How are Nvidia GPUs holding up on Linux at this point? Since the author specifically praised AMD and Linux.
this post was submitted on 21 Jan 2026
94 points (92.0% liked)
Linux
11225 readers
580 users here now
A community for everything relating to the GNU/Linux operating system (except the memes!)
Also, check out:
Original icon base courtesy of lewing@isc.tamu.edu and The GIMP
founded 2 years ago
MODERATORS
Heck, you can’t even use it to give a slideshow without risking Windows Update deciding to reinstall your whole OS mid presentation! Ask me how I know.
There are several ways to prevent that. There's a lot of valid criticism to be made about windows but it's hard to take someone seriously who apparently doesn't know how to use the OS.
This time I want the credit, so I’m posting my prediction publicly while everyone still thinks it’s ridiculous: I predict that within 15 years Microsoft will discontinue Windows in favor of a Windows themed Linux distribution.
People have been joking about a Windows DE atop the Linux kernel for a while now buddy, too late for you to call dibs
Ya, I even wrote into a Linux podcast like 7 years ago making this exact prediction. It's not new.
I mean they did switch to their own flavor of their competitor with Edge already so it wouldn't be a first.
I just hope that is they do this they won't gain leverage to control the development of Linux for their own purposes in some way.
Has been predicted for almost as long as the year of Linux on desktop.
Windows NT kernel was never an issue, Microsoft would encounter exact same issues they’re facing now since the pressure to monetise remains.
Maybe. If they do it's going to represent a pretty dramatic shift away from backwards compatibility, which has always been the biggest defining characteristic of Windows. Probably not for Enterprise, but maybe for Personal?
They would keep backup compatibility using wine. Well, at least sort of.
Yeah, that's fine for handling some stuff, but there are some very old applications still in use that were coded by a programmer who died twenty years ago who counted on Windows bugs. A lot of companies refuse to spend the money to upgrade or replace those systems, and so Microsoft has maintained compatibility for them ever since. Wine hasn't reached full parity for those bugs, at least so far.
The key word being so far. A company like Microsoft would be able to ramp up wine development substantially, if they decided to. If I'm not mistaken, these very old legacy 16 and 32 bit apps have to be run on emulators running old versions of windows anyways - in these cases the OS running the emulator doesn't matter anyways.
True! If they committed back to the Linux kernel and the Wine project, this would be a huge boon. And it might happen someday, I guess; they've contributed to other FOSS projects. They've also built the WSL which is just one transposition away from a LSW, so who knows.
It's called Microslop. Won't you ever forget it
It would be in line with their usual EEE, so totally not surprising.
They don't give two shits about brand loyalty as long as line go up.
What standards would Microsoft EEE in this case? POSIX?
they might ship a proprietary lib with their os, encourage developers to use it, then license it out of being distributed
I imagine it’d be more like Android/iOS. Lock down bootloader so you can’t tamper with the OS, enforce notarisation requirement so that apps have to go through them. But Microsoft can’t do that, they don’t have any users vendor-locked to their application store. Valve on the other hand is in a much better position to do this.
The whole OS and user base.
That’s not what EEE is.
By all means do nitpick the specific meaning of EEE, might be amusing to others.
I just don’t like people throwing around terms they don’t understand, leads to weird outcomes like people saying Meta would EEE ActivityPub back when every instance decided to defederate Threads.
@0x0 @codeinabox it wouldn't be EEE in this case. Because a Linux distribution is built up around thousands of packages from hundreds of various open source projects.
For EEE to work, the entire base OS stack would need to be extended with features not becoming useful outside Microsoft's use. Such changes would first of all have a really hard time being accepted in upstream projects. And if they did, these projects would be forked if the last E phase in EEE is triggered. And then Microsoft would be alone with their Frankenstein distribution monster while the majority of the Linux users moves on to something better.
With Linux, there is no single instance of control or power. If a project takes a path people don't like, it get forked. EEE requires Microsoft to cease full control of all the related pieces and components and kill the open source aspects of it.
That's the advantage of open source licences. Once the source is out in the public, you can't retract the source code afterwards, then it just forks.
Ever heard of android?
