The Linux Thread - The Autist's OS of Choice

[moocow]

so whats cool is apple is not magic and theyre not the only ones working on arm. im expecting boutique linux workstation vendors to offer arm systems with comparble performance sooner rather than later

Yeah, no. That'd be "cool" if anyone was actually doing it. They're not. I've been "expecting" that for ten plus years and it hasn't happened. The closest anyone's come to a decent desktop equivalent (which isn't anywhere remotely as fast as even an M1-based system) is Minisforum's MS-R1, a neat little box with a 12-core CPU, 64GB of unified RAM, and a PCIe slot big enough for a half-height half-length GPU.

I have one. It's a great little Linux box for sure, but it's not nearly as fast as even first-generation Apple Silicon. Like literally everything else out there that isn't a damned Ampere, it's focused on price and size rather than performance. There is nothing ARM-based competing with Apple Silicon in any of its markets, either already available or in the pipeline as best as I can tell.

It's infuriating. Apple shouldn't have this market cornered, but nobody else is even trying.

 

[acid_ra1n]

Yeah, no. That'd be "cool" if anyone was actually doing it. They're not. I've been "expecting" that for ten plus years and it hasn't happened. The closest anyone's come to a decent desktop equivalent (which isn't anywhere remotely as fast as even an M1-based system) is Minisforum's MS-R1, a neat little box with a 12-core CPU, 64GB of unified RAM, and a PCIe slot big enough for a half-height half-length GPU.

I have one. It's a great little Linux box for sure, but it's not nearly as fast as even first-generation Apple Silicon. Like literally everything else out there that isn't a damned Ampere, it's focused on price and size rather than performance. There is nothing ARM-based competing with Apple Silicon in any of its markets, either already available or in the pipeline as best as I can tell.

It's infuriating. Apple shouldn't have this market cornered, but nobody else is even trying.

yeah i was thinking of ampere but in a desktop chip
arm development is happening at a breakneck pace apple cant keep the throne forever
just realized i should provide an example and i think nvidia is going to be the big player

 

[Susanna]

just realized i should provide an example and i think nvidia is going to be the big player

This seems like a safe bet, because it sure as fuck won't be Qualcomm.

 

[moocow]

yeah i was thinking of ampere but in a desktop chip
arm development is happening at a breakneck pace apple cant keep the throne forever
just realized i should provide an example and i think nvidia is going to be the big player

NVidia's ARM-based offerings so far (for desktop) are along SBC lines (Jetson Nano, etc.) outside their super-expensive DGX Spark line (they start at $3k and hit hard on local AI workloads but not general desktop and gaming workloads). ASUS has a new model, the Ascent QN10, with an 18-core Snapdragon X2 Elite CPU and 32GB of RAM ($1,700 for that model) that even runs Windows (for ARM) that's getting good reviews. None of these come close to Apple Silicon's total memory bandwidth.

Ampere remains the best option for "dense ARM server on a desk," not as a desktop computer. System76 and ADLINK sell pre-built systems, and the ASRock Rack ALTRAD8UD server board has already been mentioned here as a DIY option.

An AI says: "If the goal is 'buy a fast ARM desktop that is not a Pi and not a Mac' in late 2026: QN10 for Windows productivity, DGX Spark / N1X for local AI, Ampere only if you specifically want 64–128 cores and expansion. None of those yet erase Apple Silicon as the default high-end ARM desktop."

Shit sucks, man. As others have noted, Apple Silicon is genuinely innovative. It's just wrapped in endless Apple bullshit which makes it undesirable for anyone besides mac fags.

 

[Dante Alighieri]

It's infuriating. Apple shouldn't have this market cornered, but nobody else is even trying.

Is it even worth trying? Mac users are their own ecosystem. People either only use Mac or they use Windows and Android or Windows and iPhone. While there is a Windows for ARM last I was aware you have to get software specifically compiled for ARM. I have no idea if there's Professional programs (Autodesk, Adobe) for ARM.

