Hacker News new | past | comments | ask | show | jobs | submit login
Steam survey shows Linux marketshare hitting 1.0% (phoronix.com)
401 points by bemmu on Aug 2, 2021 | hide | past | favorite | 354 comments



A headline in Linux gaming circles that hasn't hit the wider circuit yet (probably because it hasn't actually been delivered yet) is that Valve are working with Battleye and EAC to get anti-cheat working on Proton for the release of the steam deck.

If this happens it will be, without exaggeration, the greatest thing to ever happen to Linux gaming.

I'm not expecting a massive increase overnight, but it should mean that the marketshare will increase more likely than it will decrease.


I'm a huge fan of the work Valve is doing, both in gaming hardware and Linux support, but I'm unconvinced helping game-oriented malware run on Linux is good period, let alone the "greatest thing to ever happen to Linux gaming".


Correct me if I'm wrong but my understanding is that those are anti-cheat clients that are required by many games to be able to play online. Not being able to run on Linux means the game will never see Linux adoption.


The trade-off is that you have literal malware on the machine that is controlled by a third party company that most likely doesn't have their security practices up to snuff.


The malware is there regardless, this is about Linux marketshare. You can dislike it (and I do) without gatekeeping your favourite operating system because people want to run software on it that you don't approve of.


The problem with anti-cheat software is that it needs extremely high permissions. It wants to inspect every process, take screenshots etc. It's pretty extreme stuff.

And the stupid thing is that a lot of games install this software regardless of whether you actually play online or not! I've even had some games that insisted I enable Valve Anti Cheat and refuse to launch otherwise, even though I just wanted to play single player (I rarely play online). That check should obviously come into play only at the multiplayer menu.

But this is actually why I still game on Windows. I just have one Windows box which I use for gaming and gaming alone. I don't want this crap on my Linux and FreeBSD computers I do serious work on. So I won't be one of those Linux users in the stats :) Even though I use it daily.


But that is the nature of exploits. They often operate outside of a program modyfing its memory, applying visual aid etc.

Thus permissions like that are kind of required. Now, you can dispute if some random online lobby should perhaps have an option of "i dont care". However large group of gamers does care, especially in more competetive games as league of legends, csgo etc.


The trouble is it will be opaque and now I have to worry about one more opaque dystopic thing when installing games over Steam. I'll consider it Valve's transgression against me if they let me install anti-cheat malware without warning me any less strongly than screaming at me.


Sure it sucks but ultimately it's the user's decision and Proton will allow the porting of other Windows apps. If you can use it to run a AAA game it could easily handle business apps. I can't tell you how many times I've heard someone say they'd switch to Linux but can't because it doesn't run [insert some Windows app here]. It's a net gain for Linux adoption.


It’s not the users decision though. They don’t even realize this creepy software gets installed and they don’t know what it does


To be entirely fair, with BattleEye it's 100% clear to the user it's being installed. They have a pop up for every game that uses it that states it's to be installed if they click 'OK'.

But not all anti-cheat software does this, even those that install drivers, and it's not always clear.


Yeah. Nobody reads terms and conditions, privacy policies... At least a decade ago I remember seeing one that reserved the right to scan my processes, my memory, my file system, take screenshots at any time as well as exfiltrate all of this data to remote servers. Basically a RAT.


Oh yeah, except that big nice pop up.


This is sadly not true, since buisness apps use a much wider variety of windows APIs while games are mostly using a few (DirectX, OpenGL, Vulcan) APIs for most of what they do. You can see this with custom game launchers like League of Legends uses. The Game runs with basically native performance and without issues but the launcher is barely usable.


Plenty of key business apps use DirectX, OpenGL, Vulcan like CAD applications and games still use other Windows APIs. It's open-source, people can build off Proton to add the rest. These games for generally designed for Win10 now which support for is somewhat lacking in Wine. Concentrating on the latest version of Windows and APIs is a huge step in the right direction.


You're inverting the problem. The problem isn't that those things are used; the problem is they use lots of other things that aren't DirectX, OpenGL or Vulcan.


Don't get me wrong, such applications obviously profit from better 3D support in Wine but Proton won't make an impact on how well microsoft office or adobe cc products run in general.


IT all depends on how, where and when does this malware run. If they can be installed along with the games under a gaming only account on the Linux machine, that is, with zero permissions on system and other users files around, that's fine with me. However if they require to pollute system directories or, much worse, be installed as kernel modules or system level daemons, then it's way different. Security conscious gamers would probably put their gaming machine behind a firewall that screens their entire LAN (other hardware, NAS etc), and forget doing anything but gaming with it.


> If they can be installed along with the games under a gaming only account on the Linux machine, that is, with zero permissions on system and other users files around, that's fine with me.

Completely ineffective. It will be trivial to circumvent those protections. Might as well not bother.

The truth is cheaters are merely excersising their computing freedom. It's my computer, the game is merely running on it. If I want to read the game's memory and adjust aim or automate boring parts, it's my prerogative.

In order to prevent any of this, the game company must take over my machine. They must literally own my computer. Anti-cheating software is virtually indistinguishable from malware and there is no situation where this is acceptable. The game company's "needs" are irrelevant.


I think there is an underlying assumption that you’re entitled to play the game, therefor anti cheat is violating your freedoms.

But if we remove the assumption that you are entitled to play any game how ever you please i think this argument somewhat falls apart. If you don’t like their using anti cheat, then just don’t play those games which use it.

I just don’t think enough people agree with you that it will work in your favour unfortunately.


It's not entitlement to want to play the games I paid money for without my computer getting owned. This is basic respect, not some insane notion. Why are game companies entitled to complete control over our computers?


It depends on if you're talking about single player or multiplayer. Notably EAC and battleye are both only really used on multiplayer based games. Cheating on those games isn't "exercising your rights" because you aren't entitled to ruining others experience. Multiplayer is more than a simple "buy and own everything inside" experience because you are buying the ability to use their servers according to their rules and TOS. I support requiring the ability to run your own server without anticheat and full ownership and ability on those but you don't have the right to abuse multiplayer services just because you bought a license to play it.


Then don't give these companies your business. I certainly don't. There are enough games that don't pull this kind of shit to keep me entertained.


True. Nowadays I play single player games because of this. I simply can't trust these companies. The situation should never have gotten to this point.


Sometimes the malware is not disclosed and a game may start to use a malware after you bought it.


And the needs of other players to have a fair game are also irrelevant then?

After all, you're the top rank in CS:GO not because of your skill but because you simply exercise your freedom to have everyone else play against your computer.


Correct, they're completely irrelevant. Of all the affronts to computing freedom, anti-cheating is the worst. The truth is the online gaming model where we play with untrustworthy strangers is completely broken. We should all be playing with friends we can trust instead.


I wouldn’t install anti-cheat malware personally, but I think many of the people who want to play these games care more about not having cheaters in their games than having ring 0 control.

They’re extremely passionate about their skills and want nothing more than a fair playing field. Playing only with friends is not a real solution.

Computing freedom means you have the freedom to give up control, too. It’s your choice.


> Computing freedom means you have the freedom to give up control, too. It’s your choice.

Absolutely. As long as they don't force this malware on the rest of us who do care, it will be fine.


Completely agreed. Where is this "right not to be cheated in an online game" coming from anyway? Sounds absurd when you put it in proper terms, that is someone else taking control over your computer.


From the same place that real-life competitions tend to have rules about cheating that are enforced.

Why do people think that they get a free pass because they can't see the opponent face to face?

If you want to play with friends, plenty of multiplayer games offer that without anticheat (Arma lets you even uninstall the anti-cheat and play on anti-cheat disabled servers).


Does my local basketball competition requiring all competitors to submit negative doping tests before competing? Do they get to monitor me for 24 hours before the game?

Because that's the level of seriousness of most gaming and the proportionality of using a rootkit to prevent cheating. Sure, the Olympic athletes (or I guess in this analogy, eSports pros competing online) get the full "we're going to make very sure you're not cheating approach" but "my local basketball league" or "unranked ladder" doesn't justify these measures being routine



But if you play online with strangers, you're not really participating in a local basketball competition are you?


Why? Do I get to pick the people on the opposing team in the local basketball competition? I didn't specify a local club for a reason. Just because it's some competition organisers rather than a matchmaking algorithm doesn't change my perspective.


> Why do people think that they get a free pass because they can't see the opponent face to face?

We don't think that. We think anti-cheating rules aren't important enough to justify shipping literal malware to people's computers and taking over their machines.


It's not literal malware though, unless you have proof that it does something malicious like encrypt a users directory or exfiltrate sensitive data and that every single anti-cheat does that.


> Where is this "right not to be cheated in an online game" coming from anyway?

No idea. Who cares if some random guy cheats really? It's nothing compared to losing our computing freedom. Actual important stuff that matters. Same fundamental issue as DRM which nearly everyone agrees is bad.

Our computers are ours and game companies just need to deal with it. If people are gonna cheat, then so be it.


> Who cares if some random guy cheats really?

The kind of people who care enough to spend money on an online game.

> It's nothing compared to losing our computing freedom.

Then don't play online games with anti-cheat. Freedom doesn't mean your decisions don't have consequences—just you have the freedom to choose between the privacy of not submitting to an anti-doping monitoring regime or participating in certain competitive sports.

> Our computers are ours and game companies just need to deal with it. If people are gonna cheat, then so be it.

Game companies online services are theirs, and your just going to need to deal with it. If firms are going to require anti-cheat, then that's the cost of playing online. You have the freedom to not participate.


Don't wanna have anti cheats installed? Don't play (online) games, noone is forcing you, for me I'm really happy i have the option on Linux in the future.


You have the freedom to not install it.


What would be the point of a competitive game if you could just cheat? Racing games like Trackmania live off verifying people don't cheat because it's a game where people compete for leaderboards. Should they not be allowed to prevent you from cheating yourself to the top of the leaderboard.


They should recognize the fact that it's impossible to prevent cheating and stop shipping malware to users in an fruitless attempt to do so.


But it's not impossible to prevent cheating. CS:GO has a very low occurence rate of cheating as most cheaters are caught extremely early by anti-cheat or later by other anti-cheat efforts with more complex functionality.


CSGO actually have a lot, overwatch or valorant are way better examples.


Games like Trackmania should have SERVER SIDE data input validation instead of pushing rootkits.


They do in fact verify server side that everything is where it should be. But it doesn't stop people because ultimately the inputs of a high level player in Trackmania are nearly indistinguishable from inputs of a bot playing the track.

CS:GO faces the same issue where inputs by high level players are indistinguishable from cheating.

Speedrunning tends to have similar issues, where people perform tricks that are so close to TAS inputs that it can take years to spot an issue. Or in case of Dream, it takes an entire statistics paper to explain why they cheated. In those cases it's even worse because you cannot do server side validation on video data.

Being against anti-cheat is effectively being against competitive gaming or speedrunning.


"Online games are bad" is a hot take.


Yup. If you expect to be able to play with nice trustworthy people, it's horrible.


Not that I disagree but if you accept the EULA that states otherwise, doesn't that mean you don't have the prerogative? It sounds like you want to skip over the EULA, isn't respecting licensing models a huge thing for Linuxheads? Honest question.


> doesn't that mean you don't have the prerogative?

It means I've agreed not to cheat. I have the power to do it, I just agreed not to use it.

I won't let them take my power away though.

> isn't respecting licensing models a huge thing for Linuxheads?

Personally I'm not a fan of intellectual property in general. Licenses included. They all depend on copyright which should be abolished.

The only thing that makes any sense to me is the boundary between my computer and their servers. I should be able to do whatever I want on my computer. They should be able to do whatever they want on their servers. If they can't detect cheating server side, that's just too bad because they aren't gonna be doing it on my machine.


A lot of cheating just can't realistically be detected on the server.

Feel free to not play those games that requires anti cheats and let the ones of us who likes having it enjoy a game with less cheaters.


Nobody read EULAs.


The malware you’re describing would never be used. The designers of this stuff run it at ring zero on Windows with unlimited permissions and zero transparency


While true, and most of us here at HN will probably agree, tons of people don't care, and that's clear by how many people play games that require it.


