Hacker News Reader

Comment by neverkn0wsb357 | Hacker News Reader

Given the current momentum, it feels like (to me) the adage of “Windows is for Games” is going by the wayside.

If you look at Steam, and OSs like Bazzite it’s clear the consumer-side is finally shoring up. But that aside, from an economic incentive, game providers (for example Amazon Luna), don’t want to be paying the licenses for running Windows machines for Video Game Streaming on Demand. In fact, at my time there one of the major thing I worked on was figuring out how to stream the games using Linux + Proton + Vulkan so we could use the AMD machines.

Honestly the biggest hurdle was (and probably still is) Anti-Cheat and BattlEye.

At any rate, I’m personally happy to see this trend as I haven’t had a Windows OS since Windows 7.

View on HN

>Given the current momentum, it feels like (to me) the adage of “Windows is for Games” is going by the wayside.

I think that games have been a strategic priority for Windows for a very long time. Going all the way back to DOS/4GW on Windows 95. But the impression I get from Microsoft is that they kind of don't want the hassle of maintaining a desktop OS anymore, and they would be happier if everyone went elsewhere.

The last few years it has felt like that, Microsoft is more than happy to sell to everyone while also having Windows.

I mean Windows is still a huge cash cow for them and is THE desktop OS but the actions they are taking with it sort of makes it feel like a second class citizen.

> and is THE desktop OS

Part of the problem seems to be that desktop OS use as a whole is cratering as more and more folks who grew up in the smartphone era enter adulthood. Outside of tech circles, I meet a lot of folks who have a phone + tablet but no actual computer...

What's a computer?

It's pretty clear from the context that I meant "general purpose PC". We don't have to get all semantic about everything

I guess this was a quote from this iPad ad: http://youtu.be/Ixy2ql7g3-I

I'm not sure. They put a ton of effort into things like DirectX, were outright anticompetitive against OpenGL, and there was an atypically high degree of competence and vision throughout. It probably wasn't about the games but about tying people to Windows. People didn't make games for Linux because there were no gamers on Linux. There were no gamers on Linux because people didn't make games for Linux.

On top of this, gaming used to be (and probably still is) the main reason to cycle through PCs. If you're just going to browse the web, use relatively low resource software, etc then a PC or even laptop from a decade+ ago is 100% fine. The reason consumers upgrade is going to be heavily weighted by games. And each of those upgrades often comes with new OEM software that was licensed and other economic benefits to Microsoft.

---

As for modern Microsoft, I agree with you from an outsider's perspective, but I'd bet internally it's a different game. Microsoft seems to be having major issues with labor competency, on both the implementation and management side, and it's making their entire ecosystem collapse. Anything that has major outward visibility (like desktop OS) is going to make the circus most immediately visible. I have little doubt they have the same stuff going on internally with their other offerings.

I think that's the key explanation. Gamers are no longer a mass market user group for microsoft - they target all the casual computer users. Gamers actually want full control of their computer, pay attention to details and care about performance: that's a niche user base, and it's clear why their interests overlap with the linux community: user needs trump everything.

I don't know if I'd agree with the adage about gamers wanting full control. A subset of gamers, absolutely.

But this excludes the entire console population. This arguably excludes most Steam Deck customers, who picked it because Valve made the Linux experience seamless, so they don't have to pay attention to the details. This excludes many of the PC gamers I know, that do not care beyond whether their computer is capable of playing the games they want to. They won't even reformat their Windows to remove OEM bloat.

I think you are way overestimating how technically inclined the average gamer is. Most of them just want to play a couple rounds of fifa or madden or call of duty to blow off some steam after work, not to spend extra time and effort optimizing their gameplay experience besides maybe buying a nicer keyboard on Black Friday.

You don't tend to hear them online of course. They are the silent majority keeping the AAA industry alive.

> the adage of “Windows is for Games” is going by the wayside.

