Windows 7 BSOD issues

[LiquidSnak]

Follow Hopalong's advice, and you will have done it the proper way. As he said, raising voltage can have serious ramifications.

Before messing with ANY of the voltages, please test each stick, one at a time, to rule out the possibility of bad RAM. Be sure to test all sticks, even if you find a bad one.

 

[Len19]

Yes, i am in the middle of running memtest on each individual stick.

 

[CarlTR6]

Thanks. That eliminates that as a problem. Follow Mike's advice about setting the voltage. But first I would make sure the RAM is good and the slots are good; then I would work with the voltages.

 

[Len19]

Anyways i think it definitely is one of the sticks that has gone bad, memtest86 on one stick, 30 passes 0 errors.

 

[CarlTR6]

Very good. Now test the other stick. Then test each slot with a known good stick.

 

[Dave76]

Hi Len19, Welcome to Seven Forums.

Be sure to test all the RAM cards individually as CarlTR6 has mentioned.

RAM errors can be from bad RAM or incorrect settings.
Sometimes changing the timings and/or voltages will get the RAM running error free.
This information can be confirmed on any RAM forum, Corsair has a good one.
Corsair forum:DDR3 Configuration and Tuning

It's generally not a good idea or suggested to increase the RAM voltage too fast, small increments are the safest way to proceed, usually 0.05v increases are recommended.
If your RAM is at 1.5v now, try raising it to 1.55v and then run the memtest+86.
Sometimes when you are running multiple RAM cards you will need a NB, North Bridge, voltage bump.
Check in your BIOS and tell us what the NB voltage is.
The NB voltages must be done very carefully, it will require a small increase and must be kept within 0.5v of the RAM voltage or you can permanently damage the CPU.
Please get confirmation from someone here before increasing the NB voltage.

The best way to get some important information on your RAM is to download CPUZ.

When you run memtest86+ you are looking for bad RAM cards, bad motherboard slots, so you need to follow the testing procedure in the tutorial.
Sometimes errors from these tests can be corrected by adjusting the settings.
RAM - Test with Memtest86+

Can you download CPUZ and post a snip of the CPU, Mainboard, Memory, and SPD tabs.

For posting the information , enter 'Snipping tool' in the Start button > Search box, hit enter. Select the area you want to post and save to a convenient place.
How to Post a Screenshot in Seven Forums


Have you been checking your temperatures?

Use Speed Fan or Real Temp , run one of these and check the CPU temps when at idle, under load and/or testing.
Let us know the minimum and maximum temps you get.

 

[Len19]

On a note oddly enough i've only crashed once while gaming, nearly every crash is while idle, using firefox, or any random windows program. For instance i just got a bsod using the snipping tool, NTFS error.

Also including idle temps with speedfan

 

[Dave76]

Good info.

Temps look good, check them next time you get the system under load, post the load temps.

The RAM timings SPD info is a little strange.
Is this RAM listed as compatible for your motherboard?

The JEDEC, JEDEC is the standard that RAM companies use to list recommended and tested frquencies and timings, are a bit low.
I don't usually see a frequency of 444 MHz listed for 1000 MHz RAM.

Have you done any overclocking? CPU or RAM?

I'll look up info on your mobo and RAM, then post back.

Are all the RAM settings in BIOS on Auto?
Have you changed any of them, if yes tell us which settings.

 

[Len19]

I did make a attempt to overclock the cpu by increasing the core multiplier until the bsod's came. The Ram settings are all on auto at the moment, after the first few bsod's i reset my bios settings to default so everything has been at stock clocks. The ram is clocked at 1066mhz at the moment.
I'm sure the mobo/ram should be compatible, i have looked it up on the Corsair webpage.

I'm not sure if it's related but i have 2 hard drives in raid 0, i don't think it would cause this kind of problem but might as well get it out.


memtest on my 2nd stick of ram, 10 passes 0 errors.

Going to try the 3rd and last stick, all testing in the same dimm slots when testing all 3 with memtest86

 

[Dave76]

If the last stick passes, it means the RAM settings need adjusted.

Let's get it running stable at stock settings, then you will have a base line to go back to when OC'ing.