I care, but what is the alternative? I play on a computer that I use for little else, but most people out there can't do that. Most competitive games would just not exist on PC without anti-cheat software and highly competitive games need intrusive software. It is really ridiculous that it is necessary but people suck.

Although there are all sorts of attempts at server-side and even player anti-cheat hardware devices, to my knowledge there is no option that can realistically replace client system anti-cheat.


South Korean approach, bind your real id to the game, cheat and end up with a fine irl.


People can't care because people don't adequately know or understand.


I hope it is a 2 way discussion between each side where they can make suggestions. On windows it turned into install this antivirus level of access program that has barely been tested and no guarantees it wont install malicious software as it auto updates every boot of pc. At a minimum it should require 3rd party(valve could step in) validated signed patches to minimize threat of individual employee selling off right to deploy ransomware to millions of machines.


There is no discussion. In order to prevent cheating, the game companies must stop my computer from doing what I want it to do. They must own my machine. The only difference between their software and malware is their software comes with a terms and conditions license that nobody reads.


So whats your solution? Let cheaters run around?


Anticheat without malware has been around for ages: simply don't trust the client. Even Minecraft servers manage it.

Instead of playing a constant game of cat-and-mouse trying to prevent forbidden code from running when you can just plug in a USB rubber ducky, put the logic server-side. Ban problematic behaviour, as opposed to methods that might result in problematic behaviour. It takes some ingenuity, but what are gamedevs known for if not ingenuity? (Well, apart from withstanding abusive working environments for long enough that they can be profitably discarded, traumatised, after they can no longer function as effectively as a green new intern… but let's focus on the ingenuity.)


As a Minecraft server operator, I can tell you there are plenty of client-side cheats that are hard to stop server-side. Also many players end up complaining about false positives typically caused by movement errors from internet lag on their end.

Also, all these anti-cheat methods are not supported by Mojang but are third-party plugins. Stock Minecraft Server hardly tries to prevent many of the hacks.


> Stock Minecraft Server hardly tries to prevent many of the hacks.

It would require stuff like server-side chunk culling, which doesn't jibe with the way Minecraft is designed. However, there's nothing stopping a new game from being designed like that.

I named Minecraft because it's fundamentally flawed for server-side anti-cheat (you're limited by a client that requires a lot of theoretically-unnecessary information, and doesn't send enough requests for you to provide this lazily), and people manage to catch enough cheating to keep things fun for everyone else. If you control both the server and the client, you don't really have much of an excuse.


> However, there's nothing stopping a new game from being designed like that.

Modern games do network culling. UE4 supports it out of the box. The _millisecond_ that information is replicated to the client, (which has to be before the client views it on screen to allow for hardware latency), a cheat will have access to it and will be able to act on that information. Even if you have the fastest monitor known to man, and no rendering buffering there's 6ms before you can draw the player on screen, or 1ms to poll your mouse/gamepad, so the cheat wins.


So don't let clients react on information until 8ms after they get it? If it's a game where you can't do that, collect probabilistic evidence on players until you have a lot of evidence that a player is cheating, and add them to the “we'll ban later, but let's mitigate the damage for now” list.

Cheating players are statistically distinguishable from really good non-cheating players most of the time – and when they're not, we're at the level of cheating where it's unlikely your anti-cheat will help, because they'd just fork out the €50 for cheating hardware.


> So don't let clients react on information until 8ms after they get it?

This just delays the problem by 8ms - you still have all the same problems, except you've introduced 8ms latency. The cheats are still perfectly accurate, and are responding to perfect information. If you could only delay the information to the cheat, then you wouldn't need to do any of this because you could reliably just boot out the cheaters.

> ollect probabilistic evidence on players until you have a lot of evidence that a player is cheating, and add them to the “we'll ban later, but let's mitigate the damage for now” list.

That's what anticheat providers already do already, and yet we still have that problem.

> We're at the level of cheating where it's unlikely your anti-cheat will help, because they'd just fork out the €50 for cheating hardware.

Hard disagree here.

1) if you have an algorithm which can statistally distinguish cheating players from non-cheating players enough of the time, I'm guessing you have an algorithm or model that you can point to that does that? Because as far as I'm aware, they _help_, but don't fix.

2) People aren't forking out for dedicated hardware to spoof mice en masee, but they _are_ forking out for kernel level cheats to bypass the userland detection.


Aimbots need no trust from the server. As long as the client knows things the player doesn't, and there isn't a chain of custody from the physical mouse to the server, you can cheat.


Why would a client need to know of such things? If the player can't / shouldn't know of something, then the client shouldn't either.

There are some interesting repercussions of this train of thought though. Take an enemy hiding in brush in an FPS game. The player would have line of sight to their hiding spot, and current games (I think) just leave it to the player's eyeballs' ability to see them through the brush. That's kinda cool in that it raises the skill ceiling and it's obviously immersive.

However, there's a whole host of reasons it's bad. First, there's the aim bots and other cheating it allows for. Second, what if the player has bad eyesight? So there's an accessibility issue there. Third, you get people doing a... lighter form of cheating where they turn their graphics settings down to the lowest setting (sometimes even editing INI files to lower them down below what is possible via the GUI) that would lower foliage quality, making it easier to see the enemy. Probably some other things too.

There are perhaps many interesting ways to fix this. D&D has their own way with random chance, but I can see players of skill-based scoffing at this, and I don't blame them. But I feel the current approach is lazy game design, and perhaps something interesting could be thought up.


because the alternative is server-side rendering of every frame, which would introduce unacceptable latency to input and also likely significantly reduce graphical detail, given the existing huge investments in consumer rendering hardware that have not yet been duplicated serverside

google stadia exists. nobody playing competitively uses it, and that's exactly the demographic that cares the most about anti-cheat

and, it still doesn't prevent computer-assisted input


Speaking as an avid Stadia user, I think this has a lot more to do with Stadia's library being garbage compared to the non-Stadia options. I can't imagine a reason an e-sport pro would want to go from the entire competitive ecosystem as a choice to just what's on Stadia. Even for casually competitive players, there's a massive lack of mainstream competitive titles on Stadia. Maybe the latency input is the problem but it seems like a big assumption with how miniscule the library is.


I'm very much a competitive fps player, i have low latency to the closest stadia server and I'll tell you: It's absolutely unplayable for me due to the input lag. I'm normally in top 1-2% and on stadia I'm closer to top 30% it's a huge difference.


> Why would a client need to know of such things? If the player can't / shouldn't know of something, then the client shouldn't either.

Think about the case of seeing an enemy onscreen and headshotting them instantly.


I think you misunderstood my post, and considering the surprising number of downvotes on it, I wonder if others did too?

Can you explain further though? I don't understand your case. It sounds like an aimbot, which is what would be prevented by not sending the client more info than is needed.


The thing about aim bots is that they're quicker than you and they don't make mistakes. A monitor and a rendering engine will introduce up to 3-5 frames of latency (double buffering, refresh rates). Your client likely has the next 1-3 frames of data ready to go so it can do client side prediction/interpolation/smoothing, which means the cheat has access to that data. All of the data that's required to make modern games look and _feel_ good is the data you're talking about throwing away here


I guess it is the case where the info begs to the player, the player reaction time is in tens or hundreds of milliseconds (time to render the image, display it on the screen, player to see, move the mouse to aim and shoot) versus an aimbot that does not need to wait and can react in a few milliseconds. This is cheating if the cheater's aimbot is using an advantage normal players don't have (unless everyone is using aimbots and this is no longer a human-played game but a bot game).


As far as I know an aimbot can work just with pixels that it can see.


Alternatively, you could provide the geometry without providing the semantic information. I'm not sure how viable that is, but it'd make aim-bots a bit harder.

> First, there's the aim bots and other cheating it allows for.

You can make an aim bot by connecting the output of your monitor to a hardware-emulated mouse. Unless the server can detect this, kernel-level anticheat is building a really tall wall with the door unlocked. (If it worked I wouldn't object so much.)


> Alternatively, you could provide the geometry without providing the semantic information.

Yes - cloud rendering. It's _a_ solution but comes with tradeoffs. Namely, it introduces large amounts of latency, and costs a dumpster truck of money.

> You can make an aim bot by connecting the output of your monitor to a hardware-emulated mouse.

Anticheat is a cat and mouse game, and raising the barrier to entry improves the experience no end. There's a _big_ difference between downloading and running an exectuable, and buying specialist hardware to cheat at a game.


> Yes - cloud rendering.

Only a very small part of the rendering needs to be done on the server; notably, the camera could still move, and the client could still make the usual heuristic predictions about where things are, so it wouldn't feel laggy.

> and buying specialist hardware

If you have a streaming rig, you already have nearly all the hardware. Just add a Raspberry Pi Zero or something, and you're good to go. (But yes, the barrier-to-cheat is still higher… at the cost of also increasing the barrier-to-play-at-all.)


> Only a very small part of the rendering needs to be done on the server; notably, the camera could still move, and the client could still make the usual heuristic predictions about where things are, so it wouldn't feel laggy.

This is just nonsense.

The data that the client uses to make those heuristic predictions is the same data the cheats, so once you send that extra data over you've undone the benefit of not sending the client any extra info it needs, _plus_ youve added an intermediate network hop for your remote rendering, _and_ you've now required that your game has essentially a hardware DRM in the remote computer that you're rendering.

> If you have a streaming rig, you already have nearly all the hardware. Just add a Raspberry Pi Zero or something, and you're good to go.

The number of people out there who are spending thousands of dollars on secondary rigs to hook up emulated mice via a raspberry pi to run custom cheats has got to br vanishingly close to 0 for even biggest games in the world. Given that _these_ are the measures youre suggesting to bypass the anticheat, it sounds like they've sufficiently raised the bar to me.

When people writing AI driven hardware mice is a common cheat vector let's talk, but right now the cheat vector is people running cheats installed inside the kernel.


> The data that the client uses to make those heuristic predictions is the same data the cheats,

No, it's not. https://en.wikipedia.org/wiki/Lag#Make_clients_extrapolate

> plus youve added an intermediate network hop for your remote rendering

I never said anything about an intermediate network hop. It's not that expensive to cull a few tris, diff the geometry and then send it – especially if you can compress the information, which you can, because you control the client and server. (In fact, you likely don't even need to go down to the tri level to get the effect; all you need to do is turn locating obscured players into a computer vision problem.)

I don't know where this “intermediate network hop” comes from.

> spending thousands of dollars on secondary rigs

Think €60. An HDMI splitter, a USB HDMI capture card, a Raspberry Pi and a USB cable. Not that expensive.

> but right now the cheat vector is people running cheats installed inside the kernel.

It's functionally equivalent. Your “anticheat” is taking control of other people's computers, stopping a lot of people from being able to play the game entirely, because it was easier for the developers.


Thanks for a wikipedia article on networkiong basics. Believe it or not, I do have some experience in this area, I've worked on networked games on physics and gameplay for almost 10 years. The information you use to extrapolate (position + velocity) is the same information a cheat is going to use. If you have enough information to draw _a_ thing on screen, a cheat has enough information to say "fire at that thing".

> It's not that expensive to cull a few tris, diff the geometry and then send it

Multiplayer games are doing this aggressively already. This is an out of the box feature in UE4 and Unity. The second the client gets this information, the cheat has the information too.

> all you need to do is turn locating obscured players into a computer vision problem. > I don't know where this “intermediate network hop” comes from.

You either send the positions to the client, along with an association for what it belongs to, or you render the _entire_ scene on a remote "client", which means streaming the video back to the end users device. The video streaming introduces an extra hop on the network as you now need to go from end user to streamed client to server, rather than end user to server. If you don't render the whole thing, at some point the client gets sent the information that it needs, and the cheat has it.

> An HDMI splitter, a USB HDMI capture card, a Raspberry Pi and a USB cable. Not that expensive.

Apolgoies, when you mentioned a "streaming rig", I thought you were talking about having a secondary gaming PC hooked up. Just because it's _theoretically_ possible doesn't mean it's actually happening though.

