I just upgraded to an RX 6950 XT and have been experiencing driver timeouts that result in games crashing (older games could run well, but Alan Wake 2 cannot run for more than 30 seconds unless I use a low resolution). The game will freeze up and Epic will let me know there was a GPU driver issue. AMD says there has been a driver timeout. Once the game closes, my computer is fine and I can reboot the game briefly before it crashes again. Does anyone have any thoughts on how to fix this?
(full build here: https://pcpartpicker.com/list/B2PqWt)
CPU: Ryzen 7 5800X
CPU Cooler: Thermalright Peerless Assassin ARGB
Motherboard: MSI Mag B550 Tomahawk (updated the bios a few week ago)
RAM: Teamgroup 2 x 16 GB DDR4-3200 (using XMP profile from my bios for 3200 MHZ)
Storage: WD Black SN770 2TB M.2-2280 (in first slot, this is where games are installed), Crucial P1 500GB M.2-2280 (the drive from an old build, installed on the lower M.2 slot), WD Black 6 TB 3.5" 7200 RPM HDD
GPU: AMD Radeon RX 6950 XT 16GB
PSU: Corsair RM750e 80+ Gold
Wifi Card: TP-Link Archer TX3000E (in the lowest PCIe slot)
Case: Corsair 4000D Mid Tower
Monitor: LG 27GL83A-B 27"
What I've done so far:
Use DDU to delete the GPU driver and reinstall it (I recently switched from an 2060 Super and had deleted the NVIDIA driver too)
Test RAM with HCI Memtest to about 350%
Test RAM with MemTest 86
Disable MPO in the registry
Extend driver timeout in registry (https://helpx.adobe.com/substance-3...s-crash-with-long-computations-tdr-crash.html)
Other notes:
Temps appear pretty normal (80ish for GPU, 60ish for CPU)
In Task Manager, I noticed the GPU gets to 99-100% under "3D" but the GPU memory use doesn't get anywhere close to 16 GB
I'm wondering if there's a power spike that makes the GPU unable to collect enough power, since I only have a 750W PSU. Could that be causing my issue?
(full build here: https://pcpartpicker.com/list/B2PqWt)
CPU: Ryzen 7 5800X
CPU Cooler: Thermalright Peerless Assassin ARGB
Motherboard: MSI Mag B550 Tomahawk (updated the bios a few week ago)
RAM: Teamgroup 2 x 16 GB DDR4-3200 (using XMP profile from my bios for 3200 MHZ)
Storage: WD Black SN770 2TB M.2-2280 (in first slot, this is where games are installed), Crucial P1 500GB M.2-2280 (the drive from an old build, installed on the lower M.2 slot), WD Black 6 TB 3.5" 7200 RPM HDD
GPU: AMD Radeon RX 6950 XT 16GB
PSU: Corsair RM750e 80+ Gold
Wifi Card: TP-Link Archer TX3000E (in the lowest PCIe slot)
Case: Corsair 4000D Mid Tower
Monitor: LG 27GL83A-B 27"
What I've done so far:
Use DDU to delete the GPU driver and reinstall it (I recently switched from an 2060 Super and had deleted the NVIDIA driver too)
Test RAM with HCI Memtest to about 350%
Test RAM with MemTest 86
Disable MPO in the registry
Extend driver timeout in registry (https://helpx.adobe.com/substance-3...s-crash-with-long-computations-tdr-crash.html)
Other notes:
Temps appear pretty normal (80ish for GPU, 60ish for CPU)
In Task Manager, I noticed the GPU gets to 99-100% under "3D" but the GPU memory use doesn't get anywhere close to 16 GB
I'm wondering if there's a power spike that makes the GPU unable to collect enough power, since I only have a 750W PSU. Could that be causing my issue?