Black screen detected, but 3070TI is working fine. Any driver problems?
Black screen detected, but 3070TI is working fine. Any driver problems?
Previously I used VR on a headset with my 1070 GPU, everything seemed normal. Recently I upgraded to a 3070 Ti and noticed some problems. Initially there was no obvious difference. Over time, my computer would freeze while the GPU handled more tasks. Both screen and headset displayed black screens. I had to power off and on again to use it. This never happened with my 1070, even when playing the same games. When I played Half-life Alyx, the 3070 Ti caused severe lag and crashes after a few minutes. On the other hand, my 1070 ran smoothly without any issues. This is puzzling since the 3070 Ti should be more capable. I started with the game on my 1070, then switched to the 3070 Ti – it worked fine except for some loading stutters. Eventually I completed the whole game. Today I played DCS in VR on the 3070 Ti, but it crashed after a short time again, though FPS stayed stable. This seems odd because a month ago the same game also crashed on the 3070 Ti without problems on my 1070 either. I suspect a driver-related problem. I attempted reinstalling Windows Mixed Reality and Steam VR, but nothing helped. I tried updating GPU drivers (GeForce Experience) without success. It’s worth mentioning that during crashes the 3070 Ti reaches around 83°C. Also, when not in VR, it performs better than the 1070 in terms of FPS, though black screens still occur occasionally. Anyone know what might be going on? Also, is it normal for Nvidia not to warn me about driver updates when switching GPUs? Are the drivers for the 1070 and 3070 Ti compatible? When comparing, only the GPU changed; CPU, RAM, power supply remain the same.
This model is the Fox Spirit GT-850P V2, featuring an 80PLUS Platinum rating and delivering 850W of power.
Based on the quick search, the PSU appears to be a decent unit with a reasonable price. However, I suspect another issue is at play since the screen went black during shutdown. The GPU reached temperatures close to 83°C, which is likely its maximum safe operating point. Running at that level shouldn't damage performance but could affect stability. First, perform a fresh installation of the GPU drivers. Use DDU to remove any residual drivers, then apply the new ones.