Dont hold your breath. It’s just initial support. It’s still opt-in and I can’t see Valve using it with Proton by default unless they start supporting native Wayland clients in Gamescope
I swear it is my machine or something, but despite CachyOS claiming being faster and more optimized I have yet to benchmark it as faster than the stock kernel for things I play around with. I wrote an application in rust to process a large text file and it both compiled and ran slower on CachyOS. I play around with llama.cpp and again it compiles and runs slower on CachyOS. I want to like Cachy, but right now all I can see is a bunch of window dressing to stock Arch with KDE and a couple of themes that I would rather change to default.
Also, why in the hell am I being asked to make a wifi password encryption key with the damn USB installer? CachyOS is not the only one. A lot of KDE using distros pop up the encryption window when you setup WiFi on the install image. Why? You want me to temporarily encrypt my wifi password on a temporary live image??? I just slows me down.
Anyways, I’m sure I’m crazy and clearly it is fast for somebody, but I can’t even get games to benchmark higher.
Did you test with different kernels? Them using a custom scheduler that prioritizes desktop applications might cause background things to run slower.
Plus, the use of ananicy (cpu/ram limiter) limits stuff like that as well.
I use cachyos because they set up zram, anf uksmd by defualt. That’s ram compression and deduplication, and it’a pretty powerful in my experience. If you’re using cachyos, then uksmdstats and zramctl can give you an idea of how much you are saving.
I used the default v3 kernel that Cachy installs by default. My guess is the workloads I have are Ram I/O bound and that just doesn’t mesh with the scheduler. I’m literally rooting for it to be faster because I want caring about scheduler and optimization to matter, but freaking stock Linux Mint ran the loads faster.
Could you describe the kind of glitches you are getting?
As a first test (and only as a test) I would try holding space bar during boot, then pressing E while focusing the Pop!_OS option, and removing quiet and splash from the line on the bottom, then pressing enter to boot.
General memory-corruption artifacting, such as pink checker-boxes around elements that are animating on the desktop. Swapping from fullscreen games would sometimes lock the system with a white screen (visually) until I restarted GNOME.
I’ve swapped over to KDE-Neon to see if Wayland might help, and it seems to have cleared up the artifacting but I can’t be sure until I test some games.
There’s also a possibility that the System76 power daemon was causing issues. I noticed that artifacts disappear altogether if I’m running any kind of GPU test, and I’ve seen some suggest that the GPU was running at a too low frequency. Would make sense if the daemon was the cause as that’s the other big change between Pop and KDE-Neon.
I’ll likely be back on Pop! when they bring out CosmicDE so I’ll try again then, but it might be that I avoid it for now.
I have the same chip in my mini PC (7840HS) and it works fine for me on Linux, but then again I use Arch + Wayland. Maybe you could try a couple of different distros on a Live USB or something (you could create one using Ventoy and then put a few different ISOs on there to play around with). I’d recommend choosing a distro with a recent kernel and updated graphics stack, for eg Arch or Bazzite and see how it goes.
But the artefacts you describe sound more like a hardware glitch to me. Have you tried running the Lenovo hardware diagnostics from the system boot menu? (IIRC you need to press F12 or something to get the menu and then choose the diagnostics mode).
I ran the diagnostics and they all came up clean, which I mentioned to Lenovo (who were pretty chill with getting the mobo replaced).
It seems to happen more when the graphics are clocked lower, or transitioning from high power state to lower power states. That being said, I’ve so far had it work pretty good on KDE-Neon on Wayland. Haven’t test X11 yet tho.
I had the same issue (on Pop!_OS), and I fixed it by tweaking the boot options to change IOMMU settings for my GPU.
I would try testing without the splash option, as that will change when/how GPU drivers are loaded and it might fix the glitches issue (but might still cause other issues).
I don’t like Ubuntu because of their forcing method to use Snap package manager.
