Question Afraid that i'll overheat my RAM during MemTestx86

miknepa

Reputable
May 20, 2019
33
1
4,535
So,

I have had a bluescreen which i already had 1 Month ago with the stopcode: WHEA_UNCORRECTABLE_ERROR
This is 99 Percent an Hardware issue so i've decided to stress the hell out of my components to detect which one is causing the issue. I ran chkdsk already, planning to run Prime95 and FurMark for CPU and GPU but i wanted to start with MemTest because i think my RAM is the issue.
I'm running an i9-9980XE with 128GB Corsair Vengeance DDR4 RGB @ 3200Mhz, GPU is a RTX 2080Ti running 11GB VRAM.

Let's get to the point: I am scared that running MemTestx86 for 4 Cycles on my RAM will fry it. Research i've made so far tells me that RAM does not throttle or shut down the system when its overheating as it's a "Critical component"

If someone has some more advice or info on this please share.
 

iMatty

Reputable
Mar 14, 2019
1,211
141
5,390
Well that is the only option you have, you could do them one by one but use like 8 or something passes and should be more than enough.
Or you could wait for @Colif to see this, he and gardmen knows a lot about BSOD's.
Just give it a bit and he'll probably see this and tell you what to do.
 

Colif

Win 11 Master
Moderator
running memtest on multiple ram sticks at a time is a waste of time, as if you do get an error, you can't tell which stick had that error. So yes, 4 passes (since thats what memtest free now offers) on each stick is only way to know it isn't ram)

Can you follow option one on the following link - here - and then do this step below: Small memory dumps - Have Windows Create a Small Memory Dump (Minidump) on BSOD

that creates a file in c windows/minidump after the next BSOD
copy that file to documents
upload the copy from documents to a file sharing web site, and share the link here and I will get someone to convert file into a format I can read

just check dumps and make sure we can't see any candidate.
remove any overclocks, remove any overclocking software as well.
Can be heat as well, so make sure evertything clean
WHEA errors can be any hardware which always makes them fun.
 
  • Like
Reactions: iMatty

TJ Hooker

Titan
Ambassador
Testing them one at a time would see if any individual stick is faulty, but even if every stick passes they won't necessarily pass when you put them all together.

Did you buy all 8 sticks as a single kit?

DDR4 doesn't really generate a lot of heat. As long as you're running it at a normal voltage (e.g. 1.35V), there's nothing to worry about.
 

miknepa

Reputable
May 20, 2019
33
1
4,535
@TJ Hooker Yes they came in one kit
@Colif I wanted to run them all at a time to see if they make any errors, if they do then maybe invest some more time in the ram

Edit: Yes i had overclocks and removed them quite a while ago, thermals are pretty solid for my hardware (GPU: 50-55 idle, CPU 30-45 idle) Windows didnt make a dump because i hard shut down the system during the bluescreen, it was running at 0 percent for 30 minutes
 
Last edited:

TJ Hooker

Titan
Ambassador
I really think it makes more sense to test them altogether first. If they pass that way, you know they're good. If not, you can start testing one stick at a time.

Doing it that way, best case scenario you only have to run the test once (worst case is 9 times). If you start with running one stick at a time you will automatically have to run the test 9 times, even if your RAM ends up being totally fine.
 

miknepa

Reputable
May 20, 2019
33
1
4,535
@TJ Hooker That is exactly what i had planned. Im doing this with a friend on discord willing to help me out and he said exactly what colif said. And my dumps are indeed activated but how i mentioned earlier:
"Windows didnt make a dump because i hard shut down the system during the bluescreen, it was running at 0 percent for 30 minutes"
thing is, scared that i'll fry my ram, but i will check my voltage on the ram and run it overnight
 

miknepa

Reputable
May 20, 2019
33
1
4,535
@Vic 40

The GPU is not completely "idle" im running youtube, steam and discord.
PSU is a Thermaltake Toughpower iRGB 1250W
oh and the gpu is a 2080ti and i think the 2080ti has some higher temps
oh and (second edit) its an gigabyte aorus 2080ti extreme edition 11gb
 
Last edited:
Test all your ram together first.
You should be able to complete a full pass with NO errors.
If you get a failure, then is the time to drill down to find the bad stick or sticks.
Or, possibly to identify a bad motherboard ram slot.

Do not worry about overheating your ram unless you are overclocking well past advertised speeds.
If you have any doubts about sufficient airflow over the ram, take the case covers off and direct a house fan at the innards.
Ram heat spreaders are largely marketing and only useful to those who are trying for record overclocks at insane voltages.
If you are really paranoid, reset the ram to default speeds for the first test.
 
  • Like
Reactions: TJ Hooker