Memtest errors and ram instability

[rotwolf38]

Hello guys, so because i´ve got a blue screen with the MEMORY_MANAGEMENT error, ingames stuttering and System CPU usage spikes, i´ve tought to better run a MEMTEST, so i´ve run my test for 10+ hours over the night, and it has found 1 error here is a screenshot:

The ram is running at 2400mhz, i tired to use XMP and it is working fine i can overclock it at 3200mhz pretty easy, but i´ve brought it back to normal, maybe i should with a threadripper 1920x.
The ram is even on my motherboard QVL site. When i run a cinebench r15 it works just fine.

Specs:

CPU : Amd ryze threadripper 1920x
MB: x399 ASRock Taichi (Changed with a faulty x399 msi gaming carbon)
RAM: HX432C16PB3K4/32
GPU: Nvidia gtx 980 ti

 

[bignastyid]

That screen shot shows 33 errors 32 on test 6 and 1 on test 9. You either have bad ram or a bad memory controller, try testing the ram 1 stick at a time and see if you can narrow it down to a failing stick.

 

[rotwolf38]

That screen shot shows 33 errors 32 on test 6 and 1 on test 9. You either have bad ram or a bad memory controller, try testing the ram 1 stick at a time and see if you can narrow it down to a failing stick.

so i´ve tested the ram 1 stick at a time and i´ve found that only one stick is causing errors here are 2 test that i did on the same stick :

 

[bignastyid]

If the ram is under warranty rma it. Since it is part of a kit they will replace the whole kit.

 

[rotwolf38]

If the ram is under warranty rma it. Since it is part of a kit they will replace the whole kit.

Ok thank you, sadly this is the second time that i will replace the ram the first kit had the same issues.

 

[bignastyid]

Wow that's some bad luck. It almost certainly the ram. If it was the memory controller it would have presented itself on all the dimms not just 1. Just to be sure were all the dimms tested in the same slot?

 

[rotwolf38]

Wow that's some bad luck. It almost certainly the ram. If it was the memory controller it would have presented itself on all the dimms not just 1. Just to be sure were all the dimms tested in the same slot?

Yes since i´ve bought the Threadripper i´ve only run into problems with ram.. maybe i am going the CL14 from gskill they say that kit is the best for threadripper.

No, the dimms(you mean the ram sticks right?) were tested on their slots, now because i still have the old same kit i´m trying to run a test on the same motherboard slot with an old stick and then i´m gonna try to mix the ram.

 

[bignastyid]

Just to make sure its not a bad slot test the stick that was showing errors in a slot where another dimm passed.

 

[rotwolf38]

Just to make sure its not a bad slot test the stick that was showing errors in a slot where another dimm passed.

So i ve tried to put the "bad ram" on another slot and it showed no error at all. When i ve putted the bad ram on the "bad slot" the error reappeared, i even put another ram on the "bad slot" and an error came, so now the motherboard slot is the problem i guess.

 

[bignastyid]

Could be a bad slot, bent pin in the socket or even possibly the memory controller on the CPU.

 

[rotwolf38]

Could be a bad slot, bent pin in the socket or even possibly the memory controller on the CPU.

If it is the memory controller on the cpu i should get the CPU replaced right? so sad..

Should i first try to put the memory on other slots? And see what happens? Do you have any suggestion for making sure that the memory controller is the problem?

 

[bignastyid]

Could be a bad slot, bent pin in the socket or even possibly the memory controller on the CPU.

If it is the memory controller on the cpu i should get the CPU replaced right? so sad..

Should i first try to put the memory on other slots? And see what happens? Do you have any suggestion for making sure that the memory controller is the problem?

Assuming it not a bent or broken pin in the socket the only way to know if it is the motherboard or CPU that is is to have another motherboard and/or cpu to test with.

 

[rotwolf38]

Assuming it not a bent or broken pin in the socket the only way to know if it is the motherboard or CPU that is is to have another motherboard and/or cpu to test with.

You know i have another MB (msi x399 carbon ), but it gave me even more (couldn´t start the pc even with 2x8 gb) the pc was going on and off in a boot loop. So i even replaced the motherboard with a new solid ASrock taichi and the ram is working with 32 GB but it is giving me this problems. Maybe it really is the CPU

 

[bignastyid]

Very possible its a bad cpu. I'd contact AMD and see what they say.