Quoting: basedFirst time hearing of a graphics chip that can do Vulkan but not OpenGL.

Reminds me of one 2003 XP laptop I had which couldn't do OpenGL on Win7 yet DirectX games ran fine with it, but that was a driver issue

It's not so much that they _can't_ do opengl, and more like they didn't bother writing a driver for it. The premise of a layer like Zink is exactly that it lets vendors stop caring about opengl and focus on vulkan.

For the record, the samsung xclipse gpus don't have an opengl es driver and use ANGLE.

View this comment - View article - View full comments

NVIDIA is testing the waters with open source, and I'm sure they'll have more to come. I can only hope they have a good enough experience that they won't feel it's too hard and give up!

View this comment - View article - View full comments

Quoting: Fremen

Quoting: Eike

Quoting: FremenThis is bad news. Can likely lead to the Noveau Driver not being developed properly anymore. Perhaps its Nvidia's way of "shutting down" development on the open source driver with seeming like it is.

Why would they send lots of patches then?

I also lean to them enabling the free driver at least for good AI performance. Don't know if they care for graphics performance here in the same way though.

To be fair this is something that companies like Microsoft are famous for. It's even got a name, "Embrace, Extend, Extinghuish"

https://en.wikipedia.org/wiki/Embrace,_extend,_and_extinguish

Come on, this patch is coming from a previous maintainer of this very code. So what, he was an open source hero, then went to nvidia and continues working on the same project but now suddenly it's all a conspiracy to thwart open source? That's nonsense.

View this comment - View article - View full comments

Are we no longer too few to ignore? :)

View this comment - View article - View full comments

Somebody should make a physical robot that looks at the screen and handles the mouse with super precision (not an android of course, it can be a simple camera and a couple actuators) and play the game as if it was a super skilled human. There is no amount of os introspection that can prevent that kind of cheat.

Maybe **then** will they realize they are approaching cheating wrong.

Seriously, what if instead people voted if someone is pleasant to play with or not after a match, and you match pleasant people together and unpleasant people together. That also solves the toxicity problem with the same mechanism.

View this comment - View article - View full comments

Quoting: Ehvis

Quoting: ShabbyXFor example, imagine a cheat that takes the app's wall rendering shader, and makes it semi transparent. Now the cheater can see through walls. They could still be playing with mouse and keyboard, no scripting involved.

That example is still a server problem. It's giving the client information that it doesn't need.

It's not as simple as that though. Even if the server doesn't disclose players / items that should not be visible, it can't do that very precisely for the simple reason that latency exists. You can't afford to reveal players _just_ as they become visible, because to the client it looks like they pop through doors. So taking latency into account and revealing another player earlier with a margin of error, that can still give an edge to such a cheater.

View this comment - View article - View full comments

Quoting: kaktuspalmeClient side anti cheat will never work. I don't get why no one tries server side cheat detection. I think AI in server side cheat detection might be a very useful thing.

Unfortunately it's not that simple. You kind of need both. Server side is needed, obviously, but there are ways to cheat that can be entirely invisible to the server.

For example, imagine a cheat that takes the app's wall rendering shader, and makes it semi transparent. Now the cheater can see through walls. They could still be playing with mouse and keyboard, no scripting involved.

View this comment - View article - View full comments

Quoting: fagnerln

Quoting: ShabbyX> Half of anti-cheat is making sure the environment hasn't been tampered with, and this is extremely hard on Linux by design.

If anyone ever claims 'windows is secure now", show them this statement. These sort of invasive actions is impossible to do on Linux (and easy on windows), for the same reason we don't get viruses (but windows does).

I dislike windows and never installed on my PCs since 2014 (well, I have an old HDD with it because I needed to install an AMD application just to receive a steam key). But I believe that nowadays it's secure if the user uses it "moderately", just don't install any bullshit on the OS (prefer FOSS apps) and don't visit weird websites and you're good to go.

"Be careful what you do" is not security. There is no way anyone, even security researchers, could "just know" the website they visit is malware free, etc. browsers go to great length to sandbox websites for this very reason. They go to greater length than windows does to secure applications!

Anecdotally, my father in law constantly had things breaking on his windows laptop (because he clicks on everything). We switched him to Linux four years ago, not a single problem since (just an old battery that needed changing^). So an OS *can* be good, windows just isn't. You shouldn't have to take security into your own hands.

^ Ironically, the guy who changed his battery told him he should use windows, it's better :|

View this comment - View article - View full comments

> Half of anti-cheat is making sure the environment hasn't been tampered with, and this is extremely hard on Linux by design.

If anyone ever claims "windows is secure now", show them this statement. These sort of invasive actions is impossible to do on Linux (and easy on windows), for the same reason we don't get viruses (but windows does).

View this comment - View article - View full comments

Quoting: PublicNuisanceLet's apply some logic. You get caught not paying your ticket. They ask you to but you refuse. They call the police. The police tell you to but you refuse. The police issue you a ticket or straight out arrest you and you refuse to comply. They go to take you by force and you resist. They end up killing you in the struggle.

The flaw in this logic is that, the reason this person gets killed in the struggle is not that they didn't follow whatever law started this. The reason is that they attacked another human being and the other acted in self defense. The scenario is changed, and the matter of the original law is no longer relevant.

You can go to supermarket، <some conflict happens like you bump carts>, threaten people with a gun and get taken out by the police.

You can walk on the street، <some conflict happens like you step on someone's toes>, threaten people with a gun and get taken out by the police.

You can chew gum، <some conflict happens like someone says you're chewing too loudly>, threaten people with a gun and get taken out by the police.

We don't consider that we must be able to have people die to have supermarkets, walks outside, or chewing gum. That's also why we don't consider that when making (most) laws.

To your original point, I don't think anyone is arguing with you that the choice of software shouldn't be enforced. There was one comment, yes, but we generally all believe in that choice.

View this comment - View article - View full comments