This is the last major reason for anyone to use Windows nowadays, with the exception of legacy applications.

Windows' days are numbered.

You’re going to be counting for many many years

The progress is vast, but at the same time, it feels overstated.

Linux still is not a great daily driver for video games in many circumstances, unless you're on a specialized device like the steam deck that gets extra attention to smooth out the rough bits.

On my gaming PC I haven't found a single game that runs noticeably faster in Linux. Most run considerably worse often while suffering various glitches (sometimes game-breaking).

Sometimes, with work (different versions of proton, startup options, configs, or even new kernels or compositors, etc) you can get around those problems, but... it takes work. Work that you just don't have to do on Windows.

> Most run considerably worse often while suffering various glitches (sometimes game-breaking).

That's an interesting experience, I'd be interested to hear more. There certainly are games that do not work well, no question, but as far as I'm aware it's a pretty small minority. To my knowledge, the two biggest issues are anti-cheat and video codecs, both of which are business/legal problems, not technical issues. Are those the main problems you're seeing? If not, are you possibly running fairly niche games, or on a niche distro or specialized hardware setup?

Re-reading that, "most" might be too strong a word - "many" would be more accurate. A few recent examples of games I've tried:

- Borderlands 4 was basically unplayable on my hardware (9800 X3D, 3080 TI) - though I didn't care enough to try and fix it.

- Dune Awakening was decent, but noticeably less performant, stuttery, etc. Probably fixable with some settings tweaks and other stuff, but the experience was markedly worse than windows out of the box.

- ARC Raiders runs fantastic - but even still, it had noticeable visual issues particularly with shadows

General issues:

- It seems to vary by desktop environment how confused steam and/or the games were as to which monitor to play the game on

- Steam itself required some futzing to get big picture to use hardware rendering (software rendering is very laggy)

- Multiple games seemed confused what my native resolution was

- Mouse issues with multi-monitor setup in several games (though sometimes this is an issue in windows too)

"Many" is definitely more in line with what I'd expect, yeah :) Sounds like most of your issues are multi-monitor & windowing related, which isn't surprising.

Games make a lot of assumptions based on how Windows's one-and-only window manager operates, stuff like windowing message and focus event sequences, effects of various windowing states on window sizes and chrome and mouse cursor behavior, and so on. Linux WMs don't match Windows's behavior or even other WM behaviors, so it's a nightmare trying to get every WM to align to how every game expects Windows's WM to behave. Then multi-monitor adds another layer on top of that, for things like reporting resolutions, cursor behavior, window focus, etc.

We focused on the big 2 (Gnome & KDE) on X11, and personally I use multi-monitor XFCE on X11 so I was quite motivated to get games working well there, too. Plus SteamOS's compositor/manager on Wayland, obviously. But there's so many combinations affected by so many things (I didn't even mention graphics driver behaviors on any of the above...) it's just really hard to get right as you add more little edge cases. And as you said, many games get it wrong on Windows, too. We'd often reproduce bugs on Windows just as they were reported against Proton.

All that is to say, yeah, I believe that has been your experience now that you've explained a bit more :)

Ah wow I see you did work for Code Weavers. You guys are rock stars!

Once upon a time I was a paying customer (like in the early early aughts). Glad to see them still doing their thing.

Nah.

Linux is still too bloody awful for power users, never mind the median gamer.

Most Linux usage is SteamOS which only barely counts.

It’s a great hedge that keeps Windows almost honest. But we’re a long long long long long <breathe> long long long ways from the median gaming PC being Linux.

Linux is for power users. Windows 10 with Powertoys and WSL for stuff like yt-dlp is a fine stopgap, but you can get the same workflow on Linux with a leaner system.

I never installed Windows 11 on any of my PCs, there's no place for it in my work or gaming regimen. If Linux is supposed to keep Windows honest, then some dev at Microsoft must have a Pinocchio nose.

