3 ms·

But why do we then aim to replace X11, which is from a user perspective almost feature-complete?

by joebiden2 4y ago

smoldesu 4y ago
x11 gets some stuff plain wrong. It's a trooper with decent hardware support and an enormously long legacy behind it, but along with that comes really poorly-designed hacks and deeply ingrained desktop failures. Like, the x11 framebuffer for example: add a second display, and it just extends the framebuffer resolution to the largest monitor. If your refresh rates are different, you're also wasting GPU cycles. It can be worked-around in software, but it pretty significantly hobbles the integration possible with a fully-featured desktop. Or, look at how x11 views synchronization primitives: it doesn't! You need to implement a compositor with an opinionated hack to make it work, which still doesn't work consistently across configurations. Want to use FreeSync? It's blocked by the model x11 uses to abstract displays. Entire classes of usage (eg. gaming, responsive gestures, video editing, HDR pipelines, color calibration, etc.) are inherently limited by what x11 offers. I do miss the old DEs and their feature-completeness, but getting them to compete with modern MacOS and Windows was like oiling up Hulk Hogan to go fight Floyd Mayweather. Truly the only competitive way forward was abandoning the xorg model and letting DEs define their desktop.
- joebiden2 3y ago
  It would be fair if you also compared Wayland to X11 in the areas wayland really lacks, which are the ones I originally posted. It is a strict regression in all of these. In my mind, a modern framework should embrace security first, then accessibility, then architectural beauty. Wayland gets this, from my humble point of view, exactly wrong (architectural beauty, security, accessibility). Maybe it's time for another re-approach of linux DE's.

3 ms·

But why do we then aim to replace X11, which is from a user perspective almost feature-complete?

by joebiden2 4y ago

But why do we then aim to replace X11, which is from a user perspective almost feature-complete?

smoldesu 4y ago
x11 gets some stuff plain wrong. It's a trooper with decent hardware support and an enormously long legacy behind it, but along with that comes really poorly-designed hacks and deeply ingrained desktop failures. Like, the x11 framebuffer for example: add a second display, and it just extends the framebuffer resolution to the largest monitor. If your refresh rates are different, you're also wasting GPU cycles. It can be worked-around in software, but it pretty significantly hobbles the integration possible with a fully-featured desktop. Or, look at how x11 views synchronization primitives: it doesn't! You need to implement a compositor with an opinionated hack to make it work, which still doesn't work consistently across configurations. Want to use FreeSync? It's blocked by the model x11 uses to abstract displays. Entire classes of usage (eg. gaming, responsive gestures, video editing, HDR pipelines, color calibration, etc.) are inherently limited by what x11 offers. I do miss the old DEs and their feature-completeness, but getting them to compete with modern MacOS and Windows was like oiling up Hulk Hogan to go fight Floyd Mayweather. Truly the only competitive way forward was abandoning the xorg model and letting DEs define their desktop.
- joebiden2 3y ago
  It would be fair if you also compared Wayland to X11 in the areas wayland really lacks, which are the ones I originally posted. It is a strict regression in all of these. In my mind, a modern framework should embrace security first, then accessibility, then architectural beauty. Wayland gets this, from my humble point of view, exactly wrong (architectural beauty, security, accessibility). Maybe it's time for another re-approach of linux DE's.