Announcement

Collapse
No announcement yet.

Test results

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Test results

    Hi, I am trying to run diagnostic on my hardware that has been misbehaving recently and I decided to come here for help.

    I am running i7-12700k with 16 gigs of 6000 mhz ddr5 ram from kingston.
    For a month now I have been having random crashes in certain games with blue screens citing memory issue or kernel trap. Certain game crashes cited memory couldn't be written to a specific field.
    Today I ran a memtest86 on my ram in the following order: Stick A and B were ran first resulting in 46 different errors and took 38 minutes to do, then I ran stick B and A (swapped positions) it took 42 minutes to do and only 16 errors occurred. Following that I ran only B stick where 0 errors occurred and it took 20 minutes to finish, and then I ran only A stick where unexpectedly 0 errors occurred but the memtest time ended up being 50 minutes long.
    Is it safe to assume that the issue is in fact in the stick A due to how long it took to run the test, or is there an issue somewhere else, for example my cpu or motherboard, or could the speed of stick A impact the way it functions when paired with the stick B that causes the errors?

    Apologies for suck a rumbling post, I have no idea where to ask for any help and the PC being new gen I don't have other components anywhere to swap and test with.

    Thank you all for any information you could provide!

  • #2
    Sometimes RAM only throws an error in dual channel mode (with two sticks installed).

    I don't think the test duration is an indication of RAM stability. But there shouldn't be such a time difference (20min to 50min) when testing 1 stick at a time, of the same capacity. Maybe the test settings were different, or maybe the BIOS clocked them at different speeds. Maybe the internal ECC function in DDR5 was hard at work slowing things down. The execution time of the row hammer test can also be variable if some problems were detected in the 1st pass (it does additional deeper check sometimes). Further testing / investigation would be required to get to the bottom of this.

    Sometimes reseating a RAM stick cleans up the gold contacts. And then the better electrical contact fixes the problem.

    See also this page
    https://www.memtest86.com/troubleshooting.htm

    Comment

    Working...
    X