Video card failure occurs shortly after launching a game.
Video card failure occurs shortly after launching a game.
Hi, I own an ASUS 2070 RTX and every time I start a game it crashes with a black screen. The rear fan on the GPU stops working unless I install MSI Afterburner and force the fans to run above 80%. Under normal browsing (just web use), it drops around 30% and then stops immediately when a game starts. The machine is about seven years old, and the graphics card was replaced in 2019. I haven’t experienced any problems until recently. I cleaned the fans, dusted the CPU cooler and radiator, but everything else seems fine except for a bit of dust. I removed and reinstalled drivers, even performed a factory reset, yet the issues persist. There’s no unusual heat or plastic smell, just strange behavior. I’m guessing there might be a problem with the power supply or the GPU itself. Any advice would be greatly appreciated. Also, when I run MSI Afterburner and force the fans to 80%+, the game runs smoothly with over 100 FPS, no tearing or stuttering, but monitoring software shows the GPU is barely active—this is confusing, especially since I’m not an expert.
CPU: i7 7700K LGA1151 (Kaby Lake)
GPU cooler: Cyber Power Asetek 550LC 120mm Liquid Cooling
Motherboard: MSI Z270 Gaming Pro Carbon ATX
RAM: 8GB x 2 Gskill Trident z
Storage: WD 1TB Blue Series SATA 3
GPU: Asus RTX 2070 Super 8G
Power supply: Corsair 850M original from 2017
Case: Cyberpower
Operating System: Windows 11
Monitor: Omen 27 (primary) and Samsung U28E590D (secondary)
Welcome to the forums, newcomer!
I own an ASUS 2070 RTX.
Asus offers several RTX2070 models in their collection—what’s yours? Attempt to uninstall all GPU drivers (Intel, AMD, Nvidia) using DDU in Safe Mode, then reinstall the newest driver from Nvidia's support site with administrator privileges by right-clicking the installer.
Additionally, when sharing troubleshooting threads, it’s a good practice to provide complete system specifications. Please list your details clearly:
CPU, CPU cooler, Motherboard, RAM, SSD/HDD, GPU, PSU, Chassis, OS, Monitor, including the age of the PSU and current BIOS version for your motherboard.
This suggests an issue with the GPU's fan speed controller. When using Afterburner, I limit the GPU's maximum power to 95% for better stability during rendering, keeping the fan speed at Auto. This makes sense since GPU usage can spike from below 30% while browsing to over 80% during intense gaming, increasing heat and instability risks. It could be either a software or hardware problem. The easiest solution is to replace parts until the issue resolves. The RM850 (specific model and year?) appears to be a solid power supply. The RTX 2070 may show signs of wear. Ensure the PCIe power connections are securely connected to the GPU. The LGA1151 CPU and motherboard are aging (many boards have 1150/51/55), so don't dismiss them. Look for bulging or leaking components on the motherboard, and consider checking RAM or CPU health first.
When I get home I'll attempt to limit the power to 90% and let the fans run freely to observe the results.
The usage remains stable when starting a game; it drops and stays below 8%.
I can't recall the exact year, but it seems to be around 2017 or 2016 since that's the original model I installed. After cleaning the GPU, I reinserted it and checked the plugs—everything was fine. I also unplugged all the leads from the PSU, blew air through them to clear debris, and inspected; everything looked good. I wish I had replacement parts, but I don’t have any to exchange or test.
The motherboard appears in excellent condition. There were no unusual signs such as bulges, CPU grease leaks, or oxidation. It’s been stored in a well-maintained environment with careful temperature and humidity control, protecting my collection from damage.
uninstalled and reinstalled drivers for NXZT software indicate no overheating, and I’m not excluding the power supply. However, why would increasing the power to the GPU fans help if the issue wasn’t with the power source? Pushing more power to the fans on the GPU seems to improve performance but doesn’t show any temperature rise.
Despite the software indicating no overheating, there remains a chance that overheating is contributing to the issue. A malfunction in the GPU might cause it to fail at a temperature lower than expected. To identify this, replacing it temporarily could help with troubleshooting.