Será possível adicionar um configuração onde a gpu Intel (gráficos híbridos Nvidia) fique a funcionar no wayland, enquanto a Nvidia só no xorg?
EOS currently does not support Wayland as a display server protocol. Eventually in the future, there will be a switch from Xorg to Wayland. In other Linux distributions which support Wayland already, you have the option to choose which session to start at logon - either Wayland or Xorg, so i think this will also be the way Endless goes.
I don’t think i got your question Wayland and Xorg are two different approaches on how to display graphical objects onscreen (atop of the Linux graphics stack, see (1)). For any Wayland session there’s normally a XWayland also running, which allows legacy applications (which use the X11 protocol, which is implemented by Xorg) to run in a Wayland session.
(1) https://www.reddit.com/r/archlinux/comments/6la6n5/trying_to_understand_drm_dri_mesa_radeon_gallium/
Vou tentar reformular o que eu tinha dito
As gráficos híbridas da Nvidia usa tanto a Intel, tanto Nvidia
Quando o wayland for iniciado a partir do gráficos híbridos, o endless iniciaria com Intel
Quando fosse iniciar com o xorg, o endless mudaria automaticamente para Nvidia
OK, i’ll understand. And i think this is possible in general, but not with Endless OS, as it probably would take a good amount of fiddling around with various configuration files and scripts located in read-only places on Endless OS. Additionally, it would be more of a proof of concept than a solution for daily use.
IMHO a good integration of Bumblebee (NVidia Optimus for Linux) would make more sense. Run the Desktop (either Xorg or Wayland) with the Intel GPU and let individual applications run on the dedicated card as needed. This is implemented in Linux Mint 20, and it works perfect. Take a look:
Under the hood, the application is simply run with optirun
prepended .
Seria bom se isso fosse implementado num futuro próximo
Mas para isso acontecer teria que a Nvidia dar suporte ao ostree
No they don’t need to. They just support Linux fine with their proprietary drivers. It’s just a matter of implementing it. And IMHO there’s a reason it’s not already implemented in EOS by Endless - Manpower. Every technology you ship has to be supported in the future, and every new technology costs resources in form of manpower.
Endless will absolute for sure do the calculation on where to invest its resources, based on various parameters gained by knowing it’s userbase (gathered by Telemetry).
Se for esse o lado dos desenvolvedores do endless creio que investir um pouco nisso valeria a pena
Pelo menos, os usuários são os únicos que podem se beneficiar disso
This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.