The problem is that Windows power users arrive on Linux and think they know what they're doing, when in reality they don't know up from down. Very basic things like norms about user confirmation and warnings are frequent stumbling blocks.¹

Windows power users expect their habits and instincts to be right and treat the system as broken wherever they aren't. After all, they "know computers"! So when one of them hits a snag, even if it would have been avoided by heeding a system's warnings, reading the documentation, or adhering to its norms, they declare (for others to repeat) things like "Linux isn't (ready) for power users".

1: Windows power users arrive to Linux with a mixture of incredible fatigue from pop-ups and blindness to all interruptions. They are used to mindlessly batting away constant notifications and distractions. They are also used to a host of familiar warnings that they know are bullshit, and reflexively ignore. But the warnings on Linux systems are not the warnings they know. They don't actually know what they mean or which are safe. To the point that their blindness to warnings becomes outright comical, as in this infamous example: https://i.imgur.com/J39WfLK.png

That imgur link is just sending me to the image template for the "I'll Fuckin' Do It Again" Goofy meme. If that was intentional than what youre referencing isnt as infamous as you thought as I have no idea what it means in terms of windows users on Linux.

Lmfao, thanks. Damn indistinguishable imgur URLs on my clipboard. Fixed.

Yes. And Linux power users are also terrible at Windows.

All the operating systems are very different but also largely the same. There are a ton of dumb little things you need to learn for each of them. It’s annoying. Most people don’t spend the time and just go back to what they and works.

I don’t think your comment is particularly insightful or interesting. It’s bias way yonder too overwhelming.

The insight is that power users for another OS are a bad measure how good an OS is because power users are inherently specialized and thus often have more difficulty using a different system.

I'd bet even more Linux usage is ChromeOS which even more barely counts, and certainly both are dwarfed by Android which simply doesn't count.

I am really curious: what is your definition of a power user.

Double clicking an icon real fast and be the last 10 players consistently in a public match of fortnite. That's a true power user.

What do you consider a power user? Because I'd consider an OS that refuses to let you do what you want, and constantly reverts your customisation, the opposite of power user friendly.

We're a long way not because Linux cannot do it. We're a long way because publishers refuse to take it serious.

If you call yourself a power user and cannot use Linux properly, you are not a power user.

This is the kind of gatekeeping that proves the point. "You're holding it wrong" doesn't work when we're talking about mass OS adoption.

How are these statements compatible?

Like if most linux usage is SteamOS that suggests its good for gamers right?

And that all any other distro has to do, is target SteamOS in terms of gaming usability?

SteamOS is Linux.

Online multiplayer games keep trying to allow linux users in and keep having to lock them out because there's an instant influx of cheaters.

The Nintendo Switch (which runs Linux) was a favorite of cheaters after jailbreaks came out.

When anyone can compile and run their own kernel with god knows what for modifications, that makes it substantially easier for cheaters and substantially harder for anti-cheat. I don't see that ever changing.

You can't rely on server-side detection either, because some of the cheats are so advanced they go to great lengths to "behave" like a highly skilled human player would with their aiming

The Nintendo Switch runs a custom operating system codenamed HorizonOS.

I was going to say. There's no way Nintendo would be caught dead with GPL anything, even GPLv2.

>The Nintendo Switch (which runs Linux) was a favorite of cheaters after jailbreaks came out.

If you're saying the Nintendo Switch system software is Linux-based, I don't think that's correct. It's a proprietary system based on a microkernel architecture.

I think it's a FreeBSD variant, if I remember correctly.

It's definitely not. Completely bespoke OS.

ps* is based on freeBSD, Nintendo is proprietary (though use OSS libraries)

It's proprietary, but they kit bashed a bunch of existing components (iirc, Android's surface flinger, the Nvidia embedded linux drivers, and the freebsd network stack)

The status quo's days are numbered. Online chess shows how.

An AI will play these games like a human but better. The AI can be totally separate from the windows box wearing anti-cheat ankle bracelets just as your brain a separate thing to the windows box when when you play. It can interact with the box via keyboard, mouse or controller.

