Posted by haunter 3 days ago
It helped that the DOS executable format was the same as the CTOS format - because we had traded Bill Gates our linker (which produces executables) for his BASIC compiler.
Thanks for sharing, never heard about it before. What was kernel programming back then? Briefly checked the wikipedia and looks like CTOS was kinda big in the government space back in the 80s.
The kernel was in Intel ASM86 but the rest of the OS was written in PLM86. When I joined it was 2MB of code on a 128K 8086 cpu. By the time I left it was 9MB of code running on an 80386.
What does this mean? System calls?
https://www.geeksforgeeks.org/operating-systems/traps-and-sy...
https://www.nxp.com/docs/en/reference-manual/M68000PRM.pdf see page 292, also see page 629 for the table of "exception vectors" (addresses for code to handle each specific trap/exception/interrupt)
Most processors support both "interrupts" (an external peripheral is banging on the CPU's interrupt pins... but also invocable from software; software interrupts; SWIs; INT instruction on x86) and "exceptions" (e.g. divide by zero, bus error, illegal instruction). Depending on the processor, accessing the "privileged" mode can be done either by software interrupts, exceptions, or both. An operating system should pick one and stick with it.
Other uses for interrupt/exception/trap vectors include hardware breakpoints: don't try and single-step the CPU, overwrite the code with an illegal instruction and control will flow to the illegal instruction handler where you can see all the registers then execute the real instruction that was meant to be there and return to where you left off. Some CPUs have a formal "BKPT" type instruction for that.
One other use on the 68000 is that any unrecognised instruction that started $Fxxx triggered the F-line handler; all the floating point instructions were in the form $Fxxx, so if you didn't have an FPU, you could put a software emulator for the FPU instructions in the F-line handler and software wouldn't know the difference. Traps/exceptions don't have to be a jump from unprivileged to privileged, they can just be utilitarian.
Practically most will support access via both, but for different reasons. For example, page faults (which the software cannot possibly predict) are going to be exception-mediated, but syscalls (which the software asks for) are triggered via an interrupt.
Later on the 386 Intel added virtual 8086 mode which trapped to the kernel privileged instruction exception also for certain instructions that had to be virtualized, among them INT.
We used a set of INT instructions in well-known low memory addresses that all jumped to the same place. We had an ASM file that you linked with, that had sixteen different address combinations for each.
The common entry point would look back on the stack and calculate from the return address which entry point had been called, and run the appropriate kernel call. We called it the CS:IP hack.
In the context of this post, the DOS INT10 and INTx(I forget) required the caller to load registers with the desired system call number, then perform the trap instruction in their code. Fortunately CTOS didn't need those particular software interrupts, so I could implement them for my purposes.
The routine at 0xFFD00 could then enter protected mode and use the code segment to build the index into a table of entry points: FFD0 goes to index 0, FFCF goes to index 1, and so on. But for extra kicks, the address isn't actually pointing to valid code. It points to a random "c" character in the BIOS, which is an ARPL instruction - which in turn is invalid in v8086 mode and therefore invokes the undefined opcode exception handler. The exception handler then takes care of doing the 32-bit call.
mov ah, 2
mov dl,7
Ahhh.. probably my first program. Don't forget the int 20 at the end! It was beeping great. Still never unlocked the mysteries of those TSR programs though.
So, what TSRs would do is overwrite one or more interrupts to point to a routine that would check if the system call in question was one it wanted to handle (eg, to add a hotkey it would grab the keyboard handler and check for a special set of keys before passing control back to the normal handler). Once that was fine, it would call the TSR system call and control would be passed back to the OS with the hook still in place
Traps typically also result from exceptional conditions (like divide by zero or page fault).
An architecture may or may not provide non-trap paths for less-privileged code to invoke more-privileged subsystems (call gates, "syscall" instructions, etc.).
Traps typically need some way to preserve all userspace-accessible registers (otherwise resuming from a page fault is .. hard). Dedicated syscall instructions may only need to restore a subset of registers.
In some implementations, processors may discover that an instruction must trap after it starts irreversibly changing architecturally-visibile state; in cases like that, the processor needs to leave enough breadcrumbs for the OS to allow either a clean unwind or a resumption of the interrupted instruction. My understanding is that the original 68000 somewhat famously got this wrong.