> It's functionally equivalent.

I fundamentally disagree. One of these (kernel level cheats) are readily being sold on the internet, available to anyone with a debit card/paypal/btc and a computer, and the other requires dedicated hardware, along with the knowledge of how to write these cheats in the first place.

> stopping a lot of people from being able to play the game entirely, because it was easier for the developers.

The numbers of people who are legitimately stopped from being able to play the game entirely are very small, but those people do exist, and it sucks for them. Your blame of `it's easier for "the developers"` however is untrue. It's not because it's easier, it's because (at least right now) it's necessary.


And how server-side detection of aimbot can look like? If the latter is not written by a retard, you're very quickly down to statistical detection. Which is also cat-and-mouse, but the cat is crippled.


Fine by me. Windows games secretly install malware with unlimited permissions to spy on you without your knowing how it works. Sometimes it collects browsing history etc like that red shell system. If steam releases this on Linux, we should all uninstall steam


Just write software that somehow pretends the anti-cheat module is running even though it's not.


As a person who both knows how easy cheats are to make, and what anticheat entails - There's no way I'd play without it.

In CSGO for example, many players _want_ a more invasive anticheat solution, and are literally asking or even paying for it through third parties like ESEA and Faceit.


The alternative is requiring publishers to remove their DRM, and use Valve’s, and they (the publishers) most certainly would not do that. I’d bet that if Valve tried that, there’d be an antitrust lawsuit against them real quick. It sucks, but the alternative is getting the DRM to work on Linux, so that’s what they chose.


The arms race between game publishers and the small minority of gamers who actually cheat is harmful to everyone else. Stub the DRM APIs to usefully return nothing quickly rather than enabling voyeurs who think scanning and logging the contents of people's hard drives on their servers is acceptable.


I've heard of some online-only games - such as Fall Guys - being basically killed sales-wise by insufficient anti-cheat protection.

After all, if another player can run faster or fly or has perfect aim, everyone else is basically playing an unwinnable game.

This isn't helped by the current fashion for automatic match-making and 100-player battle royale games - which make one cheater impact a lot more players.


> if another player can run faster or fly or has perfect aim, everyone else is basically playing an unwinnable game.

Not a gamer, so forgive me if the proposal is stupid, why not just treat cheaters like they do on (li)chess? If I am playing with someone that I believe to be cheating, I flag/blacklist that user and simply don't play with that user afterwards. That by itself is enough for me to play against only honest players, but if you want to be more strict you could even submit the game session for review, and in case it is ruled there was cheating, the cheater is banned and every game you lost to the cheater is awarded to you.


User reports are indeed a common tranche of anti-cheat systems.

But in chess terms "Middling player occasionally made suspiciously good moves" is a pretty weak cheat report. But if you can tell they had stockfish running and changed windows to it several times in the course of the game? Well, now you've got a very clear cheat report.

More broadly, many gamers are chasing the dopamine rush of winning, complete with slot-machine-style flashing lights, music and dancing character animations. Being awarded a win by e-mail a day or two later doesn't really compare.


What if the cheater is just running stockfish on a separate laptop, is the anti-cheat system also going to turn on the webcam and track eye movement?

> Being awarded a win by e-mail a day

That is only half of the story. The other half would be in banning the cheater and making them lose access to their account, i.e, impute a real monetary cost to being caught cheating. Make it expensive for those caught cheating and equally expensive for those making false reports and I'd venture that we would see some sort of equilibrium where cheating is not worth it.


>Make it expensive for those caught cheating and equally expensive for those making false report

Well now you've tipped the scales too far. Nobody is going to make a report, even when it's justified, if they're risking their account to do so.


The cost can (should) be progressive for repeated offenses, and it can be divided if multiple people make the same (wrong) call. Say you get one "free" report per month, but you don't lose it if either you were correct and/or more than 5 people reported the same person.

The important thing is not about being "perfectly provably fair", it is just to curb abuse (on either side) so that people still can have fun playing without having to accept such invasive software on their machines.


Overwatch briefly had a system like this, where you could mark players in your matches positively or negatively, so they'd be more or less likely to show up in your future matches. They had to remove it after a top level player could no longer find fair matches. He was waiting 10-20 minutes before getting matched with much lower skilled players. He was by all accounts a pleasant person, but he was so good at the game that many people would mark him negatively simply so they would never have to play against him.


You fix that by adding handicaps. Not only it solves the problem of lower-skilled players still being interested in playing better opponents, it gives top-level players a way to continue improving their own game even when the opposition is weak.


That absolutely does not solve the problem? A grandmaster chess player is going to get tired of playing handicapped games against 1000 Elo noobs.


> going to get tired of playing handicapped games against 1000 Elo noobs.

False dichotomy. What about playing 1800 players but starting with only one bishop? Or how about playing the same 1800 players get to use some assistance that warns of blunders/mistakes, but does not give away the better moves? It's only "cheating" if it is not agreed beforehand...

Also, bear in mind that your Overwatch example gave the impression that the good/honest player was so above the others that even without the opponent flagging system, the above-average player would be matching with "boring" games most of the time. Removing the opponent rating system does not solve his problem and still leaves the cheating issue open.


> Stub the DRM APIs to usefully return nothing

If breaking DRM was that simple, DRM companies wouldn’t exist; Games would be cracked on day 1 (or earlier). It’s way more complicated than that, and each DRM system is different.

That’s why Wine doesn’t do well with DRM. For example, some systems rely on undocumented behavior in Windows’ APIs that Wine doesn’t implement.


> the small minority of gamers who actually cheat

The small minority of gamers who cheat impact the experience for a much larger number of customers who don't. So each cheater might cause dozens of lost subscriptions. (Contrast to software piracy, where each pirated copy only represents a probability between 0 and 1 of a lost sale.)


I don't fully trust the vendors of anti-cheat software, but I don't think its fair to classify it as "malware". The "mal" in malware mean "malicious". I don't think most anti-cheat software operates with malicious intent.


Most malware is developed to serve some profit motive, not as a result of actual abstract malice toward the victim. I really don't think that's enough to disqualify it. Merrium-Webster defines malware as "software designed to interfere with a computer's normal functioning"[0]. WebRoot defines it as "any type of software designed to infiltrate or damage a computer system without the owner's informed consent"[1]. Red Hat's definition is a bit broader: "malicious software, including any software that acts against the interest of the user"[2]. Most anti-cheat software would fit all of these criteria, with perhaps some quibbling over what counts as "informed consent".

[0] https://www.merriam-webster.com/dictionary/malware

[1] https://www.webroot.com/us/en/resources/tips-articles/what-i...

[2] https://www.redhat.com/en/topics/security/what-is-malware


If you think of 'great' in the 'big impact' sense rather than 'unequivocally good' sense, then it holds.

Most complaints I hear about using Linux for gaming are that it can't run games. With anti-cheat it will be able to run games (all of the top ten and most of the top 100).

The principles of allowing Anti-Cheat at all are a valid discussion, but I don't think we should be sabotaging Linux as a viable gaming platform just because some of the trends in software at the moment aren't in line with absolute ideals.

After all, if you don't want anti-cheat you can just continue to not play the games that require it.


Its about choice. You aren't required to run EAC or battleye but you can choose to if you want to play video games with your friends. Also anticheat on linux is easier to work with than anticheat on windows. Running a linux gaming VM / dualboot is going to be much easier than a windows gaming vm / dual boot.


People here are obsessed about it, while I don't understand what the problem is. Just run a separate Linux (with kernel-level anti-cheat and trusted computing validation, whatever) for gaming only, that's all?


https://partner.steamgames.com/doc/steamdeck/faq

> We’re working with BattlEye and EAC to get support for Proton ahead of launch.


My brother wanted to play CoD Warzone with me and of course it doesn't work with Proton. After attempting and failing to dual boot into Windows (how hard could it be to create a bootable Windows installation usb key?), yesterday I decided to try vfio. It actually was quite straightforward and way simpler than one would expect from the very intimidating archlinux wiki page. I spent more time getting a nice grub vfio menu entry than actually setting up the VM.

Of course now I expect I'll be eventually banned for running under a VM.

Proton support for anti-cheats can't come soon enough.


I don't think Activision bans VM players. I've been playing Warzone in a VM since it was released without issues.


I've played overwatch for years in a VM now, no ban here


That's what I have been reading online. Good to have another data point, thanks.


vfio works great for games (and general "Oh cr*, I need this random Windows only thingy), and the added bonus is if you install to another nvme drive, you can actually dualboot if you ever wanted to do that :-).


Umm, is there anti-cheat that doesnt need a locked down OS and hardware? Dont understand what you are cheering for.


Steam deck could conceivably use a signed bootloader to provide some of the guarantees that an Anti-cheat is looking for. But regardless, Valve sounds committed to describing the steam deck as "still a PC [that you could put windows on]".

For non-steamdeck Linux users, at least running on Linux shouldn't be a disqualifier for the Anti-cheat software.


There is efi secureboot and lockdown lsm. Maybe it will require both.


Yes, and the ones that don't have crazy requirements already work fine.

Unfortunately a lot of people don't care about running malware (see: the % of people who still use Windows). Lots of those people would be more likely to switch if they could run their malware on GNU/Linux.


How does the anti-cheat malware suddenly stop being malware when run under Linux?


It doesn't. User freedom means freedom to run malware, just as all choice implies the option to make a bad decision. In this case, it might increase adoption and so be a bad means to a good end.


They don't. People who are conditioned to want to run malware will be more likely to switch off they are able to run their malware of choice though.


I'm cheering because this will mean more people can use Linux as a daily gaming driver. I see that as a good thing, albeit with trade-offs.


EAC already exists on Linux for Linux games. Only the Windows version is incompatible.


Client side anti-cheats are getting disrupted anyway. You can already buy ML and external hardware cheats that are obviously impossible to detect using the same old methods. It can work for games being streamed locally or over the internet, Stadia, game consoles, anything with a screen.


I really hope the work Valve is doing with Proton will help to get Office (no, the browser versions aren't a substitute), Adobe software and Touchdesigner working on Linux. I don't game, and would love to switch to Linux but I need those apps to work on it.


Microsoft could at least stop actively trying to break O365 on Wine, which I believe it is doing. I have heard rumors (albeit unsourced ones) that MS is trying to bring Outlook over first.


Outlook is a progressive webapp now, so that makes some sense.


Outlook has a PWA, but the thick client is still here and will be for awhile to keep legacy plugin infrastructure running.


I think that's quite unlikely, most games use a fairly small subset of the Windows APIs, if non-game applications currently don't work then it's unlikely that the situation for those applications will be improved by better game compatibility.


Doesn't all of that stuff have great Wine ratings?


I really wonder how they're gonna do that tho, like implementation wise. I suspect either they are gonna have some sort of deal where VAC can be swapped in, or they're working on some secret sauce kernel module that implements Battleye and EAC.


It'll certainly be the jump point for me -- I'm even holding back a much needed hardware refresh until BattlEye is supported on linux.


Interesting question on how to create same level of protection. On Windows I think the general idea for a bypass is find a vulnerable signed driver and exploit that, but Linux doesn't have the same paradigm for kernel modules as the Microsoft-signs-everything approach on Windows.


I'm only waiting for it to nuke out my windows install.


Please no. Keep all those poorly-written proprietary privacy-invading vulnerability-enabling anti-cheat kernel modules as far away from Linux as possible.


That's ok, you don't have to install the games if you don't want to.


Okay, give me my money back then. Let me count how many thousands of dollars my Steam account is worth.


I think I'd just recommend you not play games with anti-cheat.

I'm not sure why you'd buy games with anti-cheat if you don't like anti-cheat though.


I bought those games decades ago. Anti-cheating software used to be a lot more tolerable back then. They were still obnoxious malware but at least they were not literal drivers with vulnerability-as-a-service ioctls that give the games and everything else free access to ring 0 from user space. They were also optional, I could find game servers that didn't require the software.


Which games did you buy decades ago that now use Battleye and EAC?


No idea honestly. Been a Steam customer since 2006 and I have almost 300 games. That's why I said I want them to keep this kind of software as far away from Linux as possible. Who knows what their next update will bring into my machine?


