Color banding w/390.12 (randomly happening)

Computer: HP Z820
Gpu: Quadro K6000
OS: CentOS 7.2
Driver: 390.12
Monitor: Eizo CG245

Hi,

First post here, new to the forum.

Sometimes the screen shows color banding. Nvidia-settings shows 24 bit color depth but the screen has banding instead of smooth gradients, on the desktop as well as in applications.

Killing Xserver with CTRL-ALT-BACKSPACE will bring the screen to normal (no more banding).

Killing Xserver again and the problem can come back randomly. Most of the times it’s fine but let’s say 1 times out of 50 it will have the color banding behavior.

Seeing this with driver 390.12 (not seeing with 346.35 which is the former driver that was used on this machine).

Not sures what causes it and if it is possible to prevent from happening ?

Thanks for any input.

This kind of thing can happen if the cable or monitor is flaky. Can you please run nvidia-bug-report.sh in both the good and bad cases and attach them to your post?

I don’t know why 390.12 would be more sensitive to link training flakiness than 346 would be.

Some troubleshooting done:
On another machine, the problem is not randomly there but always present with 390.12 (384.98 also acts the same.) As soon as I downgrade to 346.35 the color banding is gone.
This machine is using an Avenview DisplayPort extender cable (model FO-DP-25-MM) and that seems to be part of the problem. (issue with the extender and some versions of the driver (384.98 and 390.12 are the 2 I tested that showed the problem)

To sum it up:
346.35 + Avenview Extender = No problems
390.12 + Avenview Extender = Color Banding (also seeing with 384.98)
390.12 + regular displayport cable = No problems

I’ll bring this up to Avenview for their consideration as well.
I’ll look into attaching the logs.
Thanks!

Can confirm this issue on a System76 Oryx Pro laptop with a GTX 1070M. However, it isn’t random, but happens 100% of the time.

mmstick:~# inxi -xSGI
System:    Host: pop-os Kernel: 4.15.0-10-generic x86_64 bits: 64 gcc: 7.3.0
           Desktop: Gnome 3.26.2 (Gtk 3.22.28)
           Distro: Pop!_OS 18.04 LTS (Bionic Beaver)
Graphics:  Card: NVIDIA GP104M [GeForce GTX 1070 Mobile] bus-ID: 01:00.0
           Display Server: x11 (X.Org 1.19.6 ) driver: nvidia
           Resolution: 1920x1080@60.02hz
           OpenGL: renderer: GeForce GTX 1070/PCIe/SSE2
           version: 4.5.0 NVIDIA 390.25 Direct Render: Yes
Info:      Processes: 316 Uptime: 15 min Memory: 4248.1/32139.2MB
           Init: systemd runlevel: 5 Gcc sys: 7.3.0
           Client: Shell (ion) inxi: 2.3.56 
mmstick:~# glxinfo | grep OpenGL
OpenGL vendor string: NVIDIA Corporation
OpenGL renderer string: GeForce GTX 1070/PCIe/SSE2
OpenGL core profile version string: 4.5.0 NVIDIA 390.25
OpenGL core profile shading language version string: 4.50 NVIDIA
OpenGL core profile context flags: (none)
OpenGL core profile profile mask: core profile
OpenGL core profile extensions:
OpenGL version string: 4.6.0 NVIDIA 390.25
OpenGL shading language version string: 4.60 NVIDIA
OpenGL context flags: (none)
OpenGL profile mask: (none)
OpenGL extensions:
OpenGL ES profile version string: OpenGL ES 3.2 NVIDIA 390.25
OpenGL ES profile shading language version string: OpenGL ES GLSL ES 3.20
OpenGL ES profile extensions:

http://www.mediafire.com/file/k9v3ocq3iv7odqc/nvidia-bug-report.log.gz

mmstickman, can you switch the laptop to hybrid or intel mode to rule out a general hardware issue?

This system doesn’t have the Intel iGPU available as an option in the BIOS, and disabling the NVIDIA driver will simply switch to LLVMpipe software rendering. lscpi verifies that an Intel iGPU isn’t available:

mmstick@mmstick-oryx-solus ~ $ lspci | grep VGA
01:00.0 VGA compatible controller: NVIDIA Corporation GP104M [GeForce GTX 1070 Mobile] (rev a1)

I have noticed, however, that using the NVIDIA X Server Settings control panel, and switching dithering to Disabled, and then back to Enabled fixes things, but the process has to be repeated each time you reboot.