Quoting: Dunc

Quoting: MayeulCAs for CRTs... I definitely agree on there being something to their quality. It could be multiple things, like the non-square grid pattern (some recent phones have this), but what I really think makes a lot of difference is the ultra low persistence of the display.

I think it's more to do with the fact that they're analogue. There's no native resolution, so scaling and (to an extent) AA are, effectively, done in hardware for free. I've never used a 4K monitor for any length of time, but from the experience of my (now rather antiquated) phone - 720p in about 4" - I think the extremely high DPI levels we're beginning to see in LCDs will help. Although obviously you run into the sort of problems that we see in the OP. (Yay, back on topic! :D )

Well, it really depends what kind of screen you are talking about... You are right if talking about monochrome CRTs (or the old "vector") displays (though there is definitely a resolution limit imposed by the beam width), but look closer at RGB CRTs, and you'll see color stripes, that make "RGB" pixels. So you have a horizontal resolution at the very least (though pixels could blend a bit into the neighbouring ones if the beam is too wide, some LCDs also have this issue). And most monitors I've seen also have those grouped in horizontal lines (sometimes arranged in a non-square grid, as I wrote above):

Spoiler, click me

To protect your privacy, external media requires approval to load.

Source: upload.wikimedia.org

View cookie preferences.
Accept & Show Accept All & Don't show this again   Direct Link

So there's also a discrete vertical resolution. What's more, I don't recall the specifics, but I am pretty sure that the beam is actually controlled by the monitor (with vblanks, hblanks, etc to give it time to go back at the start of the screen/line). The only analog thing might be the intensity of the electron beam, and monitors can already approximate this quite well.
Another difference might be that only OLEDs can surpass the blacks of a well-tuned CRT (the phosphor layer will give something of a permanent fluorescence) :)

This video [External Link] neatly illustrates low-persistence of a CRT, and there are a couple that track moving objects on screen to illustrate persistence influence on motion blur. You can easily do so yourself by running glxgears and tracking the edge of a cog with your eyes. A lot easier to do on a CRT! (Or for an illustration, the big gear is already easier to track as it's turning slowly, thus persistence has less of an effect on it).

View this comment - View article - View full comments

Quoting: Linas

Quoting: MayeulCI'm considering buying a monitor with strobed backlight to imitate this (AKA blur reduction, or ULMB).

Is this something like the flickery backlight dimming on some panels? Because that messes me up quite bad. Cannot look at it for more than a few seconds. :sick:

Yeah, that could be it, though it should be imperceptible, depending on the frequency. Some specific conditions can make this worse (freesync with huge spikes in framerate, special content). Do you see lights flicker at 60Hz? (Though the waveforms are different, 80Hz seems to be the point where flicker starts to disappear to our eyes [External Link]).
It can also be unrelated, with some crappy pulse-width-modulation applied to the backlight to dim it.

View this comment - View article - View full comments

Quoting: FaalagornI think the same can be achieved with gamescope [External Link]? This DXVK discussion [External Link] led me to this idea, though I didn't manage to figure out how exactly achieve this on my Wayland (sway) display, but maybe I didn't try hard enough (sadly gamescope lacks documentation).

It would be trivial to add a mechanism for a wayland compositor to upscale a given window. Maybe we should ask for a feature request?

I'm using xpra almost daily. My use-case is that I have a piece of software (a big EDA tool I won't name) that I have to access over the network with X forwarding. Fine for LAN, but a lot less so over long distances. Xpra allows me to tune the quality/responsivity balance to my needs, and handles disconnections quite well, whereas X11 would just close the client application. So much for "network transparency"... Xpra makes X forwarding closer to vnc or wayland, it's a bit like GNU screen for X.

As for CRTs... I definitely agree on there being something to their quality. It could be multiple things, like the non-square grid pattern (some recent phones have this), but what I really think makes a lot of difference is the ultra low persistence of the display. I'm considering buying a monitor with strobed backlight to imitate this (AKA blur reduction, or ULMB). High framerates help, to an extent (it reduces the ceiling on the persistence value, but not as much as CRTs or strobed backlights).

