Software Engineer, Linux Enthusiast, OpenRGB Developer, and Gamer
Lemmy.world Profile: https://lemmy.world/u/CalcProgrammer1
- 0 Posts
- 17 Comments
Hopefully they can find a new home. I am ashamed of GitLab. I used to love it but they get worse and worse by the day. Maybe Codeberg would be a better home. Nintendo can’t kill this, there will always be new places to host software and it’s open source.
It’s absolutely ridiculous they took it down even though Nintendo didn’t DMCA the Suyu project directly. Shitty corporate cover-our-ass behavior at its finest.
Awesome news! I’ve been testing it on my RTX 3070 laptop for a month or two now and most games seem to work with it now, but performance is quite lacking vs. the proprietary driver. However, the proprietary driver is borderline unusable in a lot of games even though it pumps out high frame rates because something weird happens with render offloading and I end up with really bad frame tearing and out-of-order frame pacing that makes stuff jitter around. GNOME Wayland with mutter-vrr, AMD Ryzen 5900HX with integrated graphics powering the laptop’s 165Hz VRR panel. With NVK, there is zero tearing or weird display pacing and VRR seems to be working so once the performance is better it would hopefully make gaming on my laptop viable in Linux. Tired of having to game in Windows on that PC when all my other systems I can game just fine in Linux.
Personally, I think anticheat should be optional. At lower ranks, cheaters and smurfs are often indistinguishable, but smurfing is commonplace and usually unpunished. I don’t really care if I’m getting stomped by a sweaty tryhard on an alt account or by an aimbot, it sucks just the same. The solution is to move those players up the ranks until the cheaters and the tryhards are in the top ranks. Then you get the people who actually stand to truly benefit from cheat detection, and they tend to be the ones who would want anticheat even at the expense of privacy and system integrity so let them optionally enable it.
Us casuals at lower ranks should not NEED to run anticheat to play games at a casual level. It’s a freaking game, nothing is at stake. I’m not competing in tournaments. I care more about my system being free of kernel rootkits than whether some guy who keeps headshotting me from a mile away is a real player or an aimbot. Matchmaking should be able to deal with the discrepancy regardless.
Why does the entire Linux community assume that sandboxed apps are something we all want/need these days? I have no interest in sandboxed apps tbh. It makes sense for certain situations but I’m happy without them. I don’t like how Flatpak isolates all apps’ config files off into their little sandboxes and makes editing config files annoying. I just want stuff maintained in a central package manager and I want to use software that’s trustworthy enough that it doesn’t need to be sandboxed in the first place.
I use Wayland, but mainly because VRR support is better (except having to keep rebuilding mutter-vrr every time GNOME updates) and I don’t get screen tearing. Couldn’t care any less than I do now about sandboxed apps or unnecessary forced security. I hate that screen capture gets broken on a lot of programs running in Wayland and that global keybinds get messed up because of “designed with security in mind” bullshit. An operating system’s job should be to provide software with the features it needs, not to restrict said features.
Awesome that we now have cheaper options for the Deck. With competition from other handhelds starting to encroach on the Deck’s lower price territory (at least the 512GB model) it’s definitely good to see. I was hoping they would refurbish all the RMA’d Decks and not just recycle them. I had to send my original Deck in because of a firmware bug that made the CPU lock to like 1W TDP but it was otherwise in flawless condition. They replaced it with a brand new unit. Hopefully units like that are ending up as refurbs now.
It’s more of “NVIDIA bad” than “AMD good”. AMD does what is expected in the Linux world, to make open source drivers that are part of the Mesa project. That shouldn’t be an amazing feat of awesomeness, that should just be standard procedure. However, when the competition is so horrifically bad at drivers on Linux, following the standard makes AMD look amazing. For what it’s worth, I have an Intel Arc A770 on my Linux setup and it works great. Intel also follows the standard procedure of making their drivers open and part of the Mesa project. However, AMD has been in the graphics card (and driver) game for much longer and their drivers have a lot more optimization, plus Valve has put work into making AMD’s drivers better for gaming workloads over the past several years (especially given the Steam Deck runs an AMD GPU). Hopefully Intel gets more performance parity with AMD in the Linux driver world as time goes on. It’s definitely gotten much better since launch already.
As for NVIDIA, maybe NVK can make them even sort of useful without the nasty proprietary drivers but reverse engineered drivers are always going to take longer to get anywhere near the same performance of ones written based on actual official documentation.
tl;dr Alyssa Rosenzweig. Amazing news, she’s been a driving force behind reverse engineered GPU driver project Panfrost as well as had part in the Apple M1 GPU driver. I was surprised to hear that she was leaving Collabora but it makes sense that she’d end up at Valve. Hoping for great things in the PC GPU space from her!
Same, as long as they continue to test and verify the game works on Proton and provide support for Linux users using it, the game still works on Linux. Wine/Proton and DXVK/d3d12 have gotten really damn good these days, and I don’t mind that games are relying on them to work well on Linux if they actually hold true to officially supporting Proton as a platform.
You can also use NVIDIA’s Pendulum G-Sync demo in Wine/Proton. Despite the name it does work for any VRR capable display/GPU and I’ve used it to test VRR on AMD and Intel graphics on Linux. As much as I dislike NVIDIA, it’s a pretty decent VRR test tool.
https://www.nvidia.com/en-us/geforce/community/demos/