Live data from Hacker News

The X.Org Server Is Abandonware?

phoronix.com

301–310 of 714 posts

Re: The X.Org Server Is Abandonware?

#301
post #67

Sure X.org is abandonware but there are serious problems Wayland still hasn't solved that X has. 1) Wayland is really slow. I don't know if it's the compositing or what but it's unusable on lighter hardware that X ran fine on. 2) Widget toolkits handling window decoration is awful. Before the large number of toolkits just meant some controls were a little different but now basic behavior changes based on how programm…

I am forced to use ms teams for work. Every time I attempted to share my desktop, teams would crash. I thought it was just because teams sucks. As it turns out, it seems to be that running under wayland was actually causing it. Everything works fine under x.org. After I was running with x.org for a couple days, I realized there are quite a few little oddities that I just lived with and didn't realize it was because w…

Yes, Teams uses a function that is unsupported by Xwayland (XGetImage, if I remember the backtrace correctly), so it crashes. I worked around it with a rather convoluted setup that involves running Xephyr under Xwayland, then running Teams under Xephyr (via setting `DISPLAY`), and running vncviewer under Xephyr connected to a VNC server (wayvnc) sharing my screen on the parent wayland instance. Then I tell Teams to share the "whole screen" (ie the entire Xephyr window) and quickly maximize the vncviewer window (within the Xephyr window).

To be able to maximize vncviewer you do need a compositor running under Xephyr too. I picked i3 to be the most minimal. Again, i3 can be told to run in the Xephyr window by setting `DISPLAY`.

One caveat is that I still use Teams under Wayland to join calls for the audio, so this setup means I need two Teams instances to join the same call. This works for meetings but not for individual calls. Of course if you want to use the Xephyr'd Teams to do audio too there shouldn't be any problem; I just prefer having easy access to the Teams window to mute myself, etc instead of having to reach into the Xephyr window and unmaximize the vncviewer window.

People on the #sway IRC channel on Freenode suggested an alternative might be to use v4l2loopback to create a "camera" device that is sourced from the screen and then have Teams use it as a webcam, but I couldn't get v4l2 to work on my distro (it kept insisting my distro's ffmpeg couldn't encode video even though it could) so I didn't investigate further.

Re: The X.Org Server Is Abandonware?

#302
post #194

Earlier quoted context omitted.

I routinely use modern X applications running on a remote (25 miles) machine, all day long as part of my job. It works flawlessly if you have a reasonable amount of bandwidth and reasonable latency. Sure, it doesn’t work over dial-up.

Honestly, X forwarding doesn't work that well in my experience, unless you have a very stable connection, with quite a bit of bandwidth (~1Mbps at the very least). I've had more success using xpra for forwarding, as I'm often connecting over Wi-Fi (hostel, campus rooms...). It's also rather complicated to set-up on the server side (xauth, magic cookie, etc). waypipe, on the other hand, was a breeze to use, even thoug…

To me, it’s been a while since 1Mbps was “quite a bit of bandwidth”, even for WAN connections.

Re: The X.Org Server Is Abandonware?

#303

Earlier quoted context omitted.

Honestly, X forwarding doesn't work that well in my experience, unless you have a very stable connection, with quite a bit of bandwidth (~1Mbps at the very least). I've had more success using xpra for forwarding, as I'm often connecting over Wi-Fi (hostel, campus rooms...). It's also rather complicated to set-up on the server side (xauth, magic cookie, etc). waypipe, on the other hand, was a breeze to use, even thoug…

> It's also rather complicated to set-up on the server side (xauth, magic cookie, etc). you mean "ssh -Y user@host..."?

No, that is the client side. I went back and looked at the documentation, I was wrong and conflated two ways of doing it:

- X forwarding over SSH: this only requires changing X11Forwarding in OpenSSH sshd's config

- Plain X over network, which is secured with `xhost`, insecure, and needs transfering the magic cookie or other authentication information

So, not nearly as complex to setup as I recalled, though it's much simpler to run a nested wayland compositor (which waypipe does) than a X11 server (which xpra does). The difference between X11 and Wayland remote access thins when xpra is involved.

https://stackoverflow.com/questions/37157097/how-does-x11-au...

Re: The X.Org Server Is Abandonware?

#304
post #261

Earlier quoted context omitted.

I'm saying that we get your point. But you seem to be missing ours. Nobody cares about benchmarks that a) don't exist or b) are not indicative of what folks will experience.

Obviously? The disconnect is that you seem to think that we can draw conclusions from nonexistent benchmarks.

You are asking us to literally draw conclusions from hypothetical benchmarks where the opposite results will exist.

