When Kernel Level Anti-cheat was a requirement for League and I couldn't play the game, I got a girlfriend, started a job, touched grass more often and balanced my life... I couldn't believe it.
@@Thebigbean114 I have no idea whether to upvote or downvote your comment! XD On the one hand, it was very funny and there's a lot of fun and cool things to do online. On the other hand, I lived my life entirely online for 10 years and towards the end of it... well, I won't go into my health, but I'll just say that when I built my health back up to the point where I could go out on my e-bike most days, my body could stand a whole lot more bacon than before. A little regular exercise is worth it for the bacon! ;)
It doesn't mean anything because the article was a complete fabrication and the author didn't read the blog post which explicitly said 3rd party kernel level access is not going to be removed
@@CodeAsm I don't think LLL actually even researches things. IIRC he was completely wrong about what caused the CrowdStrike outage in the first place. Talks WAY too confidently about things he doesn't understand. Fake it til you make it I guess.
@@cooltrashgamer i think he does "research", as brodie stated in his reaction video, probably skimmed over the original blog if at all, and read the notebook blog as truth. And building ontop of precious acquired knowledge and assumptions. Basicly what i do personally alott, but scared to confidently vlog about. Theres always more knowledgeable people around, but this was a somehwat easy to read spurce blog and notice the mistake. A slip, redact or correct yourself after more indepth research and contact ppl that are in the know. Clear example of journalistic echo chamber sadly
Client side anticheat is a software engineer saying "oh, we don't need to sanitize our database inputs, that's handled in the javascript frontend" and *_not_* getting fired.
Pretty much... And it goes hand in hand with encrypting the data in flight. That's 20% not wanting people to write their own clients and 80% "we take whatever we get sent verbatim, no checks". The amount of games, esp MMO's, that just keel over and die once you break the encryption scheme is just... mind bending. Many are so bad that you basically have RPC endpoints to do "whatever" once you get past the encryption.
@@thedave1771 The problem with RFC 3514 is that it raises a 'moral' dilemma. If we assume that somewhere in flight data can pass through an evil actor, and being that evil is as evil does, it might have changed the evil bit, thus making it dishonorable. To err on the side of caution we now must drop ALL traffic. Amusingly, that in fact would make the Internet 100% secure through non existence...
What is this take? Client side anti-cheat is superior to server side anti-cheats in almost every way from a detection stand point. There is a reason all server side solutions suck (even with heavy ML training over crazy amounts of data). At the end of the day its hard to detect a cheat running on the client without being on the client yourself and searching for it. You can also write cheats that are 'humanised' enough to never trigger server side cheat detection. I haven't even mentioned false positives in which server side anti-cheats struggle with all the time which are basically non-existent on the client side.
Good thing it is hyperbole then? The economy is just all the exchanges of goods and services that happen. Just because all computerized financial transactions are prevented for a few hours doesn’t mean that *all* exchanges of goods and services stop for that period of time.
That article misunderstood the source (the Microsoft blog post). They aren't removing kernel mode anticheat - just providing more features in userspace.
Fact check: valve does not have a kernel level anti cheat, in fact valve has been extremely anti-kernel level anticheat to the point that it is costing them money to hire contractors to go over VAC/player reports.
Valve is a lazy company, and you can gather why they don't want a kernel level anti-cheat. They couldn't even fix anything, take a look at CS2 and TF2, still filled with cheaters using decade old cheats.
@@AlfaPro1337 Eh, a kernel level anti-cheat would've been the "lazier" route for Valve to take. Not going that way turned out to be a more herculean task until the initial momentum and enthusiasm they had taking it ran out.
Read the source article, MS is not kicking 3rd parties out of kernel space. They don't even say that. They say restricting. Details matter. Plus, even if anti cheat were kicked out of kernel mode, either A) the devs will still block linux with user space anti cheat or B) (Most likely outcome) the new security API MS is talking about implementing for them to use will also not be compatible with linux.
The new security API if used from userspace should be manageable to implement into WINE. Definitely much more manageable than kernel-level stuff which, as the name suggests, requires the kernel.
@@rtzgf67games7 security api even used from userspace will have a kernel component. That said we don't know yet, what we know is MS isn't kicking them out of the kernel and EAC and BattleEye already have an easy way to implement wine/proton support and actively choose not to. Not to far fetched to think they would even let the Linux community play even if we did implement the security api (which we can't without the kernel side of things)
Tell me you dont understand the promlem with linux gawing without telling me you dont understand the problem with linuxgaming... Hint: Its a lot easyer to emulate a few API's than half the kernel.
@@bas1858 easier =/= possible. Publishers want kernel level anti cheat. If wine was able to emulate or simulate the new api and make the anti cheats function that way then game publishers would likely start "actively" blocking wine from running their game. Look at Fortnite, uses eac and battle eye. EAC and BE both work in Kernel mode in Windows, user mode in Linux. Because Linux does not work in Kernel mode, Epic actively blocks it from running on Linux. New API is not going to change this.
Trust factor system is the only thing actually separating cheaters and legit players in CS2, anyone developing external or internal cheats for cs2 has a trivial time avoiding VAC's detection vectors without ever interfering with the anti cheat modules themselves
The Microsoft blog post explicitly states that they are not removing the ability for vendors to write kernel-level code of this nature, rather that they are looking into adding more security features for vendors to access in user space. The original article is misleading if not downright clickbait. There's no reason to think that this will result in less kernel-level anticheat and the hype around this news is a dud. EDIT: I misread a quote from one of the security vendors, ESET, as being from Microsoft. Microsoft were not the ones that stated that it was “imperative” that access to the kernel remains open for vendors. However, still nowhere did the blog post imply that this access was being locked down or removed, but rather that they are looking to add additional options in user space for vendors to utilize. Sorry for the misread.
Well... Microsoft will probably have to maintain some kind of compatibility for a while... does that mean they won't deprecate these methods eventually ? And even make it impossible some day...?
While the article is misleading, I think this is a move towards fewer kernel level anticheats. In one of their dev blogs, Vanguard developers mentioned that if Microsoft provides a framework for verifying kernel integrity, they would be more than happy to restrict Vanguard to run only on game launch and in user mode. My prediction is that Microsoft will have a setting to block unsigned kernel modifications, and games will require you to have that setting turned on to play. OEMs and vendors that require kernel extensions will go through Microsoft similar to how firmware updates and the like are signed through Windows Update.
@bjornarnols5905 The problem with Linux is that it's for very tech literate people. The average person cannot work with Linux, and few can do troubleshooting. Grandma considers windows confusing. She's not going to use Linux.
@@the11382if your grandma just uses a web browser it isn't a huge issue. Most distros come with a GUI based system updater anyways, and from my experience the likelihood of a failed system update is significantly lower than on Windows.
@@the11382 I agree Linux does need to become more user friendly (as always), but also counterpoint: People who aren't tech literate probably do less complex things with the tech. Grandma probably just needs a browser and MAYBE a file manager, both of which Linux offers plenty versions of. - A Fedora user
- "Client-side data should never be trusted." - "Understood! Let's run it on the client even harder - In kernel mode! What could possibly go wrong?"
4 หลายเดือนก่อน +13
it still on client tho, u can't manipulate the anti cheat using a usermode cheat, but a kernel leven cheat can access and manipulate the anti cheat data
Client side data isn’t trusted. That’s why you don’t see hackers noclipping in every game you play. They can check for noclipping server-side, so they do. Theres a ton of game-ruining cheats that you can’t detect server side. Most notoriously being aimbot. The client looking at enemies and shooting is completely valid client side input.
That tweet at 10:46 is unbelievable bs. Using Linux is not the equivalent of running away, it's the equivalent of vote with your wallet. We are still here and part of this society. We just don't want spyware, ads and security/safety issues in every single aspect of our lives.
Whats funny about that specific tweet is that its made by Tim "Epic Games" sweeny. The guy accusing steam of a monopoly while using monopolistic practices to try and goad players onto their store instead of giving a good product. The tweet literally comes with an asterisk of "*please use our store instead of that big meanie Gabes!"
Wow, since when does windows stand for freedom? And why is Windows’s one sided bs considered politics? They are shoving unwanted features down our throats without any discussion. Switching to Linux is voting with your wallet (PC) about what you think is right and better for you.
You should've read Microsoft actual blog post, they're not removing Kernel level access. They're going to make accessing it 'easier' and safer, with better tools and ways of doing so. This is to prevent another Crowdstrike.
Afaik valve actually dont have a kernel mode AC, because it would prohibit their own linux based devices (the steam deck) from playing their own games.
I'm pretty sure they _could_ write kernel level stuff for Linux, but they don't because 1. they don't want to, Valve has gone on record that they don't agree with the philosophy behind kernel anti-cheat. And 2. they'd have to maintain the two anti-cheats independently of one another, as the Windows kernel and the Linux kernel are obviously vastly different beasts.
