how to do hardware troubleshooting ?

ioverclockers

ioverclockers

ioverclockers

Associate
Joined
29 May 2020
Posts
67
Hi, after finalizing my first build :
MOBO: ROG CROSSHAIR VIII HERO WIFI
CPU : AMD RYZEN 9 3950X
GPU : AMD SAPPHIRE RADEON VII
RAM: 32Gb G.SKILL TRIDENTZ 3200MHZ (F4-3200C14D-32GTZSK)
FANS:3 BEQUIET! SILENTWINGS 3 120MM HIGH-SPEED PWM
3 NOCTUA NF-A14 140MM PWM CHROMAX BLACK
FAN HUB: THERMALTAKE COMMANDER FP 10 PORT HUB FOR PWM FANS
HEATSINK: Zalman CNPS20X
HEATSINK GPU: RAJINTEK MORPHEUS II CORE
and had it run windows 10 for a few weeks, i'm facing major problems... W10 doesn't want to boot anymore. It loads for a few seconds, says it encountered a problems and gets in a never ending rebooting loop...
I have to admit i tried out the ClockTuner for Ryzen (CTR by 1USMUS) setup and probably didn't finished everything properly. So did that screw up the bios ? I reset the bios, tried to reinstall W10 on another internal SSD etc, but to no avail.
Is there any way to do hardware troubleshooting to see what is the culprit ?
 
How did you reset the BIOS?
I would try clearing the CMOS (usually a jumper on the motherboard). If that does not work then try clearing CMOS with the battery removed and the power unplugged.
 
Strip it down to the bare minimum, mb, proc, 1 stick of ram, gpu (if needed) hdd, try and install windows, if it works or not switch ram sticks and check again, then if it works add the second stick, then if it's still working add back one thing at a time.
 
well, i cleared the CMOS, removed its battery, and that made windows boot, but only to get itself again into a never ending rebooting loop.
So i stripped down everything, swapped the RAM, and - for the moment - that's what made things work again (i did have to completely erase my NVME disks though). Adding the original RAM sticks to the mix, got everything stuck again. Removing them made the system work... So obviously it was the RAM.
I'm a bit upset though. I mean, i got the best RAM i could get on the market : 32Gb G.SKILL TRIDENTZ 3200MHZ (F4-3200C14D-32GTZSK). I did a mild overclock, being careful with the voltage, so was that what caused the problems ? Anyway, is there a way to "reset" RAM to factory defaults so to say, or must i consider these sticks ready for the trash can ?
 
Last edited:
Usually just resetting bios back to default would work, you could try 1 stick of the problem ram at a time, might be completely a coincidence that 1 of the stick may have failed.
 
BIOS reset also drops memory voltage back to JEDEC standard, so did you remember to raise it to 1,35V after that?
 
Try each stick alone in different sockets. Also try running memtest from a usb stick and fully test the ram, it may well still be in warranty and worth checking,

From the webpage:

Limited Lifetime Warranty
All G.SKILL memory products come with a limited lifetime warranty and the G.SKILL technical team is always ready to provide consumers with complete technical support via online forum, telephone, and email.
 
Unfortunately PC hardware is so complex that its really difficult to diagnose. It's largely a question of experienced guesses backed up with interpreting error messages and trying replacement parts.

When you run RAM at the XMP settings you are in fact overclocking parts of the CPU and motherboard. Overclocking can fail at any point in the history of a PC so the first thing I would do is make sure you are not running the RAM on XMP settings, but just let the motherboard default to its lower, default settings. This is done in the BIOS or is the default after the BIOS is reset.
 
You must log in or register to reply here.
Back
Top Bottom