after having some issues with my igpu (7800X3D), i decided to disable it in uefi.
i’m running win10 (22h2, 19045.4651).
as for the description what i was seeing:
- i disabled the igpu in uefi
- reboot
- crash of OpenRGB upon starting up/scanning devices
- i deleted %appdata%/OpenRGB
- i ran the .exe as admin
- as soon as the scan went over the gpu (gigabyte gtx1080 ti aorus master) the programm CTD’d again
- disabling all detectors in OpenRGB.json allowed me to launch the app again
- in the settings i was able to reenable the mobo detection without issues and passed rescanning the devices
- in the settings reenabling all gigabyte device detections triggered the CTD upon rescanning again
at least it was able to log something to that point:
what i noticed in the logs is the following weird message:
for some reason my mobo (an ASUS TUF GAMING X670E-PLUS WIFI) is being identified under gigabyte:
7325 |Debug: [Gigabyte RGB Fusion 2 SMBus] FAILED - 'TUF GAMING X670E-PLUS WIFI' was not found in the JSON list. Do NOT enable if this is a USB based board.
to me it seems that disabling the igpu somehow has messed up the enumeration of the devices/ids/slots, which led to the mobo being somehow wrongfully pushed up to the gpu slot, or something along the lines.
edit: before any questions arise, i’m running version 0.8, but upgrading to version 0.9 yielded no change in behaviour, which is why i downgraded again.
You can run OpenRGB.exe with --nodetect to skip detection IIRC, then disable all devices in the settings menu. You can then enable them one by one to track down the offending detector. The Gigabyte thing looks to just be a debug message where that controller sees your motherboard is (rightfully) not on its list, so I don’t think that is the problem.
narrowed it down to the “Gigabyte GTX1080ti 11G” detector. all except that one can be left enabled. as soon as i add that one for my graphics card into the mix, openrgb crashes.
the confusing part here is that only after disabling the igpu this happens. if i reenable the igpu, things are back to normal with all detectors being able to correctly identify all their components.
any idea why that might be?