I don't know OG x86 (cuz, ewww) but on 68k this was generally the way. On my Atari ST a syscall was performed by filling your registers and stack as expected, then executing one of the TRAP opcodes and that would get the CPU To save PC etc & jump to the handler but in supervisor mode, where your syscall could then read state perform accordingly, and then return back to you.
I think x86_64 has just formalized this into a specific SYSCALL instruction?
ARM variants call it SVC (supervisor call).
Same difference.
Some older operating systems just implemented their syscalls as ordinary subroutine jumps, though, and everything ran in supervisor etc. I believe AmigaOS was like this, you just went through a jump table. Which, I think, shaves some cycles but also means compromises in terms of building for memory protection, etc.
Decades ago I ported some games to linux but I do think proton is the correct approach now. One underappreciated advantage is you get most of the mod environment too. In ESO for instance, there is an addon (tamriel trade center) which lets you download item prices, but it requires a windows client exe to do that. That client works on proton.
I also do some modding myself and can cross compile my rust code to windows with cargo xwin, and run it right away in proton, which is fairly amusing to behold.
I actually don't mind windows generally (been a MS user since DOS 5), but Win11 is a game changer, pun intended, and not in a good way.
1. An equivalent of kernel level anti-cheats. Cheating really sucks. It ruins online games. Kernel level anti-cheats aren't perfect, but they're much better than user-space or server-side anti-cheats. Maybe in the future AI solves this, but inherence-based anti-cheats are likely going to be a cat-and-mouse game. Valve have stated they are working on this problem and I think if anyone is going to solve it, it's them.
2. Immutability. Right now distributing games on Linux isn't distributing games "on Linux." It's distributing games to 12 different distros with a hundreds different configurations and a thousand customisations. This is impossible to support. When SteamOS gains traction, developers will be able to target exactly one distro with fixed configurations and limited customisations. Valve will set the standard for other distros.
3. An enforced equivalent of .exe. One of the most wonderful parts of Windows is the near universal acceptance and use of the executable installation method. You just double click the file and install it. Linux is an absolute clusterfuck of installation manuals and scripts and competing app stores with their own repos and permissions and packaging methods. If Valve were to mandate the use of, for example, flatpaks in SteamOS, that will become the universal standard. I think this is one of the most frustrating parts of using Linux for regular people.
4. Better hardware support. My Fanatec peripherals don't work well in Linux. Fanatec doesn't offer drivers and open source options are limited in functionality (and stability). There are many products for which drivers support sucks in Linux. I think AI will solve many of these issues over the next few years. Unless the manufacturer has gone out of their way to encrypt of obfuscate the communication layer with the product, you can basically point Codex at the peripheral and tell it to build an interface driver. Within a few years, I imagine operating systems will have this kind of functionality built in. If the OS encounters a peripheral it doesn't recognise, it will just build its own driver on the fly.
I am more optimistic about all of these than ever before. Linus Torvalds famously said it will take Valve to fix this fragmentation problem for us, and that looks like where we are heading. No doubt there will be Linux fans who lament the loss of diversity and competition, but I think we end up with a true competition to Windows for gaming. That's when I will make the jump.
Steam has already solved that problem. You target steam (not steamOS) and all other distros will do the work for you.
For the OSes runtime side you can depend on SteamOS / Apple's Game Porting Toolkit / Crossover / Proton / DWProton / Wine / and Android's Winlator/Gamehub/Gamenative .
For DirectX compatibility you can depends on Apple's D3DMetal , DXMT , VKD3D , DXVK and WineD3D .
I was playing Project: Gorgon recently, I was about to refund because it ran terribly on my machine (despite the low end graphics), when I noticed it was using the native build, switched to Proton and got a 200% FPS boost.
As long as I can play on Linux, I don't care what translation layer it goes through.
[0] Arch Linux, btw, because that must be mentioned.
I have been using Linux for nearly 30 years now, including running CS (HL1) via Wine with better performance and stability than Windows 9x on a LAN party. Good times.
Sometimes native ports don't get updates, while Windows port does. If you can then run it via Wine, you may have a more stable/less buggy experience.
Note I use both Wine and Proton. BG3 I run with Proton. But Proton is 'just' a fork with (neat) improvements which also partly got backported.
Oh, and I have to mention, I don't use Arch Linux.
So if you force it to run on your Linux desktop you don't get an experience commensurate with your hardware.