Valve doesn't have a kernel-mode anticheat In fact, Counter Strike is completely unplayable above a certain rank and everybody goes to faceit instead Counter Strike's current state is that competitive integrity is completely absent when a kernel level anticheat is absent too
I don't understand why your t-shirts get washed and not destroyed in the dish washer but when I put my dishes in my laundry machine they do get clean but also are broken.... There must be a little discrepancy somewhere, I just don't know where.
@@DarkblooM_IO is it? Seems like a pretty understandable grammatical error. It that qualifies as "crazy" for you I'd suggest getting out of that bubble you live in.
Bringing in a security module that verifies the boot state and provides attestation for applications has serious implications on control. It sounds great on the surface, but this is exactly what android is doing and manufacturers use it to lock down their devices, making it so you cannot modify your device or install custom firmware on it, which limits control and turns it into e-waste in some cases. This sort of thing would pave the way for desktop computers coming with locked UEFI's where the only operating system you're allowed to install is Windows for example. You cannot trust companies to do one thing and one thing only with this sort of control. Instead, we need to encourage companies to invest more in server-side protections instead of being lazy with their security and instead wanting to force users to run their apps in an environment where this sort of thing is not possible.
I've never heard of a kernel level anti-cheat causing major problems, but it could happen. If you are seriously concern, play games on a spare computer. If it does go boom, you're good. And you aren't slowed down by the 30 different gaming applications constantly running.
Well, don't think so. Rockstar just broke GTA 5 Online for everyone on the Steam Deck (and any Linux distro) by borking their implementation of BattlEye, which has native Linux support; but they didn't bother.
It makes sense that they don’t bother, even if I really want them to care. I just checked Steam’s hardware survey, and they show that 96.78% of users use Windows, 1.3% use Mac and 1.92% use Linux (interesting side note , arch linux is the most used Linux distro among steam users) Sometimes it takes a lot of man hours to make programs and games compatible cross platform. It’s honestly not worth it in certain scenarios to put in effort to support Linux. My biggest hope for Linux gaming is honestly Steam OS. Valve announced that they will allow other handhelds to use the OS (like ASUS ROG Ally). If PC handhelds becomes the affordable and portable option for PC gaming, then SteamOS and Linux might achieve a big enough market share to force developers to respond to the market.
Found the jif sadboi. Honestly though, it's crazy to me that there's a TH-cam creator that hasn't seen/been conditioned by like 17 million videos sponsored by them. Seems like they sponsor half the content on here 😩💨
I'm generally happy, that I'm not a competitive gamer, so I simply don't play any games which have Kernel Level Anti-cheat. So Linux for gaming works 100% for me.
At this point we can't trust any employer let alone game developer if they wanna go this route and want access to your personal data time to start living in tents to get away from all of it and turn on Minecraft survival mode in real life.
@@bigbossgaming7663 it is on android 15, they are trying to discourage people from installing apps that are already on the Play store, so it blocks you from installing it, that's what I've seen so far
The definition of kernel that you gave is a modern redefinition of the term. DOS had a kernel. In MS DOS it was MSDOS.SYS. A kernel is simply the core functionality of an OS, i.e. the part that handles system calls and interrupts. The part about memory and process separation wasn't possible on x86 until the 386. Redefining words is all the rage these days.
Reconciling the entire packaging system differences and distros would be nasty asf but gaming will most likely get some common standards up, here's hoping Steam gets us there as Linus Torvalds (jokingly) imagined
There's going to have to be a unified, supported platform first. The "anything goes" nature of Linux is in direct contrast to being able to manage third-party software distribution. I.e., nobody is going to publish software for an ill-defined target, and that is practically the definition of Linux. Don't get me wrong. I love Linux, and I use it every single day. But there are several reasons it hasn't taken over the desktop, and the main one is ... nobody knows "Linux"... they just know _their_ Linux.
@@nickwallette6201I agree. Linux suffers from issues that BSD and other FOSS OSes don't because of lack of unification. Linux has a unified kernel with stability guarantees, but no unified userland, just trends among distros which constantly change. With how big Linux is, it would be really advantageous to have an "official" distro to target ABI and library-wise, and anyone who wants to do something else is on their own (alpine, NixOS, etc). The freedom of Linux would be preserved, but new users wouldn't have to deal with BS like trying to run Davinci Resolve with janky scripts because it targets CentOS and they're using Pop
I really don't think anti-cheat is the big reason why many developers don't push their games to native Linux. I'd argue that the main issue with gaming on Linux comes down to historically poor driver support by the major GPU manufacturers, noteably Nvidia. In recent years there have been more and more ways to work around that limitation and now Nvidia seems to finally be coming around to the idea of open-sourcing their drivers which should eventually blow the gates wide open for gaming on Linux, but the lack of decent drivers for graphics cards by THE single most popular GPU company was/is almost certainly the biggest hurdle to native Linux support in gaming.
You are spreading misinformation as is your tech-bro source. Microsoft never claimed this. Please do your research if you go hands up claiming "I was right" and don't just rely on a single source.
@@wellgun3 As i've mentioned in another guys comment, this dude is clowned on in malware development communities which are full of people who are known for creating some of the best C2's and exploitation frameworks out there currently. He appeals to people who have 0 knowledge on the topics but has absolutely no substance when talking to people who actually know their shit and do it daily for a living.
The fact that a company named "crowdstrike" did the equivalent of going on a work crowd strike to protest kernal access by accident is really funny to me
@@heatnup hating kernel level anti cheat doesnt mean one is cheating. Kernel level programs in general should just not be so easy to change thus break systems (see cloudflare)
People who non-ironically think the drivers of a mouse or keyboard equates in seriousness or risk to an online game should legally not be allowed to vote...
@@eekee6034 Not sure what that has to do with anything, if somehow they can inject a keylogger into the keyboard driver that is already passing every possible security measure. The only solution to the problem is to literally have no connection between the PC and the keyboard. But we can't have that, can we? And Keyboard drivers are less regularly updated than a videogame, and their anti-cheat. And of course the more games need it, the more folds do these issues increase. So no, you can't compare one neccessary piece of software for keyboards and mouses to function which you should have very few of, to an unnecessary piece of software that you need to have multiple of, and gets updated very regularly by companies which writes these software on the side, not as a main focus.
@@katamas832 Thanks. That is a fair point with having multiple keyloggers which don't need to be there, and too-frequent updates can be trouble. I'm evidently a bit out of touch when it comes to driver security measures. On driver quality though, I'm pretty sure most hardware manufacturers don't write the best drivers. They certainly didn't in the 90s. Keyboards are probably all right though. The driver for a standard USB-HID keyboard should be high as it can be coded by the OS developer and one driver can cover almost all keyboards. I'm not quite so sure about keyboards with extra features.
Problem is that cheats are going the same route. Game devs don't want to spend time and money on anti cheat, it's the gamers requesting it indirectly by wanting to have a fair game. Reality is that cheats are kennel mode or even hardware side. Without deep access to the OS it becomes virtually impossible to detect them. The alternative is that games will be full of cheaters. Pick your poison.
It is worth noting that in the original microsoft blog ESET said that they'd like if there was still some level of kernel access for security solutions, and that they never explicitly said they were removing any security functionality from the kernel, only that they were adding more outside of it.
Userland business logic has no business in kernel mode. Full Stop. 1.) Too volatile a logic framework 2.) Too many cooks in the kitchen 3.) Too much risk of code eats dog + world 4.) Too Much Separation of agency from responsibility -- Microsoft was the ultimately responsible entity, even when the 3rd party wrote & shipped crap kernel code.
I still remember how Riot Games said "the reason we created Vengard is that Windows doesn't give us better tools. Which they can, and it is probably a matter of time"
It is the right move. Letting anti cheat for a video games to have full control of your system is the same as giving your key to a suspicious stranger to clean your house without limit.
The biggest thing about the whole thing is that the Linux CrowdStrik errors a few months before the Windows outage was an eBPF program that crashed the kernel... Something that eBPF was suppose to prevent. So even an eBPF solution on windows could suffer the same way.
Says, "i was right, i love being right" in the first 2 seconds yet he was still wrong XD love to see people getting jebaited by people who have 0 reading comprehension.
Kernel authors slowly learning about the advantages to micro-kernel designs... If there was an API for interacting with the kernel and detecting these things without running in kernel mode, then this use-case could still be supported, for the most part. eBPF and the ideas behind that are promising, and I like how standardized it seems to be. Maybe "kernel-level" anti-cheat can be cross-platform if the kernels exposed a common interface. Although I don't cheat, it wouldn't be too hard to compile a kernel with an eBPF that would lie to anti-cheat, and swapping kernels on Linux-based operating systems is very easy.
Having a game to run anti cheat software is like the power utility company to forces you to put a camera on the meter, because they're afraid that you might cheat, but it's in my home I don't like it. Also, buy separate private VIP servers and play games without cheaters, forcing everyone to have something that will eat more computer resources just to check if every client request is legit is a nonsense, mainly because this is not a banking software it's just a game.