 

[acid_ra1n]

Is it even worth trying? Mac users are their own ecosystem. People either only use Mac or they use Windows and Android or Windows and iPhone. While there is a Windows for ARM last I was aware you have to get software specifically compiled for ARM. I have no idea if there's Professional programs (Autodesk, Adobe) for ARM.

microsoft and google are very interested in making usable arm desktops even though googles idea is just make android tablets a desktop and microsoft doesnt give enough of a shit to do it right

 

[Hakase Shinonome]

Is it even worth trying?

Not if you’re a power user or really need to use x86-based programs. ARM remains somewhat of a novelty for now, or a power-saving alternative for others.

microsoft doesnt give enough of a shit to do it right

“Just call it Copilot + PC — No, Johnny, I also don’t know what it means, it’s something upper management thought of”.

 

[Kokonut Kangs Klan]

Is it even worth trying? Mac users are their own ecosystem. People either only use Mac or they use Windows and Android or Windows and iPhone. While there is a Windows for ARM last I was aware you have to get software specifically compiled for ARM. I have no idea if there's Professional programs (Autodesk, Adobe) for ARM.

I'd speculate MS already has a translation layer like Rosetta 2 in the works, but doesn't want to launch it unless the market shifts that direction first. I don't think they have much incentive to influence the hardware market themselves, but the business is still sensitive to the market's hardware choices in the long run, so it makes sense to prepare for this. Doesn't count as "trying" though.

Outside of MS we'll have to see how well FEX works, it might make ARM practical for arbitrary Windows applications. Valve definitely is trying here, although it won't be a huge market anytime soon. It seems ambitious and all of this is very of course.

 

[moocow]

Is it even worth trying?

What do you mean, non-x86 computing? Yes, it is. ARM is very competitive, especially on performance-per-watt, and as it turns out when you feed a good ARM chip a lot of power (i.e. on a desktop or server model) it's just as fast (or faster) than x86 on a shit ton of workloads. There's also other more "toy" architectures like RISC-V and friends, but there are literally billions of ARM CPUs out there in the field.

Not to mention the software stack is so good these days even mid-tier ARM CPUs can actually run Windows+x86-only video games on Linux reasonably well. I'm not joking when I say I literally have an ARM mini PC (Minisforum MS-R1) sitting on my desk with a GPU in it running Kubuntu 25.10 that has Steam installed and can run a fair number of Windows-only x86-only video games. It's not my daily driver anymore (it was great as a non-gaming workstation and -- as usual for ARM -- it's great for a home server too now) but it's a fine little box and surprisingly capable. Dual 10 gigabit NICs, too (that it can actually keep saturated).

"You have to get software for it" is not a downside for alternative architectures. The only time an alternative architecture is potentially more trouble than it's worth is when the conversation begins with shit like "how do we crack this fucking bootloader?"

ETA:

Outside of MS we'll have to see how well FEX works, it might make ARM practical for arbitrary Windows applications. Valve definitely is trying here, although it won't be a huge market anytime soon. It seems ambitious and all of this is very of course.

It works very well. It's near-native performance in many scenarios.

 

[Betonhaus]

I'd speculate MS already has a translation layer like Rosetta 2 in the works, but doesn't want to launch it unless the market shifts that direction first. I don't think they have much incentive to influence the hardware market themselves, but the business is still sensitive to the market's hardware choices in the long run, so it makes sense to prepare for this. Doesn't count as "trying" though.

Outside of MS we'll have to see how well FEX works, it might make ARM practical for arbitrary Windows applications. Valve definitely is trying here, although it won't be a huge market anytime soon. It seems ambitious and all of this is very of course.

They can barely keep Azure online there’s no way the jeets at microslop can pull off a translation layer.