I have a couple more things to figure, I need XBox authentication to work for Halo Infinite and Sea of Theives, among others, and I need to figure out some solutions for some ancient software I have to run, which will probably end up being a Windows 11 VM. But as for my daily driver OS, I am so excited to get off Windows once and for all.
The stow approach is something that I considered but ultimately rejected for a couple of reasons around handling conflicts of game-installed files as well as how to ultimately handle the symlink lifecycle (eg wrapper to make the "non-running" state always clean or to let it always persist and then need to run manual cleanup/update steps). But if you're interested in that approach when I was applying for Nexus Mods approval I discovered https://github.com/Marc1326/Anvil-Organizer in the overall list of mod tools which I believe uses that strategy (though I haven't really looked too closely)
But basically my original idea to just install the files directly into the game directory stems from the fact that when I switched to linux for gaming and not having success with MO2 that's literally what I was doing. I would download the mod from nexus and unzip/tar it into the game directory manually. When I wanted to uninstall or update I'd find the original archive list the files in it and then delete them from my game directory. After doing this too much I realized that I was basically missing the functionality of a standard linux package manager (eg apt, pacman, etc)
So if you need to persist changes into the lower layers, I think you may need to do tricks like taking snapshots and then swapping the bind mount (maybe with some diffing logic) or some other offline methods.
I don't want to discourage you, but what's wrong with helping MO2 and Vortex get ported to Linux?
I should add that it's a CLI tool only (I may add a TUI later but it probably won't ever have a GUI if that matters). Anyway if you check it out and have any feedback whether positive or negative that would be cool
I would say custom modding and online multiplayer anti-cheat systems are the last real hold outs, and even then it doesn't affect every game.
My point is, you may find the one or two games holding you back won't be missed much.
Heroic is a launcher aggregate/wrapper I think? for 'Epic, GOG and Amazon Prime Games' It's either Steam, native/standalone or arr for me. for non-steam stuff I use umu.
* I should add that I am launching a steam purchased copy of SoT, not the one from Xbox store/gamepass or what have you, so the process is likely different, but maybe not cus you are likely going to see the same auth popup served via wine/proton.
It didn't use to be complicated, but an update messed stuff up a few months ago (halo infinite).
And, assuming your are doing x86, you probably already have an EFI partition so even doing motherboard bios updates isn't much of a big deal. You just drop the update in the FAT32 EFI partition, reboot, and point the motherboard at that location. Some motherboards even support just doing that as part of an online update.
That said some Linux distros can do the same now though I've used so many the last few months I don't know which.
It's the same tool the person you were replying to was pointing at via the Arch wiki. It's pretty standard. I'd expect most distros to support it by now.
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.
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.
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 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.
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...
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
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?
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.
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).
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.
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.
Have you even played an FPS vs an aimbots before?
I don't want to beat a computer, I want to beat another person.
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.
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.
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?
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.
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.
We're a long way not because Linux cannot do it. We're a long way because publishers refuse to take it serious.
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?
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.
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
It is the absolute best for backwards compatibility: it runs 16 bit apps.
Now it is getting faster for gaming.
And the reverse relation is also true: In Linux, the best backwards compatible stable API is WinAPI.
I can play 30 year old Windows games in Linux. They just work and run better than ever.
For the same project (https://www.descent2.de), I can not even install the dependencies to compile it in a modern distro, as every library is deprecated and removed from the repositories. The precompiled native Linux binaries also can't work.
People say this a lot about Linux and it somewhat rubs me the wrong way - sure, the Windows binary works if you install its library dependencies (wine). Likewise (OK, ever since libc5/glibc2 changeover in 2001) the Linux binary should work if you have its library dependencies (SDLv1 it looks like?). So what's really the problem? Your distribution stopped distributing the dependencies, making them harder to find? "DLL Hell" was a thing too.
I didn't see any binary downloads for Linux on that website, only source code.
I gave it a try anyway, the dependencies were actually not a problem for me, Debian has libsdl-{net,image}1.2-dev, libglew-dev, and so on, and if your distro does not have SDLv1 there is libsdl1.2-compat. But after the dependencies, there was a problem with the source code doing something involving bitfield packing that does not compile on x86_64.
I do see the source code has lived on, i can `apt install d2x-rebirth` on Debian 13 which has a comment about https://www.dxx-rebirth.com/ ...is that helpful?