No windows kernel module is useful in detecting and deterring chess cheating no matter how fanciful or factual the vibrating "device" stories are.

Anti-cheat by kernel module, it's day will be entirely done very soon if it isn't already.

"Any time you beat a computer at a game it let you win." Are we there yet? If not, how long?

> "Any time you beat a computer at a game it let you win." Are we there yet? If not, how long?

I don't want to beat a computer, I want to beat another person.

It never was fair to play vs computers in reaction games or skill games.

IE: Quakebots and Fighting games have perfect reaction times and perfect combos. They can simply block perfectly and counter attack perfectly and never drop a combo.

You act like cheating is new to video games??

--------

We never wanted bot in these games. Still don't want them today, and it's a big reason that playing on public boxes (ex: at an arcade or eSports tournament) is still a thing.

Defeating an opponent in a tournament is a big thing for fighting games. The risk of cheating online is always there so online tournaments are simply never taken as seriously (ie: as much $$$$ risked as real life tournaments).

> You act like cheating is new to video games??

No, I think the point is that with AI the existing anti-cheat measures can simply be avoided by letting the AI play through the same interface as a human. Therefore anti-cheat kernel modules will no longer be useful, and will no longer be a reason to stay on Windows.

It seems like what this needs is the return of video arcades.

Fill a room at the mall with Linux boxen with midrange GPUs and fiber internet and the sort of keyboards you can clean with pressurized water. Charge an entry fee and then sell pizza, cheetos, coffee, soda and beer. Open at 11AM and close at sunrise.

Then publish the public IPs used by the arcade-owned machines at each location in the chain and use different public IPs for the customer WiFi. No DRM nonsense, just a way to know you're playing with someone at the arcade where the management doesn't allow cheats on their machines.

Yes exactly but you do not go far enough with your plans. What is the point of any game if we can not determine who has memorised the meta best and who’s fingers twitch fastest. We need to out law general purpose computing in society and first it must be slowly phased out. Humans have shown they can not be trusted with open platforms they will always cheat and scam each other to gain an advantage. We will also need eye tracking devices to determine if they are cheating by reading notes off paper nearby. I think your plan comes to perfection if we chip everyone in case someone else plays for them on the locked down device.

No one is going to use LLMs if aimbots are available.

Have you even played an FPS vs an aimbots before?

Chess anti-cheat now relies on looking at your moves and spotting mistakes. Not even grandmasters play tactically perfect games so this works pretty well for finding cheaters. In theory FPS games could do the same to detect aimbotting.

and will have to do something along those lines for online play.

I still don't understand why we aren't using server-side gameplay analysis for cheat detection. You can have some obvious inhuman-level gameplay heuristics for real time kicks/timeouts during matches and post-game analysis by AI to flag for review or outright automatically ban gameplay that deviates from normal high-level players.

Games very much are using server-side statistics analysis for cheat detection. Valve made a presentation about it and Epic has an API for feeding game state data to ML anticheat for aimbot detection (game-specific and in addition to their existing anticheat measures)

It’s just that it doesn’t work.

But why doesn't it work?

Either everyone on Earth who’s working on this has a skill issue (which is probably hubris?) or there’s not enough differing humanized enough aimbot from human aim (note: Valve manages to screw up even here, with cheaters in Premier basically rage aimbotting these days IIRC)

In addition, there’s not much these things can do against subtler stuff like ESP.

So now we're using an AI cheat snoop to detect the behaviours of AIs, which means the cheat AI will need to learn to avoid the tell-tale patterns the AI cheat snoop looks for and avoid them, which mean the AI cheat snoop will need to....

It won't close the skill gap bump, but the more an AI aimbot degrades itself to mimic a human to beat cheat detection the less advantage it will give to the players using it.