I am sympathetic to the idea that things needed to start over. I'm annoyed with the lack of honesty and self critical approach. As framed by you, Wayland is above criticism. Which immediately raises my suspicions.

Re: The X.Org Server Is Abandonware?

#305

Earlier quoted context omitted.

Honestly, X forwarding doesn't work that well in my experience, unless you have a very stable connection, with quite a bit of bandwidth (~1Mbps at the very least). I've had more success using xpra for forwarding, as I'm often connecting over Wi-Fi (hostel, campus rooms...). It's also rather complicated to set-up on the server side (xauth, magic cookie, etc). waypipe, on the other hand, was a breeze to use, even thoug…

> It's also rather complicated to set-up on the server side (xauth, magic cookie, etc). you mean "ssh -Y user@host..."?

that would be the client side

Re: The X.Org Server Is Abandonware?

#306
post #265

Earlier quoted context omitted.

I've seen libdecoration, and yeah, that seems like their intended solution to this. It doesn't seem like a terrible solution, but last time I looked at it at least, libdecoration was a long way off being production ready, and GNOME with Wayland is shipping _right now_ and must be supported.

You can use a library like that or you can draw your own decorations. There unfortunately is no other option. I don't see it as being likely that GNOME will support SSD any time soon, it just isn't designed for that.

> There unfortunately is no other option.

Gnome could stop being unreasonably obstructionist, but that indeed seems unlikely.

Re: The X.Org Server Is Abandonware?

#307
post #251

Earlier quoted context omitted.

I get your point. But until those benchmarks at least correlate with behavior users see, they are not worth much.

Protocol limits what is possible. Legacy defines performance. Gnome is X11 first. Wlroots in Wayland first. It should be possible to optimize Gnome. Wayland was created because of horrible X11 performance [1]. It is not Waylands prime time yet but X.Org still works and maintained. Phoronix.com should have checked contributions [2]. [1] https://www.youtube.com/watch?v=GWQh_DmDLKQ [2] https://github.com/freedesktop/xor…

These are appealing arguments. But our industry is flooded with appealing arguments losing to pragmatic compromises.

Re: The X.Org Server Is Abandonware?

#308
post #294

Earlier quoted context omitted.

I was trying to be clear here: There is a reason and it's that it's not really possible architecturally for them to support SSDs in Wayland; doing so would require a redesign of significant parts of the compositor's code. I'm sorry if that was misunderstood. If you're trying to say they should have anticipated this and made a different architectural decision years ago, maybe that's true, but that also offers no pract…

I am trying to say that maybe they should have made the architectural decisions which wouldn't prevent non-GTK applications from working in GNOME. That's all.

I don't know what you're trying to get at. We can sit here and argue and say they should have done this and they should have done that, but that doesn't help anybody who wants support for them at this current moment.

And please don't exaggerate, non-GTK applications do work in GNOME. Qt5 has native CSDs that will show up in GNOME. If the application is a game or something that doesn't support CSD then the experience is somewhat degraded but they still work. You can resize and move any window by holding Super and Left/Middle clicking. I won't pretend the situation is ideal but I also don't believe it's totally unusable or on a bad trajectory at the moment—like I said the other libraries are working on getting CSD support too.

Re: The X.Org Server Is Abandonware?

#309
post #304

Earlier quoted context omitted.

Obviously? The disconnect is that you seem to think that we can draw conclusions from nonexistent benchmarks.

You are asking us to literally draw conclusions from hypothetical benchmarks where the opposite results will exist. I am sympathetic to the idea that things needed to start over. I'm annoyed with the lack of honesty and self critical approach. As framed by you, Wayland is above criticism. Which immediately raises my suspicions.

I have not said that Wayland is above criticism. I have said that the criticism which has been raised thus far is largely invalid, and that the benchmark you pointed to is flawed. If you insist on using flawed benchmarks as evidence for the inferiority of a technology simply because no less-flawed benchmark exists to provide a counterpoint, you are wrong.

Re: The X.Org Server Is Abandonware?

#310
post #167

Earlier quoted context omitted.

> Wayland is a protocol. It has nothing to do with multi screen support. You're talking about the compositor you used. The one I use handles multi-screen setups quite well. Which compositor do you use?

I use Wayfire [0], a customizable compositor based on wlroots, the same base as sway. It's quite involved and not absolutely perfect yet, but it has some features I haven't yet seen on other DEs like being able to swipe horizontally on your touchpad to smoothly switch workspaces (following your fingers) and the satisfaction of having it properly configured is pretty high. There have been 3 issues I've had regarding i…

>XWayland apps are unresponsive in the upper half of the second screen (4K at 1x scaling)

Seems to be a bit of a dealbreaker for people who want to use 100 percent of their screen instead of only 75 percent.

Post reply on HN