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

G-Skill Ram Unstable

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

kmo_9000

Member
Joined
Apr 10, 2012
Location
Ontario, Canada
Hey, so I've been having constant system freezes and recently BSOD's since I built my comp in January. Everything that has firmware has been updated and I've spent a lot of time troubleshooting things.

I think I've narrowed the problem down to memory issues. I made a thread here with g-skill that has more info about the problem.

BSOD's show errors 0xA1 and other codes that are commonly associated with memory errors.

The thing is I've set the ram to 1333mhz and it seems stable. Arkham Asylum has been running all morning, where last night it would crash every 30 minutes or so and then the system would freeze (with 1600mhz).

I just want to get a second opinion about this. It seems like this kit might actually be a 1333mhz kit with XMP 1600mhz slapped on. A number of newegg reviews of this kit show the exact same instability issues. In addition, the heat spreaders on one of the sticks just came off on its own when I was putting them back in yesterday.

Considering I have to pay for shipping, I'm having second thoughts about RMA'ing it. I'm concerned that they might deny my warranty if they find the heat spreader was taken off (even though it fell off on its own). Also, I think they may be overheating at 1600mhz because they get very hot to the touch, and this might be why memtest doesn't always show errors.

All things considered, should I RMA the ram or just bite it and pick up a Corsair kit instead? Would really appreciate a second opinion on this.
 
Last edited:
So at DDR3-1600 w/ the timings manually adjusted to 9-9-9-24 (all sub-timings on Auto) and DRAM voltage at 1.50V, does Memtest86+ pass or fail? This w/ all other voltages on Auto, and the CPU not overclocked. And I assume you've already tried enabling the DDR3-1600 XMP profile in the BIOS, and then run Memtest86+ to check the RAM for stability? And the 0x000000A1 error code / value relates to a PCI_BUS_DRIVER_INTERNAL bug check.
 
I ran memtest86 on each stick three times this week. The first time one of the sticks got errors, second and third time they all passed.

All primary cas timings were set manually (9-9-9-24), voltage to 1.50, cpu is not overclocked and speedstep and turbo turned off.

Also I mistyped, the error was 1A and not A1, it said it was related to MEMORY_MANAGEMENT.
 
A 0x1A MEMORY_MANAGEMENT STOP error doesn't necessarily mean there's a problem w/ the RAM. What were the results of enabling the XMP profile? And without having to read through the thread you linked to, or the other three links in that thread, what voltages have you adjusted and what were the Memtest86+ results after each change; more or less errors in the same length of run-time? Have you tried disconnecting all of the HDD's, and / or any cards / peripherals not absolutely needed by the OS? Basically list everything you've tried in order to resolve the BSOD's / system freezes. You might also want to post a BIOS template of all your current non-overclocked settings, including voltages, strap, and DRAM settings at 3.6GHz.
 
These are the current settings.

o1OPD.jpg
9AQWh.jpg
IdDjK.jpg

At first I thought it was a problem with my Crucial M4. I updated the firmware, tried 3 different sata cables. The motherboard BIOS is updated to the latest. Tried XMP on, and tried standard with XMP settings set manually. I've upped the VCC to as much as 1.185, and the DRAM voltage to 1.565. Non of this seems to do anything.

I've also ruled out my Xonar Essance ST by using custom drivers which did nothing. I've ruled out crossfire problems as it freezes regardless, and the 6950's were running fine in my old system.

This guy in this thread here seems to be having the exact same problem though.

I also don't think it's related to the HDD's, I've reinstalled windows at least 5 times since January. Once on my 1tb mechanical drive and it still had these problems. But what tells me there's something wrong with the ram is that it is not having any problems right now set at 1333mhz.

EDIT: I should also mention that when one of the sticks got errors, I put the three that passed in tripple channel. It crashed shortly after. I then tested them all again and they all passed.
 
Last edited:
Either Load Optimized Defaults or reset CMOS, then besides changing only those settings required for a successful POST and eventual boot into the OS, either enable the XMP DDR3-1600 profile or manually adjust the DRAM frequency, timings (only the first four), and voltage. Then boot to Memtest86+ installed to a thumb drive and post the results.
 
I went through something similar with gskill and asus. It actually ended up being more than one thing. Gskills website showed my ram being compatible with my mobo but 2 sets in a row couldn't run at xmp settings. Getting a set of mushkin ram fixed that. I set it to xmp profile 1 and so far so good. Also, my asus essence stx seems to crash max payne 3 and possibly arkham city but it was hard to say with arkham city because of having 2 issues at once. All I know is when I disabled the stx and switched to onboard audio, no more crashes since. Haven't seen a new asus stx driver for over a year and I'm wondering if it's going to happen. A program called goldmemory showed memory errors when memtest ran all night without issues.
 
Either Load Optimized Defaults or reset CMOS, then besides changing only those settings required for a successful POST and eventual boot into the OS, either enable the XMP DDR3-1600 profile or manually adjust the DRAM frequency, timings (only the first four), and voltage. Then boot to Memtest86+ installed to a thumb drive and post the results.

