|
https://github.com/trunkmaster/nextspace/issues/228
|
#
?
Feb 13, 2020 01:17
|
|
|
|
| # ? Jun 9, 2024 21:14 |
|
|
The_Franz posted:attn mods: posts from 2008 are somehow being mixed in with current ones please look into it mods im also getting this, debug output attached: BobHoward posted:disabling selinux, which breaks all kinds of things out of the box, is cool and good
|
|
#
?
Feb 13, 2020 01:35
|
|
|
mystes posted:Yes, that's definitely the reason this is crazy. i don't begrudge people their nostalgia-driven hobbies, I'm just saying at least use Swift or something so you don't hate every moment of your existence.
|
|
#
?
Feb 13, 2020 04:10
|
|
|
Sapozhnik posted:i don't begrudge people their nostalgia-driven hobbies, I'm just saying at least use Swift or something so you don't hate every moment of your existence. think about the extremely hairy C library interactions a window manager requires i wouldn't want to deal with that in swift
|
|
#
?
Feb 13, 2020 04:26
|
|
|
Notorious b.s.d. posted:i'm not sure what kind of person would even bother with 3d on linux if they are allergic to proprietary drivers Did you forget the discussion we had 6 or so months ago where everybody showed you that AMD works great with the vanilla kernel and this line of thought you keep spewing is utter horseshit?
|
|
#
?
Feb 13, 2020 04:36
|
|
|
BobHoward posted:disabling selinux, which breaks all kinds of things out of the box, is cool and good BobHoward posted:selinux is a good idea implemented incredibly poorly Also incorrect.
|
|
#
?
Feb 13, 2020 04:36
|
|
|
edit* Double post.
|
|
#
?
Feb 13, 2020 04:39
|
|
|
ratbert90 posted:Did you forget the discussion we had 6 or so months ago where everybody showed you that AMD works great with the vanilla kernel and this line of thought you keep spewing is utter horseshit? protondb disagrees
|
|
#
?
Feb 13, 2020 04:59
|
|
|
Notorious b.s.d. posted:think about the extremely hairy C library interactions a window manager requires you can literally just include c library headers into a swift program
|
|
#
?
Feb 13, 2020 05:17
|
|
|
psiox posted:good news, nerds, linux finally has a great desktop window manager https://github.com/trunkmaster/nextspace but can it run quakeed
|
|
#
?
Feb 13, 2020 05:28
|
|
|
Notorious b.s.d. posted:protondb disagrees Feel free to provide actual proof.
|
|
#
?
Feb 13, 2020 06:02
|
|
|
is a ratbert like a lambert
|
|
#
?
Feb 13, 2020 06:08
|
|
|
ratbert90 posted:Feel free to provide actual proof. pick a game, any game, look for red dots, note "amd graphics"
|
|
#
?
Feb 13, 2020 06:53
|
|
|
rjmccall posted:you can literally just include c library headers into a swift program last i checked swift got made about things like unions the x libraries are extremely bad and hairy
|
|
#
?
Feb 13, 2020 06:53
|
|
|
ratbert90 posted:Did you forget the discussion we had 6 or so months ago where everybody showed you that AMD works great with the vanilla kernel and this line of thought you keep spewing is utter horseshit? haha, yeah, "everybody". and for that matter, haha, yeah, "discussion".
|
|
#
?
Feb 13, 2020 09:30
|
|
|
i remember it, and ratbert is right. nbsd is searching protondb for anecdotes so he can protect his amd-drivers-will-always-be-bad-forever mantra
|
|
#
?
Feb 13, 2020 09:33
|
|
|
Notorious b.s.d. posted:think about the extremely hairy C library interactions a window manager requires You literally need 0% C in a window manager. I've written programs that talked to the X server through an X11 implementation in Common Lisp. And even if there's no X11 client protocol implementation in your language, the Xlib functions are pretty FFI-friendly. I've wrapped many of them in Haskell. And the point of XCB is that you can generate the client code based on XML descriptions of the protocol, although I'm not sure anyone did (except libxcb itself). Accessing it in not-C is probably one of the few things that X11 does OK.
|
|
#
?
Feb 13, 2020 09:43
|
|
|
Athas posted:You literally need 0% C in a window manager. I've written programs that talked to the X server through an X11 implementation in Common Lisp. Why the hell would you do *any* of this?
|
|
#
?
Feb 13, 2020 10:52
|
|
|
James Baud posted:Why the hell would you do *any* of this? I have a similar question regarding your post history
|
|
#
?
Feb 13, 2020 12:52
|
|
|
Tankakern posted:i remember it, and ratbert is right. nbsd is searching protondb for anecdotes so he can protect his amd-drivers-will-always-be-bad-forever mantra I remember NBSD also comparing TI2080s to RX580s and using that as proof that AMD is worse than Nvidia.
|
|
#
?
Feb 13, 2020 12:54
|
|
|
Notorious b.s.d. posted:pick a game, any game, look for red dots, note "amd graphics" Ah the ol: “I make a claim and then have you provide evidence”tactic.
|
|
#
?
Feb 13, 2020 12:55
|
|
|
this is fun, can we do x11 forwarding vs wayland again next
|
|
#
?
Feb 13, 2020 13:03
|
|
|
I'm sure that the next will be somebody claiming they got wifi or printing working on Linux. lol, or sound.
|
|
#
?