@@DamjanDimitrioski Completely different examples, the game running the anti-cheat is completely fine, its the fact that some of the anti-cheats are kernel-level which is not. When you buy/play a multiplayer game you expect a fair game for everybody, no cheaters. Private VIP servers have a whole host of problems, cheaters don't mind paying money for things like this. Take CS2 Prime as an example, you have to play a considerable amount of time or pay for Prime, but there are still so many cheaters in Prime because people don't mind spending money for an account. It's also just outrageous to have people pay to not play with cheaters, if you've already bought the game why do you have to pay more to not play with cheaters?
OS developers have all known the advantages of microkernels since the 70s. Somewhat like the Rust Evangelism Strike Force in recent years, it has at times been hard to escape microkernel proponents :) The trouble is it's really difficult to design a microkernel architecture with good performance. Not that you really have to design one, you can use the Mach design as IBM Apple and the GNU Hurd project have. There's a bit of a funny story about Hurd. It's older than Linux; it was going to be the GNU kernel, but it wasn't ready for use when Linux appeared. Years later, Linux was running 70% of Internet servers and Hurd was still not ready! I thought I heard they couldn't agree on design details or something, maybe they wanted an independent design rather than Mach; I don't remember, it was too long ago. When it was finally released, it was just another Mach design. But people do design their own microkernels, usually without taking quite as long as Hurd. :) Managarm is a POSIX microkernel OS that's been looking pretty good for a couple of years now, supporting Wayland and X11, but I don't know if it's reliable enough for everyday use. There's also ToaruOS -- if you've heard of PonyOS you've heard of ToaruOS ;) -- but its author doesn't really want it to be used seriously. He made it to learn about OS development.
@@hashcrash9729I think the only truly POSIX compatible shell is the shell BSD distros use. Bash is not completely POSIX compatible and you need manually or consciously exclude convenient GNU features for your script to be POSIX and run everywhere even MacOS. Let alone ZSH and FISH which are way away from POSIX.
I think windows (and linux if they want to support security paranoid game companies) having a launch mode that tells the kernel that this program should be protected from tampering by the kernel not exposing memory in any way etc etc then that seems entirely reasonable. "I need this program to run entirely isolated" sounds like a valid request whether it's for something a little frivolous like a game or putting in more kernel protections for a password manager process.
You don't want ANY program reading ANY other programs memory, or especially writing. If this was just a mode you could add it would be the default, and so, uhh, it kinda is the default. cheaters or whatever are purposefully getting around all the defenses, Microsoft can't really add new ones, or theyd have done it already
@@turtlefrog369 vanguard *can* do it but no way would they open themselves up to so much risk. cheaters on the other hand dont exactly have to abide by that ruleset
Kernel-level anticheat can still say "no get lost linux" but at the same time, not being allowed to make kernel-level anticheat on windows does mean that linux can in theory be more malleable for a kernel-level anticheat, since you can freely distribute kernel modules.
One big risk is if Microsoft decides to abuse this and perform cryptographic OS attestation under the guise of an anti-tamper API (like Google similarly abuses on phones), making it so games will simply refuse to run on an OS that doesn't provide Microsoft's cryptographic signature...
Valve would totally take that to court in, say, the EU, saying that Microsoft implemented a feature that is being unjustly anti-competitive towards SteamOS by preventing Valve's platform from being used as an alternative for gaming. Microsoft could very well be considered a "gatekeeper" by EU laws. Just my conjecture.
@@ToxicMothBoi what are you even on, have you ever played valorant? now compare the number of cheaters to cs2, it's not even close. if they take away kernel level anti-cheat, I will stop playing
i saw an article that said this change might actually negatively affect gaming on linux; instead of simply being unable to play due to kernel level crap, there'd be other hurdles to overcome as emulating all of the new crap might be difficult. and even then, companies could just add even more ways to detect linux and ban users. Not sure how accurate it is, but it definitely raises some concern for me, i want linux gaming to thrive as much as possible
Mostly don't give an F. Everybody can argue but it's a Result Theory. That's what all matters, result. Other than that is anxious paranoia which can be all in the tech landacape and technicality. Vanguard for example does it's job very well and brings out result.
"so I steal money" ->Which is illegal, Vanguard isn't. Your point? Oh that's right you don't have one. I could do a better logical counter-argument against my post in my sleep.
hmm IDK, dude.. if you think about it, it's gonna be having a high security code installed by default, run by microsoft, waiting to be called by apps, with intrusive functions that will be required by devs like those from riot, ea... Up to now, it was optional. You can create a dual boot to install vanguard and that stuff in a different installation that you'd normally use. Now it's gonna be built into the same OS. Sounds too much like what happened with Android: an open os closed to a security OS to satisfy what devs asked for safety with, let's say, money operations and stuff.
Finally, locking something down being a good thing. Wanted to ask... when you put clothes in the dishwasher, do you put fabric softener, and regular clothing detergent? What about bleach? And where would you put it before starting the wash?
A bit late now, but apparently they are only discussing the addition of new options outside the kernel. They wouldn't lock vendors out, but give them options to stay outside the kernel. Many vendors might still prefer using their own anticheat. It is possible that some are going to move to such potential alternatives.
@@Raphipod genshin impact.... literally single player, with just their shitty little multiplayer implementation that could be compared to inviting your friend to your self hosted minecraft server, its about the public i'd say
@@Raphipod That's stupid shortsighted thinking. DLC capability is provided by distribution platform. You don't need highly intrusive anticheat or always-online tether for that. Yes you can release paid expansions for the game, nobody will mind! Or you can do sequels and sidequels. Why it's stupid, is because a person can spend 50-80€ per game for 6 games a year easily, a typical person. Then they'll have them finished in maybe 12-25 hours each, and then they'll want another game. All these live-service games... they rely on sucking your time completely dry every day-ish, every free minute, and rely on people becoming addicted and starting to whale and spend untold amounts of money on it. But most of your audience is spending barely nothing. Live service games are a big bet with high gain, high loss potential, a major risk, and there's a limit to how many such successful games a market can absorb. There is no such hard limit to self-contained games. So it's smarter to partition the product space to where they won't compete. Many people will NEVER play a second live-service game, they'll play one, they'll become grated by it after a while, they'll find predatory monetisation abrasive, find it's not how they want to live, and abandon it, and will be wary further on out. They want self-contained experiences tailored towards them, where the spending decision is frontloaded and separated from moment to moment game. Single player games are relatively low-profit but also relatively low-risk if you know your audience. Another self-contained single player game is much less likely to hurt your vaguely similar game's sales prospects than a live-service game is likely to hurt other live service game's sales prospects, since every live-service game wants all of your time, while a self-contained game only wants a manageable chunk of it.
@hagnahugna2948 It would help if you said which information you disagreed with. What you said is almost impossible for him to realistically respond to, and just dilutes genuine criticism to the detriment of the community.
@@memes_gbc674it could work a bit, some gta v cheat devs aren't willing to go to make a kernel driver to continue their business already, it's gonna take some time and cheats are gonna be paid now
This does nothing for Linux, because it's simply not possible to add such a security sensor to it. What should that sensor attest? That no processes run with kernel access except for the kernel? No problem, just patch and recompile the kernel itself then. There's nothing a TPM chip could do here except maybe as a non spoofable id to hardware ban people?
All of this over cheats!?!???? Even more reason for me to never play online games. This is just too much to ask for and i don’t trust game corporations, or microsoft but game corp are even worse
i feel like this mean literally nothing most of the commonly used kernel level anti cheat software works just fine on linux, and game devs just disable the ability for their games to connect to client using linux because.... reasons??? anyways the barrier is not anti cheat software developers not wanting to support linux they already do the problem is game devs not trusting linux and this doesnt fix that.
Anti cheat protected games aren't coming to Linux any time soon even if kernel is removed. 'Emulating' windows to pass all their checks is never going to be feasible or reliable.
Have you ever heard of wine and/or proton? Lots of games have linux anticheat, and usermode anticheat can translate your honest linux setup into an honest windows setup. The problem is you can't virtualize or lie about kernel mode, CPUs aren't designed that way, but usermode is entirely beholden to what the kernel says, and a kernel level wine extension is an option.
Easy Anti-cheat does have a fully functional usermode AC for Linux system. But because Linux is so powerful in installing kernel modules for cheating lots of games straight up refuse to support Linux at all. Even when there is a potential to ship games on Linux if "Microsoft actually implement it" then uh it still doesn't sound viable if there isn't a way to verify if the Linux kernel ship with a dedicated module for cheats
@@aominnomad -1 isnt a "real" ring but its referred to when talking about hypervisors since they have more control over the system than normal kernel code
6:19 Most drivers are user mode these days anyway. So your mouse is not using the kernel mode driver. That is the best thing that Microsoft did to the Windows, they tried to push as many drivers to user mode.
i dont trust games that want kernal acess so i got a separate pc for my personal and one for gaming with no persnal data but i just recently started becoming aware of my digital privacy
The solution is load and store ports of the CPU to have encryption capabilities and putting in the decription table of the process and threads the keys for what it is supposed to access only. It would make the memory opaque for all other threads, with the keys relocated every 300ms like what windows does already for the "GDT" like data structure.