I don’t like Manjaro because of its poor dependency management. Many dependencies are not declared, so that if you update a package, it won’t update the undeclared dependency and it won’t work any longer. You have to update everything or nothing, and when disk space becomes low, updating everything at once is impossible.
partial upgrades on distros without hard linked dependencies is a disaster caused by the user.
You should never have a system with less than 20% free space, but I mean system, not /home, not /var/cache/ of /var/cache/pacman,
Make partitions and mount things separately, especially /home
In a pinch you can live without man-pages remove /usr/share/{doc,man,html}/*
and on /usr/share/locale/* keep just the ones you use
I assume that Manj follows #Arch and doesn't improvise on sys dependencies. Definitely not poor.
Arch-archives by date, means you can build a system exactly as it was fully upgraded on a specific date, and the system works just like it used to.
Other systems that may carry 3 versions of the same library because different sw use different versions are the ones with the problem. Except for redundancy and space the system is not very coherent..
Pipewire is a modern audio server that can drop-in replace the mess that is alsa, pulse audio, jack, etc.
Wayland is a communication protocol made to be a replacement for the X11 protocols, with it’s compositer implementations being the replacement for Xorg.
What is your actual personal use case, all you mention is a terminal, which every distro will support, likely with many different choices as to terminal options?
Beyond the usual browsing I’ll mainly be doing tinkering with hardware, gateware, firmware, CAD, art, projects that I may or may not finish, and the like. It’s going to be my “everything but playing video games” machine.
I’m assuming you’re running a Ryzen 7040 series then. No kernel support for the FPGA yet.
Honestly, I wasn’t aware that they had included a fabric. That’s really awesome, whether it is supported yet or not. I have a couple of dev boards and intend to build a board with a previous gen Xilinx chip that can fit in the expansion bay.
Also, Linux is great for gaming. Not sure why you’d limit yourself there.
100% agreed. However, I already have a Steam Deck and console, so, it’s more that gaming is already covered by other devices than thinking the system is not capable. I’m intending to take advantage of the modularity to turn the laptop into a platform more physically spacialized for tinkering.
Not quite your setup but I run Debian stable KDE with KVM.
I am also using distrobox to run applications in containers. It’s nice having arch/ubuntu/fedora/gentoo software running in a container and the application gets exposed to you stable environment. Another option is Bedrock Linux to look at
That’s absolutely my thought. Having a rock solid system close to the metal that doesn’t really get touched is something that I’ve become used to from work. It gives a lot more insurance against having to do as many re-installs and maximizes compatibility.
So I did miss that Linus is in the article, but the reference to him says he was awarded the title, which makes it sound like an honour rather than a hierarchical system. I don’t believe that he’s ever been anything other than the projects owner/founder but I’m happy to learn if I’m wrong.
Yes, that’s just how open source works. Of course they always serve at the pleasure of the community, otherwise forks would happen. Nobody said otherwise. As the “Usage” section of that article implies, the “benevolent” bit comes from the feedback loop of a happy community supporting their dictator-for-life.
I mean how the community refers to him. I’ve never read a thread where someone called Linus a BDFL, I have with python. If they do, they do. Just haven’t seen it myself.
Free software doesn’t have owners. If someone else did a better job of being the “benevolent dictator” of a fork of Linux, everyone would start using that fork. Arguably this is a more free-market system than non-free software.
You can fork it, sure Linus is very respected and his decisions are considered very important but you can fork it and change however you want so it’s still compatible with Anarchism.
Linus’ power doesn’t come from Ownership, but respect. Anyone can fork it and do what they want, but because Linus is respected, everyone else follows suit.
Anarchism would function in a similar manner, it wouldn’t be a bunch of opinionated people doing whatever they want, but people generally listening to experts who don’t actually hold systemic power.
Many times what? Most forks die within a few months. Especially for big and well known projects. For example, io.js was a fork of NodeJs. Didn’t last long and was killed by NodeJs. All the Firefox forks are pretty much dead as well. Linux also had plenty of forks by people who disagreed with Linus and where are they now? I bet you don’t even remember their names.