The real problem here is the tooling. New versions of automake and C++ compilers are stricter than they used to be.
Indeed, for long term conservation, you need to vendor your dependencies AND your tools to make sure that the project is sustainable without intervention. But with the magic of LLMs now, this is not a problem anymore. Claude was able to fix it in less than 10 minutes.
And having the source code beats playing the original game. You can recompile in high def and make all the modifications you want to make it feel like a modern game (or not!) and sill get you your dose of nostalgia.
I can't prove it, but the Steam Deck has probably torn down a lot of barriers for mainstream use among the crowd that care about the game more than the OS. Getting some of the other games (League, Vanguard, Warzone, BF6, etc.) or whatever is popular in those segments onboard might be the critical mass that justifies fixing all the rough edges that get fixed when a big pile of users are represented.
If you want statistics, Linux’s gaming market share is 2x that of MacOS.
The barriers to gaming on Linux have never been lower. They’re certainly much lower than the barriers to running windows games on windows were back in the Win 95 - XP SP2 days (when I jumped ship).
I don't think you'd need to block multi tasking though, but the kernel would need to prevent or tamper root access so it couldn't modify the game memory.
Userland anti cheats can work (and do) on Linux if the developers want to. Most of the third party ones the developer buys/licenses already do.
But reality is that only the kernel level ones seem to work to some extent. Difference in the amount cheating between counter strike and valorant is just massive (both free to play games)
In other words, no one is going to refuse to use Linux out of loyalty to Windows, as long as all the games they want to play work.
This NTSync stuff is very impressive, but I haven't seen a lot of end-to-end numbers versus Windows. The last comparisons I saw showed pretty much every distribution on the order of 5-30% behind Windows, varying on the game. And Nvidia GPU support was still not great.
I WANT to swap. Please give me cause to do so. I'm sitting here with my finger on the button waiting for it to finally get good enough to make sense.
Just do it. Swap and let go of objectivity. Let your subjective experience guide you.
For me, the subjective joy of not having to fuck around with Microsoft's bullshit was worth multiples of having to mess around with technical crap to get a game working (spoiler: I nearly never have to do that because I play single player games, Dota and CS). I couldn't give less of a damn if my FPS in some random title is 10% slower than it would be in Windows. So long as it's playable, I benefit in spades from the trade-off.
I think most of the people who really care about game performance aren't people playing games like you do. They are either playing AAA games where the graphics quality is paramount, or competitive games where performance is useful for being competitive.
To give a particular example, I started playing GTAV on Windows after building a new PC since I had no spare drives. After finally installing Linux I decided to try GTAV on Linux just to see how well it would run. And it runs amazingly well, and yes, it runs a few percent points slower than Windows, but the only tradeoff I did was slightly increase FSR4 and the game still looks amazing. I didn't really notice any graphics issues, especially not during actual gameplay (if I stayed at the same place and started to nitpick I could notice differences).
I select games based on ProtonDB. There’s always constraints and I’m cool with the limitations that this brings. BF6 is a no-go due to its anticheat tool, no problem. Got lots of other choices.
Happy to not give EA any money if they're so set on shoving online play and therefore anticheat down the players throats.
Grateful that Steam allows easy refunds.
What other goal is there than maximizing your subjective enjoyment of the game?
Sure if you're a professional streamer, your feels are maybe less important than engagement metrics but if you're just a casual?
Dude just play what feels good. It's literally the best and only metric.
If Linux was measurably 5% slower on all benchmarks, would that mean you wouldn't do it even if you wanted to? Is every single nanosecond of performance really that important to you? I switched 10 years ago when things were a lot rougher than this, and in the end everything still worked well enough that I never cared to swap back.
But the issue is that it is many multiples of that, especially on the most common PC gaming hardware (Nvidia GPUs), often more than a 25% difference in framerates. Not so important at 144fps, but very important at a 60fps baseline and for genres like fighting games.
A lot of people don't mind, say, an extra 5 frames of input delay. They don't notice it. But a lot of people do notice even an extra 2 or 3.
I do think that frame pacing issues kinda do have a critical thin threshold where it's either bearable or an unacceptable difference. And the native windows version can often already be riding right on that line. So while it's not fair to the Linux version to demand better, it is unfortunately the case that it might tip over that line.
I've long since decided that buying the latest top end hardware is just spending a lot of money to be upset by buggy drivers or not being able to get 5000 fps in a benchmark but has no real gains in how fun games are.