Vanguard once thought an Asus driver was suspicious and just blocked it. My motherboard is Asus. It completely broke my OS and had to format my PC in order to get it to work again.
A washing machine use (somewhat moderate) foaming detergents for suspending dirt out of fabric surfaces, with no or very mild enzymes to not destroy the fabric integrity. Their pumps are designed to work with and tolerate a significant bit of foam - they use a propeller'esque or water wheel style impeller on a regular axle, driven by cog or belt by power of an aircooled motor. Dishwashers are made for hard surfaced materials, and their "detergent" is _hella corrosive to organics._ It will utterly destroy the integrity of any organic fabric fiber, and most polymers used for clothing. Dishwasher detergent work by that corrosive mechanism, and reuse most of the same water throughout the cycle - it's not like a traditional (foaming) soap. Their pumps are submerged, impeller type cooled by the water and *_DO NOT_* tolerate foam/air. So dishwasher detergent with clothing is a no-go, and if you put a foaming soap/fabric detergent in the dishwasher, the submersion dishwasher pump will burn out. Like a watercooling pump without water.
Sigh, isolation and formal OS kernels predate linux and windows by decades. Hilarious that the industry is still arguing about issues that were well understood in the 70s.
@@geekswithfeet9137 your misinformed the first result on google isnt always right, valve said they will NEVER use kernal level anti cheat your misinformed
@@roborob347 Nope, it's only stat bans and not root level like Vanguard, every anti cheat is technically kernel level (otherwise they are useless) but most of them aren't ring 0, the new security updates coming to windows only affects root level anti cheats like Vanguard, Richochet, Genshin and what ever the new EA one is called.
edit: I misspoke about valve having kernel-mode anticheat, meant to say valorant. my bad. haha subscribe plz
cant lol (i already am)
I was just typing a comment that hey hey hey, valve doesn't do that kind of stuff :P
Y r u gae?
vac being hilariously bad:
@@LowLevelTV damn boi isn’t a real gamer, using stolen fps valor
When Kernel Level Anti-cheat was a requirement for League and I couldn't play the game, I got a girlfriend, started a job, touched grass more often and balanced my life... I couldn't believe it.
@@tranthien3932 Damm sorry to hear that
@@Thebigbean114 I have no idea whether to upvote or downvote your comment! XD On the one hand, it was very funny and there's a lot of fun and cool things to do online. On the other hand, I lived my life entirely online for 10 years and towards the end of it... well, I won't go into my health, but I'll just say that when I built my health back up to the point where I could go out on my e-bike most days, my body could stand a whole lot more bacon than before. A little regular exercise is worth it for the bacon! ;)
Im sorry to hear that you are a normie now
I'd rather be a normie than grind LoL 12 hours a day @@Spiritwonder420
It will never last, things always fall apart. I only give this hard truth to those who gloat.
Idk about the dishwasher thing...
If it cleans a dish, why not a shirt?
It is legit, just use top rack so as to not burn cloth on heating element
@@sccur I did test it accidentally by leaving an a torn bit of an old shirt in the dishwasher and it was surprisingly very very clean
It's secure because others don't know your next move
and why not also use our clothes washer & dryer to clean our dishes?
It doesn't mean anything because the article was a complete fabrication and the author didn't read the blog post which explicitly said 3rd party kernel level access is not going to be removed
@@BrodieRobertson your reply needs more upvotes and maybe a colab to straighten up these facts 😎🤭 brodie you made me learn tons more
its crazy how many tech youtubers spread misinformation and yet being watched by so many people. Shame
@@CodeAsm I don't think LLL actually even researches things. IIRC he was completely wrong about what caused the CrowdStrike outage in the first place. Talks WAY too confidently about things he doesn't understand. Fake it til you make it I guess.
@@cooltrashgamerso i guess you could say he learns at a low level
@@cooltrashgamer i think he does "research", as brodie stated in his reaction video, probably skimmed over the original blog if at all, and read the notebook blog as truth. And building ontop of precious acquired knowledge and assumptions. Basicly what i do personally alott, but scared to confidently vlog about. Theres always more knowledgeable people around, but this was a somehwat easy to read spurce blog and notice the mistake. A slip, redact or correct yourself after more indepth research and contact ppl that are in the know. Clear example of journalistic echo chamber sadly
Client side anticheat is a software engineer saying "oh, we don't need to sanitize our database inputs, that's handled in the javascript frontend" and *_not_* getting fired.
Pretty much... And it goes hand in hand with encrypting the data in flight. That's 20% not wanting people to write their own clients and 80% "we take whatever we get sent verbatim, no checks". The amount of games, esp MMO's, that just keel over and die once you break the encryption scheme is just... mind bending. Many are so bad that you basically have RPC endpoints to do "whatever" once you get past the encryption.
@@felixjohnson3874 you can do both. Sanitize at the browser and the rest api...
In fairness, it’s way easier to write efficient and performant code if you simply require that malicious users mark their traffic with the evil bit.
@@thedave1771 The problem with RFC 3514 is that it raises a 'moral' dilemma. If we assume that somewhere in flight data can pass through an evil actor, and being that evil is as evil does, it might have changed the evil bit, thus making it dishonorable. To err on the side of caution we now must drop ALL traffic. Amusingly, that in fact would make the Internet 100% secure through non existence...
What is this take? Client side anti-cheat is superior to server side anti-cheats in almost every way from a detection stand point. There is a reason all server side solutions suck (even with heavy ML training over crazy amounts of data). At the end of the day its hard to detect a cheat running on the client without being on the client yourself and searching for it. You can also write cheats that are 'humanised' enough to never trigger server side cheat detection. I haven't even mentioned false positives in which server side anti-cheats struggle with all the time which are basically non-existent on the client side.
1:20 "Turned off the economy for 8 hours" is not a sentence you should be able to say without hyperbole
I just Linuxxed it and got on with life....
@@dillbourne He probably missed a flight and the rest is being American.
Good thing it is hyperbole then? The economy is just all the exchanges of goods and services that happen. Just because all computerized financial transactions are prevented for a few hours doesn’t mean that *all* exchanges of goods and services stop for that period of time.
I'm glad the internet is powered by Linux
The Economy didn't stop. A lot of snowflakes melted down though.
That article misunderstood the source (the Microsoft blog post). They aren't removing kernel mode anticheat - just providing more features in userspace.
No, they understood it fine. They just wanted more clicks on their article.
Fact check: valve does not have a kernel level anti cheat, in fact valve has been extremely anti-kernel level anticheat to the point that it is costing them money to hire contractors to go over VAC/player reports.
check pinned message, theres a correction for that
Valve is a lazy company, and you can gather why they don't want a kernel level anti-cheat.
They couldn't even fix anything, take a look at CS2 and TF2, still filled with cheaters using decade old cheats.
@@AlfaPro1337 Eh, a kernel level anti-cheat would've been the "lazier" route for Valve to take. Not going that way turned out to be a more herculean task until the initial momentum and enthusiasm they had taking it ran out.
@@kyle-amnk Nah, Valve is too lazy to make anything new, and fix bugs within their game.
Steam has a driver. They already have kernel level access.
And who needs a dryer if you have a microwave
this guy gets it.
I like how relevant this ended up being for the video. Who needs Riot in the kernel, if you already have Microsoft in the kernel?
How am I supposed to dry my cup of coffee?
@@mgancarzjr There's no other way to get that crunchy ceramic shard flavor we all know and love. Am I right?
Some of us never moved beyond a clothes line
Read the source article, MS is not kicking 3rd parties out of kernel space. They don't even say that. They say restricting. Details matter. Plus, even if anti cheat were kicked out of kernel mode, either A) the devs will still block linux with user space anti cheat or B) (Most likely outcome) the new security API MS is talking about implementing for them to use will also not be compatible with linux.
The new security API if used from userspace should be manageable to implement into WINE. Definitely much more manageable than kernel-level stuff which, as the name suggests, requires the kernel.
@@rtzgf67games7 security api even used from userspace will have a kernel component. That said we don't know yet, what we know is MS isn't kicking them out of the kernel and EAC and BattleEye already have an easy way to implement wine/proton support and actively choose not to. Not to far fetched to think they would even let the Linux community play even if we did implement the security api (which we can't without the kernel side of things)
Tell me you dont understand the promlem with linux gawing without telling me you dont understand the problem with linuxgaming...
Hint:
Its a lot easyer to emulate a few API's than half the kernel.
@@bas1858 easier =/= possible. Publishers want kernel level anti cheat. If wine was able to emulate or simulate the new api and make the anti cheats function that way then game publishers would likely start "actively" blocking wine from running their game.