Forks don’t work unless the original project is dead.
So mass adoption is your answer, and I’d say you’re misguided. The purpose of FOSS isn’t to make a profit, but to satisfy uses and needs. If a few people have a need for a fork and use it, then it’s a success.
You’re judging FOSS software by popularity, rather than use, as though it’s for profit.
Most new businesses fail as well. Maybe we shouldn’t be starting new businesses either? Or perhaps this more about people being unprepared and out of their depth whether it’s starting a new business or forking a code base. And not the individual actions themselves.
This is incorrect. It’s true that most (in fact, I would say almost all) forks go nowhere but that doesn’t mean forking isn’t incredibly valuable. Even the example you cite, “original project is dead” isn’t just incidentally useful, it’s critical to open source. Other examples include:
project’s core team is part of a for profit org that is moving the project in a bad, profit motivated direction:
project’s leader suddenly and dramatically loses respect (maybe he killed his wife or something);
project’s leader dies without leaving a digital will regarding who controls the core repo;
project continues to direct effort into features while falling to address major security concerns;
project is healthy and useful in every way but there is an important use case not being addressed, and the fork would address it.
Even if 99% of forks fail, that’s irrelevant because 99% of original projects fail in the same ways. Forks are critical to open source.
I would say we should just let unjust societies fail so just ones can take their place, but that seems to be the natural course. We’re seeing that right now.
Nextcloud is a FOSS fork of OwnCloud. Both projects are great in their own way, hugely successful and serve a lot of people very well. They just moved in different directions.
This is just one example of many. Ability to fork is super important to ensure that projects stay open source, like in this example.
I would disagree and say it’s more akin to a philosopher king hence less anarchy and more monarchy. It’s all good until the king dies and let’s see who succeeds them.
I meant that as a reply to the second paragraph which generalised anarchism; including the non-Linux world.
I also disagree that this isn’t an issue in the broader Linux community however. See for example the loud minority with an irrational hate against quite obviously good software projects like systemd who got those ideas from charlatans or “experts”.
I know, I used Linux as an example. Just like not everyone needs to be a weatherman to trust weatherman that can recognize experts among themselves, so too can engineers recognize experts among themselves, and so forth.
Skilled programmers can see that Linus is an expert. It works in tech. It probably works in any professional environment - anywhere where skilled people are picking someone highly skilled.
For the average person, we have clearly seen average people suck at picking expert leaders, though it works fine in small groups
There’s a word for this, the promotion of leaders based on merit instead of popularity - Technocracy. And it’s not a distinct ideology but a syncretic one that has been adopted by many groups with differing politics. The most prominent example would be the Technocratic faction of the People’s Republic of China, which was opposed to the Maoists back in the 50s and 60s; they argued for society to be led by experts instead of Democratically with a strong emphasis on Peasant participation (the standpoint of the Maoists). China today follows a moderate path taking from both factions.
In the West, however, Technocracy is mostly associated with Liberals; however, I would argue that the modern Liberal view of Technocracy is fundamentally flawed, since it relies on Capitalism distributing wealth meritocratically (which Socialists understand is not the case).
my take is that you really cant get big compnies to port to linux in the way he describes, and a lot of linux users wont use it anyway cuz its not foss
also afaik cosmic isnt just gnome in rust? its more like a realised knome (mix of kde and gnome, april fools 202x)
big companies will move with userbase, and cosmic being developed wont hurt the userbase growth of desktop linux. jeez that last part about foss evangelists just like no
honestly this man just seems a bit fustrated by not having a latest popos release?
also : people create clones of software all the time, not just in foss projects
overall, id say i dont really agree with him. imo cosmic is fine and the big companies really arent that interested anyway, i don’t think giving them money will help tbh, id much prefer foss alternatives being given funding
linux
Active
This magazine is from a federated server and may be incomplete. Browse more on the original instance.