I mean I guess that's the peril of using an automated update service.


This is not a good thing...


If it increases Linux Market share and the viability of Linux as a gaming platform, I see it as a good thing.


i’ve been using linux as my daily driver for just over a year. steam on linux has been great, and proton (wine) has been indispensable. big title games i’ve played with _no_ issues at all:

* Cyberpunk 2077 (ok that has issues, but it has the same issues for windows users..)

* Dying Light

* Titanfall

* The Division

along with _many_ other indie/small games.

honestly the only thing left is titles that include invasive anti-cheat. when that day comes, i don’t know if most of my friends will have any reason to remain on windows.

and who could blame them? finally, they won’t need to deal with a company that collects data from them without their knowledge, runs ads in their taskbar, demands they have a useless microsoft account, restarts their machine on a whim for updates, etc.

i hate that i’m going to say this out loud, but could 2022 really be the year of linux on the des- why does it feel so wrong to say that? :)


Same here. Switched 2 months ago after using WSL all the time & being annoyed over many things, including "failing updates" on windows. Did a dualboot but haven't used windows since. Game to add to your list: - CS:GO


well, strictly speaking CSGO isn’t using proton, it’s a native port, which is why i didn’t count it.

but i play that, too. probably too much..


FWIW I get about 10 more FPS in Ubuntu vs Win10 for CS:GO.


I’ve played Cyberpunk, Skyrim, Fallout 4, Halo Master the chief Collection (games 1, 2, 3 and Fall of Reach) plus EUIV and Stellaris just in the last year. I haven’t used windows since the vista era


One issue i have with Proton is that it seems to run these popular games fine, but when faced with something more obscure there are issues left and right.

For example recently i installed Arch on my GPD Win 1 (which is essentially a much weaker Steam Deck that comes with Windows 10) and tried:

* RoboBlitz - it needed Ageia's PhysX installer which always failed to install at startup

* Clive Barker's Jericho - Didn't start at all

* Cherry Tree High Comedy Club - Worked fine (but it is a very simple 2D game so i kinda expected it)

* Mars: War Logs - It started but 3-4 in the game (actually during a cutscene but was rendered using the game engine) always crashed at random points

* Blood Knights - Worked fine

* Marlow Briggs - Didn't start at all

* Chaser - Didn't start at all

All of the above work under Windows 10 fine which makes me suspect that Valve is just going to make game-specific patches for the next Proton in Steam Deck. After all being able to play "the entirety of your library" sounds implausible when you can't do that even under Windows :-P (many games - especially older games - need custom fixes, etc).


The Outer Worlds is one of the big AAA titles I ended up playing on Linux. I've historically played a lot of them on Windows out of pessimism but The Other Worlds was dying in the opening cutscene on Windows for no explicable reason so I gave it a go on Linux and so far it's worked perfectly.


99% of games I play work with Proton, or Lutris, which is great most of the time. However, it's the small amount of time when something stops working or needs small config tweaks that really annoy me, as they cut into my gaming time.

Also, setting up games can be a little bit of a faff. For example, wanting to play FFXIV, well, nothing worked at all. I had to go into the config file and skip the cutscene, and also change the launcher to the old launcher. Now, that doesn't seem like much, and it isn't, but the first time I had to do that took me hours to figure out.


Setting up FFXIV parser is also annoying on Linux


ACT works by poking around in the memory of the ffxiv process, I imagine it's tricky for the same reason anti cheat is.


Ay, fellow Titanfall player o/

Noticed any difference between Windows and Linux for that particular game?

It's on the list of my most played games so making sure it runs properly is really important for me to make the switch ^^


well, Titanfall is essentially a source engine game (although the engine is heavily modified), so i'm a bit annoyed there isn't native linux support - but afaik all source engine games do well through proton (this includes apex legends, but EAC is its only undoing).

i've never played it on windows, so i can't give you a 100% answer. i've never had any issue with it, and i don't think you'd be able to tell it wasn't a linux native game.


speee is life!


> The Division

Division 1 or 2?

Division 1 was built in a way that would be very hard to run on Linux, Division 2 however was ported to Stadia which is basically Ubuntu.

If it wasn't for Stadia that effort would never have been spent, I was a huge advocate for linux on those projects for years.


https://www.protondb.com/app/365590

> Division 2 however was ported to Stadia which is basically Ubuntu

the division 2 isn't on steam, because ubisoft. and hence - as far as i know - isn't generally available on linux. there's some irony there i think.


Division 1 works fine.


That’s really heartening to hear!

I had assumed, if nothing else, that easy-anti-cheat would have prevented it.


The only thing I'm missing: streaming with audio from discord. If someone at discord could fix this I would be so happy. I'd actually pay for nitro.


i've used discord without any noticeable issue (except that their "magic" noise cancellation algo is missing from the linux version) - what do you mean by streaming audio, out of interest?


During the pandemic I've been playing puzzle and story games with friends by streaming the game to them over discord. Discord cannot capture the audio of a game in Linux for some reason.

On windows it can capture the audio from that specific window.


You should be able to work around this yourself using a pulseaudio null sink and some loopbacks. You send the game audio to the sink, and loop the audio from the sink back to your main output device. I've done this in the past for OBS, but you should be able to add your microphone input as well using another set of sinks and loopbacks I think.


Sadly this causes discord to noise cancel the game audio and it sounds horrible


Yeah, also the audio codec for voice is really low bitrate so even if you turn off all the filtering in the UI it's still pretty bad, while the proper streaming setup has better audio


Steam also has a bunch of titles with native Linux support. I believe Dying Light is one of those titles.


What hardware have you had that experience on? Probably going to build a new Linux box this year, and I'd like to pick hardware that's as hassle free for wine/proton as possible.


So far I’ve had no problem with my Ryzen+Nvidia 1660x.

I’d maybe try to find an AMD GPU but I bought one of those first but need tensorflow and AMDs GPU support for ML was almost non-existent.


Personally I've had no end of problems using an nvidia GPU on Linux.

For example, some driver packages support Steam but not CUDA, while others support CUDA but not Steam [1, 2].

And I found some instructions online that let me "fix" it to work with both - but thereafter, any time I used CUDA I would lose audio when I next resumed from suspend.

[1] https://www.reddit.com/r/linux_gaming/comments/ocuqtm/can_i_...

[2] https://github.com/ValveSoftware/steam-for-linux/issues/5778



what distro? pop_OS?


i went for ubuntu, actually. i want to give arch and pop a try though.


FWIW, I'm running Steam on Arch and it's working great. I remember that there were occasional issues with library version mismatches when I first started out with it, but I have not had any problems in the last... 3 years at least.

However, it should be noted that I don't play much AAA stuff. Most of my playtime is actually on the Nintendo Switch. The playtime that is in Steam is mostly in Linux native games (Tabletop Simulator, Cities Skylines, Unrailed, etc.), but the few Proton games that I've tried have all worked great.


used ubuntu for a longer while & gave arch a shot, the initial setup might seem ugly, but its not that hard and the distro is just gold, just need some time to set everything up


Why would it matter?


Because if you have laptops with Nvidia Optimus, distros like pop_os make your life much easier since nvidia doesn't give two shits about Optimus on linux, hence the famous "nvidia, f*ck you!" quote from Linus Torvalds.

Or if you have a brand new laptop with the latest Ryzen CPUs and the latest Radeon 6800M GPU with switchable graphics then you absolutely need a distro with the latest kernel to have a good experience on this new hardware.


Maybe I underestimate how difficult some distros make it to have some semblance of control, but my model of Linux is that distros are nothing more than bundles of packages as a starting point from which you can add or remove whatever you want.

So I asked why it matters if its 'pop_OS' or not since if that happens to bundle some driver, or Steam, or whatever, I assume I can continue not using 'pop_OS' and still install whatever it is if I want?

It just seems like an irrelevant detail, (and potentially not even a valid one? the more you do subsequently the less it 'is' that distro) outside of saying 'everything worked out of the box, didn't have to install a thing', which may or may not have been true, but wasn't what was claimed.


If you're using Ubuntu LTS, for example, you're going to have the kernel version it shipped with and the Nvidia driver version is shipped with. When Nvidia releases a new driver with fixes for specific games, as they do every couple of months, there's no guarantee that this driver version is going to be compatible with the relatively fixed version of the kernel that Ubuntu LTS is using so you'd have to wait for the next release to play those newer games.

Of course you could forgo the packaged kernel and build your own then grab the Nvidia driver from nvidia's website, but then what is Ubuntu buying you?

Ubuntu LTS with its 2 year minimums lifecycle is kind of a worst case here among desktop distros, but different distros among the 6 month crowd can run into this depending on how conservative they are about putting the newest core packages into each release.

There's also the question of the method of obtaining third party software. There's basically a divide here between the ports method (see the AUR for the most well known example) where users distribute a build script to other users that in most cases should be pretty simply cloning the upstream source and running their build script, or the third party package repository approach where you download a binary package and install that. The port has the advantage of being very easy to audit the packager's code (if not helping for the developer's) and generally not needing it to be rebuilt every time the dependencies do, but the third party package repositories have convenience and you could argue that relying on launchpad's moderation is not any more or less safe than relying on Nexus Mods moderation which Window gamers happily do.


Pop OS power management is great. If you’re on compatible hardware it’s fantastic.


> restarts their machine on a whim for updates

That's the thing, it's not their machine if they are running Windows on it. It's Microsofts. They can even wake up your machine to apply updates whenever they want.


For hardcore gamers, the added stability of Linux and choosing when to update, could be a big plus. Provided, hardware compatibility (there's always a caveat!).


I'd say Windows 10 VS NixOS unstable, Windows is more stable, has better functioning drivers. I still use NixOS to host a Windows VM because I like the userspace more, but the "WiNdOwS iS L3Ss StAbLe" meme doesn't hold very true for consumer hardware. Servers will be more stable on Linux because the manufacturers incentives aligns with yours, you buy our hardware to run Linux, we make sure it runs well on our hardware.


Mint works consistently for me now. Of course, there were fixes I had to implement myself. But I get to decide my own upgrade cycle, and can always timeshift everything back (like System Restore). The OS convergence of features are high, so differences are small today. Hopefully Pulseaudio and such stuff that often break can become better.


I would like to recommend pipewire over Pulse, it's compatible but actually really good, also ALSA compatible


Just to be fair and balanced I'm a hardcore linux user and gamer and I've tried gaming on linux and I've hit many issues. 365 games on steam and while some windows games work I've hit probably just about every possible issue you can hit with other games. Graphical artifacts, freezing, stuttering, and even full OS reboots.

Sometimes I read this thread and scratch my head. I get enthusiasm for FOSS, but enthusiasm to the point of delusion? Honestly, Linux is not up to par with windows in terms of running games that are basically designed for windows. Thus the experience compared to windows for these things is definitively inferior.


Linux is surely not up to par with windows, but it is definitely getting better. Not only supporting more games but becoming more accessible.

Linus tech tips did a video recently on installing and gaming on Linux using pop_os which I think is great service for the mainstream user.

So yeah, long way ahead, but we are living the first step, becoming mainstream. If this trend continues, you won't have to be a hardcore Linux Fan to game in Linux.


> I get enthusiasm for FOSS, but enthusiasm to the point of delusion

I see your point but consider this: 20 years ago you could make the same argument for all aspects of linux, not just gaming. Why use some hobby level OS for serious work when you have Solaris, HP-UX, AIX?

Today we have a very usable Linux desktop and pretty much total dominance in the server arena, only because people were at one point "delusionally enthusiastic" about making it work.


Propriety Unixes always had very poor CLIs, to the point people would pay the FSF* to get to install GNU utils. No surprise people would flock to the (offering practically better experience) free OSs.

* Payment for CDROMs. Net distribution wasn't as common then.


Only because the owners of Solaris, HP-UX, AIX decided to outsource part of their UNIX development costs to Linux on one side, while others decided they wanted a free beer POSIX implementation to avoid paying for UNIX.

