An article from this weekend that seemingly got buried by soundbites about the Steam Machine price in the same interview, but given that we have no information on price, this seems way more interesting to me. I mean…I basically self-select games that don’t use these kinds of anti-cheat at all, but this is important information for a lot of people, especially if you’re looking for an off-ramp from Windows and still want to play some of the most popular live service titles.
It affects only the most cancerous type of anticheat that’s been bypassed for a decade and introduces huge risks to your PC - Kernel level anticheat. People should stop playing any game that has such anticheat.
People should stop playing any game that has such anticheat.
Another reminder that markets don’t self regulate.
I’m with you, but you’ve got a lot of people to convince. A lot. The people playing those games make up the majority of the market.
Umm, actually. All proprietary software is unethical and introduces huge risks to your PC. People should stop playing any game that has such closed closed source.
Dude. This is such a dumb argument. Kernel level anti cheat means you got code running at kernel level. See the whole crowd strike fiasco.
Having a third party updater attached to a kernel that could be hacked or tampered by malicious worker at any time (see Crowdstrike) seems riskier than just running teams, but fair point
Yes, it is riskier. That’s not my point, though. It is that telling people “just don’t do that” isn’t helpful.
This is a huge deal for Linux gaming.
Please make this optional. I’d rather not have any third party kernel modules mucking around in my OS. I don’t use anything the requires this.
It is optional what games you purchase and install
Well yeah of course it’s optional already. If you don’t want that then you just don’t buy those games.
Currently kernel-level anti cheat isn’t available for Linux, so games that are released with multiplayer support don’t require it (e.g. games that enable Linux support in EAC).
If kernel-level anti cheat is supported by Valve, many of those games will start requiring it. So if you don’t want kernel stuff, there’s a real chance this development will reduce the number of available games in the future.
Except people are going to still buy those games, still complain for something to be done and when the potential resolution is there, they’ll go “I DON’T WANT IT!” and just cycle through.
Fuck sakes, some people…
The article only mentions TPM not 3rd party modules. I’m guessing the idea is that you will have to run an approved kernel.
Kernel modules can be installed, loaded, and run without a reboot in Linux. TPM support would just ensure that the firmware/kernel and modules loaded at boot are expected.
Edit: basically, TPM support wouldn’t really do what a game dev would want for a kernel that can be modified at runtime, unless I’m missing something
Yeah, unless valve starts restricting root access when secure boot is enabled, you’re probably right.
It kind of bothers me that people are putting the responsibility on valve for this, when the companies themselves have purposefully not enabled compatibility in most cases.
30% cut from developers. Steam machine. Valve is working together with anticheat devs on this, not alone
They haven’t enabled it because they don’t get the same level of protection on Linux as they do on Windows, so Valve is trying to address that.
In case this is serious, kernel-level AC has been shown to not be particularly effective. There were people with hacks for BF6 before it released, for example. Them blocking an operating system doesn’t prevent cheaters. It only prevents consumers from having options.
Of course I’m serious. “Not 100% effective” is not the same as “not effective”. And to be clear, I hate it and do not endorse it. I will not buy any game that goes as far as to use that kind of anti-cheat. But developers use it because it’s more effective at catching cheaters than not using it. All downvoting me does is cover your ears to what’s actually going on. There are a number of big live service games that once enabled Proton and have now disabled it after cheaters took advantage of the more lax security. They would not cut off a portion of their customer base if they didn’t have to because user space in Linux was somehow just as effective as the Windows variant that lives at ring 0 in the OS kernel.
In this instance not effective is 100% not effective.
Both kernel and non kernel anti cheat are equally effective in actual practice. In both cases your preventing kids, lazy and low knowledge users from cheating. But anyone who is willing to spend any amount of money to cheat can easily find someone who will provide them with a bypass.
In both cases the anti cheat is only as good as the on going support from the devs of both the anti cheat and the game.