Every time I get something mid range or second hand I feel good about what a good deal I got, and how I'm getting 98% of the features for 40% of the price, and how realistically as soon as you stop pixel peeping screenshots, you won't even notice your settings are on High instead of Ultra. You just take in the story, the sound design, and the actual game.
Your definition of great performance is not mine, but it’s fantastic to watch Linux users continue to hand wave away real issues whilst continually claiming the same or better performance across the board, which is provably false.
> but has no real gains in how fun games are.
It absolutely does for me. Modern displays are absolutely dogshit. I won’t play at anything less than 144hz, as much as I can I aim for 200hz and I want that with consistent frame times.
The game story, gameplay elements, and such have become secondary to the real hobby of consumerism. If people could have fun gaming 20 years ago, there is no reason it isn't possible to have just as much fun gaming on low to mid range hardware today.
The hobby of optimising your gaming desktop is a related but different hobby to actually playing games.
It's much harder to step back and realise you don't need the new thing most of the time. Sure if you have a 15+ year old desktop and you can't run the new games at all then an upgrade could be good, but I'd guess most hardware purchases come from people who already have great hardware.
what is the source of this non-determinism?
I kept running into issues that took me time to solve. I understand that the only reason it took me time to solve these issues is because I'm new to it and that people who have been gaming on Linux for years already know how to solve them all. But what would happen was is I would sit down to play a game spend maybe an hour or two fixing issues and then after that I ran out of time to play the game. I kept this up for a couple months but honestly at some point I just gave up. Now I'm playing games on Windows again.
To be clear, I'm a huge proponent of Linux gaming. I just unfortunately am too busy these days to spend the time to get it to work.
Although, everyone probably says that about whatever distro they happen to use lol.
I have no idea why people recommend this to people who aren't actually deep into tech and linux already.
Most egregious problem is that steam games start in a strange window rather than full screen and you have to press a weird key combo to fix it.
Nvidia based Acer nitro FWIW your mileage may vary
Unfortunately the install process is always going to be at least a little bit technical. I wish it wasn't, but idk how you'd do that without making the os like an emmu chip that you can swap out, instead of a thing you write on your drive.
I know you framed this as a negative, but this is something I yearn for; It's the one of the best games, imo. I often wish I ran into more issues, but for the most part, things _just work_^TM.
Gaming moved for a lot of us from 'now I have 5 hours or whole weekend to gaming if I want to' to mere blips here and there, which need to be as frictionless as poasible.
Which is great - it means we are doing something actually meaningful and more worthy in our lives. But it also means I will never have enough time for such fiddling. I am fine with it, as much as I can be, but lets be honest to ourselves here.
Where are the hordes of kids like us back then who were content with the afternoons, evenings and wee early morning hours of endless fiddling? What I realize now is those years spent fiddling sharpened our debugging senses in both ineffable and tractable ways.
A larger proportion of the juniors I see coming through the corporate halls these days than I remember from even 10 years ago do not have that knack for fiddling, nor history when it comes up. And it shows in their debugging temperament. LLM's are making this worse.
For all the issues people claim to have with iOS or Android, they really "just work" compared to the shit we had to deal with back in the day. And I don't even mean bugs, but UX just wasn't as sleek.
I can find a pdf of the TTRPG I'm playing that's hidden deep in an iCloud drive by simply opening spotlight an typing the approximate name. And the same works on my iPhone. Apps that create documents for me hide their file structure, because it's all abstracted away from me. It works, and I don't have to think about it as much.
You still have kids that start fiddling with tech, but only out of clear interest. Not as a necessity.
Absolutely not. It works, it doesn't "just work". Tuning is absolutely required for a lot of games to get them working. Random crashes, "oh multiplayer doesn't work? singleplayer does?", random glitches, random performance issues, etc.
I still prefer dealing with some issues over dealing with Windows, but it doesn't "just work".
Some of my favorite games that I play don't work on it, though, so I need to keep my PC. My issues are not performance, but inability to play at all.
For me personally, the biggest game that keeps me from only using Linux for gaming is EA FC (used to be called FIFA, it is the soccer game). It requires Windows to play online. The same for PUBG, which is another game I play with friends.
As long as I can't play those games, I have to keep my windows gaming PC.
