• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

PNY Anarchy?

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

knoober

Member
Joined
Mar 18, 2015
I bought a set of these sticks (2x8gb) last summer when memory was cheap and was running them in my NAS. I never had a problem with them in that set up. I recently switched around a bunch of hardware -making my sig out of date a bit- and have been getting errors that I think are due to the memory. here is a quick rundown of what I have been seeing:

1) Gaming crashes. The error message from the game stated to check the RAM. I ran through memtest86+ and couldnt make 2 passes. When that happened I bumped the voltage and manually set the timings to the XMP value listed iin CPUZ mem tab and ran 12 hours with no errors. I kinda thought the problem was solved here.

2) Getting random blue screens, 1 or 2 a day, from the kernel (ntoskrnl.exe) and alot of the reading about that error says to question the memory.

Before I go and run through hours upon hours of memtest86+ again (which I will likely do anyway, but I will put that off for as long as possible) is there anything else that I should check? Are there any red flags about this ram that I havent already run across? Bad chips or something? I guess my biggest fear is that these issues are somehow related to the Kingston/AMD "not playing nice"-ness that I have read about. Most everything I could find about it was customer reviews on amazon and newegg stating that the heatspreader was improperly placed and made for poor insertion into the mobo, but I havent seen any issues like that with my sticks. Either I got a different revision where the problem had already been solved, or I just got plain lucky and havent had that issue.

Here is a link to the exact kit. TIA
 
Bump-sy? :)

8.5 hours into memtest86+ now. That's almost 3 passes for 16gb.
 
I got nothin knoober. Typically the issues arose with higher speed ram not the 1600 kits. You could try raisn the CPU_NB voltage a bit if the problem persists. Could also be a driver proble considering it's gaming.
 
They could have called it alot worse for the prices it was selling at :) I went back again and again for 16gb kits in the $60 range. 32gb 1866 for the NAS and 16gb 1600 for the daily driver. The anarchy has better timings, but thats to be expected because of 1866/1600 speeds.

Just to keep current with the troubleshooting and problems: 9 hours memtest was all I could take. If there is anyone who truly believes there is something to be revealed with a longer run, I will set time aside in the future. I am calling it good with the 2 runs I have done (1x12hr run 1x9hr no errors).

I have also reverted to a previously stable OC and the symptoms remain. Blue screen has happened 2x now while web browsing and nothing else. Reimaged OS and I have only had one BSOD since (but am still searching for the issue). If I put the CPU-NB one notch higher I get a noticeable decrease in performance so I think Ive got it dialed in, but I still welcome any advice. Thanks again guys :)
 
It certainly seems like the memory is good after that thorough a memtest run(s). Is it possible the problem is with the IMC on the chip?
 
It certainly seems like the memory is good after that thorough a memtest run(s). Is it possible the problem is with the IMC on the chip?
Keep in mind that 12hrs is only 3.x passes with 16gb. Most reading I have done says "as long as you can handle, but 7 is standard/kinda minimum" . Im just saying I hate memtest and how it steals my machine for all that time, necessary evil it may be :)

Unless the IMC suddenly failed (and it could), I would err to the side of not. I had this chip running 32gb 1866 (only ran @ 1600) 3.8Ghz with realistically midrange voltages (im not burning up the chip or ram with high voltages anywhere). The IMC seemed strong enough until I switched the 32gb to NAS and threw the 16gb into daily driver. :shrug:
 
The IMC may not have "failed", but maybe it stumbled with the new load? I know memtest likest to use up everything until the cows come home, but realistically, 3 full passes should be about 12 times the stress typical day to day operation puts on it. Will the mobo handle ECC memory?
 
Web browsing could be a problem with the browser/java/ HW acceleration or a GPU driver.
 
Back