Is it happening that the display frequently turns off unexpectedly on my latest model?
Is it happening that the display frequently turns off unexpectedly on my latest model?
I recently assembled a new computer and it suddenly powers off while browsing the internet or engaging in other activities about 10 to 15 minutes after turning it on. The machine remains on during these incidents, but the screen goes dark as if it had lost power and shuts down. I've been looking for solutions for more than a month and haven't found any relevant information. I'm extremely frustrated because I've invested over two thousand dollars in everything, and I can't even use it anymore. I've updated Windows, the graphics and system drivers, but nothing seems to work. If anyone has suggestions or additional details, please let me know—I really need help figuring this out. Thanks.
I've confirmed the updates are in place. My power supply is a Thermaltake/Toughpower GF1, delivering 850 watts. The screen shuts down with an error message, indicating it can't find the input signal. It's going to sleep, but my computer remains active.
When boosting your RAM beyond the DDR4 JEDEC limits (usually around 2,133MT/s or 2,400MT/s), consider disabling the XMP/EXPO/DOCP overclock in the BIOS. If your T-Force Xtreem memory shows slight instability at speeds like 3,600 or 4,000MT/s (or according to the Asus DOCP setting in SPD), adjust the RAM clock rate and look for a fully stable configuration.
To verify stability, download MemTest86, load it onto a USB stick, boot from USB, and perform a comprehensive test lasting several hours.
If errors appear, lower your DOCP overclock in Asus BIOS—such as reducing from 3,600MT/s to 3,200MT/s—and rerun the full MemTest86 pass.
Note: The Ryzen 5950X memory controller supports up to DDR4-3200. Speeds above 3,200MT/s may work but could cause instability. By reaching 4,000MT/s, instability becomes likely.
Integrated components and memory channels: 2 total, supported speed 3200, ECC available.
I tried this and still face the same problem. I began with DOCP off and then increased to 3200MTs to 3600MTs. When I attempted 3800MTs my computer would not boot and would just turn on and off. After each PASS (four times) I would restart the system, leave a YouTube video playing and it would shut down. I set a timer to monitor how long it took before the screen went dark. Here are the outcomes for each setup:
DOCP off - PASS - 11 minutes
3200 DOCP - PASS 11 minutes
3400 DOCP - PASS 12 minutes
3600 DOCP - PASS 13 minutes
The RAM I'm using is DDR4 3600 CL14-15-15-35 with the standard BIOS configuration.
It seems altering RAM speed doesn't impact the issue. Because the YouYube video stops in a similar time frame, I question whether it's related to heat buildup reaching a critical point. Playback shouldn't strain your CPU or GPU, but could you verify their temperatures while it runs?
Did you review Event Viewer, Windows Logs, and System for any alerts with a red exclamation just before the screen becomes unresponsive? It appears unusual that the issue recurs at specific times after powering on.
Yeah, I looked at the event viewer last night and noticed a few warnings for distribution com, but no red alerts. Just a couple of critical messages about the computer shutting down suddenly. I tried undervolting and also did a fresh install last night to see if that would help, but it didn’t. I had to run the install from USB because my computer kept freezing when trying to set up the reinstall.
Most of my red Event Viewer alerts aren't critical, though occasionally they highlight issues. I wonder if you could exchange the Xtreem RAM for another model from a different PC, a friend, or a shop? If it's not available on the QVL, it could cause problems, but I haven't encountered any issues using RAM not listed there. I usually stick to low-speed Crucial or Kingston products, though I haven't tried TeamGroup yet.
It might be useful to take out the CPU from its socket and inspect the pins for damage. I tend to feel uneasy about removing the CPU unnecessarily in case it gets dropped and some pins bend.
If your cooler is tightly secured, the board could flex as it heats up, causing the CPU to lose contact with one or more pins. Try loosening a few screws on the four cooler mounting points by half or a full turn. I know it's not an LGA1700 setup, which is more likely to flex, but who knows?