEye anti-cheat. I've never looked into customizing my kernel but perhaps there are real features to be found there.

  • source
  • parent
  • hideshow 7 child comments
  • [–] 2 points 17 hours ago (1 child)

    Yes, ideally there would be some other way of handling anti-cheat. But you should have the choice if you want to allow some company's anti-cheat in exchange for playing their game. Nobody is forcing anyone to enable it, but the choice is there. At least it can steal more customers from Microslop.

  • source
  • parent
  • hideshow 1 child comment
  • [–] 2 points 16 hours ago

    Likewise, I have the choice to advocate against people choosing it.

    If I buy a copyrighted book then I can't legally redistribute copies but it's still my book. I can rip pages out or edit it and if a company could take away my book because they don't like my edits that would be a clear injustice. Likewise we aught to consider our purchased games running on our computers to be "our games" to make company's EULA over-reach clearly visible.

    We probably have most games working on Linux by now. Maybe bending over for a few more popular games isn't worth our time - instead we could promote open source alternatives.

  • source
  • parent
  • [–] 1 point 14 hours ago (4 children)

    Does an EAC kernel do more than restrict what the user can do under the guise of preventing cheating?

    Other than give them the option to play the game? Does this really need an explanation?

  • source
  • parent
  • hideshow 4 child comments
  • [–] 1 point 9 hours ago (3 children)

    If a shopkeeper require customers to carry a large bag of potatoes to be permitted entry it's not the potatoes giving you the option to enter the shop.

  • source
  • parent
  • hideshow 3 child comments
  • [–] 0 points 7 hours ago (2 children)

    In this scenario Cachy would just be the guy handing out potatoes outside the store.

  • source
  • parent
  • hideshow 2 child comments
  • [–] 6 points 1 day ago (4 children)

    will Linux try to emulate those?

    To be clear, "Linux" isn't doing any of this, this is entirely Epic's EAC.

    But as far as having an "OS designed with anti-cheat in mind", it depends on what the device is for. Is it my PC that has access to my personal data? No, I want control over my PC, and I don't want random programs having kernel-level visibility. Is it a game console hooked to my TV that doesn't have access to my data, and is used to play games that are only fun with effective anti-cheat? Sure, I'd like the choice to allow that.

    But I do fear a future where MSFT coerces applications to use some form of hardware attestation in order to even run, or Epic pushing EAC as an anti-cheat standard on Linux, and now suddenly I have to be running a specific kernel and giving more access than I'd like in order to just run a random steam game.

  • source
  • parent
  • hideshow 4 child comments
  • [–] 1 point 18 hours ago* (last edited 18 hours ago)

    You're right, Linux isn't doing anything of this. I've tried to edit my comment to not give that impression.

    It's wise to seperate trusted computing and untrusted gaming but yeah we shouldn't need to do that. I know there are selfhosted Valve game servers without anti-cheat but I assume most games with anti-cheat do not give users that choice.

    I have not liked game consoles since they stopped being easier than PCs in terms of plug-in-and-play. Modern consoles are basically a custom PCs; requiring updates to locked-down software/hardware. It would be better if users had more control over them (and the "smart" TVs people use with the consoles which is creating and sharing data about them).

  • source
  • parent
  • [–] 0 points 3 hours ago (1 child)

    No, I want control over my PC

    You get it back when you reboot to your unsigned kernel

    and I don’t want random programs having kernel-level visibility

    You don't have to give it, you just have to boot into a signed kernel with signed modules while playing EAC games. You don't need to run a kernel driver for EAC on Linux.

    It honestly seems to be less invasive than the Windows version in the format they've currently announced.

  • source
  • parent
  • hideshow 1 child comment