I've played before anticheat was a thing and it is meaningless. Cheaters are going to cheat. The best anticheat systems are voting. The game kicks a winning vote total and then that server sends the rest of the servers it's results. Then that account is flagged as a cheater. The only way a cheater can exist is they hide and don't cheat and are obvious in it
Such a system can be used against non-cheaters as well. How would you feel if you played fair and got permanently banned from playing a game you bought simply because of a few bastards?
Kick votes alone should never lead to bans or suspensions. Instead; this is when we should switch to server-sided analytics of gameplay and lean heavily on human moderator input to determine if a player might be cheating.
Anti-Cheat should be on server side anyways; not the client side.
I can't wait for the rise of motherboards with hyper-customized TPMs that can fake legitimate ones and totally gaslight the OS and any anti-cheat running that everything is alright. /s
Maybe a second step of reviewing but also if you bought a game permanent bans should never be a thing on less it has an offline mode, if you bought something you should permanently own it they shouldn't be able to take your access permanently
Votekick really doesn’t work as an anticheat, especially without good playback analysis option and even then good gamesense looks like wallhacking in shooters to new players for example when all you are doing is tracking sound and have good crosshair placement. If you can’t review replay and there is no blatant cheating like speedhack or spinbot or teleporting, what are you voting on? The fact that you are getting stomped?
To be clear I’m not saying invasive anti-cheat is the way, but IMO voting is not the way to go.
[The meme shows two fanart images of the character Sayori, from "Doki Doki Literature Club", with text to the right of each image.]
[In the first image, Sayori is wearing sunglasses and scowling, with her hand up in a blocking gesture. The text reads:]
Anti-Cheat
[In the second image, Sayori has her head up high, looking pleased, with a finger pointed to the right, where the text reads:]
Kernel Level Surveillance
I am a human who transcribes posts to improve accessibility on Lemmy. Transcriptions help people who use screen readers or other assistive technology to use the site. For more information, see here.
As long as there are people playing a game, there will be cheats. However, I decide what happens on my device, not a game or software developer. When the developer thinks he can set requirements, he is barred.
Not a single piece of software is worth risking my device for.
But that won't happen. It's against their very nature/purpose. Also kernel level anything is iffy. There is a reason modern OS are trying to push things into user space, like file system. Prevents an error from crashing the whole system.
My recommendation now that computing devices are reasonably cheap is to not do anything requiring a high level of trust on a daily-use device that runs a wide variety of software, such as a gaming PC with lots of games. Instead I'd do those things on a device that only runs what you trust and nothing else.
Anti-cheat isn't compatible by nature with Open-source. Client side anti-cheat is based on the premise of security through obscurity, it's not a canidate for open source and the ideologies behind most of them aren't really compatible with freedom either.
Also DRM isn't really compatible with that idea either because it is by nature anti-freedom.
I wanna know they have to have low level shit making these checks on my device in the first place. Why can't the checks by on the god damn server, checking against what the developer knows is and isn't possible to do without cheating?
Edit: Er... I guess you wouldn't really be able to tell if they used walls or aimbots that way... 🤔
A lot of anticheat methods are not to catch the people with the proper, premium cheating software. It's to stop little Jimmy downloading an exe for Fortnite because he loses too much and making a new account as f2p is now the norm. As such, a lot goes into making it hard to have a cheat hide itself without significant effort from the user, be that running a custom kernel module yourself or some sort of emulation techniques. The kernel level anticheats can naturally be bypassed, but you have to do more than just running an exe most of the time which is about as far as the average kid who downloads their cheats from a YouTube video is capable of. The result is you catch 99.9% of what would be cheaters, and that's a much bigger improvement to your player base than catching the 4 players at the pro level who pay thousands a month for custom software which doesn't really matter in the grand scheme of things.
You can still detect that stuff on the server. It's all about will and competence.
The real reason is anticheat allows easy surveillance. The Ven diagram of people with Tencent anticheat and essential IT personnel overlaps a lot. This is a big problem talked about but not solved in sec ops.
Detecting the angle, acceleration and speed at multiple points of a shot in an fps is trivial, and developing a check to see if it's human movement or computer movement is easy after that.
Aimbots are easy too. Is the camera following someone without having vision? Oh no aimbot. A bit more complex than that... But not by much.
It's easy enough to do it right, but then you don't get that sweet sweet surveillance
I definitely get it. Have you seen the state of Team Fortress 2 for the past few years? It's repugnant. I don't know how people are still playing it.
Free to play multiplayer games are at the highest risk of cheaters, since they can just make a new account. I would rather pay for a multiplayer game without microtransactions than a free one which gatekeeps and facilitates cheaters.
League of Legends is fine and it runs via wine. Whatever they are doing seems to work. I won't believe for a second that you absolutely need kernel level malware for Valorant. Not that I care about shooters.
I'm wondering if the reason why it's taken (taking) the anticheat companies so long to implement support on linux is just that they're scared that some linux kid will manage to completely circumvent the entire anticheat with some FOSS sandboxing tool.
They should just start doing more server-based anticheats instead of scanning on the client.
Yup and to top it all off most online gaming communities are really toxic when it comes to any discussion of Anti-Cheat in this regard (it's considered the same as cheating to put down anti-cheat), and will usually defend it and the company behind it. I've even heard some of these chuds try and say linux is evil or used by criminals (you can tell these people are either kids or very immature) and that they should outright ban Linux users altogether.
I guess people are much more rational on Lemmy and fewer of the fanboys who stick up for those companies (or those people just avoid Linux communities like the plague). Either way it's much better than Reddit was.
Some games have a complicated enough Anti-Cheat that it'd be an absolute pain to get it working with Linux. Honestly, it's not worth it to gain 3% more players. I'd recommend just running a Windows VM in Linux.
I've spent quite some time setting up KVM with GPU passthrough and modifying qemu and my kernel as to circumvent VM detection of anti cheat software. While it worked in principle, overhead from virtualization and reduced core count meant that some resource-heavy games ran extremely poorly (while they would have run just fine without virtualization).
Running a VM would imply dealing with VFIO. For a recently converted casual Linux gamer it's better for them to dual boot than deal with that headache.
I didn't fully transition to Linux until I simply decided I needed to (thanks, Windows 10!) and committed. Using Windows as a crutch removes a lot of incentive to learn how to get it all running under Linux.
I get your point in dual boot being less of a headache, but learning some libvirt/qemu and running your own virtual machines is a lot of fun.
I went for a virtual Fedora Workstation with VFIO and a dummy plug. Then I use Sunshine/Moonlight to stream my gaming session to whichever device I feel like using.
Anytime I wanna try something I feel might crash my Fedora I simply backup the virtual machine files and go to town on it.
If I fail I roll back and try again.
I run my servers the same way, as virtual machines that I can easily backup and experiment with, and I do think it makes learning a lot quicker.
I mean Leagur of Legends works fine on Linux and I never had issues with obvious cheaters. If Riot Games can get it done, then so can everbody else.
(I feel the need to state that ingame runs fine, the client is the worst piece of software I have ever had to run on my machine)