Except gaming culture is all about IP, something that FOSS don't seem to grasp on their quest for freedom über alles.


Video games are a particularly interesting subsection of software in the context of free software.

Broadly speaking video games are an expression of software and multimedia solely to be used for consumption, and intended to be consumed exclusively within the canonical form presented by it’s developer.

In that way IP is very important, because often there’s more than just the investment in underlying game engine and third party middleware involved; it’s a combination of artwork, audio, video, bespoke scripting, and architecture, to make a whole work. It could be reasonably argued that those pieces should be aggressively defended to ensure the investment is not wasted as these pieces may be of limited utility outside of the complete whole.

I’m certainly all for increased transparency into the inner workings of this package. But it’s an interesting topic and I don’t know enough to even form an opinion on where I think the correct answer lies.


Proprietary Unix systems usually required equally proprietary hardware, which, twenty years ago, was quite a bit more expensive, and also quite a bit slower, than x86.

Also, those proprietary systems usually sucked from the usability point of view, due to being business-oriented.


Less than 20 years ago, Windows was the hobby level OS compared to Solaris and AIX. Windows Server 2005 was a complete and utter joke compared to Solaris. The company I worked for in 2006 was moving their Solaris servers to Linux. Windows wasn't an option because it wouldn't even meet the specs in real life.


I think you may thinking of Windows Server 2003. I also suspect you've never run Windows Server 2003 at enterprise scale either. No operating system is perfect, but Windows 2003 was a reasonably capable workhorse, even in 2003.


My old company made a DIS box (UK gov-specified integration server that would talk to the central Government Gateway) as Java on Solaris in the early noughties, and doing a performance test that one Solaris machine took down the entire Windows-based Government Gateway test environment. Draw your own conclusions :)


It's difficult to draw any conclusions without knowing the environments, what's running on them, were they configured properly, etc etc.

Don't get me wrong, I'm not ideologically bound to any platform, but stuff like this just sounds like another tired old-man Windows vs Linux/Unix pissing contest.


Well, regarding config, it was run by Microsoft, so it's hard to say how well it was set up.

It was more of a fun story than anything though. Didn't mean to offend anyone who identifies strongly with MS.


> Graphical artifacts, freezing, stuttering, and even full OS reboots.

I have seen similar on Windows. For example when Bioshock Remastered Freezes there is no way to exit it, windows still responds, but there is no way to close the game because everything you open is hidden by it. Also the amount of Graphical glitches I encountered in Skyrim is just hilarious. DotA 2 seems to sometimes glitch out when you hit alt tab while it loads. That is just the games I played the last few weeks, I think there is not a single game that isn't somehow broken on Windows either.


For situation like this, make sure you have at least 2 virtual desktops available and then alt+tab and then ctrl+win+arrow key to switch to an empty virtual desktop and open task manager.


this is the way, you don't even have to have 2 desktops open, just win + tab and then open a new one for taskmgr.


In the past (ie Win7 days and before..) I always made a point to set games to "Windowed Fullscreen" wherever possible as that vastly improved Windows' stability.

It was probably bad for graphical quality/performance/latency but being able to temporarily exit games without worrying about a crashing desktop was more important for me.

Apparently that's still an issue then..


I've found using borderless "fullscreen" to actually improve latency. I had a lot of trouble with deflecting on Sekiro when I was running in fullscreen mode that I found immediately went away once I started running it borderless.


For this kind of situations xkill is marvelous.

I've used Windows xKill [1] for years, but I can't find a download link for which you don't have to sign in...

Alternatively, SuperF4 [2] looks decent, also with a separate CLI only xkill found in the github issues [3].

[1] https://www.deviantart.com/suprvillain/art/Windows-xKill-100...

[2] https://stefansundin.github.io/superf4/

[3] https://github.com/stefansundin/superf4/issues/39#issuecomme...


Sysinternals’ process explorer gives you a crosshairs to select a window’s process — that crossbones pointer has more style though.


Many tiling wm offer something like that out of the box (and KDE among the mainstream desktop managers)


Worst case, on most Linux setups just C-A-Fnum to a TTY, log back in, and `kill -9` the offending PID from the command line.


> I have seen similar on Windows.

Sure, but Windows is supported, so instead of going to a forum and being unhelpfully told to try a different distro, you can take up your issue with the developer and they're much more likely to pay attention. Granted, some of them still have shitty support, but you're a lot less likely to be dismissed out of hand.


> you can take up your issue with the developer and they're much more likely to pay attention.

Outside of indie, is this something people actually (and successfully) do?

For everyone I know the default assumption when a game doesn’t work is you’re SOL and either refund or hope for patches.


> is this something people actually (and successfully) do?

No. You fix it yourself, move on with your life or spend a few hours getting shit on by customer support and then move on. Videogames are an industry where you need to expect to be disrespected, because those companies do not care a whit about you.


And where do you think the patches come from?


I'm not a hardcore gamer but with some games that work reasonably well on a mid-class PC without an SSD drive I've noticed one thing in comparison. Windows keeps doing things that are hard to trace down while Linux (Mint in my gaming case) keeps its feet still. That is an important issue for me because some random System process may freeze up my games randomly. Not nice...

Overall I get a better performance on Linux at the cost of some graphics performance, which is a good deal for me. I'm really not playing the most current games to be fair. And I'm running on a 5400rpm HDD


I think at this point most of the issues (not all) that the average gamer experiences are caused by incompatible libraries and the like, it will be interesting to see how good linux gaming gets when the Steam Deck/SteamOS 3 release on which Valve will have better control on what users run.


To be fair, Valve already has a way to control mostly what is used by Steam and the game. They have their own runtime that uses (or can, can't remember if it's always the case) containers to create a controlled environment.

I have had issues sometimes with library/drivers versions on my rolling release preventing me from running everything natively, but each time using the runtime version of Steam allowed me to play.

They have information about it all over the place, I can't find exactly the page I wanted, but here[0] is some more information about it from their repository.

[0] : https://github.com/ValveSoftware/steam-runtime/blob/master/d...


Agree. Updated os (and probably drivers) to support specific hardware seems easier to maintain.

Windows get the first class on drivers and it's update due to market share. If this one success we may find vendors to start focusing drivers for steam os after windows.


I don't understand why people insist on this shoehorning either.

It's like some obscene tribal topos that refuses to die.

KDE 1.0 came out in 1998 - 23 years ago (O god I'm old).

The discussion of "Linux to replace windows desktop" is over two decades old.

In computer age this is something geological. It's like... well, Macintosh came out in 1984. Xerox Alto came 1973. If we go back 5 years we reach Englebart's Mother of All Demos in 1968 which I think can be considered the intellectual precursor of those.

So there is 5 years from a tech demo on high-end research platform to a (more or less) commoditized consumer offering - even though Xerox had no idea what to do with it. Steve Jobs visits Xerox 1979 and five years later they deliver Macintosh.

So, with engineering talent PLUS business drive they copy the idea, implement their own hardware and software stack and are instant hit (well, let's say for the sake of this discussion they are a hit).

In FIVE years.

Linux is trying to copy the software stack, of an existing platform, and has been "attempting" this for two decades.

This is not an engineering problem. This is not a community problem. It's a "lack of business interest problem".

Honestly, the Linux desktop is quite usable. I'm quite sure two decades are enough for the open source software stack to find some local optimum for the desktop offering.

But really, copying and supporting a continuously moving target needs real capital and real business drive to sustain the boring, mind numbing support work that is needed to actually sustain an industrial quality platform.

Linux is fantastic in lots of things.

I'm not sure reverse engineering Windows stack on Linux is very effective way of spending our civilizations engineering resources.

I appreciate masochistic Rude Goldbergish feats of engineering as much as the next geek, but I just don't see the value of individuals detached from the corporations that are implementing the master stack trying to reverse engineer everything on top of a third party platform.

Native Linux support? That would be nice. Native drivers and all? That would be nice.

If it works for someone that's very cool and satisfying - but I still think reverse engineering based gaming stacks for modern platforms that are alive and well are not perhaps the best way to spend engineering effort.


Games push an OS's limits. So a compatibility layer like Wine is more like trying to implement a browser that implements all of Chrome's API without forking it.


> I'm not sure reverse engineering Windows stack on Linux is very effective way of spending our civilizations engineering resources.

I don't think it is a question of a pool of resource that can be reallocated, however, as in a business.

The people who work on reverse engineering Windows presumably do so, because they are interested in doing so. They might not be interested in building something new.

The choice might therefore not be between doing this, and doing something new, but between doing this and not doing anything at all.


We don't collectively own engineers time, we are not under full blown socialism, yet and people are free to do what they want.

If individuals enjoy reverse engineering windows and making it work on Linux, that's great, actually I really need that. I wouldn't spend my time on it, but I'm sure I have equally suboptimal hobbies (from a "civilisation" point of view).

You don't get to spend someone else engineering effort.


"We don't collectively own engineers time"

Fully agreed. My intention was not to signal any entitled presumption of ownership of said resources - just that they are not maybe applied with maximum impact, which does not imply I presume to benefit from them anyway.


Writing these comments is also not applying your time with maximum impact to our civilisation. This metric feels perfect for armchair analysis.


There isn't any scenario under 'full blown socialism' where society collectively owns engineers' time, either. You may be thinking of Communism?


To add a datapoint, the bioshock games would crash while loading/saving on win 10, which seems to be a common issue. None of the suggested adjustments worked for me. Very little problems running on Ubuntu, there might've been graphical glitches but at least the games were playable.


I've had a similar experience. I would love to stop using Windows but I have found that ProtonDB wildly exagerrates the compatibility of games. I had game breaking issues with 50% or more of "gold" rated titles and even some with a "platinum" rating.


ProtonDB is driven by reports/feedback from individual users, right? Did you submit your experience so that the ratings are more accurate?


ProtonDB tells you that someone, somewhere at some point managed to get it working. It doesn't directly help you, but knowing it can be done can be an incentive to experiment instead of giving up.


To draw a parallel: message boards are 99% observation and 1% participation. There’s no need to goad your fellow HN readers.


That's unfortunately not unexpected. "Linux Desktop" is a ludicrously diverse not-platform for which Valve already has to bring its own runtime of basic libraries along just to ensure some minimum level of compatibility even for native Linux games.


Just dual boot and only use Windows for games? That's what I do but I haven't booted to Windows in months now.


Windows and Mac are not great for productivity and not easily customisable so if I want to be productive I need to work from a Linux environment and a Linux UI.

I don't game that much and rebooting / having another dedicated machine is a PITA.

I honestly don't remember experiencing that many issues with proton or wine for that matter. Maybe having to follow a wiki or installing something and a few minor issues. Proton automated most of that.

Sure, it's not perfect but for an occasional gamer like me it's less than the hassle of rebooting.


How Linux is more productive than Mac? I think it's true only if you use a lot of customization, like a custom shortcut to switch to a specific app/virtual desktop, and do some stuff. That would be extremely time-consuming but also really cool. I'm thrilled to try but with how prone I'm to falling into rabbit holes I'm really scared how much time it could consume. I already spent too much time on creating my own fork of Colemak and a lot of custom shortcuts with Karabiner elements.


After the hype related to the Steam Deck I decided to give Pop_OS with a GPU passthrough setup another chance. I got a single card setup functional but could barely get a dual-card setup working. Problems included:

* Needed a USB input switcher and two video outputs to use it correctly. I tried Looking Glass but it doesn't work well with NVIDIA cards unless you have one of those $15 HDMI dummy plugs which are all made by sketchy Chinese vendors... * Because I only have one USB controller, all USB 3 ports got stolen by the VM, leaving only USB 2 for Linux. * Weird audio pops that I thought I fixed but would come back on reboot. * Difficult to monitor certain temps. * Random crashes during long sessions, even on low graphics settings. Hard to debug - you have to pick through both Windows and Linux logs. * Random FPS drops, especially in LoL. * Lived in fear of getting banned from some games (Also LoL) * NVIDIA drivers would sometimes crash trying to rebind the cards when I closed the VM, so I had to reboot the computer anyways, might as well have dual-booted. * Setting up OBS takes extra work.