Look at Fortnite, uses eac and battle eye. EAC and BE both work in Kernel mode in Windows, user mode in Linux. Because Linux does not work in Kernel mode, Epic actively blocks it from running on Linux. New API is not going to change this.
Everyone complained about VAC while Valve has slowly been building up the most robust non-rootkit AC solution for over a decade
Vacnet 3.0 🤞
@@Hexnano vac sucks but it's getting okayish with every new update
Trust factor system is the only thing actually separating cheaters and legit players in CS2, anyone developing external or internal cheats for cs2 has a trivial time avoiding VAC's detection vectors without ever interfering with the anti cheat modules themselves
The issue is less VAC and more that... Valve is slow.
Vacnet 3.0 is blind to cursor pixel level detection written in less than 10 lines of python code
The Microsoft blog post explicitly states that they are not removing the ability for vendors to write kernel-level code of this nature, rather that they are looking into adding more security features for vendors to access in user space. The original article is misleading if not downright clickbait. There's no reason to think that this will result in less kernel-level anticheat and the hype around this news is a dud.
EDIT: I misread a quote from one of the security vendors, ESET, as being from Microsoft. Microsoft were not the ones that stated that it was “imperative” that access to the kernel remains open for vendors. However, still nowhere did the blog post imply that this access was being locked down or removed, but rather that they are looking to add additional options in user space for vendors to utilize. Sorry for the misread.
Thank you for this precision, bc I see most people over-interpreting this microsoft post
That's one important detail the changes everything. Thank you.
Thank you. You just saved me 11 minutes of my life and now I get to dunk on my idiot friends who are more driven by hype than facts.
Well... Microsoft will probably have to maintain some kind of compatibility for a while... does that mean they won't deprecate these methods eventually ? And even make it impossible some day...?
While the article is misleading, I think this is a move towards fewer kernel level anticheats. In one of their dev blogs, Vanguard developers mentioned that if Microsoft provides a framework for verifying kernel integrity, they would be more than happy to restrict Vanguard to run only on game launch and in user mode. My prediction is that Microsoft will have a setting to block unsigned kernel modifications, and games will require you to have that setting turned on to play. OEMs and vendors that require kernel extensions will go through Microsoft similar to how firmware updates and the like are signed through Windows Update.
"I was right!" - guy who's wrong
REAL
CrowdStrike playing 4d chess to boost Linux desktop adoption confirmed
@@Skidmaster180 it’s funny because crowdstrike had the same issue on Linux a few months before it happened on windows. We just didn’t hear about it.
@bjornarnols5905 The problem with Linux is that it's for very tech literate people. The average person cannot work with Linux, and few can do troubleshooting. Grandma considers windows confusing. She's not going to use Linux.
@@the11382if your grandma just uses a web browser it isn't a huge issue. Most distros come with a GUI based system updater anyways, and from my experience the likelihood of a failed system update is significantly lower than on Windows.
@@the11382 Same with Chrome OS. "Change is hard"
@@the11382 I agree Linux does need to become more user friendly (as always), but also counterpoint: People who aren't tech literate probably do less complex things with the tech. Grandma probably just needs a browser and MAYBE a file manager, both of which Linux offers plenty versions of.
- A Fedora user
- "Client-side data should never be trusted."
- "Understood! Let's run it on the client even harder - In kernel mode! What could possibly go wrong?"
it still on client tho, u can't manipulate the anti cheat using a usermode cheat, but a kernel leven cheat can access and manipulate the anti cheat data
Did you notice the quotation marks? He wasn’t approving of the behavior of the second person in the fictional conversation he depicted.
opposite, kernel level anticheats exists, because kernel level cheats started to exist.
Client side data isn’t trusted. That’s why you don’t see hackers noclipping in every game you play. They can check for noclipping server-side, so they do. Theres a ton of game-ruining cheats that you can’t detect server side. Most notoriously being aimbot. The client looking at enemies and shooting is completely valid client side input.
@@ItsAllEnzynes I think you replied to wrong person.
thats fake, just double check info please, not just read notebookcheck and suppose all written there is truth
whats wrong about this?
That tweet at 10:46 is unbelievable bs. Using Linux is not the equivalent of running away, it's the equivalent of vote with your wallet. We are still here and part of this society. We just don't want spyware, ads and security/safety issues in every single aspect of our lives.
Whats funny about that specific tweet is that its made by Tim "Epic Games" sweeny. The guy accusing steam of a monopoly while using monopolistic practices to try and goad players onto their store instead of giving a good product.
The tweet literally comes with an asterisk of "*please use our store instead of that big meanie Gabes!"
What the fuck does he mean "the freedoms we have", my brother in Christ, have you read the Microsoft TOS?
Wow, since when does windows stand for freedom? And why is Windows’s one sided bs considered politics? They are shoving unwanted features down our throats without any discussion. Switching to Linux is voting with your wallet (PC) about what you think is right and better for you.
Would you expect anything more from one of the biggest Windows fanboys on Earth?
It's shill language, the guy is financially invested in keeping the status quo, and will say anything to keep it that way.
You should've read Microsoft actual blog post, they're not removing Kernel level access. They're going to make accessing it 'easier' and safer, with better tools and ways of doing so. This is to prevent another Crowdstrike.
He doesn't seem to have addressed this error yet.
@@thefearlessgeek Because people who have a hard on for hating kernel level AC actually don't care about being informed
he's terminal
@@caracentral there's a reason on why it's hated lol
@@mauriciobenitez5460 yeah because people are braindead sheep? Having A reason doesn't mean it's a GOOD reason
Afaik valve actually dont have a kernel mode AC, because it would prohibit their own linux based devices (the steam deck) from playing their own games.
This was a misspeak he corrected in the pinned comment, he meant valorant
Oh they could definitely make a kernel module or something that people would have to install. But they don't want to (thankfully)
I'm pretty sure they _could_ write kernel level stuff for Linux, but they don't because 1. they don't want to, Valve has gone on record that they don't agree with the philosophy behind kernel anti-cheat. And 2. they'd have to maintain the two anti-cheats independently of one another, as the Windows kernel and the Linux kernel are obviously vastly different beasts.
It's absolutely possible to run kernel anticheat on Linux, just not Windows kernel anticheat. If Valve wanted to do it, they could.
Correct answer wrong reason, many anticheat do support linux
DO NOT PUT YOUR CLOTHES IN THE DISHWASHER. No one cared about the clothes because the dishwasher broke lol.
(((Uses heatgun as a hair dryer and a soldering iron as a hair curler)))
Valve doesn't have a kernel-mode anticheat
In fact, Counter Strike is completely unplayable above a certain rank and everybody goes to faceit instead
Counter Strike's current state is that competitive integrity is completely absent when a kernel level anticheat is absent too
I don't understand why your t-shirts get washed and not destroyed in the dish washer but when I put my dishes in my laundry machine they do get clean but also are broken.... There must be a little discrepancy somewhere, I just don't know where.
I feel like you can probably write a unit test to help you verify what the issue could be. PM if you need help
@CallousCoder you only can use plastic in the washing machine.
Why does the dog bark when the doorbell rings but the doorbell doesn't ring when the dog barks?
@@StrawberryKittenSounds like a bug with the doorbell, is dog barking UB?
@@kuhljager2429 oh I always throw away my plastic cutlery as it says single use plastic 😜
"Jenshin Impact" is crazy
might explain why they remove kernel level access. game works on linux for awhile now
😂😂
I was looking for this comment
Gen-Z Impact 😅
@@afjer me too lmao
@@DarkblooM_IO is it? Seems like a pretty understandable grammatical error. It that qualifies as "crazy" for you I'd suggest getting out of that bubble you live in.
6:27 device drivers should probably also be running in ring-1 instead of ring-0 to allow the OS to recover from failing hardware
Based Crowdstrike
they succeed at failing
I will thank them for that
Bringing in a security module that verifies the boot state and provides attestation for applications has serious implications on control. It sounds great on the surface, but this is exactly what android is doing and manufacturers use it to lock down their devices, making it so you cannot modify your device or install custom firmware on it, which limits control and turns it into e-waste in some cases. This sort of thing would pave the way for desktop computers coming with locked UEFI's where the only operating system you're allowed to install is Windows for example. You cannot trust companies to do one thing and one thing only with this sort of control.
Instead, we need to encourage companies to invest more in server-side protections instead of being lazy with their security and instead wanting to force users to run their apps in an environment where this sort of thing is not possible.
all of this to reduce the likelihood of cheating in online videogames. not at all worth the tradeoff imo
Locked UEFI? Please don't give them ideas. I always wiped the whole OS and install Linux anyways
They've tried that before with Windows RT devices and it did not work.
Kernel mode anti-cheat is like blowing up your house to kill a mouse in the basement
I've never heard of a kernel level anti-cheat causing major problems, but it could happen.
If you are seriously concern, play games on a spare computer. If it does go boom, you're good. And you aren't slowed down by the 30 different gaming applications constantly running.
