|
Background
I have had + Show Spoiler [this computer] +
Mobo: Asus P8P67 PRO
CPU: Intel Core i7-2600K @ 4.5GHz (My friend overclocked it from 3.4 to 4.5GHz a year ago)
RAM: Patriot Gamer 2 series 8GB (2x4GB) (DDR3 in dual channel at 9-9-9-24 and 666 MHz I think)
GPU: Sapphire HD 6970 2GB GDDR5
PSU: Corsair TX850
HDD: OCZ Vertex 2 120GB (Windows and some applications and things)
HDD: WD 1.5TB
0x1A: MEMORY_MANAGEMENT (most frequent)
0xA: IRQL_NOT_LESS_OR_EQUAL
and once a message about dxgmms1.sys.
*Also while writing this post and getting BSODs a bunch, got 0xA1 with no message but http://aumha.org/a/stop.php says that's PCI_BUS_DRIVER_INTERNAL*
They occur frequently when I try and run a lot of things at once (video on one monitor, game on primary, manage other things at same time on third monitor) and they occur often but seemingly randomly when I do less intensive things. (maybe Opera 15 related, more on that later)
Analysis
After looking at a few websites and existing TL threads on BSODs, I assume the problem is being caused by either:
a. Virus/malware
b. Videocard driver
c. Hardware issue (RAM probably)
d. Software issue
a. Virus/malware
I have Microsoft Security Essentials installed and updated and I run it very often. I also just downloaded http://housecall.trendmicro.com and ran it without finding anything. I really, really doubt Viruses would be the cause.
b. Videocard driver
The dxgmms1 error lead me to think it might be a videocard driver issue. I hadn't updated my videocard drivers for months so I downloaded and installed the newest one. I still get BSODs so I doubt that's it.
*Maybe worth noting the 0xA1: PCI_BUS_DRIVER_INTERNAL bsod I got while writing this post?
c. Hardware issue (RAM probably)
This seems like the most likely issue based on what I am doing on the computer when the BSODs happen. I ran memtest86 overnight for 7.5h and here's what it gave me:
+ Show Spoiler [Image of Memtest86] +
I don't know how to read it but I assume any error is bad.
d. Software issue
I don't install very much software so the main thing I can think of is Opera 15. I never got a BSOD when I was using Opera 12. I installed Opera 15 maybe a week ago and it has page crashes a ton and generally crashes like a mofo. I thought it might have been the Adbock extensions I had installed but I tried disabling things and that didn't help.
Often applications will not respond (Skype, sometimes games, Opera), which I think was happening before Opera 15, so the Opera 15 issues might be a symptom and not the cause.
What next?
I am not very good at figuring out these computer issues so I would like some advice on what to do next to further diagnose the problem. I can probably upload some dumps from my BSODs. Should I try removing some RAM and running memtest a bunch more times? Should I try uninstalling Opera 15? Should I try wiping my videocard drivers and reinstalling them? Re-seating videocard or RAM? I am opening the case now to clean out dust and stuff.
Thanks for any help!
|
United Kingdom20326 Posts
Make sure your RAM is set to the correct timings, voltage and frequency, if you still get errors in memtest, try bumping it down a frequency multiplier. If you're getting errors in memtest, you have a problem before the OS or any drivers are loaded
|
Any errors in RAM is BAD. Look at that error count. Assuming you haven't tried OC'ing them or anything, replacing them should fix all the BSOD.
|
Well, the RAM is at the appropriate timings (9-9-9-24) and the Target DRAM is 1333MHz which is right. The voltage is set to auto and is currently at ~1.5V. I'm not sure what the voltage is supposed to be for this RAM though...Newegg description here says, "You'll appreciate the 1.65V low voltage for lower internal temperatures and easier overclocking."
So...what should I try changing before running memtest again? Then, if that still errors, buy new RAM?
Edit: Although CPU-Z says the DRAM Frequency is 667 MHz. I don't understand this stuff. >_<
Edit2: Took one stick out and going to run memtest again. On the stick it actually says 1600MHz and 1.65V...
|
On August 16 2013 06:11 Durak wrote:
Well, the RAM is at the appropriate timings (9-9-9-24) and the Target DRAM is 1333MHz which is right. The voltage is set to auto and is currently at ~1.5V. I'm not sure what the voltage is supposed to be for this RAM though...Newegg description here says, "You'll appreciate the 1.65V low voltage for lower internal temperatures and easier overclocking."
So...what should I try changing before running memtest again? Then, if that still errors, buy new RAM?
Remove the faulty RAM and run with 4GB. If the crashes are gone, it is that.
|
There's an "XMP" profile that comes with the memory sticks. If you enable that in the BIOS, it will use timings and speed and voltage as described by that profile. It might set wrong settings so you might want to double check.
There's also a bunch of default profiles and that 1333mhz at 1.5v you see might be one of those.
If you again run into errors in memtest86, pull the two sticks and test them separately. It might also be one of the slots that has problems so also try different slots.
If you find out one of the sticks is broken, a lot of brands actually have life time warranty for RAM, so definitely go and ask them what to do with your broken stick.
|
United Kingdom20326 Posts
Faulty RAM? Where? 
Just manually set it to 1.65v.
DDR = double data rate, 1333 as advertised will read as 667mhz.
If it doesn't work @1600, 1.65v, try 1333, 1.65v, before you do anything else.
No need to jump to pulling out RAM or buying new yet
|
I tried 1600, 1.65v and 1333, 1.65v and I got a blue screen with both.
Now I test both RAM sticks separately with memtest86? Then try switching their slots?
Edit: I guess I should also ask which settings I should have the RAM on when testing.
Edit2: I just got a new BSOD too when Windows was loaded and idled. 0x19: BAD_POOL_HEADER
aumha.org/a/stop.php:
A pool header issue is a problem with Windows memory allocation. Device driver issues are probably the msot common, but this can have diverse causes including bad sectors or other disk write issues, and problems with some routers. (By theory, RAM problems would be suspect for memory pool issues, but I haven’t been able to confirm this as a cause.)
|
Viruses:
1) Download MalwareBytes Pro (trial run) and run some scans.
2) Download Kaspersky TDSS Killer. It is an anti-root kit scanner that is very helpful.
3) Run a scan from Bitdefender's online scanner.
4) Check your processes and start up programs, take note of programs that you have no clue what it does.
5) Run CMD > Type "netstat -b -n" > Look at the programs and the following external IP addresses > Take note of the ones that look suspicious.
6) *OPTIONAL* Download and run ComboFIX (DO THIS ONLY IF YOU KNOW HOW TO USE IT OR HAVE PROFESSIONAL HELP).
Hardware issue:
As for the RAM, just set everything to default and see if that fixes the issue. If it doesn't, then you can check your CPU overclock (if it is overclocked).
Software issue:
Very low chance, but it could just be a corrupted file somewhere in your Windows files.
|
It will definitely crash a lot with all kinds of errors as long as memtest86 doesn't run fine. So looking up Windows blue screen codes or whatnot isn't important until you get those memtest86 errors to go away.
|
Okay. I figured it might help you guys diagnose it.
I'll run memtest86 (I don't have memtest86+ but I assume both programs are fine) on the RAM sticks separately. So far, it seems like only one of the sticks is showing errors but it hasn't been going very long.
Edit: After 30 minutes of testing one stick, it says pass complete and no errors. The other stick has a ton of errors after only a couple minutes. Does that tell us that the one stick is the issue?
|
United Kingdom20326 Posts
If you tried them in the same RAM slot, i think so
|
|
|
|
|
|
|
|
EPT![[image loading]](http://i.imgur.com/yPKgDKm.jpg)