Feb 13, 2020 13:08
|
|
|
amd drivers were absolute dogshit two years ago, and i will not bother doing a lengthy new experiment on the guarantees from the kinds of posters who claimed btrfs was production-ready twenty huge data loss bugs ago. if it hurts you badly to hear someone say your os is a pos (whether super-factual in every particular or not) you may have continued troubles ahead. e: and hopefully they are indeed improving briskly. the bigger thing is of course the continued (and arguably increasing) irrelevance of linux on the desktop, but that's way more depressing than poking fun at details. Cybernetic Vermin fucked around with this message at 13:41 on Feb 13, 2020 |
|
#
?
Feb 13, 2020 13:28
|
|
|
Cybernetic Vermin posted:amd drivers were absolute dogshit two years ago, and i will not bother doing a lengthy new experiment on the guarantees from the kinds of posters who claimed btrfs was production-ready twenty huge data loss bugs ago. To be fair, I have never advocated for BTRFS. It loving sucks. Edit: https://www.phoronix.com/scan.php?page=article&item=linux-54-radeon&num=3 FlapYoJacks fucked around with this message at 13:52 on Feb 13, 2020 |
|
#
?
Feb 13, 2020 13:50
|
|
|
ah, bring up the btrfs-is-never-production-ready meme too, the perfect trifecta
|
|
#
?
Feb 13, 2020 13:52
|
|
|
klafbang posted:I'm sure that the next will be somebody claiming they got wifi or printing working on Linux. lol, or sound. sound? never heard of it
|
|
#
?
Feb 13, 2020 13:59
|
|
|
klafbang posted:I'm sure that the next will be somebody claiming they got wifi or printing working on Linux. lol, or sound. Wifi and printing sure. But let's not go crazy with the sound thing.
|
|
#
?
Feb 13, 2020 14:00
|
|
|
James Baud posted:Why the hell would you do *any* of this? Misspent youth. I even ended up designing an entire programming language for writing small X11 tools before I realised how crazy it was getting. I now compensate by focusing solely on GPUs that do not have monitors attached at all.
|
|
#
?
Feb 13, 2020 14:03
|
|
|
are fonts on linux better than on windows because theyre v bad on windows
|
|
#
?
Feb 13, 2020 14:47
|
|
|
poty posted:are fonts on linux better than on windows because theyre v bad on windows not really, but there is some measure of taste involved. the correct solution to fonts is a hidpi screen, pretty much all font rendering looks pretty much the same with enough pixels to work with. unless truly hilariously broken.
|
|
#
?
Feb 13, 2020 14:51
|
|
|
Athas posted:You literally need 0% C in a window manager. I've written programs that talked to the X server through an X11 implementation in Common Lisp. unfortunately if you want to do any GL, the GLX/EGL bindings are based around Xlib's Display*, so you're stuck using Xlib. Athas posted:And even if there's no X11 client protocol implementation in your language, the Xlib functions are pretty FFI-friendly. I've wrapped many of them in Haskell. And the point of XCB is that you can generate the client code based on XML descriptions of the protocol, although I'm not sure anyone did (except libxcb itself). unfortunately it sacrifices this by making it terrible to access in C. my favorite is all the 32-bit data that's passed as "unsigned long*" which can't be changed for ABI reasons, so if you ever deal with pixel data from Xlib you have to write a copy loop to copy it from 64-bit back to 32-bit lol libxcb is weird and has its own problems too... they made error handling worse, for some reason.
|
|
#
?
Feb 13, 2020 16:34
|
|
|
those systemd / at what cost unix talks were great i think i read them in this thread
|
|
#
?
Feb 13, 2020 16:45
|
|
|
finally, somebody said the magic word "systemd", now I can shitpost about systemd again
|
|
#
?
Feb 13, 2020 16:48
|
|
|
systemd is a great: - init system - cron replacement - container manager - it's an OK mount manager I guess systemd is a loving terrible: - logging framework - dbus - dns resolver - login manager - home directory manager - hostname manager - network manager - time/date manager unfortunately distros basically turn systemd into an all-or-nothing choice, and the "all" keeps getting bigger and more lovely Poopernickel fucked around with this message at 16:53 on Feb 13, 2020 |
|
#
?
Feb 13, 2020 16:50
|
|
|
does anyone know a good ubuntu based distro using systemd+wayland+btrfs that runs well on amd cards?
|
|
#
?
Feb 13, 2020 17:06
|
|
|
Zlodo posted:does anyone know a good ubuntu based distro using systemd+wayland+btrfs that runs well on amd cards? It's Ubuntu, op! You won't even need to disable selinux.
|
|
#
?
Feb 13, 2020 17:08
|
|
|
Zlodo posted:does anyone know a good ubuntu based distro using systemd+wayland+btrfs that runs well on amd cards? Fedora friend. Never use Ubuntu.
|
|
#
?
Feb 13, 2020 17:11
|
|
|
last time I checked, ubuntu had better default font smoothing than fedora. like last year ago.
|
|
#
?
Feb 13, 2020 17:38
|
|
|
|
| # ? Jun 9, 2024 21:14 |
|
|
ratbert90 posted:Fedora friend. Never use Ubuntu. it even supports btrfs in the installer, never choose lvm on a desktop
|
|
#
?
Feb 13, 2020 19:34
|
|