Microsoft can't do funny things. Microsoft is a bloodsucking sociopath in a clown dress.
A little crazy, but not a lot crazy. ARM adoption may provide the spark necessary to ignite this fire.
"A Windows themed Linux distribution" can't really be "Windows themed" without some closed-source blobs and maybe even some backdoors sparkled here and there
Android is a Linux distribution like that and once notarisation requirement is implemented it’ll be indistinguishable from iOS. SteamOS will also likely become gradually more like iOS to appease creators of popular multiplayer games.
@mmmm @codeinabox Sure it can.
Microsoft can distribute their Windows Desktop Environment for Linux in a repository they control. And that repo can even contain just the binaries, with their own proprietary licence. These packages can further have dependencies to other open source packages.
If Microsoft ends up with their own Linux distribution or just mirrors an existing distro (like what Alma/Rocky does with RHEL, or Ubuntu with Debian) ... That depends on how much control they want over the Linux distro base OS.
But they certainly have the possibility to add a Windows experience as an alternative to GNOME, KDE, Cinnamon, XFCE, etc, etc, building on top of a shared base OS layer. As well as providing WINE like layers to make existing Windows programs run in that environment.
I think if they were planning this at all, we'd see linux support for maui. currently their linux use has been very focused on only for devs and servers
Not what I want to happen though.
I want them to release Windows as free and open source software. Windows is on a purely technical level not a bad OS. If it were FOSS, we would get dozens of excellent distributions of it.
Windows IS a horrible OS at the technical level. It seems okay-ish from a non-technical perspective. Everything just below the surface is absolute junk. There's a reason it's the most exploited piece of software in history.
I remember reading a post from an intern there about why they switched to using a browser for the start bar and windows settings UI: the code was just layers upon layers of friction requiring conversion from C++ to some intermediate format into another one and back. Making a simple change would take days if not weeks.
Ad now that they want to rewrite millions of lines of C++ into Rust using AI to rewrite all of winblows in Rust, it's going to get much worse.
I'd be very, very nervous about using AI for a project that big.
One of the biggest issues with AI, when it "works", is that it's incredibly hard to catch its errors. We have layers of cognitive biases around content that appears "written well" is most likely authoritative.
That means LLMs sound very authoritative, since correct structure is something LLMs are very good at. LLMs can write text that uses advanced vocabulary and complex sentence structure correctly. Similarly, it can write well laid out code with detailed comments and include an explanation for how and why the code works.
It all looks really convincing. It takes significant attention and effort to critically analyze LLM output to notice the errors it makes, and it's very easy to forget that the LLM is categorically incapable of understanding your prompt or its output—LLMs don't understand anything! They're just advanced word (token) prediction machines.
So, on a project the scale of refactoring the entire Windows codebase, with pressures to do the work quickly, the results are very predictable. They are going to introduce a lot of errors. And end up with incompatible spaghetti code that's incredibly inefficient.
(Somewhat technical example: Anyone who's studied algorithms/computer science knows how easy it is to write functional code that scale terribly because of missing a small logical step that makes one factor of code operate in linear time instead of logarithmic time. Multiply that by a few other factors in the same function and your code balloons from n or log(n) time to n² or even n³ time.)
We're witnessing the beginning of the end of Microsoft as an OS company.
We're witnessing the beginning of the end of Microsoft as an OS company.
Keep going. The faster the better. Imagine if billions went into Linux and opensource every year. It could be way better. The only thing I'm afraid of is companies starting to decide the future of linux. Everything would die in committee or they'd try and enshittify it to the max by adding proprietary kernel modules, filesystems and whatever else in order to get an upper hand.
There are some good ideas in there (the main one IMO is structured objects as the main form of IPC, instead of text), but the execution is really bad after XP and maybe 7. They stuck to a lot of legacy garbage in the name of "compatibility", but in the end it's an inconsistent mess of dozens of frameworks, and yet many old programs run worse than they do in WINE on Linux. Now with the advent of vibecoding the core components, it's pretty much over unless they literally start over from an old checkout.
MS already has their own Linux distribution...
And their own Unix before that. And arguably their own Linux compatible kernel until they dropped that ball.
Watch them be the ones to make Loss32