You can’t control what a client does end of the day
I wish you the best in convincing devs with the data in front of them that there’s no difference between the two, but they seem to have data that indicates that they see fewer cheaters with ring 0 anti-cheat than when they let Linux players in with user space anti-cheat. If it were true that there’s no difference, surely Valve’s engineers could convince them of that, too, but that doesn’t seem to be happening.
The reason they do it usually is because some executives hear the Linux is less secure and that it’s only a small segment of users. It isn’t because it’s effective. The games that blocked Linux are almost all some of the games with the worst hackers. Guess what happened when they blocked Linux? Nothing. The number of hackers that were on Linux were near zero.
The issue is they cant be bothered to put the actual money/work to create a solution that’s effective. Instead they signal to their audience that they’re doing something by removing Linux, which doesn’t cost them anything and makes a show that they’re actually trying. It doesn’t fix the problems, but they get to make a show out of it.
Please cite sources for any of that. Game companies aren’t in the business of losing money. If they could make more money by supporting Linux customers, they would do so, and I’ve never heard of a gaming company’s executive ever mentioning anything about Linux except for Gabe Newell, openly or behind closed doors. If they wanted to make a big show of getting rid of cheaters, they’d never have enabled cross play between consoles and PC in the first place. They openly tell you why they don’t enable anti-cheat on Linux, in a way that’s beyond just being plausible, and you refuse to believe them. You’re only going to be surprised when this continues to happen even though the answer is right there.
I have to cite sources but you don’t? One example is Rust, a notoriously hacker filled game.
Of course they’re trying to make money. I literally explained that. The executives see Linux as not providing value, and it’s extra effort to support it. They’d rather instead use it as a symbol of how they’re actually trying really hard to fight hackers, but it’s a lie. It’s just a convenient excuse.
You haven’t heard an executive say almost anything. They run companies. They don’t publish their every decision. They are the ones making the calls. They’re the ones responsible. They’re also largely technologically innept. They probably don’t even know what Linux is. They just know what they’ve been told.
You’re only going to be surprised when this continues to happen even though the answer is right there.
There are like two major companies doing this. There’s EA and Riot. There’s a tiny minority of minor players, like Rust. There’s also a lot of Chinese companies doing it. (China is infamous for having hackers, so yeah, didn’t solve that problem did it?)
I can’t tell you the last time I booted up a western game and it didn’t work on Linux. (I think it was Squad44, which then added support, and support in the main Squad game has been in for a long time.) Everyone is moving toward supporting it, not away. The only places it’s an issue are large slow companies where the executives have too much control.
Your explanation is bordering on conspiracy theory, so yes. Rust cited why they cut support, as did Apex Legends, as did GTA Online. The rest often don’t even bother with supporting it in the first place because of how it always plays out. The existence of hackers at all doesn’t mean that Linux anti-cheat is equally effective, and you’d know that if you read the write up from the Rust team.
guys this comment is clearly satire, why did yall downvote?
Poe’s law, maybe
I’m not sure what there is to gain by pretending that downvoting me changes anything.
Irrelevant to me personally but I’d like to see it cause more windows users to jump ship
TBH, I kinda get the feeling that’s what most of the hype surrounding the Machine is. People hoping it sells well, but not necessarily people planning to buy one for themselves.
I jokingly told a few redditors that they are doing A LOT of the marketing work for the Steam machine. They didn’t like that at all, lol.
lol, they should be proud to champion what valve is doing for the Linux world
That being said I actually don’t have a desktop and would totally buy a steam machine if the price is right
I have no need for Steam products. Gaming emulators have been all I needed for years now. I don’t play the latest games so it’s totally fine for me too.
Speaking of irrelevant
I have this feeling that even if valve makes it work, rootkit anticheat devs will push updates that intentionally make it not work again. Probably with more claims like the majority of cheaters being linux users
If you’d read the article, Valve says they’re working with anticheat devs to come up with a solution together. This can only happen with their cooperation, if Valve somehow could bypass it on their own that would represent a vulnerability that should and would get patched.
They’re not just making that up. Cheaters migrated to Linux because it was easier to bypass the anti-cheat protections there. If the anti-cheat is equally effective in both operating systems, they’ll have no reason to cut off a portion of their customer base.