It is scary how nuanced the cheating tools already here. Here is a video promoting cheat software explaining how nuanced their aimbot system can be made to mimic real play: https://www.youtube.com/watch?v=hrBohlkHMjU

> existing anti-cheat measures can simply be avoided by letting the AI play through the same interface as a human.

Great. Now we are going to get “secure cables” for mouse and keyboard and bluetooth device attestation.

You're already one step behind. The interface the human player is using isn't USB, it's genuine keyboards and mice.

I generally agree with this conclusion but there are a class of cheats that don't just improve the player's speed/aim/reaction time/etc but actually give you information that you shouldn't have but needs to be available locally to hide network latency like players on the other side of a wall.

China solved this years ago with mandatory ID verification to play video games.

https://en.wikipedia.org/wiki/Real-name_system_for_online_ga...

This was to prevent children from getting addicted but also leads to real life penalties for cheating in video games.

This a new and exciting use of the word "solved." I don't think I'm completely comfortable with it, possibly due to being raised on a diet of anti-soviet propaganda that spoke of such thing as "freedom" "justice", not sure if people care much anymore but there it is.

In the time span of Nixon to Trump and Mao to Xi, America got caught by China.

The idea of Mao's face or Trump's face on the global reserve currency feels really off.

Your argument appears to be nihilistic in nature: "don't bother fighting the cheating because it's inevitable." Forgive me, but I won't be giving up that easily. No anti-cheat is perfect, and we're not aiming for perfect. We're aiming for a reduction, and the harder we make cheating, the fewer cheaters there are. If cheating requires special hardware to mimic mouse and keyboard input, that significantly cuts down how many cheaters one will encounter in a given day. I have no doubt that the threat vector here widens and deepens as AI becomes more integrated into our operating systems. That does not mean we should give up or accept cheating as inevitable.

Comprehension issue.

"Quotes" of something not said directly that you are quoting is always bad. Quotes containing a summary that is entirely false are even worse. Don't do that.

You can try to deal with cheating as chess.com and others _do_, or you can do something you know will not work. Kernel mods for chess.com would be stupid. Their anti-cheat strategy involves zero windows kernel mods. That is how it will go for all online gaming if it hasn't already. So use something else.

Windows kernel modules won't work much longer if they even still do. Pretending they will, doing nothing to stop cheating is a a nihilistic, give up, faintly ridiculous attitude.

Using quotes to intentionally misrepresent someone is of course bad. I feel I accurately represented your position and nothing you have written refutes that. In fact, you double down here by claiming that chess.com's method is a reasonable solution, and vaguely gesture at "something you know will work." Chess.com relies on inference, which is wildly inaccurate, with many false positive and false negatives. There have been many scandals over the years, and you would know that if you had spent a few minutes on Google prior to this comment.

If you disagree with my interpretation of your nihilism, please go ahead and provide a workable solution for significantly suppressing cheating without the use of kernel level anticheats. Right now you're talking about of both sides of your mouth.

Your reading has serious issues.

“Something you know will NOT work” Is what is written right on the page. You quote it without the “not” wTH? Kernel mods won’t work much longer if they still work at all. Gotta find something else. What? Let’s see. Won’t be kernel mods.

Doesn’t make much sense to see this level of non comprehension of plain English. There’s no point talking against it what i say makes no difference if you read the opposite. Maybe this is convo with AI? Dunno.

Best.

> You can't rely on server-side detection either, because some of the cheats are so advanced they go to great lengths to "behave" like a highly skilled human player would with their aiming

Shouldn't that be the goal of anti cheat? That cheating is indistinguishable from expert gameplay? Seems to me like these companies are just trying to avoid implementing proper infallible server-authoritative gameplay by offloading the cheat detection to the untrustworthy client, and then trying to lock down the client to make it trustworthy.