Gaming in Linux natively was, for all intents and purposes, the same. The games I play (mostly online multiplayer) are either unavailable or unplayable.

After over two weeks of my computer being semi-functional and my friends asking me for the fifth time when I was going to be back online, I decided it wasn't worth it. I see so many people in these threads praising Linux gaming, saying it's now functional and simple to set up. They must either only play single-player indie games, know something I don't, or are being dishonest.

Still want a Steam Deck though.


Very likely the former: the majority of proton users, by necessity, play only single player games (I'm one of them).

I'm trying VFIO at the moment, and so far (but it is very early) seems to be working (I also have an integrated video card which greatly simplifies things). I didn't pass through the whole USB controller, just the specific USB peripherals (mouse+kbd and bluetooth controller for audio). I assume this is not done by actual passthrough but there is some layer of indirection; what would be the disadvantage? more latency?


Latency and the ability to switch between the VM and the host. I have a cheap USB input switcher which I've owned for ages for this purpose.

IIUC unless you're running Looking Glass or some other alternative method, passing through each device individually locks them while the VM is running.


Yeah, people are definitely trying to oversell how well gaming works on Linux, especially now that the Steam Deck is coming out. I've had games rated Platinum and Gold on protondb be completely unplayable due to graphical issues, or fail to even launch. I've also run into Linux specific issues even for games that have native ports, for example, holding down a single key for a long time randomly causes the OS to stop sending signals for it until you release it and press again. It's gotten me killed quite a few times in fast paced shooters when holding down the movement key stops working.


Do you have an nVidia graphics card?


What graphics card and which drivers? It's especially problematic with older GPUs, but with modern ones it's largely fine and none of these issues show up as much as they apparently did for you.


> full OS reboots

while it's theoretically possible for this to be caused by software, it seems more likely considering the other issues you've listed that there are some hardware issues here. linux drivers exercise the hardware differently from windows, and usually more efficiently. if your hardware is weak (particularly PSU), it can fail during intense loads. see: people blaming prime95 for crashing their computer.


What games were you having trouble with? I haven't had this experience at all.


No one is claiming all Windows games work yet. Why would you assume they did?

https://www.protondb.com


Because some people are claiming that they all work, and Valve is saying that they plan to have every game working with Proton by the time Steam Deck releases. I've lost count of the amount of times I've seen people point to the "15.5k games work" number on protondb since the Steam Deck was announced, even though "15.5k games work" only requires a single person to say a game worked for them, and less than half of games on the site are rated as platinum.


>Because some people are claiming that they all work

Who? Just go to that site. It's tells what's working and not working. It says right there on the front page only 50% of the games are currently rated gold. You can't really complain that something doesn't work when they're telling you it doesn't work.


> I get enthusiasm for FOSS, but enthusiasm to the point of delusion?

Nothing wrong with being honest. We sometimes have to keep hiding the fact that when we use something on a system that is not officially supported and comes with tons of issues or bugs, we quietly fix it ourselves with hacks and workarounds and try to move on if possible.

Unfortunately with most users, they cannot tolerate 'Graphical artifacts, freezing, stuttering, and even full OS reboots' and would simply give up and use something fully supported like Windows.

> Honestly, Linux is not up to par with windows in terms of running games that are basically designed for windows. Thus the experience compared to windows for these things is definitively inferior.

If it says it is designed and optimised to run on Windows, then that's a hint that it will perform badly on other systems. You can 'try again' on other systems but obviously the help-desk guys will say: 'Sorry, that's unsupported.'

Better to go for something that has official support and move on, not half way there. Or wait until the other system has official support.


I made Linux Mint the primary OS on my son's first laptop some 3-years ago when he was around 12, and he's taken to it like a fish to water. He's even taken to using regolith as his (tiling) window manager now, and prefers to use vim for programming (python/julia/c) rather than any IDE.

Also have Mint installed on my mom's laptop - she only uses the browser and is ok with it - she has no idea of what an OS even is.

IMO, the only thing preventing Linux's more widespread adoption is the reputation that it's only meant for developers, and OEMs not providing an option to have it pre-installed.


Any time I have a family member who asks me to 'fix a problem' with their machine, my first step is to copy all important files, and then evaluate their use of the machine. 99.9% of the time, they use it for e-mail, internet browsing, and maybe home banking.

I put mint on everyone's machines and don't look back. It's just so stupid easy, and has drastically reduced the 'tech calls' I receive.

At this point, I have no idea what's holding linux back. Seriously, mint is so user friendly that it's sort of confusing why people don't use it.


Marketing and pre-installation on devices. Seriously ChromeOS and chromebooks have been gaining huge marketshare, Linux could've been that if there was any player interested and wealthy enough to put on a similar campaign to what google did.


I did the same for my kids and now my son is 19 and is now quite the Linux pro and advocate.

He corrects me “I use Arch btw”.


Both of my nieces are running Ubuntu, it's been a big win in reduction of support calls and ease of setting things up for them.

My only regret is that I should have been braver and put them on Manjaro. I thought Ubuntu would be more "set it and forget it" and maybe have better documentation, but some of my most annoying support calls I've gotten have been dealing with driver issues that just weren't present on Manjaro at all. And I generally think the Arch documentation is a lot better than Ubuntu's anyway.

> the only thing preventing Linux's more widespread adoption

I think I disagree with this in some ways; I think absent any support at all Linux would be difficult for some people. But if you have someone around who's good with Linux, I think it'll be more stable than Windows and have a lot fewer annoyances/problems.

I know that I would be debugging a Windows install more often and that it would be more annoying to do, because I have other family members who run Windows and I know what kinds of issues they're running into.


I switched from manjaro to Ubuntu recently after an update made my system unbootable. Took me a week to find the time to sit down and debug it, and the manjaro live USB wasn’t working at all so I had to try my repairs with a live Ubuntu USB. I never had something like that with Ubuntu that I couldn’t fix in 15 mins


I think it sometimes depends. If you're doing a lot of stuff with your system, the rolling release schedule means you get fixes faster, and Arch's documentation is pretty good for a lot of other stuff. Manjaro is more likely to "just work" with things, because it has newer drivers.

On the other hand, Ubuntu is pretty stable as long as you stick with what it provides and you're not going too far off the beaten path. Ubuntu feels very consistent to me, it just also feels a lot more stubborn, and I find myself fighting with it more whenever I plug in devices or configure it in a non-obvious way.

I think I just misjudged what side of that divide my nieces would be on. Small stuff like, I got them a drawing tablet and on Manjaro, it plugs in and the Gnome settings detect it, and the config is all graphical. Ubuntu has the same Gnome settings and the tablet works, but the settings don't detect the tablet at all, so to get buttons configured and left-handed mode working I needed to wire up xsetwacom and use dbus to detect when the tablet was plugged in. And that was made harder by the fact that there wasn't a lot of Ubuntu documentation for how that stuff worked, I basically used the Arch documentation and mentally translated some of the commands for Ubuntu.

But I think that's dependent on the person and what their needs are. If I had bought a more mainstream tablet for them, maybe they wouldn't have had that problem. My regret with Ubuntu is that I didn't realize how often I would want to be able to get things like a more recent kernel without compiling it myself. Ubuntu seems a lot more obtuse to me when it does break.

Everything is relative, they're both still Linux under the hood, and both are less likely than Windows to randomly break imo. I've gotten support calls from Windows family members about drivers just randomly not working one day because something updated behind the scenes without their input and started conflicting, about buying external displays that just don't get detected at all, and I don't even know how to start debugging issues like that. Even with Ubuntu, I don't really get that many support calls, mostly the computers just work.


That's a great idea.

It may be too late for me but my sons will be ninja developers who use vim like pros, instead of that Microsoft electron behemoth.


I'm so thankful that Valve is embracing Linux. They obviously have commercial interests behind it and fear the Windows Store [1] and vendor lock-in by Microsoft, but they have made gaming on linux better, even for non-steam games and embraced the open source ideals of linux, at least for their work on steam os and proton. I'm happy I don't have to have any windows systems for gaming anymore and I pretty much have valve to thank for that (and all the other great people working on wine, lutris and co of course)

[1] https://www.bbc.com/news/technology-18996377


I'm not a gamer, but annecdotally, the pandemic has given me enough screentime that I've finally been able to switch to Linux as my primary OS. I wonder how many others are in the same boat.


I've been gaming on linux for almost a year now and I've had no issues at all. I've mainly been playing popular titles but Proton makes every game feel like it's a linux native game.


Regretfully most racing sims (and any other sims that use peripherals) are still a huge pain to get going, even with Proton. At most you can get the actual game "running" (with performance sacrifices over Windows), but you still won't get things like Force Feedback, and half your peripherals features probably won't work (or will work intermittently for no reason).

I'd love nothing more than to never use Windows again, and I use Linux for 90% of my daily use, but if I have to have Windows still installed for any major games at all then it's just easier to keep using Windows for all gaming, where gaming really does "just work", every time.


"running" (with performance sacrifices over Windows)

Has this been an issue for you? Since switching to primarily Linux/Proton for gaming, I've been amazed at how little performance penalty I have seen.


I wonder how those force feedback wheels from different manufacturers all work on Windows, is there any kind of standardised API for that stuff? I had a brief look at sim racing peripherals on Windows and even there it seemed like each game you wanted to play had to support your specific model of wheel.


Force feedback worked fine on project cars for me


i use g29 with force feedback on dirt2.0 and f1 2020 without issues, the module is built in the maintree kernel


And sometimes better (like Classic Wow).

I only ran into issues with Rust(which is getting support) and pubg because of anti-cheat and the fivem mods for gta5 (something about lack of support for shared resources).

But it's working so well and so much less weird stuff going on compared to windows and osx.


I hope Linux with this Proton compatibility layer doesn't go the way of OS/2 of being so easy to run Windows games that nobody writes Linux ports anymore and as a consequence the performance and overall experience is never as good.


It doesn't help that the Linux ecosystem makes it really difficult to build truly portable truly native binaries.

Static builds? Well, that will work... But wait, you want OpenGL or Vulkan? You'll need a dynamic linker to load system .so's, as that's the standard API available to interact with these APIs on Linux. Want to load dynamic libraries? Well, you'll need to ship a dynamic ELF [1]. Shipping a dynamic ELF with glibc now? Hope you don't intend to run on NixOS or Alpine or any distribution that has an old glibc. Maybe also ship ld-linux.so, all of /lib you depend on and a wrapper shellscript that sequences startup accordingly via that ld-linux.so. That will work, but you still need to load the system OpenGL or Vulkan .so's. Have fun implementing that to work well on every distribution! Remember to respect terms of GPL/LGPL software you now ship. Oh, and I hope you don't use Qt.

Maybe you want to use flatpak? Well, better explain to your users that they need to install and configure it, and make sure whatever proprietary Nvidia drivers they have are also properly flatpak'd. What about the alternatives, something like AppImage/Snap? Those just don't work on Alpine and other non-glibc distributions. Whoops.

Pick an audio API. Pulseaudio? Some people don't run it. ALSA? Good luck, some people run Pulseaudio with ALSA emulation disabled. OSS? Dead, unless someone runs padsp. Pray your users don't run JACK. You'll probably end up on OpenAL, which is now proprietary and/or has an LGPL fork (remember you now have to let users re-link against any other version of OpenAL!).

I sometimes make games and other small interactive demos. I run Linux, and want to send builds to friends also running Linux. The easiest way to distribute these is to build and ship a .exe on a Windows build host and get my friends to run it on Wine, that just works everywhere out of the box. And the .exe's end up smaller and launch faster, too!

[1] - Or go for hacks like https://github.com/pfalcon/foreign-dlopen


Honestly I'd argue the opposite: that because Linux Desktop is such a fragmented mess of a non-platform, "native" ports are likely to work at least as badly if not worse than those running on a compatibility layer. There's even some anecdotal evidence that this is the case with games that do have native ports and running the Windows version in proton being a better experience.


The compatibility layer for the linux native games is the steam-runtime though. So fragmentation of the distributions matters to a much lesser degree already.


