Skip to main content

Hey gang... a couple questions for our more computer-savvy members...( this isn't me... LOL, I mean, I know basic things, but not so much internal hardware stuff).

I had 16 gig of RAM installed in my PC - two separate sticks of 8 - and everything has worked fine for over a year; but yesterday I got the W10 version of "the blue screen of death", and the cause cited was "memory management". I tried a bunch of different things, and then finally pulled one of the 8 gig strips out ( it was purchased separately and installed a few months after I got the PC (new).
Apparently, this was the issue, because my PC then fired right up, no problems, everything stable.
Of course, since I needed to prove to myself that the stick of RAM really was the problem, I put it back in, and after starting, I encountered the same problem again.
So my question(s) - can RAM just "suddenly" go bad like this without warning? I didn't notice any signs or symptoms of this happening, one minute the PC worked fine and the next, I got the dreaded blue screen.
I've since removed that RAM stick - and everything is running fine again - except now I'm down 8 gigs of memory.
Is it possible that the RAM is actually fine but the slot on the MB went bad? I haven't tried putting it back into another open slot...

Thoughts?

Topic Tags

Comments

Boswell Fri, 03/18/2016 - 07:11

DonnyThompson, post: 437185, member: 46114 wrote: So my question(s) - can RAM just "suddenly" go bad like this without warning?

Yes, it can, but there could be other explanations for the problem.

One of my colleague's PCs suddenly got bad memory messages, and it turned out that a "game improver" download had altered the mobo's CPU and RAM timing. Most of the RAM was OK with the mild overclocking, but one stick was not...

DonnyThompson Fri, 03/18/2016 - 07:29

I haven't done any new software or driver installs; nothing other than the typical security updates that W10 adds occasionally.
It would be nice to know if this RAM stick really is bad... or if there's something else going on with the PC that is making the stick act weird or appear to be bad.
I've grown used to 16 gig for DAW production, and I really don't have the money to buy a new stick.

Kapt.Krunch Sat, 03/19/2016 - 03:30

DonnyThompson, post: 437185, member: 46114 wrote: I tried a bunch of different things, and then finally pulled one of the 8 gig strips out ( it was purchased separately and installed a few months after I got the PC (new).

So my question(s) - can RAM just "suddenly" go bad like this without warning? I didn't notice any signs or symptoms of this happening, one minute the PC worked fine and the next, I got the dreaded blue screen.

Is it possible that the RAM is actually fine but the slot on the MB went bad? I haven't tried putting it back into another open slot...

RAM can suddenly just go bad. Especially, cheaper budget RAM, like Kingston KVR (Kingston Value RAM), for instance. What makes it "budget RAM" is that it started life intended to be premium RAM, but couldn't pass the stress tests for that higher-quality RAM, so they lowered the specs to just where it would pass, and sold it as "value RAM", instead of trashing it. It may OK to use for really light-duty...but I sill avoid it. I definitely would not use it for a music or graphics computer, or anything where reliabilty and stability was essential.

Also, yes...the slot could also have just gone bad. China/Taiwan crap. Can't really expect quality from people who are trying to bury us, can we? Sell us crap, it goes bad, we buy more crap. Fund them...defund us.

Anyway, ideally, the RAM should be matching. Same brand, size, speed...everything. Is it matching in all aspects? Mismatched RAM may work fine, but it's more likely to cause problems than matched RAM.

First off, though...many (most) motherboards are very particular about the order of how their RAM slots are populated. If your board only has two, then it's pretty obvious that a single one would go into Slot 1, and add the other to Slot 2.

It can be more complicated with 4 or more slots, though. It may specify that they be added slot-by-slot, as you go (Slot 1, Slot 2, etc.), or that Slot 1 must come first, and Slot 3 gets the second one, and then it may specify if adding more that you must add an entire other pair to slots 2 and 4, or, if it can run on three modules, it could specify that Slot 2 then gets a third one, and Slot 4 gets the fourth one added, eventually. The RAM population chart should be in the MOBO manual.

Anyway, if that's all OK, and if they match, then, yes....run each by itself in the known-working slot. If each works alone, then it MAY not be the RAM.

Then, if the suspected bad one seems to work in the known-working slot, then pop the suspected good one in the second slot (adhering to what the MOBO manual says needs to be the 'second slot'), and see if the problem reappears.

If it does reappear, it may be the slot. If it does, and you really want to further confirm it, then put the suspected good one back in the known-working slot, and test it. If it's good, then pop the susopected bad one into the second slot, and see it the problem reappears. If it does, then it's looking more like the slot, than the RAM.

Of course, you may even find that it works with both RAM modules, each in a slot, but if you only reverse the modules, then it suddenly doesn't. That's when it can remain confusing.

Check anything in the BIOS that may have anything to do with it. if you have installed mismatched RAM, then there MAY be some settings in BIOS that was set up properly for the original RAM, but installing a mismatched module may have whacked things out. It may just not like it. The BIOS will often tell you the amount of RAM installed, and could tell you how much is in each slot. (Depends on the BIOS features of that board).

It's impossible (for us) to know what settings any particular board BIOS may have available. Get out the manual (or download it), and/or fiddle around in there to see.

I don't know if any stores in your area have anything to electronically test the RAM, but it may be worth a look. They do make "RAM Testers". We use them at work.

There are also software tests that can give more clues, like MemTest. It's a bootable program that you put on a boot disk (or USB drive, etc.), and it will run a series of tests on whatever RAM you have installed.

You must be aware that it is NOT 100% accurate and reliable, but it can often pop up quickly to say "yeah...this one DEFINITELY has some kind of a problem".

If using something like MemTest, run each alone in Slot 1, and if each passes, then run both. If it then fails one, it could STILL be either, one of the RAM modules, or the second slot.

So, to wrap it up....

1) Use good-quality matching RAM.

2) Populate the RAM slots according to the specs.

3) Check BIOS.

4) Perform swapping tests.

Hope that helps.

Kapt.Krunch

OBrien Sat, 03/19/2016 - 07:53

"" but yesterday I got the W10 version of "the blue screen of death", and the cause cited was "memory management". "

Me personally, I look backwards before I look forward. When a memory issue arises out of no where, with no indication that there is an issue, I have to look into what was it I was doing when this happened. I have not seen a blue screen in years, that said, I have to take into account how old is the computer? It also begs the question of how long has windows 10 been on this machine? And then, if I had no issues with the former iteration...is it an option to go back to it?

Memory management may or may not be a hardware issue, right? These guys have addressed all the things that need to be considered in the likely event it is in fact a hardware issue.

Just something else to consider, as always.