I personally don't mind that much, honestly. It would be nice to play on Linux for everything, but I can dual boot when I am not gaming if I want to.
> These old workarounds got subtle edge cases wrong in ways that produced occasional hitches, deadlocks, or weird behavior in specific games, which are bugs that don't show up on benchmark charts but can absolutely ruin individual experiences. NTSYNC fixes those at the source by matching Windows behavior exactly, and that means as soon as your favorite distro moves to the new kernel version, whether it be Bazzite, CachyOS, Fedora, or a flavor of Ubuntu, they all get this much-needed fix.
That's the crux of the article. NTSYNC isn't faster, it's more "correct". Most games are around the same level of performance, with certain outliers both ways. Right now there isn't anything performance wise that Linux has to do that would impact all games. Just tweaks and additions to the different layers [1][2][3] in the same way driver vendors do. Much of the poor performance is for API violations and other shenanigans.
1: https://gitlab.freedesktop.org/mesa/mesa/-/blob/main/src/uti...
2: https://github.com/doitsujin/dxvk/blob/master/src/util/confi...
3: https://github.com/HansKristian-Work/vkd3d-proton/blob/maste...
It depends on what you're using now, though. If you're just using a vanilla wine/proton install, then NTSYNC should indeed be a lot faster as well. If you're using fsync or... I forget the name of the other one... then you many not see much in the way of perf improvements.
I have no idea why this is not turned on by default.
This was true 4 years ago, but is outdated knowledge now. Nvidia used to disallow distributing drivers with distro images, but they have since made agreements with some popular distros. If the distro image you download includes drivers or you know how to install them, the proprietary drivers work really well.
Gamer's Nexus has a pretty extensive benchmark video: https://youtu.be/ovOx4_8ajZ8?si=Cx5Q1a-lMMm14H4i . They refuse to compare to Windows, and it kinda makes sense: if it's satisfactory on Linux for your demands then who cares what Windows can do?
Here's a less professional, but direct comparison https://youtu.be/Giois6VtLPM?si=XFaVUMbea3u0AmP. An extremely important thing to note: AMD GPU. I personally have no idea what NVIDIA is like, but it sounds like their drivers are still all over the place.
And kernel-level anti-cheat doesn't work, though some (e.g. EAC) run in user mode if the developer allows it. Make sure to check ProtonDB for the games you care about. I have personally never had a good experience with Linux builds of games, so I just always use Proton now - but maybe I'm cursed because others have passionately disagreed with my experience. Either way, if a Linux game is broken/bad, try forcing it into Proton.
I don't want to say, "switch now" because it still has rough edges in terms of gaming. Better for you to have a great experience and stick around, than hate it and leave for good. Only you can figure out if it needs more time to cook based on some very light (ProtonDB) research.
I last used a Windows machine about a year ago, and I can say with confidence that the average desktop experience is significantly superior to the barrage of bullshit that Windows puts you through.
Pretty much everyone? If bread and water is satisfactory for your demands then who cares about Beef Wellington?
If it was better than Windows they sure as hell would be comparing.
Anecdotally, I find that getting Linux on somewhat older or underpowered hardware is always a massive positive. Better performance as well as battery life. I'm not as familiar with modern hardware's relationship to either OS ("OS vs. some flavor of OS based on a similar or same kernel" - I know) with modern hardware. Worth a shot though!
Every supercomputer seems to do quite well with Linux kernels. Probably good enough for Crysis :)
It's an easy weekend side project, and any numbers people give you will be ballparks anyway - the performance of Linux drivers for YOUR specific GPU running YOUR specific Steam games are all that actually matter.
Just take the two hours to do it. You won't regret it.
Got it running in less than an hour.
I never would’ve been able to root cause it under windows (certainly not with builtin tools), but dmidecode on linux made the problem obvious.
Fixing the timings fixed crashes in amdgpu that windows users widely reported (with no diagnosis), and increased frame rates by 30-50%.
Anyway, if you really want to move, do yourself a favor and just go with straight AMD.
Software support is better than intel and nvidia, HW blows intel out of the water. The only exception is if you need cuda for AI dev work.
The title after the jump is "Linux gaming is getting faster because Windows APIs are becoming Linux kernel features"
Getting faster. Not at parity yet.
I ask because I feel like I can frequently play games at, say, 150fps, and losing 30% would mean almost nothing to me to switch to Linux. I worry more about general capatibility and anticheat.