For me it had the effect of me being able to run Linux at home almost exclusively. and to be able to game occasionally with friends.

Tabletop simulator/ teriforming mars/ lords of water deep and a bunch of other games just work.

Getting that Linux percentage up should help the platform. I don’t want to go use my older machine and it’s non Linux. So I seek out desktop applications to drive Linux use up.


I don't see why it wouldn't have that effect. The obvious path to follow for gaming companies is write for windows and make sure it runs fine under proton.


I switched from Mac to Linux more than 3 years ago, and I've been gaming on Linux for 2 years. I'm far from a hardcore gamer, but I got back into gaming just because I wanted to see how far Linux gaming had come. I was pleasantly surprised. I've played quite a few native games, but also some running on Proton. I was amazed when Doom (2016) ran with excellent performance. The Witcher 3 is another one with great support. I only cracked and started dual booting Windows when I couldn't get Doom Eternal running properly. It felt like such a betrayal! But since then I've decided to be more pragmatic about things.

I still prefer running things on Linux where possible, and I sometimes buy native Linux games even if I don't know when I'll get around to playing them. There are definitely some rough edges, but I think there's a future for Linux gaming. I think it's mostly due to Valve, which is why I get all my games from Steam.


What is the point of these surveys when we know how many "level farming" operations[1] are going on in China, etc.? Casual CSGO has become unusable these days. Of course, apart from farming operations, cheaters come from across the world. And AFAIK, most cheats software is available only for Windows. Valve's esports-grade games like CSGO, and DOTA2 are free. There is no penalty for getting banned in such a context. It is easy for the "farmer" or cheater to create a new account, and start playing again.

I would be more interested in knowing what percent of paying customers run whatever OS. This info would be useful to new game developers too.

[1] I won't post links because my comment might get auto-flagged as spam, but interested readers can find them easily by searching for something like "Buy CSGO High Tier Accounts".


CSGO is free but has recently cut off drops for free users and grinding-based path to Prime accounts, leaving little incentive for farming as you mentioned.


The survey has always been opt-in continuous collection. I see no reason why bots would opt-in.

> Participation in the survey is optional, and anonymous.

First paragraph on https://store.steampowered.com/hwsurvey


Level farmers and cheaters are not bots. While level farmers may be likely to ignore surveys, because it would waste their time, cheaters don't have that factor.


The bots that plague TF2 runs on Linux. No VAC on Linux either so I suppose when they get VACd it’s a manual thing from Valve’s side.


Definitely the number is gonna increase dramatically next year with the Steam deck. Unless it's going to fail in some major way.


The odd thing to me is how immediately people jumped to "you can install Windows on it", like one would need to. The other thing that was funny is people saying "with Windows, you can run emulators", as if Linux couldn't already do that and with better framerates.


Lots of people (especially younger people who are used to consoles) don't seem to understand what Linux is exactly. To many, PC = Windows and !PC means there is no software for it. Many seem to be under the impression that Linux is just a commandline thing.


Yeah, this is why like another commenter had stated that it's so imperative influencers like Linus Tech Tips distilling how to install and first-time operate GNU/Linux to the casual tech enthusiast and gamer crowds that don't focus on software or free software movements. I'm hopeful to see this be the trend, especially if it goes to non-Ubuntu-based distros strictly to avoid monoculture in the ecosystem.


I have used the popular open source NDS emulator - Citra, and it runs like 15 times faster on Windows compared to Ubuntu (NVidia GPU).


Versus like cemu which you can see people in forums suggest running through WINE to get better FPS even without an working Linux build.


It's Valve hardware, so more likely than not it'll flop and you can pick one up for $50 on the 2022 winter sale. Even then it might be relevant for enough weeks to make devs consider designing for native Linux and be in too deep to back out.

Speaking as someone with a steam link and controller floating around their house.


I wouldn't mind it flopping an having the option to get it for 50 buckaroos :). But that's highly unlikely.

Even if it fails as a portable gaming platform the hardware sounds very decent and I wouldn't expect them to have Apple level margins on those.

I've skipped the link/steam machine because "livingroom entertainment" isn't my thing, but I was pleased with the quality of the Steam controller. Although I would have liked them to be bold with the design and keep just the original prototype (without thumb stick/face buttons) (wouldn't have minded a novelty controller that works well with just a few games).


Anecdata: I finally bought GTA V on Steam Linux now that protondb reports flawless single player operation (which I’m happy to report is accurate). It allegedly runs 20% slower than on Windows, but because I’m living 8 years in the future, relative to its release date, I now have a PC powerful enough that it can run at 60fps and max settings regardless of that 20% hit.

Fortunately I have no interest in multiplayer.


Steam has defiantly been putting in the work promoting Linux based developers and developing Proton (their wine package). Over the last year I've played a half dozen AAA games on Linux through steam, all of them working near perfectly.

I'm surprised the Linux market share isn't higher than 1%, giving the resources they seem to be dedicating to Linux.


They're now introducing Linux based hardware, which looks set to be super-popular, called the Steam Deck. So they're definitely using the Linux work in their own products. That should give the numbers a boost.


Steam Deck looks like a Wii form-factor device but they had already been collaborating with Alienware and others on the release of console style devices for a few years running SteamOS (based on debian).


Nowadays OS matters pretty little.

Personally, on some level, I don't even notice difference working macOS, Linux and Windows. Other than platform specific development work, it doesn't really matter in the big picture.

Other than that, only Linux allows me true visibility to its innards. Fast to fix almost any issues, should they happen. And I say this as someone who has for example no problem debugging Windows kernel remotely. Or reverse engineering using Ghidra etc.


OS matters a lot.

The fragmentation on Linux is astounding; I have experience with maintenance of Linux-based projects.

You'll be surprised at how fragmented distributions are, even within the same family (e.g. Ubuntu-based).

The nature of Linux is a double-edged sword, and in gaming, this is a very serious downside - at least, when it comes to Linux-native gaming; Proton/Wine is another story.


Would you say that the new flat and snap format alleviates some of these issues?


Personally I'd say no, not really. They're hacky and a lot of overhead for what is actually a very straightforward and natural thing on other OSs. AppImage is a bit better, but it still has to pull some namespacing tricks to work.


I don't understand how you say AppImage is better when it doesn't even provide runtimes and SDKs. The hardest problem is to know which APIs to use, and AppImage provides zero support for the developer there.


My personal opinion below.

- SteamRuntime: A runtime environment for Steam applications. The next generation is/will use Flatpak related technologies like Bubblewrap. It's what you target.

- Flatpak: there's close to no overhead, you can try the Steam Flatpak and see for yourself. It's aimed at desktop applications and the sandbox gets better and better. There's a FUD website called flatkill that has already been torn to pieces so don't even bother linking that. You can't instantly make all closed source or old software work in a sandbox without compromises. Flatpak-override and Flatseal exists.

- Snap: nobody cares about Snap not even Canonical. Recently a former employer said he gave up after trying for years to solve issues with Snaps in the desktop. https://s1.desu-usergeneratedcontent.xyz/g/image/1600/72/160...

- AppImage: nobody cares about AppImages they had almost two decades to try and prove it can work. It doesn't. The main developer is now trying to integrate Flatpak runtimes with AppImages and that's all you need to know to ignore this project. One of the most popular video players tried to use it and it kept crashing or having issues, and they know what they are doing.

tl;dr only Flatpak matters, SteamRuntime uses Flatpak concepts and tools, "huge overhead" is FUD


FlatPak requires a package manager and repos to work. AppImage doesn't. That's what I meant by overhead.


AppImage is simpler. It requires no management infrastructure, no repos, and no package manager. You can keep multiple versions of the same application around, you can store and run them on different volumes and media, and installation and removal is identical to simple file management.


The documentation claims that flatpak allows for single-file bundles, so repository is not required:

https://docs.flatpak.org/en/latest/single-file-bundles.html

Also, you can install multiple versions of an application, but you can only run one version at a time:

Note that flatpak allows to have multiple branches of an application and runtimes installed and used at the same time. However, only one version of an application can be current, meaning its exported files (for instance desktop files and icons) are visible to the host. The last installed version is made current by default, but this can manually changed with flatpak make-current.

https://manpages.debian.org/testing/flatpak/flatpak-install.... https://manpages.debian.org/testing/flatpak/flatpak-make-cur...


Ok, either those are relatively new features or they're just enver promoted anywhere.

Still, what I like about AppImage is that it is just a file that I download, put it wherever I want, and run. then delete it when I don't want it anymore.


The fact that you're being downvoted for pointing out the objectively true statement about fragmentation in Linux Desktop says a lot about the community behind it. Which, incidentally, I think is another big reason people avoid using it.


I think it matters a lot, specially on machines that are not that new, with modest resources


Maybe in developing countries this matters more, but RAM and SSD upgrades for older hardware are surprisingly affordable these days.


I would rephrase this: The share of users who could work well with any of the operating systems is increasing. The fundamental differences between the operating systems that impact user experience are decreasing.

I think 90-99% of users only use Browser+Office+Games which could work well in any of the big operating systems. Some software preferences are superficial and appropriate replacements exist on the other platforms.

As someone working with macOS, Linux and Windows I focus my workflow on the things that work under all of these operating systems so the impact they have on me is very little. People who are deeper into the UX of a single operation system may disagree completely.


> I think 90-99% of users only use Browser+Office+Games which could work well in any of the big operating systems.

Linux Desktop evangelists have been saying this sort of thing for about a decade now, only they used to exclude "games" from that list because Linux was so terrible as a game platform. Their constant underestimating of "average computer user" is, in my opinion, one of the big reasons they've completely and utterly failed to make significant headway in the Desktop market.


Just because it works well enough on other platforms doesn't mean that users will switch. Usually they won't. If they know MS Office and MS Windows from work they don't want to use LibreOffice in private. (And MS Office and MS Windows Server is far superior in the business context in many ways.)

I stopped caring what most people use. I prefer to use Linux and OpenSource software in general for myself but I don't push it onto others. It doesn't matter if the market share of Linux is 1%, 10% or 50% as long as it works for me - and it does.


Totally, it always bugs me when people complain that Linux is too technical and fragmented and that this way it'll never be mainstream. Who cares, really. I like having all this choice. In fact I'm using FreeBSD (though it could as well have been Linux) as daily driver because I don't get these kind of choices in mainstream software. I don't care if it never gets big, if it did I probably wouldn't want it anymore.

For example Ubuntu is getting very close to that point already. They stuff lots of stuff in it that I don't want (like snaps) and make them really hard to disable completely. Exactly the kind of thing I left Windows and Mac for (as daily drivers - I still use them for my work)


> I would rephrase this: The share of users who could work well with any of the operating systems is increasing. The fundamental differences between the operating systems that impact user experience are decreasing.

Thanks. This was indeed my main point.

Of course this doesn't mean there aren't a lot of niches that require a particular OS.


I find there is a lot of OS-specific knowledge for deep-diving.

Want to get symbols to debug the kernel? That'll be the microsoft symbol server on windows, or a System.map file in Linux...

Want to make something happen with every boot of the system? That'll be a system service on windows (and by the way they have all kinds of gotchas about what kinds of stuff they can access) or a systemd service on linux.

Want to make a file creatable, writable and deletable by a specific user, but disallow reading back the file contents? Thats easy with windows ACL's, but on linux you're going to have to mess with seccomp.

Need to script something quickly? Well thats either bash on linux or powershell/batch files on windows.

It turns out regularly using two OS's requires almost double the learning time to be as effective on both - there is overlap with concepts, but little overlap with implementations.


"Want to make something happen with every boot of the system? That'll be a system service on windows (and by the way they have all kinds of gotchas about what kinds of stuff they can access) or a systemd service on linux."

On Windows there's also numerous registry keys, Windows Task Scheduler, etc (ugh). On Linux, it depends. Could be also init.d or n+1 other options, including a single custom binary running as init process.

> Need to script something quickly? Well thats either bash on linux or powershell/batch files on windows.

Or bash on Windows (WSL, cygwin, etc.). PowerShell is also available for Linux. Both come with their footguns, especially when used on a "foreign" platform.

