F5F Stay Refreshed Hardware Desktop Frequent GPU issues and crashes in games are common problems.

Frequent GPU issues and crashes in games are common problems.

Frequent GPU issues and crashes in games are common problems.

M
Mileo
Junior Member
21
12-12-2018, 04:41 AM
#1
Hello Everyone, I'm facing a persistent problem with my PC over the last few months. I'm still trying to identify the exact cause while searching online. During gameplay, the system frequently stops in several ways: 1) BSOD with error code nvlddmkm.sys (30% occurrence), 2) complete shutdown requiring manual restart (60% chance), and 3) games crashing directly to the desktop (10%). Other issues include occasional freezes and crashes. Some titles like Fallout 76 and Dragon Age Origins cause these problems after about 5-10 minutes, while others like Binding of Isaac remain stable for hours. I've tried updating GPU drivers, reinstalling them, updating Windows manually, running memory checks, and defragging drives. My PC runs a Ryzen 2600 with an Asus GeForce 1070 and dual 16GB 3200MHz RAM—never overclocked. For reference, here are the minidump files from BSOD instances: https://drive.google.com/drive/folders/1...sp=sharing I hope someone can assist me.
M
Mileo
12-12-2018, 04:41 AM #1

Hello Everyone, I'm facing a persistent problem with my PC over the last few months. I'm still trying to identify the exact cause while searching online. During gameplay, the system frequently stops in several ways: 1) BSOD with error code nvlddmkm.sys (30% occurrence), 2) complete shutdown requiring manual restart (60% chance), and 3) games crashing directly to the desktop (10%). Other issues include occasional freezes and crashes. Some titles like Fallout 76 and Dragon Age Origins cause these problems after about 5-10 minutes, while others like Binding of Isaac remain stable for hours. I've tried updating GPU drivers, reinstalling them, updating Windows manually, running memory checks, and defragging drives. My PC runs a Ryzen 2600 with an Asus GeForce 1070 and dual 16GB 3200MHz RAM—never overclocked. For reference, here are the minidump files from BSOD instances: https://drive.google.com/drive/folders/1...sp=sharing I hope someone can assist me.

C
CuzIm100ms
Member
127
12-14-2018, 12:12 AM
#2
Evga experienced problems with their power delivery for the 1070 model. It’s possible you’re facing the same issues. Check the HWINFO64 status of your 12V output—if it’s unstable, using a multimeter to test the voltage could help confirm if power delivery is the cause. A re-paste might be worth considering.
C
CuzIm100ms
12-14-2018, 12:12 AM #2

Evga experienced problems with their power delivery for the 1070 model. It’s possible you’re facing the same issues. Check the HWINFO64 status of your 12V output—if it’s unstable, using a multimeter to test the voltage could help confirm if power delivery is the cause. A re-paste might be worth considering.

N
N4ox
Member
122
12-15-2018, 07:08 PM
#3
Sure, I'd be happy to clarify. When checking hwinfo, focus on the relevant sections such as hardware status, memory information, and system logs. Since I'm not familiar with a Multimeter, avoid using it for diagnostics. Regarding the re-paste, it seems the issue might not be software-related.
N
N4ox
12-15-2018, 07:08 PM #3

Sure, I'd be happy to clarify. When checking hwinfo, focus on the relevant sections such as hardware status, memory information, and system logs. Since I'm not familiar with a Multimeter, avoid using it for diagnostics. Regarding the re-paste, it seems the issue might not be software-related.

K
Kytzis
Member
183
12-22-2018, 06:34 PM
#4
hwinfo64 offers detailed insights into your computer's performance, including temperatures, clock speeds, and voltage readings from various components. You'll notice the 12V rail voltage along with an estimate of the voltage your card receives from the power rails. This helps you assess hardware before considering software fixes. It seems the 1070 series likely lacked strong VRMs, or your power supply might not be delivering enough power. Using a multimeter would be ideal, but if you're unsure, avoid shorting the PSU—it won't damage it, though it could trigger its protection. Under load, as long as the 12V stays above 11.8V, everything should function well. The spec drops to 11.6V, but significant drops usually indicate an issue. Values above 12V are acceptable and may even rise slightly to 12.1 or 12.2V.
K
Kytzis
12-22-2018, 06:34 PM #4

hwinfo64 offers detailed insights into your computer's performance, including temperatures, clock speeds, and voltage readings from various components. You'll notice the 12V rail voltage along with an estimate of the voltage your card receives from the power rails. This helps you assess hardware before considering software fixes. It seems the 1070 series likely lacked strong VRMs, or your power supply might not be delivering enough power. Using a multimeter would be ideal, but if you're unsure, avoid shorting the PSU—it won't damage it, though it could trigger its protection. Under load, as long as the 12V stays above 11.8V, everything should function well. The spec drops to 11.6V, but significant drops usually indicate an issue. Values above 12V are acceptable and may even rise slightly to 12.1 or 12.2V.