Answer the question
In order to leave comments, you need to log in
Errors in Memtest86+ when testing 2 memory modules, one by one everything is fine. Who is guilty?
CPU: i7-8700k
MB: Asrock Z370 Pro4 (bios 1.7\1.5)
RAM: CMK16GX4M2B3000C15R ver 5.39
If both modules are installed in the memtest, errors start to appear at 4527m exactly in the 5th test, and they may not start climbing immediately, say, on the 50th run. For the test, I did 1000 runs in other memory areas - there are no errors.
When testing on one module, there are no errors, I drove every 18 hours. Slots, dual\singl channel, overclocking and bios do not affect in any way. Visually, the memory, the mother and the socket are without damage, there is memory in QVL, but it is not my version (5.93).
The problem is 3 months old, but I was only able to catch it with memtest now. the first memtest in 4 hours did not give anything, and bsods were rare. In work, it manifests itself as random bsodes with a frequency of 15 days to a couple of minutes.
Maybe someone has come across something similar? I sin on my mother, but who knows.
Answer the question
In order to leave comments, you need to log in
It really does seem to be a memory.
I sat on different modules for the test, on one in a day I caught a bsod and got errors in the memtest. Now there are errors in many tests and if you install 2 modules, the errors are already in two places. The other module works without failure.
I'll sit for a week on a working module and if nothing changes I will change the memory.
If possible, change the memory to another and repeat the test. If the effects continue, then the matter is either in the motherboard or in the process.
"floating error" - a shitty thing, that is, it is, then it is not. if possible - change so that the brain can not stand it later.
as an option to play around - put the planks in the oven for a couple of hours at 150 degrees. melting temperature of solder smd components ~ 180 degrees - nothing will fall off.
the error may "burn out" and become stable.
Another option is to heat each mikruhu with a dryer until the solder melts - maybe a non-contact has appeared somewhere.
I would check the magnitude and stability of the voltage in the memory power circuit. Does it sag, or is there no stability with two bars?
Didn't find what you were looking for?
Ask your questionAsk a Question
731 491 924 answers to any question