ASUS TUF 3090 Ti OC for gaming – black screen issues under stress
ASUS TUF 3090 Ti OC for gaming – black screen issues under stress
Observation: I'm encountering a unique crash situation likely originating from the GPU. During gameplay or other demanding visual tasks such as Heaven Benchmark 4.0, Red Dead Redemption 2, Blender rendering, etc., both monitors go dark and enter an inactive state. Audio keeps playing via headphones, but I must press the power button to close Windows and refresh the drivers before resuming video. Occasionally, after a crash, the login screen on Windows shows strange artifacts. A quick check of the BIOS resolves it without changes. When the GPU fails, a red LED appears next to the Nvidia power adapter, which I suspect indicates no power is reaching the device. The same red light activates when the system is off and stays on until booting. If the GPU crashes, a red LED turns on, remains lit until power is restored, then switches off after about 60 seconds, signaling power presence but no display driver recovery—only after pressing the power button does Windows restart. I've recorded a crash video that can be replicated; monitoring HWinfo will include a screenshot in the reply. To my understanding, temperatures are normal and no unusual voltages are present. My troubleshooting steps include: installing a fresh Windows 11, resetting BIOS defaults, repositioning the GPU in a different PCIe lane, re-seating power cables, rolling back Nvidia and BIOS versions, adjusting XMP settings, moving the M.2 drive, swapping the GPU for an older model, and updating cables. I've also disabled/changed display profiles and tested with a different GPU. I'm uncertain if my GPU is faulty and need to return it for repair or if deeper compatibility issues exist. Motherboard: ROG STRIX Z690-E GAMING WIFI UEFI Version: 1505 CPU: 129000KS Memory: KF552C40BBAK2 64GB - Kingston Fury Beast 5200MT/s GPU: ASUS TUF 3090 Ti (O24G) Storage: 2x Samsung 980 Pro M.2 2TB PSU: Corsair HX 1200 USB Devices (model/version): Ducky Shine 7 Mechanical Keyboard Logitech G9x Mouse Shure PG27USB Microphone CPU Cooler: AIO Arctic Liquid Freezer II 420mm PC Case: Fractal Design Operating system: Windows 11 64 bit (22H2) Activation status: YES Drivers installed (details below). System overclocking: Not applied.
Performed a Cinebench and Blender benchmark. Tests ran smoothly without issues. During Blender, encountered a black screen failure twice in the classroom test. Once the graphics card seemed to recover the driver, restoring the desktop after a few seconds. The score was approximately 1500. On the second attempt, the whole PC rebooted completely. Event viewer displayed a critical error: Kernel-Power, Event ID 41, Task Category 63. Research suggests the power supply might be the issue. However, I’m curious why the PC doesn’t reboot each time this occurs. The red LED near the 12-pin power adapter didn’t illuminate during the crash, or perhaps it was too brief to notice. It’s possible overcurrent triggers a shutdown. The event details could indicate driver problems or BIOS power settings. Noting that I only saw the event after a system reset, unlike crashes that record no events, might help. Here’s a list of apps linked to the crash: Red Dead Redemption 2, Max Payne 3, Apex Legends, No Mans Sky, Overwatch 2, Heaven Benchmark 4.0, Blender Classroom. Screenshots attached for reference, and an HWiNFO .csv file is included for anyone needing further insight into potential faults. The HWiNFO graph shows the final frame before both screens went dark. Crash Log 2.CSV attached for your review.
@RZO I've had Kernel-Power errors without any PSU problems, so its not really a good indication, though the suggestion power is being cut from the PCIe connector certainly is suspicious. I do see mention that NVIDIA GPUs have had this little led for some time now and it does indicate no power on the connector. Have you tried using something like MSI Afterburner and turning down the power limit on the card to see if it still happens?
I attempted to reduce the card's voltage using GPU Tweak III, but it still shows a black screen requiring a restart to restore drivers. Interestingly, the "GPU Tweaks OC Scanner" also failed on the card.
In the GPU Tweak 3 configuration, I adjusted the power target to 90% and 85%, but the crash persists. It seems there might be a bigger problem with the card. I’m planning to buy another power supply and run the system, as Asus asked me to do that to proceed with a possible RMA. I’m also happy to note Best Buy has the HX1500i ready for pickup today.
Just a note, the new power supply didn't resolve the problem. It seems I got a bad one.
I brought my GPU to the tech store, but they couldn't fix the problem. They won't process a return. I'm beginning fresh and will check again later. I'm considering PCIe cables or an adapter for the GPU. Still unsure what to do next.
I apologize for the delayed response. I managed to return the GPU via Amazon because a seller there was willing to accept it back. I exchanged messages with Asus tech support for nearly two weeks, but they were very unhelpful. I opted for the EVGA 3090ti FTW3 Ultra (4985-KR) instead, and it has worked perfectly from the start. Most importantly, I learned from this video: if the tech store didn’t attempt a return because the issue couldn’t be reproduced, what tools they used and whether they tried to recreate it under identical conditions. If your card isn’t RMA-eligible or isn’t covered by warranty, consider checking if you can replace the thermal paste pads yourself. However, I wouldn’t suggest doing that unless you’re comfortable opening your PC and getting messy.