I have already done this. The tRFC is set to 166 because the default is 128, which according to someone from the G-Skill forum is too low. However, I did run memtest at 128 and 166 without any errors. I did get memory errors once though. Thanks for the help though, I really don't know what to think at this point.

I've never had problems booting into the OS, only constant application crashes and random freezing. It doesn't seem to give me BSOD's anymore (at 1600mhz manual settings).

But does it make sense that bad memory would be able to run at 1333mhz no problem while 1600mhz causes crashes?

I went through something similar with gskill and asus. It actually ended up being more than one thing. Gskills website showed my ram being compatible with my mobo but 2 sets in a row couldn't run at xmp settings. Getting a set of mushkin ram fixed that. I set it to xmp profile 1 and so far so good. Also, my asus essence stx seems to crash max payne 3 and possibly arkham city but it was hard to say with arkham city because of having 2 issues at once. All I know is when I disabled the stx and switched to onboard audio, no more crashes since. Haven't seen a new asus stx driver for over a year and I'm wondering if it's going to happen. A program called goldmemory showed memory errors when memtest ran all night without issues.

Try this site out for a driver for your Xonar: http://brainbit.wordpress.com/category/uni-xonar/

I'm not sure if my Xonar ever caused crashes, but that might fix some of the problems you are having.

I might also try goldmemory, but you are the second person I've read about having fixed your problems by ditching your g-skill memory. I mean I opened my computer last night and the heat spreader had just fallen off, it really looks like this memory is crap.
 
Just saying, you can try switching to onboard sound pretty easily by disabling the xonar in device manager and switching a cable. Thanks for the link. I'll try it.
 
Just saying, you can try switching to onboard sound pretty easily by disabling the xonar in device manager and switching a cable. Thanks for the link. I'll try it.

I tried running the system with the Sound Card completely removed and still got the freezes. I couldn't stand not being able to use my Xonar Essance though, so I found those drivers. But since the problem may be something else I'm not sure if the xonar was ever causing problems at all.
 
So I RMA'ed the G-Skill ram, sent it out Monday so I don't know how long it will take to get there from Canada. It was airmail so hopefully not too long.

I also borrowed a old 4gb set of ram from a friend that I am running my comp on now. It crashed once, but it was the Xonar Essance GX freeze issue that apparently was mixed in with the other freezes. It would be different in that the screen would stop, but I could still press the power button and the OS would shut down. The latest Asus drivers fixes this issue related to the GX setting being on.

All other BSOD's and freezes appear to be gone now, so I decided to pick up 16gb of that Samsung 30nm ram. When I get the RMA'ed ram from g-skill I'll just keep it as backup ram (after testing it to make sure it works of course).
 
So I RMA'ed the G-Skill ram, sent it out Monday so I don't know how long it will take to get there from Canada. It was airmail so hopefully not too long.

I also borrowed a old 4gb set of ram from a friend that I am running my comp on now. It crashed once, but it was the Xonar Essance GX freeze issue that apparently was mixed in with the other freezes. It would be different in that the screen would stop, but I could still press the power button and the OS would shut down. The latest Asus drivers fixes this issue related to the GX setting being on.

All other BSOD's and freezes appear to be gone now, so I decided to pick up 16gb of that Samsung 30nm ram. When I get the RMA'ed ram from g-skill I'll just keep it as backup ram (after testing it to make sure it works of course).
Which xonar drivers did you switch to? The official ones from last year or the newest unified drivers? I only ask because the official ones gave me issues but the unified ones may have fixed it. It's also possible that the latest patch for max payne 3 took care of it. Always fun diagnosing problems when you have multiple companies screwing you over simultaneously with poor quality control.
 
The official ones from Asus. They fixed the GX issue in version 1794.

Strange that you would have errors with the unified drivers but not the official ones. I assumed the unified drivers were based on the newest ones from asus and improved upon.
 
Strange that you would have errors with the unified drivers but not the official ones. I assumed the unified drivers were based on the newest ones from asus and improved upon.

I may have enabled GX when I was installing it. I remember having trouble downloading for some reason and I may have installed an older version. I don't remember, but it's working now with the latest drivers.
 
Glad you got it sorted out. Looking forward to your results with the samsung memory. I thought about getting some just for fun and sticking on some low profile ram sinks like they use with video card heat sink kits. Might be nice.
 
Well at this point I just want working ram, having a computer that constantly freezes for 6 months can turn you off from overclocking.

I will probably run them at 1600mhz and try and get the timings down as far as I can, then enjoy my working computer.
 
Don't give up yet. The muskin ram I replaced the gskill with is working beautifully at 10-11-10-28 @ 2200mhz 1.56v and gets buttloads of bandwidth. I think it's just motherboard compatibility.
 
Don't give up yet. The muskin ram I replaced the gskill with is working beautifully at 10-11-10-28 @ 2200mhz 1.56v and gets buttloads of bandwidth. I think it's just motherboard compatibility.

I don't know, even with the bios updated it is completely unstable. On the current 4gb kit I am running on it doesn't freeze at all. Even increasing the VCC and vram voltage on the old kit didn't do anything.

I think G-Skill used some crappy memory chips with whatever batch I got. I found multiple threads in various forums with people having the same issues with the same kit.

I am expecting a lot from this Samsung memory though.
 
Back