I would love to see actual sources for this. On both sides.
Every time I’ve tried looking up numbers (usually because of a passing interest, and never any level of in depth research) I’ve come back with interesting tidbits like “the total number of cheaters banned in one month was greater than the total lifetime number of unique Linux users of the same game (sometimes an order of magnitude or more greater).” With that statistic being pretty consistent across games and time periods.
I can already tell you that you’re looking at wrong numbers and thus coming to the wrong conclusion. Of course the total number of cheaters will be bigger on windows, 94% of PC gamers play on Windows. That’s like saying Monaco is such a poor country because their GDP is only 30 million compared to the 30 trillion of the US GDP. Except Monaco isn’t poor, the US is simply magnitudes bigger than Monaco.
If you comparable numbers you should compare the proportions of cheaters in relation to the whole userbase of the platform.
So, do you have sources for accurate numbers?
I never said I have a source for accurate numbers. I said whatever numbers you’ve looked at so far are not the numbers you should be looking at and then added what kind of numbers you should be looking at.
Did you just make this up?
Of course not. Go read the reports of any developer who once enabled Linux compatibility and then disabled it.
Interesting. Pretty much no cheat developer is making cheats for Linux as far as I can see it. Tried googling for some and could not find any
Are they working on giving the worst and most useless companies kernel level access to my pc?
I would doubt it. I don’t even know if that’s a reasonable thing to do on Linux. I don’t see how it could work. Presumably they’re trying to work on something like they did for Easy Anti Cheat. That has a kernel module on Windows, but it doesn’t on Linux. I would assume they’re trying to work with EA, Riot, and maybe some Chinese companies to have their AC option work with Linux.
Perhaps. Of course, if you were able to type that sentence out, it also means you know what to avoid if that’s important to you. I will be, because it’s important to me, too.
On the one hand, I don’t give a fuck about anti-cheat, because games using the kernel-level version tend to be giant multiplayer cesspools of little value.
On the other hand, I want Windows to lose the war.
I hope Valve can find the balance between these two extremes.
This could be huge. I hope they find a decent middle ground.
They’ve worked on anti cheat support before. It still depends on the devs actually activating that support. That will always be the case whatever they do.
Surprised they (Denuvo, etc - not Valve) don’t just hard require SELinux.
SELinux protects systems from bugs in software. Not against users with full root privileges using their own hardware.
But anti-cheat is mostly about false sense of security anyway.
Still not going to convince some stubborn hold outs like the rust guy. Nothing will ever convince them.
The market will - and it’d be foolish to underestimate the forces valve will spark by making viable alternatives mass market.
I wouldn’t expect the Machine to be any more popular than the Deck, which already wasn’t enough to convince holdouts. In fact I would bet the Machine will sell much less than the Deck, since that had a more unique niche carved out for it.
I think the hope isn’t that “maybe this will be big enough”, but “maybe together they’ll be big enough”. Who knows, though. It got a lot of hype on reveal but people are fickle sometimes.
I would not be surprised if the work they’re doing here would be compatible with the Deck. It was just less of a priority for a handheld than a living room machine.
If devs want to support one, it’ll be no problem to support the other. But I doubt devs who already refused to support one will suddenly change their minds.
They refused to support the user space anti cheat. The work they’re talking about doing here is aiming to be the same sort of security they get on Windows. Low level. I have no idea how that works with Linux’s software licenses, but they said in the interview that this might be an exception made only for SteamOS.
When Linux market share hits 20% it would be a monumental achievement, and developers would probably still avoid it.
Don’t get me wrong, I moved to Linux this year. I want to see it gain traction in the gaming space.
It’s just not likely to happen any time soon. Loads of very basic use cases are a fucking shitshow because of a lot of reasons.
Just getting sunshine setup with a virtual display is a nightmare on Wayland without scripts to enable/disable displays and without being in front of the computer you want to remote to, because the simple logic of “if this display =off, then other display =on” is not a thing.
