clevo_n950tp6/multimonitor_and_desktop/xf86-intel-multimonitor

Add the following boot parameters for syslinux/grub for correct GPU handling (Clevo N950TP6)

Add these to your boot parameters, either /boot/syslinux/syslinux.cfg OR /boot/grub/grub.cfg (GRUB configuration file /etc/default/grub preferred!):

acpi_osi="!Windows 2015" acpi_osi=Linux nogpumanager

NOTE: Without this setting you may be unable to boot on Linux desktop on Clevo N950TP6 laptop.

---

This multimonitor package creates two desktop entries to turn on/off intel-virtual-output executable (can be run as daemon, too). When enabled, the program creates new VirtualGL devices for xrandr. These devices are referring to your external monitors you use with Clevo N950TP6 laptop. Thus, you can set up all needed configuration for these external monitors in your desktop settings.

Only downside with setting up new VirtualGL devices is that discrete Nvidia card must be enabled as well. At least on Linux, all external graphics output goes through Nvidia chipset on Clevo N950TP6. However, if you have your laptop plugged in to AC power adapter, you don't need to worry about Nvidia/Intel dilemma. Temperature of Nvidia card stays at ~47-53 while idle. It is higher on load, of course.

NOTE:

This package overrides the following bumblebee file:

/etc/bumblebee/xorg.conf.nvidia

and adds the following configuration file:

/etc/X11/xorg.conf.d/20-intel.conf

Therefore, you MUST install this package AFTER you have installed bumblebee. And you need to force installation with pacman parameter --force (sudo pacman -U --force package.tar.xz)

---

Usage - personal story

There are two ways I personally use Nvidia card on my Clevo laptop. They are as follows:

• 1) Multimonitor mode: enable additional display/screen output(s). This is which intel-virtual-output executable is for. I use this mode to enable one of the miniDisplayPort output, and attaching one additional monitor to my laptop. Another miniDisplayPort is enabled by default.

• 2) Nvidia "only" mode: Switch to another TTY session (without X server enabled, only CLI). Start a new X server instance with Openbox DE by issuing command: nvidia-xrun openbox-session (AUR - nvidia-xrun)

Worth noting: These modes can not be run simultaneously. For instance, I need to shut down intel-virtual-output process (1) before attempting to launch another X server session (2). Vice versa, I can't execute intel-virtual-output if I have a X server session already opened for Nvidia card (2).

In this way, I can control when Nvidia card is turned ON/OFF.

I use mode 1) for basic desktop tasks, and mode 2) for gaming (Steam, GOG, etc.)

Sometimes intel-virtual-output can complain about not finding bumblebee. You can try checking your /etc/X11 conf files.

About role of bumblebee: it is required by intel-virtual-output binary. Practically, I don't use it for anything else.