But you could just as well use some suitable scripting language, like Python, JS or whatever.

Sure, there's complexity if you really start (or need) to look for it. I do think it's exaggeration to say double, maybe 20%, at most 50%.


Windows has no epoll(), but of course that doesn't matter for games.


Windows does have I/O completion ports, which is, frankly, superior. Then again, Linux has io_uring.

Again, pointless in the big picture. Most of the time these details are hidden by libraries.


Btw Windows built an api based on io_uring https://windows-internals.com/i-o-rings-when-one-i-o-operati...


Nice! Haven't heard about this before, must be something really new?

Anyways given extremely high syscall cost due to Meltdown workarounds, batching multiple operations makes more and more sense.


well as far as I know I've read something about it either on lwn or hn and it's really really new. I'm not exactly sure if it is even possible to use.


> Again, pointless in the big picture. Most of the time these details are hidden by libraries.

Except for when they're not, or those libraries don't expose the "one little implementation detail", and a platform hack sneaks in, and _thats_ how we end up with "well it works fine on windows and we use asio so shrug". Unfortunately games are really bad for doing this in my experience (and I'm guilty of doing it on occasion too).


Windows has completion ports (IOCP), which are much better.

At least finally Linux is also getting completion based IO with io_uring.


Seems windows is getting ring buffers based io too, eventually: https://windows-internals.com/i-o-rings-when-one-i-o-operati...

Not identical to io_uring, but close enough that porting between them will probably be quite simple.


Sure, the ring buffer part of also useful.

The competition triggered part is what was already there.


> Nowadays OS matters pretty little.

> Other than platform specific development work, it doesn't really matter in the big picture.

Yes it does. It just doesn't matter to you.


Right. I can just talk for myself. I do think in the big picture what really matters for the most is really just a good web browser.

Thinking about this, I just realized I tend to pick OS according to device type.

For laptops, macOS for the best touchpad, security and speedy open-lid-to-machine-actually-usable time. But I'd sure like "upgradeability" and "repairability" over thinness, something 2012 Macbooks still had. But for now, practicality wins.

For desktops and workstations often Windows, sometimes Linux. Windows got good desktop performance and nice commercial software library. Freedom to choose your hardware, CPU and GPU. And of course gaming.

For servers, it's pretty much always Linux or possibly some BSD. Minimal, easy to admin, fast, efficient and secure if configured correctly.


> I do think in the big picture what really matters for the most is really just a good web browser.

If you prefer web apps. A lot of people don't prefer web apps and for good reasons.

> macOS for the best touchpad, security and speedy open-lid-to-machine-actually-usable time

I can't speak about the touchpad, but Mac OS isn't immune to vulnerabilities and there have been multiple questionable privacy practices. Linux tends to cold boot faster, and can suspend and unsuspend just as quick if not quicker.

> Windows got good desktop performance and nice commercial software library

And Linux has even better desktop performance. Many people would see the open source software benefits of Linux as a pro, not to mention Windows package management still being significantly behind Linux. Many people are also very happy with gaming on Linux.

> For servers, it's pretty much always Linux or possibly some BSD. Minimal, easy to admin, fast, efficient and secure if configured correctly.

The same can be said about desktop too.


> Nowadays OS matters pretty little.

Unless you use iTunes to sync your music to your computer and iDevice or you're a music producer in need of using the best DAW software, or perhaps you're a creator that uses the Adobe suite to create the best media content you need, etc, then the OS does matter very much since millions of users still use this type of software.

The most important part of this is that the software must be officially supported; and not be at some 'experimental' stage, especially since these users do not play around with their computers and they focus on getting things done.

macOS and Windows users are the 'real' big picture and the 99% of users and that matters. Not the 1% of Linux users.

To Downvoters: So it makes business sense to ignore the 99% of users on Windows and Mac and target the 1% of your userbase running Linux and support as many distros as possible to cover the many users who could be running whatever distro they have installed?

Have you considered what the 'definition of Linux support is?' and the cost of testing for all of that is?


I buy games and purely use Linux and Steam for this purpose. It works perfectly, including AAA titles! Love from here to the Proton team at Steam. Even Penguins want to have fun.


I was always dismissive of people complaining about Linux as I’ve been using it for the last 15 years without issue. But then I got an nvidia optimus laptop as a gift recently and I just gave up using Linux on this machine.

1. Suspend on lid close simply didn’t work. I’ve tried all kernels I could get my hands on. I resorted to remapping the bogus key that the lid close was sending to suspend.

2. My laptop have the hdmi output wired on the dGPU. So I have to live with a world of pain that is mixed DPIs on Linux where only wayland have proper support.

3. Wayland got some nvidia love on 477 drivers but, in my test, it was a big mess.

So yes, I stopped dismissing people complaining about Linux and went back to windows on this laptop. I assume that nvidia and high DPIs are a common enough usecase in 2021 that I would not recommend anyone to go through this experience.


My experience is that many times the (emulated) Windows version runs better than the 'native' Linux version, which can have not-so-subtle bugs or not run at all.

This especially happens for older games which haven't been receiving regular updates. Apparently Windows userland bitrots much less.


The Windows 11 announcement is what made me transition to Ubuntu in the last few weeks. So far, I made the transition on 4 out of 5 of my windows PC.

It's far from smooth enough to become mainstream (with ubuntu at least.)

A few example: Steam wouldn't launch on my PC with an Nvidia card. (At to install some x86 package to make it work).

File sharing just didn't work at all without manually edit the smb.conf file.

Ubuntu randomly ignore my router DNS settings. All my self-hosted stuff requiring local DNS randomly stops working unless I force ubuntu to use my own DNS server.

Some games just don't work on any of my PCs (even if they are gold/platinum in protondb).

And many more stuff that works out of the box with windows. Overall, despite the issues, the experience is still positive and i'm not going back.


I've started playing a lot on Linux because it is the only place I can play some of my games. A lot of old 32-bit games are starting to become unplayable on Windows 10 but are perfectly fine on Proton or Wine. Proton has been the best thing ever for me. I no longer have to keep two versions of Steam on my laptop (Windows and Linux)

On the other hand. There is rarely a moment I play any non-linux native game on Proton that I don't have to fiddle with something. Also a lot just end up not working. It's gotten a lot better recently, but stuff like controllers sometimes end up not working at all and needing a hack to work.


> "While many believe the Steam Survey is inaccurate or biased (or just buggy towards prompting Linux users to participate in the survey)"

Are there other, possibly better, metrics used to track desktop linux adoption?


Some problems mentioned on the comments on here, although I can think of other problems too than only the ones mentioned. So, I can think of another possible solution to these and many other problems (although it has limitations, but there always will be): Don't design your game for Windows or Linux; write it for Famicom or Game Boy Advance or whatever (or design your own open source VM if you must, and use that). Emulators are available for many computer systems (including the less common ones), and will do many things for you automatically too that you will not have to deal with by yourself, such as rebinding controls, zoom levels, etc. They will then also the user will not have to worry about malware as much, either.


I wonder how Epic feels about this recent surge and the massive interest in the Steam Deck, a device shipping Arch Linux of all things stock.

They have dropped support for Linux pretty hard for anything but Unreal as a build target itself, even with things they acquired. Might have been the wrong gamble.



We depend on Steam too much for this information, we need better (more verifiable) metrics given that Steam has a direct commercial incentive to promote Linux gaming.

I'm not saying they're biased or futzing the data (have no reason not to trust Valve yet), but it would be really nice to have an independent source for this (especially moving forward).


Interestingly, Linux is the dominant gaming platform. It's just that it's called Android.


Perhaps the headline should clarify that it's talking about GNU/Linux


It's always useful to differentiate between embedded, mobile and desktop/server-grade operating systems.


> Linux is the dominant gaming platform.

Given that the article is obviously talking about Desktop Linux, since you can run Proton exclusively on Linux distros (and not Android) the statistics says 96% on Windows and 1% on Linux.

So how is Linux itself a 'dominant gaming platform' here when it is obviously Windows again? It's always cute to see lots of denial of the results here.


Android isn't Linux. Linux is just an implementation detail, which could be swapped for something else and the apps wouldn't notice.


This is and has always been kind of the point of Unix-like OSs. Blackberry had a workable Android 2 runtime on top of QNX.

You can (at least could) run a Debian on top of a FreeBSD kernel. I run a lot of Linux code on MacOS too.


Android is not exactly open, and Google Play Services is not FOSS. Its games are also mostly addictive micro-transaction-ridden trash.


The parts that aren't open source are the google-specific pieces. Amazon has a thriving ecosystem (of mostly addictive micro-transaction-ridden trash) on top of their FireOS, which is mostly Android with the Google parts replaced with Amazon parts.

Still, Blackberry managed to make it easy to repackage Android (2.x) software to run on their QNX-based OS and there have been multiple demonstrations of Android software running on Linux with a relatively thin API layer. There's at least one commercial, IIRC, tool that does that for Windows.


> Its games are also mostly addictive micro-transaction-ridden trash.

Yeah, I'm not sure we should call the mobile crap "games".


Any system with a low barrier to entry will be flooded by games that are cheap to develop and quickly recover the investment. There are good games, but the platform is generally much less capable than either a desktop PC with a beefy GPU or a console.


Eh, don't blame the platform's power for IAPs.


What are the odds of Proton helping to improve Mac gaming?

Will those improvements to Wine trickle through to the Mac eventually or they too linux-specific ? (I'm not that up-to-date with all the graphics APIs they all have nowadays, and I assume that's a factor)


> Proton does not work natively on macOS, end of story. macOS is missing several APIs to make this support viable:

> eventfd syscall is Linux-specific, without good alternative on macOS

> Apple does not support Vulkan, which is needed for DXVK

> Apple deprecated OpenGL support, which is needed for WineD3D

> macOS is missing support for Python 3 OOTB

https://www.reddit.com/r/linux_gaming/comments/gt3fat/proton...


Proton's big tech under the hood is DXVK which translates DirectX calls to Vulkan. If MacOS were to add Vulkan support it probably could benefit from Proton. But Apple doesn't seem to care about Vulkan or modern graphics APIs in general.


It would be great if Linux will start to cater games natively on their OS. The number shows that it is slowly increasing; I hope that the steam deck will not flop and will surely deliver its results.


I just switched from dual booting Mac / Windows to Linux and thanks to proton I can play anything I want to play without dual booting.

On Wayland none the less! The year of the Linux desktop! \s


My exp with linux gaming is that for sim racing its not possible. To my surprise the logitech wheel did not work, and thus forced me to play Assetto Corsa on Windows.


Anyone knows if you can get absolute values for that 1%?

Couldn't find them on the steam survey page and I'm curious how many millions of PCs is 1%.


Steam has about 120 million monthly active players, with about half playing on a given day: https://steamcommunity.com/groups/steamworks/announcements/d...

Assuming Linux players are representative, that puts the number of Linux users at 1.2 million per month and 600,000 per day.


Maybe but I think the hardware survey is opt in? I get asked occasionally if i want to be counted.

So less than that.


I am in the 1 percent !


How long before we see “Steam Linux”?



Okay - I should have been more specific. Can I run regular KDE or gnome apps (a such as gimp), or perhaps something like VSCode on it?

I thought that it only ran the steam app as opposed to a real desktop?

(pardon my ignorance here - I’m vaguely aware of SteamOS, but never tried it)


from Wikipedia:

> for example, there is no file manager or image viewer installed by default. Users can, however, access the GNOME desktop environment and perform tasks such as installing other software

It's "just" Debian with some default configuration and extras (and version 3 will replace Debian with Arch)


I’m thinking Ubuntu could do it as “Ubuntu for Gamers”. It would probably go a long way to getting a common set of infrastructure for developers to build on.


……………yay…………….


Now lets join as we prepare for the 2% celebration, about 2041.


I only use Mac and Linux computers.

I don't play games but became interested in VR, specially for programming 3D worlds . Valve basically forced me to install Windows in order to use vive headsets because the linux versions were problematic.

BTW, android uses the Linux kernel, so I consider this 1% highly suspicious.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: