Recent updates to the artifactingRegistry are triggering significant GPU issues.
Recent updates to the artifactingRegistry are triggering significant GPU issues.
I primarily use my personal computer for Microsoft Flight Simulator (2020 edition). Recently, a performance tweak in MSFS forums suggested changing the NDU registry setting—specifically adjusting the `start` DWORD value in the system registry to 4 instead of 2. This adjustment reportedly improved speed across different configurations. However, over the past week, every flight I attempt ends abruptly with a crash to desktop. Sometimes the crash appears during loading, other times in the menu, or at irregular moments while flying. I haven’t flown a single simulation session in two weeks. In some instances, restarting the sim after it crashes only causes temporary monitor disconnects from my GPU, but everything eventually returns to normal. I’ve tried all the fixes recommended by the MSFS community and directly by Microsoft/Asobo. Recently, a new idea emerged: undoing the NDU registry change. I reset the DWORD value back to 2 and restarted. Once logged in, everything seemed fine, yet after a few seconds I began streaming to Discord and performing light tasks on the desktop. Heavy artifacts appeared everywhere—on both screens (photos follow)—and no game loaded properly. Using the PC became impossible. After rebooting in safe mode and reverting the registry setting, things normalized, though the MSFS CTD issue persisted. During artifacting, I checked my GPU temperatures with HWINFO64; everything was within normal ranges—core at 38°C, hotspot at 45°C, memory junction at 43°C. With the registry set to 2, no matter what MSI Afterburner settings I applied, the artifacts remained severe. My specs are: Intel Core i9-10900K (5.2 GHz, max 80°C under continuous Cinebench R23), MSI Ventus 3X GeForce RTX 3080 10GB GDDR6X, OC, 2x 16GB Corsair Vengeance Pro SL DIMMs, stock at DDR4-3200 (16-20-20-34), running MMO: Gigabyte Z490 AORUS Pro AX CPU cooler with Arctic MX-5 paste and 2mm Thermalright pads, +150MHz on core, +1250MHz on memory RAM. I’m using MSI Ventus 3X OC at 3500MHz. Storage: 2TB Samsung 970 Evo Pro. The problem persists even after changing the registry setting back to 4. Any suggestions?
I received a check and updated the latest Nvidia game driver to version 516.59. The problem continues, showing identical behavior regardless of the NDU settings. Updated July 21, 2022 by SimPilotAdamT – clarification provided...
I’m struggling to find anything else to do except switch to another GPU. It’s odd how the artifacts vanish when using NDU 4. I haven’t worked on the registry recently, so I’m unsure if that could be causing this. It might not seem like a network diagnostic tool, but who knows what other functions Windows has for it?
I checked the NDU setting recently and everything works properly. The GPU is functioning normally without any artifacts. The main change is the room temperature—previously around 35-38°C, now about 23-25°C. It might be related to the cooler environment. I’ve ordered new thermal pads and paste to see if that helps, but this seems unusual compared to past issues.