Well, don't think so. Rockstar just broke GTA 5 Online for everyone on the Steam Deck (and any Linux distro) by borking their implementation of BattlEye, which has native Linux support; but they didn't bother.
It makes sense that they don’t bother, even if I really want them to care.
I just checked Steam’s hardware survey, and they show that 96.78% of users use Windows, 1.3% use Mac and 1.92% use Linux (interesting side note , arch linux is the most used Linux distro among steam users)
Sometimes it takes a lot of man hours to make programs and games compatible cross platform. It’s honestly not worth it in certain scenarios to put in effort to support Linux.
My biggest hope for Linux gaming is honestly Steam OS. Valve announced that they will allow other handhelds to use the OS (like ASUS ROG Ally).
If PC handhelds becomes the affordable and portable option for PC gaming, then SteamOS and Linux might achieve a big enough market share to force developers to respond to the market.
arch makes sense considering steamOS is based off arch @@Ellefsen97
@@Ellefsen97 It's just a fcking check in the BattlEye settings to enable linux proton support.
@@maledict9061Apparently having one person who is paid $30/h check a box isn't worth tens of thousands of dollars in return
yeah we're not asking for a port lol
Clothes in the dishwasher can cause fires due to cloth in contact with the heating element. YOU WERE NOT RIGHT.
Big Clothes Washer propaganda!
@@thecodemachine your dishwasher has exposed heating elements? Why?
There is no heating element in ANY dishwasher!
@@MusingsAndIdeas Then how does the water get hot?
@@SianaGearz Because it gets the water directly from the water heater.
9:17 never heard it pronounced “jenshin impact”
Then you've never seen the annual game awards XD
That's how it's pronounced
@@rohithkumarbandari げん 神 しん is pronounced GEnshin like geh
at the pearly gates, he'll call God like "Jod" and will be sent to Hell for all of eternity
Found the jif sadboi. Honestly though, it's crazy to me that there's a TH-cam creator that hasn't seen/been conditioned by like 17 million videos sponsored by them. Seems like they sponsor half the content on here 😩💨
I'm generally happy, that I'm not a competitive gamer, so I simply don't play any games which have Kernel Level Anti-cheat. So Linux for gaming works 100% for me.
Sadly it's not just comp games that have this.... Which is why i really hope kernel level ACs do not become the norm...
04:35 I’d like to see a standalone video on the topic of pre-kernel OS’s and the eventual rise of a kernel OS. More of that in detail please!
why is this video still up if you found out it's 100% wrong
microsoft removes kernal bs and android tries to kill third part app installs. What a time to be alive.
What ? when did android tried to kill 3rd party app installs?
im out of the loop on android. did a quick search and didn't find anything. got more info on that?
At this point we can't trust any employer let alone game developer if they wanna go this route and want access to your personal data time to start living in tents to get away from all of it and turn on Minecraft survival mode in real life.
@@bigbossgaming7663 it is on android 15, they are trying to discourage people from installing apps that are already on the Play store, so it blocks you from installing it, that's what I've seen so far
Please stop lying about Android
The definition of kernel that you gave is a modern redefinition of the term. DOS had a kernel. In MS DOS it was MSDOS.SYS. A kernel is simply the core functionality of an OS, i.e. the part that handles system calls and interrupts. The part about memory and process separation wasn't possible on x86 until the 386. Redefining words is all the rage these days.
starting the video by saying you were right about everything and then being wrong about everything is truly crazy, common clickbaiting youtuber L
maybe he will write a note about it? but I doubt it.
11:11 none of those video match what you said 💀
every step we can take to linux being able to become mainstream is one step closer to a more consumer-friendly pc experience
Reconciling the entire packaging system differences and distros would be nasty asf but gaming will most likely get some common standards up, here's hoping Steam gets us there as Linus Torvalds (jokingly) imagined
There's going to have to be a unified, supported platform first. The "anything goes" nature of Linux is in direct contrast to being able to manage third-party software distribution. I.e., nobody is going to publish software for an ill-defined target, and that is practically the definition of Linux.
Don't get me wrong. I love Linux, and I use it every single day. But there are several reasons it hasn't taken over the desktop, and the main one is ... nobody knows "Linux"... they just know _their_ Linux.
Lol Linux hippies
@@Spiritwonder420>anime pfp
@@nickwallette6201I agree. Linux suffers from issues that BSD and other FOSS OSes don't because of lack of unification.
Linux has a unified kernel with stability guarantees, but no unified userland, just trends among distros which constantly change. With how big Linux is, it would be really advantageous to have an "official" distro to target ABI and library-wise, and anyone who wants to do something else is on their own (alpine, NixOS, etc).
The freedom of Linux would be preserved, but new users wouldn't have to deal with BS like trying to run Davinci Resolve with janky scripts because it targets CentOS and they're using Pop
I really don't think anti-cheat is the big reason why many developers don't push their games to native Linux. I'd argue that the main issue with gaming on Linux comes down to historically poor driver support by the major GPU manufacturers, noteably Nvidia. In recent years there have been more and more ways to work around that limitation and now Nvidia seems to finally be coming around to the idea of open-sourcing their drivers which should eventually blow the gates wide open for gaming on Linux, but the lack of decent drivers for graphics cards by THE single most popular GPU company was/is almost certainly the biggest hurdle to native Linux support in gaming.
You are spreading misinformation as is your tech-bro source. Microsoft never claimed this. Please do your research if you go hands up claiming "I was right" and don't just rely on a single source.
To be fair, people who pat themselves on the back proclaiming they were right, were usually not right.
his fans are actually as braindead lmao its actually unreal
It's a shame there's only a few comments pointing this out
@@Har0x That how you see this channel is mostly followed by people that dont know shit and he just farm their views.
@@wellgun3 As i've mentioned in another guys comment, this dude is clowned on in malware development communities which are full of people who are known for creating some of the best C2's and exploitation frameworks out there currently. He appeals to people who have 0 knowledge on the topics but has absolutely no substance when talking to people who actually know their shit and do it daily for a living.
The fact that a company named "crowdstrike" did the equivalent of going on a work crowd strike to protest kernal access by accident is really funny to me
The universe has a funny way of making the names of things relevant in very odd ways.
@@CD-vb9filike OceanGate and their controversy 😂
earliest I've ever been and thankfully it's to shit all over kernel-level anticheat
@@TheMCEnthusiastPlays You definitely cheat.
@@heatnup is this bait or are you serious
@@heatnup Nah - playing games should not require that I trust EA or Ubisoft or whomever with complete access to my computer.
@@heatnup hating kernel level anti cheat doesnt mean one is cheating. Kernel level programs in general should just not be so easy to change thus break systems (see cloudflare)
@@heatnupEA astroturfing bot? 🤔
People who non-ironically think the drivers of a mouse or keyboard equates in seriousness or risk to an online game should legally not be allowed to vote...
Nobody should vote, it's a scam.
Ever heard of a keylogger? I'm really curious as to why you think there's such a big difference.
so confidently wrong lmao pretentious fool calling others idiots when they are one
@@eekee6034 Not sure what that has to do with anything, if somehow they can inject a keylogger into the keyboard driver that is already passing every possible security measure. The only solution to the problem is to literally have no connection between the PC and the keyboard. But we can't have that, can we?
And Keyboard drivers are less regularly updated than a videogame, and their anti-cheat. And of course the more games need it, the more folds do these issues increase.
So no, you can't compare one neccessary piece of software for keyboards and mouses to function which you should have very few of, to an unnecessary piece of software that you need to have multiple of, and gets updated very regularly by companies which writes these software on the side, not as a main focus.
@@katamas832 Thanks. That is a fair point with having multiple keyloggers which don't need to be there, and too-frequent updates can be trouble.
I'm evidently a bit out of touch when it comes to driver security measures. On driver quality though, I'm pretty sure most hardware manufacturers don't write the best drivers. They certainly didn't in the 90s. Keyboards are probably all right though. The driver for a standard USB-HID keyboard should be high as it can be coded by the OS developer and one driver can cover almost all keyboards. I'm not quite so sure about keyboards with extra features.
Problem is that cheats are going the same route.
Game devs don't want to spend time and money on anti cheat, it's the gamers requesting it indirectly by wanting to have a fair game.
Reality is that cheats are kennel mode or even hardware side.
Without deep access to the OS it becomes virtually impossible to detect them.
The alternative is that games will be full of cheaters.
Pick your poison.
It is worth noting that in the original microsoft blog ESET said that they'd like if there was still some level of kernel access for security solutions, and that they never explicitly said they were removing any security functionality from the kernel, only that they were adding more outside of it.
Userland business logic has no business in kernel mode. Full Stop.
1.) Too volatile a logic framework
2.) Too many cooks in the kitchen
3.) Too much risk of code eats dog + world
4.) Too Much Separation of agency from responsibility -- Microsoft was the ultimately responsible entity, even when the 3rd party wrote & shipped crap kernel code.
