My gaming rig is frequently hanging in Windows 11, however I do not believe Windows 11 is the root cause of the hang. The whole system just locks up and I have to Hard Power Down and then Power On again.
Every time the system hangs, it locks up on the time/date.
In event viewer, at exactly this time and date, I have an error message.
Display driver amdwddmg stopped responding and has successfully recovered.
2 seconds later, another error message appears
Miniport driver failed to add device with status {Operation Failed}
The requested operation was unsuccessful.
.
It is worth mentioning that these issues never occurred when I had a Nividia 2080 Super GPU.
They have exclusively arose ever since I installed my Sapphire Nitro Radeon RX7900 XTX.
Prior to that, my system was rock solid and never crashed at all.
Of course, one has to go through a deep process of elimination, so over the last few weeks, I have performed the following updates on all my hardware (that is updatable).
Motherboard : Asus ROG Crosshair VIII Dark Hero. BIOS 4501 : Latest Version 23/5/2023
O/S : Windows 11 Home : 10.0.22621.1848 (Win11 22H2 2022 Update)
Processor : Ryzen 9 5900X
RAM : 32GB (2 x 16GB) : XMP DDR4-3600 (1800 MHz) Team Group TEAMGROUP-UD4-3600
GPU : Sapphire Nitro Radeon RX7900 XTX
GPU BIOS : 022.001.002.012.000001
GPU Driver : Adrenalin 23.5.2 (23/05/2023) Latest Version
Now, further testing on my Rig has thrown up a few observations.
I use AquaSuite and an internal Fan Controller to monitor all my temps.
I have 10 fans in the case. 3 pushing air in to my Arctic 420 Rad, 3 pulling air through it in to the case. I have 3 on top pulling heat out. 1 at the rear, pulling heat out.
My current temps (in Windows only, plus Word App and Edge) are
SSD : 31 C (87.8 F)
CPU Diode Temp : 43 C (109.4 F)
GPU Hot Spot : 56 C (132.8 F)
I have a sensor monitoring the air below the GPU : This reads 25.0 C (77 F)
I have a sensor monitoring the air above the GPU : This reads 26.0 C (78.8 F)
My fans are currently running 100% keeping the unit cool.
Ambient Air Temp is 22.9 C (73.22 F)
Dropping my fans to 50% speed, brings my temps to the following within 10 mins
SSD : 32 C (89.6 F)
CPU Diode Temp : 47 C (116.6 F)
GPU Hot Spot : 50 C (122 F)
The sensor monitoring the air below the GPU : This now reads 26.3 C (79.34 F)
I have a sensor monitoring the air above the GPU : This now reads 32.5 C (90.5 F)
So here is my observation…
If I leave this Rig at these temps (which are not ridiculously hot are they ?), the chance of the system hang and the Display driver amdwddmg stopped responding and has successfully recovered increases considerably.
I’ve left the Rig overnight, and next morning, the room ambient temperature is approx. 29.9 C, and the Rig will have frozen overnight. However it freezes on Aquasuite so I can see the last readings, and they are pretty good to me.
The sensor monitoring the air below the GPU froze on 22.6 C (72.68 F)
The sensor monitoring the air above the GPU froze on 30.4 C (86.72 F)
I did wonder if the mass of warm air hanging above the GPU might be causing the Rig to freeze, but those temps do not look worrying to me (feel free to correct me if I am wrong).
Other things I have tried are a complete DDU uninstall in safe mode of the AMD Drivers and Software, then reinstallations, but this makes no difference at all.
Using Sapphire’s TRIXX to monitor the voltage of the GPU shows quite an erratic up and down spiky graph. I'm not sure if that is ok ? I can’t even given an average as it moves so much, but perhaps 0.375V, with frequent peaks to 0.750v, sometimes a little higher. Is this normal ?
One last observation. My wife also got a 7900XTX Hellhound Spectral White by Powercolor for her AMD Rig (Same specs as mine). She also has random freezes and locks. She is running Win 10. Same error for her too.Before this, her Rig was also rock solid with her 2080 Super. Her error in Event Viewer is, surprise surprise...
Display driver amdwddmg stopped responding and has successfully recovered.
Again, we've done DDU uninstall and reinstall on her machine as well.
To be fair, these are our 1st AMD GPU’s, and we’re terribly disappointed and considerably out of pocket and I don’t know where to go from here. The cards were purchased in mid-April this year (2023) and I don’t feel they are fit for purpose and I am considering returning them to their vendors. We could have got 4090’s each and to be honest I wish we had, but if there is any chance we can resolve this, that would be nice !
Thank you folks for any advice you may have.
22/06/23 : Update
Further monitoring (using Aquasuite) give me some more info.
The Sapphire Nitro GPU fan normally sits at 0 rpm, when I am doing general office work (Word / Excel) and have a few tabs open in Edge.
I was in a Teams video call earlier and I noticed the GPU fan came on at about 660 rpm. I then notice at the same time, the ambient temp sensor at the top of my case (it hovers over the mobo) went from a temp of approx 22C (71.6 F) to about 32C (89.6 F). Where did that 10 C come from ? I watched it climb higher and my system froze at about 34.5 C (94.1 F) with the usual "Display driver amdwddmg stopped responding" error message.
I've just rebooted and watched the system again. As soon as the Nitro GPU Fan comes on, an extra 10 C (16.2F) of heat hits the top of my mobo.
Is the Nitro dumping a load of stored heat above my mobo in my case ?
CPU temp doesn't rise much during this process (43 C > 50C) (109.4 F > 122 F). That does not seem worrying. But this is when I am most likely to have a complete system freeze with "Display driver amdwddmg stopped responding".
Curiously, when the system did freeze on my Teams Call, mouse and keyboard hung, video froze, but I was able to talk to my friends fine. (They noted my live video had frozen).