Huh.
I wonder if they're the same people who complain that they don't have enough money to live.
Fortunately I haven't noticed performance impacts so I'm on the hype train.
If you need every last bit of FPS maybe it is lagging, but 5-30% slower is roughly on par at a large sense, it's less than the difference of e.g. one NVidia GPU generation to the next, so it makes it playable.
That's why all the data matters for all of these dimensions; game performance is much more than FPS per watt over time.
When people see "linux gaming is great now, look at the fps" it comes across as potentially disengenuous because of all the other factors that matter and should be tested. Or rather, if a reviewer is talking entirely about framerate, then I just can't trust their opinion and expertise when it comes to the state of Linux gaming.
Part of the issue is that a large part of linux gamers are saying "linux gaming is great" and meaning "linux gaming is good enough now that it is better than putting up with microsoft and windows 11"
Some people would rather put up with slightly worse frame pacing if it means no microsoft. Some linux folks are super gung-ho pro privacy, some are just super anti-microsoft but can't game on mac. There's a whole lot of reasons to wind up on linux, so the importance of specific performance details may vary depending on WHY you would be swapping.
And some people are playing games on good enough hardware that there arent noticeable frame pacing issues, so good raw FPS numbers just reinforce their views, and they just genuinely mean they are having a good experience themselves.
The user will say 'it lags' but does not mention if they have tearing enabled in wayland, what are their 1% lows, have they set an fps cap, what vsync settings have they chosen in-game. On top of that there is an ecosystem failure as not every game supports arbitrary caps and you have to configure some mangohud thing.
Linux is only a 'just click play' experience if you have no standards because I have never had this stuff be correct out of the box on a fresh install.
I'm not saying you're wrong, but the standard you've outlined excludes Windows too.
No.
> I WANT to swap. Please give me cause to do so.
If you won't put the work in, why should we help you? Just stay on Windows, and we'll enjoy our Linux gaming rigs.
In the case of my machine, I haven't observed any difference. And by observe I mean with my eyes, I haven't bothered with actual benchmarks because it seems to work about the same, which is good enough for me. I haven't booted my Windows partition in months, and I'm probably just going to blow it away next time I need storage space.
Getting reliable, consistent, meaningful performance numbers is in fact, extremely complicated:
* You need a consistent way to reproduce the exact same outputs - accounting for things like the game's RNG. You can't just walk around and snap the FPS counter in the corner of the screen and call that good.
* For Windows (and occasionally Linux) you need to ensure nothing is running that will taint the results (updates, AV scans, etc)
* Sometimes individual driver versions work very poorly with a specific game. Just because it ran badly doesn't mean you got good data, it may just be a bug in that specific driver version
* You can't just run the benchmark once. You need to run it many times, establishing run-to-run variance
* There are often a good dozen-to-hundred individual OS settings which can impact performance, and in some cases run-to-run variance. You need to know which to tweak, and which to leave alone.
* Sometimes the result of individual in-game settings differs between driver versions. Just because setting X had a big impact once, doesn't mean it always did
* FPS is not a great metric - it's an average. You need to check and see if there are huge frametime spikes. If there are, the game will have a 'good' FPS but feel horrible to play due to stuttering.
* You need to decide if you're benchmarking more GPU-heavy or CPU-heavy - those types of benchmarks require drastically different settings. If you run a CPU-like benchmark you may see a wildly different gap in framerate compared to a GPU-heavy one for the same game.
Benchmarking properly means accounting for thousands of tiny variables. Only a handful actually do it right.
>nothing is running that will taint the results
No, running background crap IS the result, because that's real world conditions, and not some artificial lab condition.
>You need to know which to tweak, and which to leave alone.
That one is easy. You leave all of them alone. Windows tweakers do more harm than good. Besides, replicating benchmark results is impossible after you do brain surgery on the OS.
>You need to decide if you're benchmarking more GPU-heavy or CPU-heavy[...]
You benchmark the games you play. Benchmarking anything else would be completely pointless.
>Only a handful actually do it right.
Rumors say that Hattori Hanzo used to work for AnandTech. I wonder what he's up to these days.
Until we get something like CoD titles being Steam Console first, linux is allways going to lag behind.