"WE DO NOT BREAK USERSPACE!"
- Angry LinuxMan
I still remember how Riot Games said "the reason we created Vengard is that Windows doesn't give us better tools. Which they can, and it is probably a matter of time"
Aka, they knew it is a temporary thing till Microsoft will make it so they don't even need Kernal level anti-cheat
3:18 Pretty sure Valve has a strict no ring 0 anti cheat policy, they want to combat this with machine learning which seems like was the right move
@@ShashaParallax lol their games are rife with hackers. What are you talking about?
and guess what ? It doesn't work! CS2 is full of cheaters!
It is the right move. Letting anti cheat for a video games to have full control of your system is the same as giving your key to a suspicious stranger to clean your house without limit.
Maybe. But for the past decade it definitely wasnt
@@DiogoIEFPhri So are the games with kernel level anticheat.
The biggest thing about the whole thing is that the Linux CrowdStrik errors a few months before the Windows outage was an eBPF program that crashed the kernel... Something that eBPF was suppose to prevent.
So even an eBPF solution on windows could suffer the same way.
Says, "i was right, i love being right" in the first 2 seconds yet he was still wrong XD love to see people getting jebaited by people who have 0 reading comprehension.
We are going places boys.... get your .iso ready its penguin time.
Kernel authors slowly learning about the advantages to micro-kernel designs...
If there was an API for interacting with the kernel and detecting these things without running in kernel mode, then this use-case could still be supported, for the most part. eBPF and the ideas behind that are promising, and I like how standardized it seems to be. Maybe "kernel-level" anti-cheat can be cross-platform if the kernels exposed a common interface.
Although I don't cheat, it wouldn't be too hard to compile a kernel with an eBPF that would lie to anti-cheat, and swapping kernels on Linux-based operating systems is very easy.
Having a game to run anti cheat software is like the power utility company to forces you to put a camera on the meter, because they're afraid that you might cheat, but it's in my home I don't like it. Also, buy separate private VIP servers and play games without cheaters, forcing everyone to have something that will eat more computer resources just to check if every client request is legit is a nonsense, mainly because this is not a banking software it's just a game.
@@DamjanDimitrioski Completely different examples, the game running the anti-cheat is completely fine, its the fact that some of the anti-cheats are kernel-level which is not. When you buy/play a multiplayer game you expect a fair game for everybody, no cheaters. Private VIP servers have a whole host of problems, cheaters don't mind paying money for things like this. Take CS2 Prime as an example, you have to play a considerable amount of time or pay for Prime, but there are still so many cheaters in Prime because people don't mind spending money for an account. It's also just outrageous to have people pay to not play with cheaters, if you've already bought the game why do you have to pay more to not play with cheaters?
Hey, a common interface for kernels does exist, it's called POSIX and almost everything that isn't windows complies with it.
OS developers have all known the advantages of microkernels since the 70s. Somewhat like the Rust Evangelism Strike Force in recent years, it has at times been hard to escape microkernel proponents :) The trouble is it's really difficult to design a microkernel architecture with good performance. Not that you really have to design one, you can use the Mach design as IBM Apple and the GNU Hurd project have.
There's a bit of a funny story about Hurd. It's older than Linux; it was going to be the GNU kernel, but it wasn't ready for use when Linux appeared. Years later, Linux was running 70% of Internet servers and Hurd was still not ready! I thought I heard they couldn't agree on design details or something, maybe they wanted an independent design rather than Mach; I don't remember, it was too long ago. When it was finally released, it was just another Mach design.
But people do design their own microkernels, usually without taking quite as long as Hurd. :) Managarm is a POSIX microkernel OS that's been looking pretty good for a couple of years now, supporting Wayland and X11, but I don't know if it's reliable enough for everyday use. There's also ToaruOS -- if you've heard of PonyOS you've heard of ToaruOS ;) -- but its author doesn't really want it to be used seriously. He made it to learn about OS development.
@@hashcrash9729I think the only truly POSIX compatible shell is the shell BSD distros use. Bash is not completely POSIX compatible and you need manually or consciously exclude convenient GNU features for your script to be POSIX and run everywhere even MacOS. Let alone ZSH and FISH which are way away from POSIX.
I think windows (and linux if they want to support security paranoid game companies) having a launch mode that tells the kernel that this program should be protected from tampering by the kernel not exposing memory in any way etc etc then that seems entirely reasonable. "I need this program to run entirely isolated" sounds like a valid request whether it's for something a little frivolous like a game or putting in more kernel protections for a password manager process.
You don't want ANY program reading ANY other programs memory, or especially writing. If this was just a mode you could add it would be the default, and so, uhh, it kinda is the default. cheaters or whatever are purposefully getting around all the defenses, Microsoft can't really add new ones, or theyd have done it already
You can't prevent cheating. You can't prevent theft. You can't escape death.
But that's not to say you shouldn't try; just try wisely.
Kernel access being taken away also would kill a wide variety of cheats.
this
Cheaters can break windows to run illegal kernel code. But vanguard can't
@@rohithkumarbandari if players can do it then so can vanguard. The installer user has more powers than an admin account.
@turtlefrog369 If Vanguard AC shipped with such a thing, Microsoft would be suing the fuck outta Riot Games
@@turtlefrog369 vanguard *can* do it but no way would they open themselves up to so much risk. cheaters on the other hand dont exactly have to abide by that ruleset
Microsoft: suffers ONE kernel level security incident
Microsoft: that's it, we ain't givin NOBODY that shit anymore
Kernel-level anticheat can still say "no get lost linux" but at the same time, not being allowed to make kernel-level anticheat on windows does mean that linux can in theory be more malleable for a kernel-level anticheat, since you can freely distribute kernel modules.
One big risk is if Microsoft decides to abuse this and perform cryptographic OS attestation under the guise of an anti-tamper API (like Google similarly abuses on phones), making it so games will simply refuse to run on an OS that doesn't provide Microsoft's cryptographic signature...
Can be bypassed but would be a pain in the ass
Valve would totally take that to court in, say, the EU, saying that Microsoft implemented a feature that is being unjustly anti-competitive towards SteamOS by preventing Valve's platform from being used as an alternative for gaming. Microsoft could very well be considered a "gatekeeper" by EU laws. Just my conjecture.
It's great to see your channel growing fast being one of your earliests subscribers, you deserve, man. Great quality content
Remember kernel level anti-cheats only exist because people think they deserve to be better than others in games without putting in the effort.
Meanwhile i know games that dont have this that still have less cheaters lol
Lets not pretend like those intrusive anti-cheats work better at all
@@ToxicMothBoi what are you even on, have you ever played valorant? now compare the number of cheaters to cs2, it's not even close. if they take away kernel level anti-cheat, I will stop playing
i saw an article that said this change might actually negatively affect gaming on linux; instead of simply being unable to play due to kernel level crap, there'd be other hurdles to overcome as emulating all of the new crap might be difficult. and even then, companies could just add even more ways to detect linux and ban users. Not sure how accurate it is, but it definitely raises some concern for me, i want linux gaming to thrive as much as possible
@ALinuxEnthusiast fair point. so yeah, im still concerned; but thanks for clarifying.
The thumbnail looks like you're being held at gunpoint while the perps are forcing you to watch as they cook a motherboard in your oven
Mostly don't give an F. Everybody can argue but it's a Result Theory. That's what all matters, result. Other than that is anxious paranoia which can be all in the tech landacape and technicality.
Vanguard for example does it's job very well and brings out result.
All I care about results so I steal money.
"so I steal money" ->Which is illegal, Vanguard isn't.
Your point? Oh that's right you don't have one. I could do a better logical counter-argument against my post in my sleep.
@@w1z4rd9 Clearly it isn't just about results. That's the point.
didnt vanguard brick network drivers once?
hmm IDK, dude.. if you think about it, it's gonna be having a high security code installed by default, run by microsoft, waiting to be called by apps, with intrusive functions that will be required by devs like those from riot, ea... Up to now, it was optional. You can create a dual boot to install vanguard and that stuff in a different installation that you'd normally use. Now it's gonna be built into the same OS. Sounds too much like what happened with Android: an open os closed to a security OS to satisfy what devs asked for safety with, let's say, money operations and stuff.
Finally, locking something down being a good thing. Wanted to ask... when you put clothes in the dishwasher, do you put fabric softener, and regular clothing detergent? What about bleach? And where would you put it before starting the wash?
Dunno about the t-shirts dude xD
Thanks, CrowdStrike. Unironically.
A bit late now, but apparently they are only discussing the addition of new options outside the kernel. They wouldn't lock vendors out, but give them options to stay outside the kernel. Many vendors might still prefer using their own anticheat. It is possible that some are going to move to such potential alternatives.
They should focus more on single player off line games....
Doesn't bring in enough money nowadays. Without DLCs a singleplayer game is basically DOA in a service perspective.