 

[guiltyspark85]

I'd speculate MS already has a translation layer like Rosetta 2 in the works, but doesn't want to launch it unless the market shifts that direction first. I don't think they have much incentive to influence the hardware market themselves, but the business is still sensitive to the market's hardware choices in the long run, so it makes sense to prepare for this. Doesn't count as "trying" though.

It already exists and is available. Windows 10 ARM can run x86 programs and with Windows 11 ARM they've added support for x64.

 

[DavidS877]

Minisforum MS-R1

I assume they got the CIX drivers to suck less in the last year or so, although I notice you're not using the on-board GPU.

 

[moocow]

I assume they got the CIX drivers to suck less in the last year or so, although I notice you're not using the on-board GPU.

Yeah, I put in a dedicated GPU (Radeon Pro W6400). The onboard GPU is serviceable for 2D GUI stuff and not much else.

 

[Pluciferous]

IIRC someone mentioned here a while back that the state of Asahi Linux is pretty good right now, there should be ways to bootstrap Gentoo and NixOS up on apple silicone from there. Needless to say the firmware is pr*prietary, but if DHH really does manage to at least partially detransition its core dev base, that might lead it to even greener pastures.

https://wiki.gentoo.org/wiki/Project:Asahi/Guide

https://github.com/nix-community/nixos-apple-silicon

asahi omarchy would probably make hector suicide ngl
imagine how extreme that humiliation would be lmao

It already exists, and it’s more featureful than the official Asahi Linux releases. Hector seems more focused on fucking his tranny partner than the state of the project he abandoned/was forced out of.

https://omarchy.org/news/2026/09/introducing-omarchy-m/

 

[United Abominations]

The native version of FXfloorboard stopped working on troonix, so I think @Slav Power has a point on Troonix needing to have better backwards compatibility to be usable for normal people tbh.

The AUR version works but its unmaintained and that is a bad idea to download

 

[Akerman]

A persistent problem I've had with daily driving Linux in the Proton age is that Proton logs are gargantuan textfiles where like... ~500ish lines at most have the information you're really looking for, and it's spread out across 25,000+ lines of repetitive noise. I've made a dent in mitigating the troubleshooting headache over the last couple of years, but it's always something I struggled with.

(...)

I did something similar recently, was using a python mess of a program that I had no desire to learn to read because I fucking hate Python. Gave an agent an instruction to add file chooser portals to it and to fix why the fuck it opened on my 2nd monitor. It did both in like 10 min. Local model. When it works and for disposable shit like this it makes using open source a strong benefit over closed stuff. For the longest time most people couldn't utilize the freedom of being able to modify your own software due to a lack of coding know how and time to learn it. Local models may end up being the key for most to finally be able to utilize that freedom afforded to them by any copyleft license.

 

[ZMOT]

The native version of FXfloorboard stopped working on troonix, so I think @Slav Power has a point on Troonix needing to have better backwards compatibility to be usable for normal people tbh.

it's open source, either wait for someone to fix it or vibe-code a fix yourself (if it even is an issue with the software itself and not the environment).
there will always be changes that need to be adapted, at least you got the option to do so. good luck with proprietary shit (on a proprietary os).

Pretty much this. They can do some great innovations. But they can do some of the most anti consumer innovations and the industry will follow in their footsteps. Like removing the headphone jack, batteries that aren’t removable, and locking down your ecosystem. It’s just bananas.

I can only blame apple partially since a big driver of that was MUH FLAT so your phone can bend by farting on it.
the retards that follow it "because apple" are much worse.

This is the way. Sadly we won't see this happen in the realm of the Linux Desktop.

https://uapi-group.org/specifications/

some distros follow those more than others. less of an issue if you stick to a (well-supported) distro and know how they handle their special flavor of autism.
most people stick to very basic requirements anyway, so there's even less need for it since they won't notice or care anyway.
the more exotic your use-case gets, the more exotic your solutions gotta be, that's just the nature of the beast.