All these games already have an authoritative server. These cheaters aren't breaking the rules of the game by being invincible, super speedy, etc. they're aim-botting and wall hacking. Those cheats can't be prevented with authoritative networking.

why can't we prevent wall hacking by not sending packets of enemy players position if the user can't see them on their screen

you run into some really difficult to solve issues when it comes to the gameplay loop, the graphics engine, and network latency trying to solve that unless you are playing some sort of turn based game where all data can be resolved before the next action

You can do it to a degree (basic room detection), but it'll never be 100% accurate because of latency and compute cost, you have to give leeway.

Sometimes user can partially see them, game client would need that position. Then user can make a mod that flashes silhouette that just appeared behind a wall for a moment.

not a trivial solution.

this is something cool done by valorant - see https://www.riotgames.com/en/news/demolishing-wallhacks-valo...

I dont know what kind of authoritative server Apex Legend uses that let the infamous "Tufi" hacker do what he did for so long. IMO it should be trivial to ban someone hitting more than 80% of their shots as headshots,, dual weilding weapons that were never supposed to be dual-weilded. Charge rifle beam permanently shooting and swapping armors from miles away

There's a big difference between authoritative networking with some security holes and non-authoritative networking. Note also that hitting 100% headshots is entirely within the rules of the game; it has nothing to do with the networking implementation.

And yet there's plenty of competitive multiplayer shooters that work fine on Linux. Rivals, The Finals, deadlock, CS2, Overwatch, Hunt Showdown, etc.

EA did a big announcement about switching to kernel level Anti-Cheat for Battlefield 6 to combat cheating, yet there's still plenty of cheaters around. It's looking more and more like an excuse in order to give the appearance of combating cheating.

Nintendo Switch does not run Linux, it runs a proprietary OS called Horizon based on the Nintendo 3DS firmware. Not sure but it might or might not have some BSD code in the network stack or something.

Pretty much everything uses BSD source code in the network stack, including Windows, so that much is a safe assumption, but there's far more that the Switch is using. According to the copyright notice, it uses the FreeBSD kernel. This tracks with reported use of BSD jails, which are part of the BSD kernel.

i am really doubting that Nintendo Switch uses Linux?... they would have to provide source code no?

Only to the kernel. And only if they patched it. Kernel modules don't count as patches. This is how myriad android vendors get out of providing source. But no, the switch doesn't use linux afaik.

What is the problem with cheating or bots really?

I feel that the solution is just to have a decent ranking/level system so that users play with other people, cheaters, bots or regular users of the same level. When I was playing mario kart with my 5y old daughter, I didn't mind she had access to helps to not run out of the road as it allowed us to play together. I don't see how different it is between say, a super skilled player, and a lower skilled player with cheat/assists. If cheating/assists system becomes so efficient, cheaters will just end up playing together and non cheater will have got rid of them and play between non cheater of similar level. Prolem solved. No?

People who are not naturally competitive (or who don't like competition) struggle to understand the drive. For many, competing against a computer is FAR less engaging or rewarding than against a person. Even if the person is less skilled than the computer. The human element - chance, variance in skill, emotions, etc - is very motivating for some people. The same people often like sports (which I do not). If you replaced every athlete in every competition in the Olympics with a robot, would you still find it as compelling? Some would, but many would not.

You should not consider Tim Sweeney's comments on the matter as a reliable source. He was veiling his true motivations behind that statement. The Switch does not run Linux either, it's a custom OS descending from the Wii's iOS.

The cheating issue isn't really a matter of being able to run custom kernel code. You can do the same thing on Windows, which is why remote attestation is a thing for some games. As someone who has developed games for Linux (and Windows / Mac), it's an endless cat and mouse game. So long as the system can execute code that is not yours, you never really are getting perfect anticheat. Ease of loading custom kernel code isn't really a hurdle to that.

I find that client and server based in combination is the robust approach. I once implemented anti-cheat in which the server lied about game state, which a regular client without cheats would act predictably on. Deviation from that behavior is a useful heuristic to build a suspicion score.

Reply on news.ycombinator.com