@@Raphipod genshin impact.... literally single player, with just their shitty little multiplayer implementation that could be compared to inviting your friend to your self hosted minecraft server, its about the public i'd say
@@Raphipod singleplayer doesn't need as much service as multiplayer...
okay lets forget about the 100 million people who play multiplayer games, that sounds completely fair
@@Raphipod That's stupid shortsighted thinking.
DLC capability is provided by distribution platform. You don't need highly intrusive anticheat or always-online tether for that. Yes you can release paid expansions for the game, nobody will mind! Or you can do sequels and sidequels.
Why it's stupid, is because a person can spend 50-80€ per game for 6 games a year easily, a typical person. Then they'll have them finished in maybe 12-25 hours each, and then they'll want another game. All these live-service games... they rely on sucking your time completely dry every day-ish, every free minute, and rely on people becoming addicted and starting to whale and spend untold amounts of money on it. But most of your audience is spending barely nothing. Live service games are a big bet with high gain, high loss potential, a major risk, and there's a limit to how many such successful games a market can absorb. There is no such hard limit to self-contained games.
So it's smarter to partition the product space to where they won't compete. Many people will NEVER play a second live-service game, they'll play one, they'll become grated by it after a while, they'll find predatory monetisation abrasive, find it's not how they want to live, and abandon it, and will be wary further on out. They want self-contained experiences tailored towards them, where the spending decision is frontloaded and separated from moment to moment game. Single player games are relatively low-profit but also relatively low-risk if you know your audience. Another self-contained single player game is much less likely to hurt your vaguely similar game's sales prospects than a live-service game is likely to hurt other live service game's sales prospects, since every live-service game wants all of your time, while a self-contained game only wants a manageable chunk of it.
No, you were not right. Check your sources. Nowhere in the Microsoft blogpost does it say what you are talking about.
@hagnahugna2948 It would help if you said which information you disagreed with. What you said is almost impossible for him to realistically respond to, and just dilutes genuine criticism to the detriment of the community.
@@superkingofdeathify the information that makes up the bulk of the video topic. he is just blatantly wrong and his source proves that
shame you didnt even read the source article
The moment this goes out, Rockstar forces BattlEye on GTA V. Good lord.
I literally just reinstalled it on my Steam Deck to play GTAO on it...
@@Duckly97 well too bad now you can't
what's confusing me is why they waited this long to implement kernel level anti cheat
@@curiousmind_ not like it'll work either lol it's a peer 2 peer game
@@memes_gbc674it could work a bit, some gta v cheat devs aren't willing to go to make a kernel driver to continue their business already, it's gonna take some time and cheats are gonna be paid now
This does nothing for Linux, because it's simply not possible to add such a security sensor to it. What should that sensor attest? That no processes run with kernel access except for the kernel? No problem, just patch and recompile the kernel itself then. There's nothing a TPM chip could do here except maybe as a non spoofable id to hardware ban people?
good to know you weren't right about this, the dishwasher is saved from flatmate's skidmarks for another day
Then Rockstar proceeds to mess up GTA V on Linux compatability
All of this over cheats!?!???? Even more reason for me to never play online games. This is just too much to ask for and i don’t trust game corporations, or microsoft but game corp are even worse
i feel like this mean literally nothing most of the commonly used kernel level anti cheat software works just fine on linux, and game devs just disable the ability for their games to connect to client using linux because.... reasons??? anyways the barrier is not anti cheat software developers not wanting to support linux they already do the problem is game devs not trusting linux and this doesnt fix that.
Anti cheat protected games aren't coming to Linux any time soon even if kernel is removed. 'Emulating' windows to pass all their checks is never going to be feasible or reliable.
Have you ever heard of wine and/or proton? Lots of games have linux anticheat, and usermode anticheat can translate your honest linux setup into an honest windows setup. The problem is you can't virtualize or lie about kernel mode, CPUs aren't designed that way, but usermode is entirely beholden to what the kernel says, and a kernel level wine extension is an option.
If you're just going to read someone else's Reddit posts at me I'll do my own searching, thanks.
I love how you explains everything as if I don't know anything. I’ve learned a lot here. Thanks!
wasn't this proven to be clickbait? microsoft has no plans to kill kernel level anti cheat...
Lol youre wrong dude. This article has already been debunked. Gaming on linux wont change.
Easy Anti-cheat does have a fully functional usermode AC for Linux system.
But because Linux is so powerful in installing kernel modules for cheating lots of games straight up refuse to support Linux at all.
Even when there is a potential to ship games on Linux if "Microsoft actually implement it" then uh
it still doesn't sound viable if there isn't a way to verify if the Linux kernel ship with a dedicated module for cheats
Hypervisor-level cheat kills kernel-level anti-cheat, SMM/ME-level anti-cheat kills the system. 😉
This. Whoever thought was a good idea to make a kernel driver to power an anticheat will think is a good idea to abuse Ring 0
@@fellipec hyper-v and other hypervisors run at ring -1 but ok
@@t_Geckowhat ring -1? There's only 4 rings in total. With ring 1 and 2 removed in x86s.
@@aominnomad -1 isnt a "real" ring but its referred to when talking about hypervisors since they have more control over the system than normal kernel code
u misunderstood everything
6:19 Most drivers are user mode these days anyway. So your mouse is not using the kernel mode driver. That is the best thing that Microsoft did to the Windows, they tried to push as many drivers to user mode.
Absolutely based! God i hate being nice to M$ but credit where its due.
i dont trust games that want kernal acess so i got a separate pc for my personal and one for gaming with no persnal data but i just recently started becoming aware of my digital privacy
@@RupertBundem wow people want to stay private and safe in the cesspol that online computing is? Crazy!
@@nikoraasu6929 You're not the protagonist, main character syndrome, nobody cares about your existence and wouldn't ever steal your information.
The solution is load and store ports of the CPU to have encryption capabilities and putting in the decription table of the process and threads the keys for what it is supposed to access only. It would make the memory opaque for all other threads, with the keys relocated every 300ms like what windows does already for the "GDT" like data structure.
Microsoft is giving developers new ways to interact with the operating system. Not removing acces to the kernel. Nice click bait tho.
Vanguard once thought an Asus driver was suspicious and just blocked it. My motherboard is Asus. It completely broke my OS and had to format my PC in order to get it to work again.
A washing machine use (somewhat moderate) foaming detergents for suspending dirt out of fabric surfaces, with no or very mild enzymes to not destroy the fabric integrity. Their pumps are designed to work with and tolerate a significant bit of foam - they use a propeller'esque or water wheel style impeller on a regular axle, driven by cog or belt by power of an aircooled motor.
Dishwashers are made for hard surfaced materials, and their "detergent" is _hella corrosive to organics._ It will utterly destroy the integrity of any organic fabric fiber, and most polymers used for clothing. Dishwasher detergent work by that corrosive mechanism, and reuse most of the same water throughout the cycle - it's not like a traditional (foaming) soap. Their pumps are submerged, impeller type cooled by the water and *_DO NOT_* tolerate foam/air. So dishwasher detergent with clothing is a no-go, and if you put a foaming soap/fabric detergent in the dishwasher, the submersion dishwasher pump will burn out. Like a watercooling pump without water.
Game companies about to make their own OS as force you to boot into that instead. I sure hope not, but I wouldn't put it past them
Do you mean consoles?
@@kaitsubaka yeah, pretty much, just even more privileged to the hardware level
FINALLY, someone else understands dual use of the dish washer!! You're a scholar and a gent.
Sigh, isolation and formal OS kernels predate linux and windows by decades. Hilarious that the industry is still arguing about issues that were well understood in the 70s.
I thought valve doesn’t use kernel level anti cheat
VAC is kernel mode, you’re misinformed
@@geekswithfeet9137It isn't, VAC is userspace.
You can even play CS2 without evem admin permissions, so it can't even install any driver.
@@geekswithfeet9137 VAC is usermode, you're misinformed
@@geekswithfeet9137 your misinformed the first result on google isnt always right, valve said they will NEVER use kernal level anti cheat your misinformed
@@geekswithfeet9137 good to know, how is that different than riot games then ?
why use a random guy on reddit as base for your argument...
Son is *WILD* with that claim about washing Ts in the dishwasher tho
I don't care if I can't play cod or rainbow 6. Anti cheat software will never get near my kernel.
That's fine, a lot of people chose differently.
Rainbow 6 doesnt have kernel level anti-cheat.
@@Gigopz rainbow 6 uses BattleEye which is indeed a kernel level anti cheat.
@@roborob347 Nope, it's only stat bans and not root level like Vanguard, every anti cheat is technically kernel level (otherwise they are useless) but most of them aren't ring 0, the new security updates coming to windows only affects root level anti cheats like Vanguard, Richochet, Genshin and what ever the new EA one is called.
@@RupertBundem Nah he is right, I dont play a single game with a single root kit anti-cheat, I would rather have a few cheaters.
How about some proper title instead of this clickbait?
Gonna just skip this vid.