View this comment - View article - View full comments

Quoting: EhvisThey should really do something about the default game though. [...] Thing I noticed is that the world was dead. No animals, no enemies, nothing happening.

Yeah, I agree, though this can also be said about Garry's mod, for instance. I was also interested in contributing, but I must say that I find some members of the community maybe a bit... childish? (I had the same issue with SuperTuxKart, though I think it's better with minetest)

Quoting: EhvisAlso noticed that there was no off hand to hold a torch in. But it turns out that doesn't matter because holding a torch doesn't actually provide light.

Is that a thing now in Minecraft? That was a bit annoying, but not a dealbreaker, and leaving with enough to build stacks of torches, as well as putting some everywhere, quickly becomes second-nature :)

View this comment - View article - View full comments

Quoting: tmtvlAnd they haven't fixed it for Wayland. I can imagine sway working well, though, as it's basically just i3 for Wayland, and i3 doesn't work like KDE and GNOME do.

Yeah, it's basically one virtual desktop per screen. I can just switch the focused one to the right/left/top/bottom screen with a shortcut (I chose ctrl-meta-arrow, meta-arrow is to move focus, meta-shift-arrow to move the window around. I like my shortcuts nice and consistent :) ).

Where it differs from i3, though is that (to my knowledge), i3's screen setup is managed trough xrandr/xorg.conf, whereas in sway it's there in the config file, which makes it quite easy and reliable. I also have shortcuts to rotate the screen in there... Though I guess you could achieve the same with xrandr, it makes it easy to apply something on the focused screen. The lack of a "primary display" hasn't bothered me at all.

Thanks for the bits on KDE and GNOME. With more time... :D

View this comment - View article - View full comments

Quoting: tmtvlThe majority of users here are on a single monitor, so Wayland usage may be decently high. Unfortunately there are some issues with WL multihead implementations on GNOME and Plasma.

FWIW, multi-monitor support is quite decent here on sway (I use it at work, dock/undock often, use projectors, etc), and I don't recall many issues on KDE nor GNOME.

Quoting: HamishTechnically my main screen resolution is 1680x1050 and my second head is 1024x768. Should we change the question to ask for your primary or main resolution? I can not be the only one with different sized screens.

Hah! I used to have pretty much the same screen resolutions. Got rid of (gave away) my primary monitor to upgrade it, but at the same time, a GPU lockup I was having quite often went away. Unsure if that was related, I've decided to do with the small screen for now ^^"

Thanks a lot for the stats, Liam! I'm eager to see the results!

View this comment - View article - View full comments

Idea (sorry, I always have one when these stats pop up): group the desktop environment and GPU stats by session type:

Show which DE is used more on Wayland, which on X11. Show if AMD/NVIDIA/INTEL users are more likely to use X11 or Wayland (plotting these graphs again for Wayland would be enough to infer the rest, but there could be some value in trying to provide an easier view, but that would be more complex).

View this comment - View article - View full comments

Quoting: Kimyrielle

Quoting: eldakingThere doesn't seem to be a common reason, it is just something that is done badly.

There is not a single good reason for these launchers to exist in the first place. It's just lazy software engineering. Nothing they do couldn't be done just as well in the main application (and/or in Steam, which most Paradox games require anyway). Their most common use case is to adjust the configuration in some fashion, which most people do exactly once after installing a game and then never again. Even if that would require a game restart, it would still be preferable to having the deal with an useless extra step every single time you start the game. Really, what's next? A launcher to launch a launcher?

Well, I also guess that the reason why companies put these launcher out is that the QA process for a launcher update is lighter than for the game itself (no one would have guessed...)

View this comment - View article - View full comments

Ah, now is Stardock could do the same and provide a regular, SDL2-based launcher, maybe we'd have less issues with steamplay. Most of their titles are plagued with launcher issues on protondb (luckily, it can be bypassed with /nolauncher).

View this comment - View article - View full comments

But... but... I just started doing that [External Link] with my xen playthrough :'(

Well, guess I'll do a full playtrough soon-ish :P

View this comment - View article - View full comments