4 ms·

It's seriously astonishing how much progress has been made in the past few years regarding gaming on Linux. When I first made the jump to using Linux as a daily

It's seriously astonishing how much progress has been made in the past few years regarding gaming on Linux. When I first made the jump to using Linux as a daily driver circa 2016, I had to keep a Windows partition to be able to play any game in the Steam library that wasn't native on Linux. With steady improvements to Wine and Valve's release of Proton however, I can play most games out of the box or with a minimal amount of tweaking. However, the games industry is still largely hostile to gaming on Linux. I've always had the best experiences with indie games or FOSS projects, and the only trouble I've had is with newly released AAA titles. I wouldn't even bother with newer multiplayer-only games or MMOs, as anti-cheat on Linux is a pretty reliable way to be falsely detected and instantly banned (not to mention requiring kernel-level access, which is utterly ridiculous).

xnyan 6y ago
>not to mention requiring kernel-level access, which is utterly ridiculous Not a troll question: how do you detect cheats run at the kernel level without kernel-level access? If you answer is "I don't want a game to have kernel level-access even if it means some cheats can't be detected", I respect that, but can you accept that others can make a different decision and not be "utterly ridiculous"?
- mhh__ 6y ago
  I think the point here is that "Others" (i.e. My non-programmer friends who play Valorant, for example) don't know what a kernel is. I wouldn't have an issue installing something like Vanguard to play Tournaments but having a single point of failure like that (and totally unverified?) just to play casual is not good. The bomber will always get through. I think the only way forward is by using a weaker anti-cheat combined with an ML and player-based (Like in CSGO) system to find cheaters. If cheaters get through, but they have to play within the bounds of what could be reasonably could be expected of a regular player, it doesn't really matter. Push comes to shove you could make a fully visual triggerbot that doesn't even draw power from the machine (i.e. HDMI->Magic->MouseInput)
  - google234123 6y ago
    People do this. But the latency can be bad and your fake mouse can also be detected. Inputs from two mice would look pretty suspicious

xnyan 6y ago

>not to mention requiring kernel-level access, which is utterly ridiculous Not a troll question: how do you detect cheats run at the kernel level without kernel-level access? If you answer is "I don't want a game to have kernel level-access even if it means some cheats can't be detected", I respect that, but can you accept that others can make a different decision and not be "utterly ridiculous"?

mhh__ 6y ago
I think the point here is that "Others" (i.e. My non-programmer friends who play Valorant, for example) don't know what a kernel is. I wouldn't have an issue installing something like Vanguard to play Tournaments but having a single point of failure like that (and totally unverified?) just to play casual is not good. The bomber will always get through. I think the only way forward is by using a weaker anti-cheat combined with an ML and player-based (Like in CSGO) system to find cheaters. If cheaters get through, but they have to play within the bounds of what could be reasonably could be expected of a regular player, it doesn't really matter. Push comes to shove you could make a fully visual triggerbot that doesn't even draw power from the machine (i.e. HDMI->Magic->MouseInput)
- google234123 6y ago
  People do this. But the latency can be bad and your fake mouse can also be detected. Inputs from two mice would look pretty suspicious