The time it takes for Steam to start off a bloody fast SSD is enough to drive a man to drink. Having said that, I got an update to the EGS launcher for Linux today, and it's still buggy as all hell to the point of being basically unusable, but at...
But we do. Server side anticheat can only go so far, and user land anticheat isn't terribly effective - VACnet and VAC are evidence of that, cheating under CS2 is out of control. Don't get me wrong, I'm by no means a fan of kernel level...
I mean that's literally what Epic is trying to soften the blow around with their eBPF implementation.
Well, I think it's a bit of a multi pronged approach - they're trying to make it as difficult up front to even attempt a meaningful attack, and...
That's 'exactly' what we don't want. We do not want to be forced to run very specific kernel builds that specifically support very specific anticheat implementations. Epic's solution is waaaaay better than such an approach.
eBPF is already...
I don't know what this is implying, it seems AI generated.
I believe it was always known just what Nvidia were doing, leaders in the Linux community just turned a blind eye until it wasn't acceptable anymore.
The use of very specific kernels just to play games is what we're trying to avoid here. Once again, this is a distinction that needs to be understood: signing kernels is not the same as locking users to specifically built kernels just to play...
You still need a shim allowing the driver to communicate with the kernel, you can't modify the kernel directly which I believe is what Nvidia were doing. IMO a full kernel level driver really is a step too far when it comes to AC, hence why I...
It secures shit more so than no secure boot in place whatsoever. This is about coming up with viable solutions to problems, while being open and transparent with leaders in the Linux community as well as Valve and other willing companies in the...
I don't believe you can actually do it anymore, the process has been outlawed, it's the reason Nvidia swapped from closed drivers to nvidia-open drivers. Under nvidia-open, the closed source driver runs in user land, it's the shim connecting the...
Chad, desktop Linux can't avoid secure boot forever. The fact Linux users are running their systems with secure boot disabled in this day and age is somewhat embarrassing.
I'm still not seeing any real compelling evidence it isn't possible.
I don't believe EAC v2 is as outright invasive as kernel level anticheat under Windows using kernel level drivers, I believe this implementation is the most realistic...
Reading through the Phoronix thread on it, it sounds like the CachyOS maintainers have reached out to epic about it because they're interested in it and waiting on a response.
So you're against signed kernels/drivers? I honestly don't see any negatives regarding either, no one's forcing you to enable secure boot. If you don't like EAC v2, don't play titles that use it, you've still got choices.
Nvidia drivers under...
Realistically? All Linux kernel modules/drivers should be signed, Linux should have supported secure boot 100% OOTB 10 years ago. A signed kernel is not a locked down kernel, lets make that distinction clear.
I run KDE Neon on my main PC, and...
They're using eBPF to verify the kernel isn't tampered with, and you don't need to run specific kernel builds, existing kernel builds simply need to be signed by distro maintainers - Epic are even providing keys maintainers can use to do so. This...
Good to see Epic realizing they better jump in, even if it is a just in case. I'm sure they realize handhelds are probably never going to really be windows based going forward. Beyond that who knows why not hedge your bets if all it costs is a...