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

RAM stick that seems to result in problems in usage passes MemTest86 whole night

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

creativus

Registered
Joined
Aug 12, 2017
I have 4 RAM sticks and one of them seem to result with problems in usage such as programs crashing or bsods but last night I tested that RAM stick alone with MemTest86, it was about 11 hours, it passed the tests for a double digit number that I don't exactly remember.

Also, when I tried to seat that RAM stick with another RAM stick on the 2nd and 4th seats it resulted with recurring beeps when I opened the PC once, it also happened when I seated all 4 of my RAM sticks once.

I'm trying to figure out what's wrong. More than a year ago I went to a computer service for a problem and he complained about the quality of electricity in the city, said there was nothing I could do about it, found fault with one of my RAM sticks, and approved my idea of getting a new RAM stick. This time when I went to him and explained the situation he said that I needed to seat the RAM sticks one by one and use the computer to see which had a fault, but I'm confused. I'm using it connected to a surge protector power outlet connected to a rather cheap UPS.

Can you help me clarify what's going on ?
 
What were the beeps exactly?
https://www.pcmag.com/encyclopedia/term/beep-codes

What slot did you put the single stick in for testing? Are you sure it isn't the slot going bad? Rotate known good sticks through the suspected bad slot. If it's a bad slot, multiple stocks should be bad.

Also how do you know you have a memory problem now? What is the bsod code? What does event viewer say happened? When the computer bsod or whatever?
 
What were the beeps exactly?
https://www.pcmag.com/encyclopedia/term/beep-codes

What slot did you put the single stick in for testing? Are you sure it isn't the slot going bad? Rotate known good sticks through the suspected bad slot. If it's a bad slot, multiple stocks should be bad.

Also how do you know you have a memory problem now? What is the bsod code? What does event viewer say happened? When the computer bsod or whatever?

I think it's continuous beeps. I tested using that RAM stick with that other RAM stick on 1st and 3rd slots and I had a problem during usage too. One of the most frequently occuring problems was Rocket League freezing and crashing and Rocket League requires 4GB of RAM, so I don't want to test with less than that during usage. I didn't notice a problem with other RAM sticks on any slot yet.

I don't know how to use event viewer for this purpose.
 
You can download the trial version of AIDA64 Extreme, run System Stability Test and check only "Stress Cache" and "Stress System Memory" in order to test a single stick / slot at a time. Are you running an XMP profile? If so the first thing you should do is disable it for testing. You can test all 4 sticks at stock profiles first to see if they work in that configuration, then you know the answer off the bat. Are all four of them matched kits? Memtest86 is useful for detecting some kinds of failures but not all of them.

You'll want to get a piece of removable tape and label each stick 1-4, then test a stick (preferably a known good one) in each slot. If each slot works, you can then test each stick one at a time. There is also the possibility of instability that only occurs when all the slots are populated d/t stress on the memory controller or the boards voltage regulation, but we gotta rule out the bad stick / bad slot possibilities first.

Also you've mentioned power, the problem you're describing could also be caused by power fluctuations, especially from an older PSU. If it only occurs while gaming, it could also be due to your GPU overclock.

Really to pin down a problem like this you'll want to methodically test everything in stock configuration one change at a time. If you don't find any problems then re-apply any overclocks one at a time and re-test.
 
No XMP, the motherboard has Performance Enhance modes turbo and extreme and it used to run on turbo but I guess it doesn't work anymore, it's on standart and auto RAM settings, memory frequency shows 667 in bios although the sticks are 800 MHz sticks. 2 sticks are a model of Kingston and 2 sticks are a model of Mushkin.

There were problems while not gaming too.

The PSU is about 18 months old, it'a cheap apfc 80 plus PSU.

How long shall I run the System Stability Test ?
 
Can you give motherboard maker/model ? What CPU do you have? Are the Kingston memory a matched set? Are the mushkin memory a matched set? <- what are their settings listed on the memory stick (IE.. DDR2-800 CL-5-5-5-15 1t @ 1.8v)
If both makers are ddr2-800 but are running @ 667 : The 2 models have different ddr2-800 timings and itÂ’s defaulting to 667 as this speed = the same timings on both models. The IMC/Memory controller can not handle the ddr2-800 speed across 4 sticks and has to down clock them to ddr2-667 for stability.
The surge protector and low-end UPSÂ’s are just there in case you house wiring decides to send 2000 volts in 1/4 sec (like lightning striking house) and they pop their built-in fuse instead of your computer frying. -> I have killed a GTX-460 video card by ESD (walking across carpet) + the ASUS Rampage 5 Extreme that the card was plugged into + the i7-5960x that supplied the PCI-e connection for the card :-(


Thank you for any additional information as this will help us - in helping you fix the problem.
 
If it happens on load, this may be the problem too..

What psu, exactly?

Rampage RMP-600-80P, there were times a bsod happened just opening a new tab on Chrome or browsing and it happened while shutting Windows down too.
 
Can you give motherboard maker/model ? What CPU do you have? Are the Kingston memory a matched set? Are the mushkin memory a matched set? <- what are their settings listed on the memory stick (IE.. DDR2-800 CL-5-5-5-15 1t @ 1.8v)
If both makers are ddr2-800 but are running @ 667 : The 2 models have different ddr2-800 timings and itÂ’s defaulting to 667 as this speed = the same timings on both models. The IMC/Memory controller can not handle the ddr2-800 speed across 4 sticks and has to down clock them to ddr2-667 for stability.
The surge protector and low-end UPSÂ’s are just there in case you house wiring decides to send 2000 volts in 1/4 sec (like lightning striking house) and they pop their built-in fuse instead of your computer frying. -> I have killed a GTX-460 video card by ESD (walking across carpet) + the ASUS Rampage 5 Extreme that the card was plugged into + the i7-5960x that supplied the PCI-e connection for the card :-(


Thank you for any additional information as this will help us - in helping you fix the problem.

The motherboard is a Gigabyte GA-P43-ES3G, the CPU is a modded Intel Xeon E5450. Both Kingston and Mushkin memories are matched sets though one of the Kingston memories is a narrower stick. In HWiNFO64 the numbers listed for Muskin are 400(Freq) are 5(CL), 5(RCD), 5(RP), 18(RAS), 23(RC), -(Ext.), 1.80(V) and for Kingston they are 400(Freq), 6(CL), 6(RCD), 6(RP), 18(RAS), 24(RC), -(Ext.), 1.80(V)

I will check the working frequency next time I use only one of the brands, right now 2 Mushkin and 1 Kingston are seated, the other not narrow Kingston seems to cause problems.
 
Back