 

[jeff7989]

Installed Omarchy on a laptop, since it's just for shitposting and youtube. I wanted to finally learn Hyprland and LLM integration is a plus. Pretty good so far, btrfs with snapshots is nice, You can probably get everything it provides yourself, but I wanted to give something that just works™ a go, since my main PC runs Arch.
Sleek, but learning keybinds will take some time. Had to patch some stuff already to fix regressions from sloppa rewrite of lock inhibitor code breaking, making it no-op when watching videos over firefox based browsers.
Looks like Hyprland people are saner than KDE folk, since picture-in-picture windows stay on top on wayland, to need for window rules or other unnecessary retardation that were not needed on X11.

 

[acid_ra1n]

https://wiki.gentoo.org/wiki/Project:Asahi/Guide

https://github.com/nix-community/nixos-apple-silicon

It already exists, and it’s more featureful than the official Asahi Linux releases. Hector seems more focused on fucking his tranny partner than the state of the project he abandoned/was forced out of.

https://omarchy.org/news/2026/09/introducing-omarchy-m/

wtf i want an m1 now

 

[Kokonut Kangs Klan]

The native version of FXfloorboard stopped working on troonix, so I think @Slav Power has a point on Troonix needing to have better backwards compatibility to be usable for normal people tbh.

The AUR version works but its unmaintained and that is a bad idea to download

Git gud xir, build scripts aren't hard to review. In this case you might be able to fix your problem by checking what the scary PKGBUILD does. This change was made back in July:

CHG: remove bundled glibc
Root cause: The binary is linked with the system dynamic loader
(/lib64/ld-linux-x86-64.so.2, glibc 2.43
  on current Arch), but its RUNPATH=$ORIGIN/lib makes that loader pick
up the bundled Ubuntu glibc 2.39
  libc.so.6. Loader and libc must come from the same glibc build, so it
crashes with undefined symbol:
  __nptl_change_stack_perm, version GLIBC_PRIVATE
       
1 files changed, 5 insertions, 1 deletions
diff --git a/PKGBUILD b/PKGBUILD
index 44aa9bdaa665..d050eae2a3de 100644
--- a/PKGBUILD
+++ b/PKGBUILD
@@ -1,7 +1,7 @@
 # Maintainer: OGrigorios <aur@giotas.io>
 pkgname=katana-mk2-fxfloorboard-bin
 pkgver=20260112
-pkgrel=1
+pkgrel=2
 pkgdesc="Editing software for the Boss Katana MK2 guitar modelling and effects amplifier."
 arch=('x86_64')
 url="https://sourceforge.net/projects/fxfloorboard/files/KatanaFxFloorBoard/Katana%20FxFloorboard%20for%20MK2%20Desktop/"
@@ -19,6 +19,10 @@ package() {
     mkdir -p "${pkgdir}"/usr/bin/
     mkdir -p "${pkgdir}"/usr/share/applications/
     cp -r Katana_MK2_FxFloorBoard_linux "${pkgdir}"/opt/Katana_Mk2_FxFloorBoard
+    # The bundled glibc (Ubuntu 2.39) conflicts with the system dynamic loader;
+    # remove it so the system glibc is used instead.
+    rm "${pkgdir}"/opt/Katana_Mk2_FxFloorBoard/lib/libc.so.6 \
+       "${pkgdir}"/opt/Katana_Mk2_FxFloorBoard/lib/libm.so.6
     cp Katana-Mk2-FxFloorBoard "${pkgdir}"/usr/bin/Katana-Mk2-FxFloorBoard
     cp Katana-Mk2-FxFloorBoard.desktop "${pkgdir}"/usr/share/applications/Katana-Mk2-FxFloorBoard.desktop
 }

So try removing those 2 files in your broken installation. You don't need to install from AUR to learn from its maintainers, you'll often be solving the same problems as them (targeting the same base OS) regardless of where you actually install software from.