Question Monitor driver anomaly

[bsmither]

This Win7 system has had a pair of ASUS PB287Q monitors, and both were running at 3840x2160. On a regular basis, the second monitor would not wake up and needed a power-cycle to be recognized. Unfortunately, this unit had developed a problem with energizing the backlight and so was replaced by a ASUS VG289Q. The VG289Q monitor was connected prior to installing the INF file for it. The video card is (XFX Pine Group) AMD Radeon HD 6950 and the driver is (Catalyst 15.7.1) 15.200.1062.1004.

Further careful observation shows:

1. The monitors receive a signal and wake from standby.
2. "The system has detected a link failure and cannot set the requested resolution and refresh rate on your DisplayPort display."
3. The VG289Q stays at standby.
4. Thirty minutes of no activity times out and video signals are suspended, monitors go to standby.
5. The monitors receive a signal wake from standby.
6. The VG289Q wakes up at 2560x1440.
7. Thirty minutes of no activity times out and video signals are suspended, monitors go to standby.
8. Goto 1.

Note that steps 3 and 6 happen reliably in that order. That is to say, at no time does the VG289Q wake up twice in a row, nor stay at standby twice in a row.

The VG289Q is running at 2560x1440 and Win7's Display Settings does not offer any resolution higher than that.

(FYI: As an experiment, the VG289Q has been tested on a Linux Mint system where the monitor's OSD reports that HDMI-1 is displaying at 2160p at 30Hz, and Linux's Display app reports 3840x2160 being sent. The motherboard is GigaByte_GA-F2A78M-D3H_v.4.0.)

I am going to (make a bare-metal backup first, then) try DDU to kill all GPU drivers and registry entries.

=======================================

Q: Is there a utility to also kill all references in the registry that describe the resolutions, multi-display positions, orientations, etc?

For example:

HKLM/System/CCS/Control/Class/{4D36E96E.../ -> 0001/Modes/3840,2160/Mode1="24.0-160.0,30.0-80.0,+,+" -> 0003/Modes/3840,2160/Mode1="160.0-160.0,40.0-60.0,+,+" -> 0006 identical to 0003

and
HKLM/SYSTEM/ControlSet001/Control/GraphicsDrivers/Configuration/ HKLM/SYSTEM/ControlSet001/Control/GraphicsDrivers/Connectivity/

and all of these:
HKEY_LOCAL_MACHINE\SYSTEM\ControlSet001\Control\GraphicsDrivers\Configuration\ACI28A346492_0F_07E0_5C*AUS28BA82184_0B_07E7_14^4FC81E0C22E2AB9BE06E7CD40EA73914 HKEY_LOCAL_MACHINE\SYSTEM\ControlSet001\Control\GraphicsDrivers\Configuration\ACI28A346492_0F_07E0_5C+AUS28BA82184_0B_07E7_14^99BFFCEC4A3B9D6C381E3BDE80303EF1 HKEY_LOCAL_MACHINE\SYSTEM\ControlSet001\Control\GraphicsDrivers\Configuration\ACI28A346492_0F_07E0_5C+AUS28BA82184_0B_07E7_14+AUS28BA82184_0B_07E7_CE^381AE4279B4595AD191CA65A0847AA22 HKEY_LOCAL_MACHINE\SYSTEM\ControlSet001\Control\GraphicsDrivers\Configuration\ACI28A346492_0F_07E0_5C+AUS28BA82184_0B_07E7_CE^909AE1F8DA77564253BE88C86BA2F315 'Stride' values are suspicious. HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\GraphicsDrivers\Configuration\ACI28A347640_0F_07E0_DB+ACI28A346492_0F_07E0_5C^564CB1B6E4BF92E92C318C41D9D8B9C2\00

=======================================


Q: Looking at the "ASUS VG289.inf" file, comparing DP with HDMI1, is there anything suspicious about the "160.0-160.0" values?

[28BA_DP.AddReg] HKR,"MODES\3840,2160",Mode1,,"160.0-160.0,40.0-60.0,+,+" HKR,,MaxResolution,,"3840,2160" HKR,,DPMS,,1 HKR,,ICMProfile,0,"ASUS VG289.icm" [28BA_HDMI1_2.0.AddReg] HKR,"MODES\3840,2160",Mode1,,"30.0-160.0,40.0-60.0,+,+" HKR,,MaxResolution,,"3840,2160" HKR,,DPMS,,1 HKR,,ICMProfile,0,"ASUS VG289.icm"

 

[Lutfij]

You might want to take the panels and hook them up to a donor system and see if the same issue is seen on said donor platform. Chances are that your discrete GPU might be making preps to leave.