2 years ago, I would have agreed with you. But so much progress has been made and lots of devs have already enabled multiplayer support, it’s really just a handful that need to be convinced, so I don’t think 20% will be necessary to get there.
60% of anticheat implementations need to be fixed. 682 total titles. https://areweanticheatyet.com/
You just need to convince developers of a handful of titles, like fortnite, apex, valorant, BF2042, bf6, rust, R6 siege, league of legends, call of duty 2025… should be easy right?
It’ll never happen. The ones who are fanatical about it like the rust guy believe carte blanche that linux support will only make cheating worse and not positively improve the community. He doesn’t care about linux sales, the windows ones throw so much dosh at him that there’s no “market force” incentivizing him otherwise.
If Valve can make wireless vr AND fix windows only anticheat, both next year, I’m going to be 10 different kinds of happy. I would love to basically never need my W11 SSD ever again.
I consider it to be a function of when I grew up with video games and how my family restricted them broadly, but I have honestly never understood the appeal of competitive online games that require intense anti-cheat controls.
I grew up playing largely single player games, and the few online games I payed were limited to ones I played in private lobbies with friends i knew.
Any game that requires this level of policing for competitive play is an instant turn off for me. I realize I’m in the minority here, but I have no problem with a console that doesn’t support kernel level anticheat- to the contrary i find it to be a huge advantage
I get what drove us here. When you find a game that speaks to you and it’s got a ranked mode with good matchmaking, it’s easy to get lost in match after match, and cheaters can take the wind out of your sails. My competitive games of choice are fighting games, which are mostly free of cheaters and this invasive anti-cheat, but I’ll be bummed if it becomes the norm, because I won’t participate in that.
Are input macros in fighting games considered cheating or accessibility tools? I like the idea of learning fighting games, but with my thumb injuries, I literally can’t do the key inputs.
I assumed key input macros would be banned, so I never looked into this as an option. I remember hearing about upset when even official pressure-sensitive input controls on Dead or Alive 3 were banned in a major tournament in that game, let alone custom input macros.
People have all sorts of custom controllers with different button layouts. There are tournament legal requirements, but you’re unlikely to violate them if you don’t know what they are, and it hardly matters if you’re playing from home.
Have you tried foot pedal? I wonder if that is allowed in tournaments.
Steam Input would work, I’m pretty sure. I think I could overload all the buttons on the back so that they do all the complex D-Pad stuff. I haven’t tested if this is possible, but I’m imagining something like: pull starting direction from the DPad, press button on back for quarter/half/full/shuriouken movement, then execute with face button. So, like, DPad-Left+R3+A does full roll starting from facing right, then A.
Alternatively, I could load a radial menu on one of the touch pads with specific moves to execute in one click, with DPad input auto-toggling switching between directionally-correct radial menus (or maybe L3 as a manual toggle or something).
There are lots of options… Maybe I’ll try and see what I can figure out.
Have you tried using a keyboard for fighting games? That’s another set up that can help reduce use of thumbs, and closer to hitboxes so easier to pull off dpad specific moves.
Something I’ve seen commonly done is wasd for movement then using the numpad rows for stuff like punches and kicks.
Here’s aiming to be hopeful…
I remember back when playing DRM video in a web browser on an open source operating system seemed like a worrying impossibility. Many sites stayed stuck on closed-source flash players for that reason alone. It was a while before we ended up with this solution I only partly understand - where the DRM decoding is handled through some kind of trusted block, that generally doesn’t have full OS control?
Do you get full HD video from streaming services these days? Last I checked, the best of them only top out at 720p without Windows.
Yeah I get 4k with Firefox on cachyOS
Interesting. Did this happen recently? When all of the streaming services starting raising prices, I started cancelling. Which ones give you full HD? Do you need to go out of your way to get there, or will regular old Firefox do the trick? Does it need TPM enabled or anything like that? I was looking to re-up Amazon Prime in the very near future, but when watching on my web browser, a show like Vox Machina was just a blur factory, and it was easier to pirate the show than it was to stream it legitimately.
I really hope it won’t be a case of requiring a Steam Machine with SteamOS on there for this to work.