Are the BSODs always happening with a cold boot?
We can try adjusting some settings but, cold boot issues are usually caused by RAM. We might have to come back to this.

I'll take a look at your mobo manual and check the RAM compatibility.

Do you have the RAM in slots DDR3_1, DDR3_3 and DDR3_5?
From the CPU, the 2nd, 4th and 6th slots.

 

[Len19]

So far the 3rd stick has no errors yet, on the 3rd pass, ram sticks are on the 2,4,6 slots i think, the cpu heatsink SCMG-2100 blocks the first dimm.

 

[LiquidSnak]

That is a nice looking heatsink

We are making good progress, keep us updated with the test results.

 

[Dave76]

There is a new BIOS FC for your mobo but, the change log says it's for SATA3 changes. This shouldn't affect your issues.
Might come back to this.

There are a few new drivers listed, Gigabyte GA-X58A-UD3R (rev. 2.0), Check if any are newer than what you have, install the Intel SATA RAID Drive 9.6.0.1014 if you don't have it already.

 

[Len19]

so far 5th pass and no errors for a single module, since it's getting late i'll probably stop at the 6th pass, all ram passes individually but has one error when using all 3, i will read into that a bit.

 

[Dave76]

This is pointing to a need for a NB voltage bump, your manual says it is at 1.15v, let's bump it to 1.2v.

I'll check your manual for the BIOS location of the NB voltage, I'll post back in a few minutes or you can check back tomorrow.

Here is the link to the Gigabyte compatible RAM list for your mobo,
Qualified Vendors List (QVL)

Your GA-X58A-UD3R rev 2.0 manual


Page 40 of your manual, QPI/Vtt voltage on auto is 1.150v.
change it to 1.20v.

In your BIOS > MB Intelligent Tweaker(M.I.T.) scroll down near the bottom you will see QPI/Vtt voltage.

Run memtest86+ with all three cards in slots 2,4 and 6 as mentioned above.
If you get an error you can stop the test.

Let us know the results.

 

[Len19]

According to my bios the qti/vtt voltage is already set to 1.200 on automatic settings.

On a note i noticed ram timings were actually set on quick, and not auto, changing that to auto and maybe it'll fix it, or not testing all 3 sticks again, seconds after i said something about auto fixing it i get a error in pass 1.

Anyways i have finally gotten rid of my old wlan usb card, cleaned out its drivers, and installed a new wlan card, that should be compatible with win 7. I've also changed the QPI/Vtt voltage to 1.20 manual, and the Dram voltage to 1.6

There doesn't seem to be any single thing that always causes a bsod since it's usually random so i can't really test stability outside memtest.

Couple random bsods again, ntfs.sys, fltmgr.sys, one while i was on firefox, the other while copying files from one hdd to ssd.

 

[Len19]

It's going to take a while to figure out what the right voltages are, tweak to 1.600, 1.200 seems to increase the amount of bsod's, have gotten 2, as well as a few while the system is trying to start

 

[Dave76]

If you're getting more BSODs, set the RAM voltage to 1.55v, QPI/Vtt leave at 1.20v, run memtest86+ again.


The quick setting for RAM allows all the RAM cards to be adjusted at the same time.
Setting to Auto should re-set them to one of the JEDEC settings, which should help, post the CPUZ 'Memory' tab again.

Try bumping the QPI/Vtt voltage, the smallest increment should be 0.01v, slowly increase the voltage by this amount, don't exceed 1.30v.
After each bump run memtest86+, if you get an error stop the test and bump again.
You need to find the lowest voltage that will run stable.

Let us know the current RAM timings.

 

[Len19]

the smallest voltage increment in my mobo is 0.02v, i'll keep trying that until it gets stable, hopefully. On another note is it possible that this issue might even be caused by a hard drive problem? very often during the bsod, the computer will make a couple loud clicks, which i've assumed was the dmp file getting written, but i've read that it might be one of the hdd's is failing.

On any note the bios shows timing at 8-8-8-20 at the moment

 

[Dave76]

Yes, that is a possible cause of the BSODs.

Go to the Seagate website and D/L Seatools, run the HDD diag.

Also check for a firmware update for your HDD.