That being said, I think we are on a precipice of AI being able to simply just rewrite games from concepts. Start with generic source code for an FPS or 3PS, then people can contribute changes in english language to tailor the game. So it won't be even copying source code, it would be copying concepts and then making a new game with it. There have been a lot of games that have very rudimentary graphics that people played in large numbers because the complexity and gameplay was quite good.
Anyone actually looking to make something genuinely fun will probably go the old fashioned way of spending countless hours honing their craft, which in turn gives them a good eye to make sure what they're making doesn't have the shovelware stink.
[1] https://www.notebookcheck.net/Sony-delists-700-PS4-and-PS5-s...
[2] https://www.gamesindustry.biz/shovelware-is-a-bigger-problem...
Fixed in Wine 11.0. Thanks to the Wine team.
Not sure if this was related to NTSYNC, but Wine's locking infrastructure definitely got an overhaul.
I'm just realizing that I can't play Battlefield 6 and I do wonder what the path is. I don't think it's ever going to be supported on Linux or Mac.
There's certainly room for improvement on the netcode sometimes (Client-side hit registration is an absolute bone-headed design), but those won't prevent aim bots.
Server-side anti-cheat relies on heuristics and can easily be evaded. At the high level, a highly-skilled player may be indistinguishable from a cheater, so you could easily get false positives.
If the vendors said: Disable anticheat and we’ll block you from tournaments / matchmaking, I’d consider that a feature, not a bug.
If some IRL friend of mine wants to be an asshole and use auto aimers / see through walls to screw with me, then I have ways to deal with it outside the game.
On the other hand, if we both want to run some bullet hell mode + cheats with physics mods and a debugger attached, then what’s the problem?
It’s none of the game developer’s business.
I’m not sure if I am in the minority or majority, but I’m not the only one with this attitude. I suspect the set of people in this boat dwarfs the 5% market share Linux currently has.
They might even get some of us to buy their games if they added support for such a mode. How hard could it be?
Cheating is endemic in BR and tactical shooter type games. I remember one f2p game was deleting 50,000 cheater accounts every month.
If the developer's winning $20 per cheater detection, and puts in extra resources when there's more cheaters, the equilibrium ends up a lot better.
And even for free games, I could imagine different ways to tie a monetary stake in in exchange for skipping invasive anticheats.
Blatant cheaters are bad in some ways, but subtle cheat are far worse imo.
The other 'state of the art' which is much cheaper, easier, and essentially impossible to detect on a hardware/equipment level, are the AI-based systems that examine the video and generate inputs via USB, emulating controllers or keyboards and mice. It's a huge problem on console right now and can only be detected via server-side analysis.
Instead of running the game in some arbitrary computer, you'd require players to buy your dedicated hardware, a black box that runs the game and nothing else.
This broke what was otherwise a perfectly normal Battlefield experience. Battlefield 4 requires Punkbuster, although it can run on Linux with no issues. You have to downgrade to an older version though, since EA hasn't updated BF4 to the latest PB AC, which causes you to get kicked.
I recently completed Stellar Blade with zero issues.
I don’t even shutdown the machine, I just hit the power to sleep it. Instantly resumes where I left off.
Incredible to see just how far it’s come.
This is not really my area, but from a quick web search, I think they mean io_uring. Here's a blog post about it: https://mazzo.li/posts/uring-multiplex.html
[1] https://news.ycombinator.com/item?id=47513667 [2] https://lore.kernel.org/lkml/f4cc1a38-1441-62f8-47e4-0c67f5a...
What do you mean? SRWLock (or the older CRITICAL_SECTION) cannot be shared between processes. A (Win32) Mutex does work across processes, but that's its entire purpose. So Windows does have different tools for different jobs.
In fact, it's really the other way round: on Linux, a futex also works across processes, but there is no equivalent in Windows. (Sadly, WaitOnAddress can only be used in a single process.)
That seems hugely useful for interprocess communication and I can immediately think of reasons to use IPC in a game. Having a separate voice process for one.
I'm saying that extra overhead from making your lock work across processes should be very tiny. That overhead shouldn't add much more than a microsecond in either latency or CPU usage, compared to an in-process lock.
What calls specifically are you talking about between windows and linux? This was started by someone talking about WaitForMultipleObjects.
But assuming reasonable implementations, the difference between those two lock styles shouldn't be more than about a microsecond, should it? So that's fine for a lock that's only used 100 times a second.
I'm not comparing windows and linux anywhere.
